Dental Materials Research

UNITED STATES DEPARTMENT OF COMMERCE • Peter G. Peterson, Secretary

NATIONAL BUREAU OF STANDARDS • Lawrence M. Kushner, Acting Director

Proceedings of the 50th Anniversary Symposium

Held at the National Bureau of Standards Gaithersburg, Md., October 6-8, 1969,

in Recognition of Fifty Years of Dental Research at NBS

George Dickson and James M. Cassel, Editors

Institute for Materials Research

National Bureau of Standards Washington, D.C. 20234

Sponsored by the National Bureau of Standards and The American Dental Association

National Bureau of Standards Special Publication 354 Nat. Bur. Stand. (U.S.), Spec. PubL 354, 238 pages (July 1972) CODEN: XNBSAV

Issued July 1972

For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402

(Order by SD Catalog No. C 13.10: 354), Price S3. 75 (cloth) Stock Number 0303-0944 ;

Abstract

A Symposium on Dental Materials Research was held at the National Bureau of Standards, October 6-8, 1969, on the occasion of the fiftieth anniversary of the dental research program at NBS. The Symposium brought together outstanding researchers in the dental materials field from throughout the world for a comprehensive examination of the present state of research and a look at future dental needs and expectations. The program covered the broad dental materials field—from an examination of the oral environment to a consideration of future needs from the viewpoints of dental practice, dental education, dental industry, and basic science. Invited papers covered metals research, new developments in nonmetallic restorative materials, dynamic methods for determining the mechanical properties of dental materials, and problems of evaluating dental materials and making such evaluations useful to clinical dentistry through the development of specifications.

Key words : Adhesives composite restorative materials ; dental materials research ; dental ; restorative materials ; future dental needs ; laboratory testing and clinical research mechanical properties ; metals research ; specifications.

Library of Congress Catalog Card Number : 70-178251 ;

Foreword

The Dental Research Section, a unit within the Polymers Division of the Institute for Materials Research, National Bureau of Standards, renders a unique assistance to efforts to improve dental health. Tlie research program is a cooperative effort both in personnel and funding involving several agencies of government and a major professional health association. The specific aim is to bring physical science expertise and instrumentation to bear on those aspects of dental research which may yield only to this approach.

The current collaborative dental research program at NBS is conducted in cooperation with

the Council on Dental Research of the American Dental Association ; the National Institute for

Dental Research ; the Dental Research Division of the U. S. Army Medical Research and Devel- opment Command ; the Dental Sciences Division of the School of Aerospace Medicine, U.S.A.F. and the Veterans Administration. Dental research was begun at the National Bureau of Standards in 1919 in response to a request from the War Department for assistance in evaluating dental materials purchased by the Government. From 1919-1922 the research staff consisted of Dr. Wilmer Souder, in whose honor the present symposium has been dedicated. From 1922-1928, the Weitistein Research Laboratory supported Research Associates as assistants to Dr. Souder. In 1928, the American Dental Asso- ciation established its Research Associate Program and this continuing collaboration has encouraged close rapport among this physical science laboratory, dental manufacturers, and practicing dentists. The achievements of the Dental Research Section during the past 50 years have transformed the practice of dentistry in many ways. Early accomplishments included the development of precision casting techniques for gold alloys. Later innovations included a high-speed contra- angle turbine drill, a panoramic dental x-ray machine, spherical particle amalgam alloy, and composite restorative materials. Current efforts are directed toward a better understanding of the fundamental properties of tooth structure, the physical and chemical mechanisms relating to initiation and development of caries, and toward the development of new and improved materials and instrumentation.

It is particularly important, in paying tribute to a half century of dental research, that opportimity be provided in this symposium to see not only where we have been but where we need to go. The response of dental materials experts from aroimd the world in meeting the challenge of delineating the research needs for the future is especially exhilarating.

J. D. Hoffman, Director Institute for Materials Research National Bureau of Standards

lU• • • Preface

This book is the formal report of the iDroceedings of the 50th Anniversary Symposium on Dental Materials Keseai'ch sponsored by the National Bureau of Standards and the American Dental Association with the cooperation of Johnson and Johnson, Kerr Manufacturing Com- pany, and Surgident, Ltd. The Symposium brought together many outstanding researchers for a comprehensive examination of the present state of research and a look at future dental needs and expectations. The papers included herein encompass the broad dental materials field—from an examination of the oral environment to a consideration of future needs from the viewpoint of dental practice, dental education, dental industry, and basic science. Invited papers cover metals research, new developments in nonmetallic restorative materials, dynamic methods for determining the mechanical properties of dental materials, and problems of evaluating dental materials and making such evaluations useful to clinical dentistry thi'ough the development of specifications.

The symposium offered an opportunity to pay tribiite to the founder and for many years guiding inspiration of the dental research program at NBS, Dr. Wilmer Souder. It also recognized the outstanding leadership of Dr. Irl C. Schoonover in the development of a unique organization for dental research which has directly involved a government laboratory, the National Bureau of Standards, and a private professional society, the American Dental Associ-

ation, in a cooperative program supported and participated in. by the National Institute of Dental Research, the Armed Services and the Veterans Administration Dental Corps. Members of the Symposium Committee wish to express their appreciation to the authors and to all participants who contributed toward making the symposium a truly memorable event. Thanks are also given to Mrs. Marion Kumpula of the ADA Research Unit at NBS and to Mrs. Ruth Davenport for their attention to the many details involved prior, duiing, and immediately after the symposium. The NBS OiFice of Technical Information and Publications under the direction of W. R. Tilley, with special help from Robert T. Cook, gave invaluable assistance in many phases of the effort. Particularly appreciated is the assistance of Johnson and Johnson, Kerr Manufacturing Company, and Surgident Ltd. in contributing to the funding of the Symposium.

Committee for the 50th Anniversary Symposium on Dental Materials Research

Gerhard M. Brauer Walter E. Brown James M. Cassel Harold J. Caul George Dickson George C. Pafl'enbarger William T. Sweeney, Chairman

Identification of some commercial materials and equipment has been necessary in this book. In no case does such identification impiy recommendation or endorsement by the National Bureau of Standards, nor does it imply that the material or equipment is necessarily the best available for the purpose.

iv Contents Page

Foreword iii

Preface iv

I. INTRODUCTORY SESSION

Dental Research at the National Bureau of Standards—Reminiscences 3 Souder, W.

Dental Research at the National Bureau of Standards—History and Individuals 7 Sweeney, W. T.

II. FUTURE NEEDS FOR RESEARCH IN DENTAL MATERIALS

The Need lor Basic Research in Dental Materials 15 Peyton, F. A.

Research Needed by the Federal Dental Services 19

Copeland, H. I.

Research Needed by the Dental Industry 23 Glenn, J. F.

Research Needed for Dental Education and Practice 27 Phillips, R. W.

III. METALS RESEARCH

Amalgams in Dentistry 33 J0rgensen, K. D.

Basic Metallurgy of Dental Amalgams 43 Johnson, L. B., Jr., and Wilsdorf, H. G. F.

Casting Alloys in Dentistry 61 Asgar, K.

Basic Metallurgy of Dental Casting Alloys 67 Nielsen, J. P.

IV. NEW DEVELOPMENTS IN NONMETALLIC RESTORATIVE MATERIALS

Dental Porcelain 77 McLean, J. W.

Dental Silicate Cements 85 Wilson, A. D.

Composite Restorative Materials 93 Bowen, R. L., Barton, J. A., Jr., and Mullineaux, A. L.

Cements Containing o-Etlioxybenzoic Acid (EBA) 101 Brauer, G. M.

Organic Adhesives 113 Alter, H., and Fookson, A.

V V. MECHANICAL PROPERTIES Page Viscoelastic Behavior 127 Oglesby, P. L.

Low Frequency Determination of Mechanical Properlies 145 Myerson, R. L.

Ultrasonic Methods for Determination of Mechanical Properties 161 Dickson, G.

Stress Analysis of Dental Structures 169 Craig, R. G.

Relations between Mechanical Properties and Clinical Behavior 177 Mahler, D. B.

VI. DEVELOPMENT OF IMPROVED METHODS FOR EVALUATING DENTAL MATERIALS

Need for Research to Develop Performance Characteristics 183 Schoenmakers, H. P. L.

Need for Correlation between Laboratory Testing and Clinical Research 187 Hedeg§,rd, B.

Biological Evaluation of Dental Materials 191 Ryge, G.

Corrosion Testing in the Mouth 201 Nagai, K.

VII. SPECIFICATIONS

International Specification Program—Australian Experience 209 Docking, A. R.

Development of European Specifications and Testing 213 Laplaud, P.

Development of South American Specifications and Testing—Brazilian Experience 217 Siiffert, L. W.

USA Specification and Evaluation Programs 221 Stanford, J. W.

VIII. APPENDIX. NBS DENTAL RESEARCH SECTION PERSONNEL

Personnel of the Dental Research Section of the National Bureau of Standards 227

vi I. Introductory Session

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 19T2).

Dental Research at the National Bureau of Standards—Reminiscences

Wilmer Souder*

Chief, Dental Research Section, NBS, 1919-1945

In 1918 dental amalgams had a high rate of failure and few data on their physical properties were available. Using the interferometer for determination of the dimensional changes of amalgam and scientific test methods for other properties, NBS began to obtain information on the physical properties of dental materials. The specification for dental amalgam developed in the early days with numerical limits for physical and clinical properties and details of test methods has served as a model for specifications for dental materials for over 40 years. Although the early results were challenged and the program opposed by some, the dental profession soon recognized the value of the work and requested its expansion. In time schools, dental associations, and manufacturers joined in commending the research program.

Key words : Amalgam, dental ; American Dental Association ; dental materials ; Dental Research Section, NBS interferometer, dental specifications, dentaL ; ;

1. Reminiscences terms. Tests to substantiate these claims in terms of physical, chemical, and engineering properties Thank you, Mr. Chairman, for the kind invita- (necessary in restoring the needed functioning of tion to attend and the liberty to address this sym- a tooth) were not found. The pioneer who first posium. More than ten years ago the Bureau said pointed out this defect was Dr. G. V. Black. He "Good Bye and Best Wishes". No attempt has been had a clear understanding of the needs and methods made to have succeeding leaders change their pro- for establishing the quality of an amalgam. His grams. I had my opportunity and am pleased with heroic efforts to create a micrometer sensitive the results. The future opportunities and prospects enough to document, with definite accuracy, the are bright. Your request for I'eminiscences is dan- length and volume changes in amalgams, and their gerous. It means looking back. One of our former dependence on chemical compositions and opera- directors, Dr. Lyman J. Briggs, had a motto tional practices, were not entirely satisfactory. His "Never look back something might be gaining on ; micrometer w^as dependent on levers and gears you". Any attempt to brag on the achievements which introduced friction in pivots plus hysteresis would flavor of conceit. To criticize would be dan- lags. This preliminary search pointed to the neces- gerous in the presence of this audience. So what is sity of a more sensitive micrometer. The interfer- left? There are some personal errors, then some ometer, having a normal sensitivity of one mil- confrontations and a rapprochement of all inter- lionth of an inch, is such an instrument. ested parties. It seems proper to document these, as they were encountered. 3. Precision Data No honest comment on the research can be made without an acknowledgement of the aid given by Mr. C. G. Peters, of the Bureau's optical labora- loyal assistants. Among the first of these were, tory, had an interferometer and gladly consented Peters, Coleman, Hidnert, Sweeney, Swanger, to measure the length changes in amalgam speci- Taylor, Isaacs, and Berger ; and later Caul, Dick- mens wliich I condensed. These tests established, in son, Paffenbarger, Lynch, Brauer, Jordon, Schoon- definite terms, the changes and defects which Black over, and Eichardson. Others are named in various had suspected. It is doubted that any dental col- publications and may be mentioned by those who lege, dentist, or manufacturer had an instrument follow on this program. approaching the precision of the interferometer. Peters and I published our findings in the Physical 2. A Dental Research Considered Review [1].^ Our data established a difference of three to one for the thermal expansions of amal- preliminary A review of the dental situation in gam to dentin; the amalgam having the higher 1918 revealed a high degree of failures in dental value. Tliis established the necessity for a positive restorations made from silver amalgams. Claims setting expansion of all amalgams to cancel any and endorsements were available and techniques separation betAveen the amalgam and the tooth were numerous and elaborate but not in teclmical

1 Figures in brackets indicate tlie literature references at ttie Present address : Landisville, Pennsylvania 17538 end of tliis paper.

3 :

when exposed to cold foods or drinks. Several give service in this field. About twelve samples brands gave setting shrinkages. The dentist is were submitted with the request. There were nu- called upon to create a restoration surpassing the merous claims for superiority and endorsements perfection of the natural tooth which has failed. printed on the cartons, but not one numerical Furthermore he must completely exclude germs, datum to support the claims. After making tests acids, and all other hazards; always ready to of setting expansion, compressive strength, and destroy his restoration. No immediate resiDonse ease of amalgamation, it was possible to rate most came from this publication. I decided to create an of the samples as defective or inferior. The Army interest by dentists, manufacturers, and the public was advised which of the remaining alloys should by writing a news item for release by the Bureau give satisfaction. This experience pointed to a containing, among other information, a statement need for a real specification. Accordingly a study "The public could better avoid the dentist who was inaugurated and a specification [2] was set drills out a cavity and fills it with a shrinking up. This specification was modeled on the pattern amalgam". The news item was sent to the Depart- of the American Medical Association's Pharma- ment for approval. The item was returned with a copoeia specifications. It gives numerical limits for statement (essentially) as follows: "There is no physical and chemical tests and refers to instru- question on the accuracy of the tests however, it ments and techniques for making the tests. These ; is not considered wise to create a situation such as two are musts for any intelligible test. Otherwise this would create unless, at the same time, a remedy quibblings are sure to erupt. I claim credit for is offered." So died the attempt to awaken the this theft or appropriation of the American Medi- public. cal Association's pattern. The limiting numbers 4. Sparrings may be changed as qualities are improved by the manufacturers. This type of specification for vari- Unfortunately or otherwise, a metallurgist, in- ous dental materials has survived forty years and terested in amalgam alloys, made some disparag- is basic for the Certification Plan adopted later by ing remarks about our budding efforts and pointed the American Dental Association. out unimportant variations in our results when he his tests. were printed made a report on own These 6. Questioned Progress in a metallurgical journal. Peters and I set up a reply criticizing his results which revealed greater The service to the Army Dental Corps was the variations than any of ours. Also we referred to first signal of a serious interest in the Bureau's his great emphasis upon a transformation in amal- work in the selection of dental materials. As might gams when heated to 80 degrees Centigrade; a be expected, some manufacturers (whose products condition not compatible with dental health. We might not meet a specification) were not enthusi- were were careful to make no reference to our astic about the plan. Deans, prominent lecturers, connections with the Bureau and paid for the and dentists whose personal endorsements were publication of our reply. We were elated at the ignored in the specifications were in a clouded neatness of our reply—until a few days after the area. We were disappointed when the Journal of appearance of the journal. We were asked to come the American Dental Association in 1920 could to the office of the Director at our convenience. We not find it possible to print our first extensive (32 were anticipating the approval of Director George page) report, "Physical Properties of Dental Ma- K. Burgess, a metallurgist of the first water. He terials". However, another agency was quite will- inquired about our work. Then he showed us a ing to print the report [3] with a volunteered copy of the magazine containing our reply and commending editorial. Later, consideration of the asked what we knew about it. We said we pre- refusal was understandable. With certain members pared it and paid for its printing and purposely of the Association lukewarm and several manu- avoided any reference to the Bureau. Then the facturers suspicious of the outcome of the Bureau's sky fell in. He chewed us out as they say in the work, they can be excused for the decisions to wait Army. He reminded us we had stooped to put our for more assurance of the purposes. When con- abilities on a level of the one we had attempted vinced of the ability and integrity of the Bureau, to expose. He said the Bureau does not encour- the American Dental Association became one of age scraps. It looked like the last day for two our loyal suppoi'ters. career Civil Service employees. Then his attitude changed. He said "You have abilities, you can de- 7. Genuine Interest velop positive approaches. Go back to your laboratories and prove it". From that day we kept After the publication of the first extensive re- within the traces, absorbing most criticisms. port, interest developed rapidly. Correspondence with dentists, schools, and associations requesting 5. Specifications lectures and clinics piled up. The Director ap- A request from the Army in 1918 asking for proved many of these requests. (The Bureau posi- assistance in awarding contracts for dental amal- tion, relating to such requests assumed that the Bu- gam alloys gave the Bureau its first opportunity to reau official would be reimbursed for travel and 4 :

subsistence costs and that no honorarium would be 1927 and the cooperative program has continued accepted.) Requests for an extension of the pro- unbroken from 1928 to date. In 1942 Dr. M. D. gram to include a study of dental gold alloys, ce- Huff, Chairman of the Research Commission, ex- ments, and accessory materials came to the Bu- pressed the feelings of the American Dental As-

reau. The Bureau felt that after having set up a sociation in the following words [8] : "It was a pattern for research in the field, it should not as- fortunate day when Doctors Barber, Brown, and sume the responsibility and expense for exten- Vol! and of the Research Commission went to the sions. However, it did ask for advice [4] on the National Bureau of Standards in 1927 to complete problem. Dr. Louis J. Weinstein, Director of the the details of the cooperative research which has

Weinstein Research Laboratory in New York, ex- become so valuable to the dental profession. . . . pressed a genuine interest in seeing the research Our associates have been wholeheartedly welcomed

continued. by the Bureau. . . . The supervision of the work, the assumption of responsibility for the data, and 8. The Research Associate Plan the complete publication of all findings by the Bureau, have given us an authoritative position Dr. Weinstein's request was simple and direct seldom possible in such fields of research." "I want to see data on dental gold alloys and accessories which will stand up when presented be- 11. World War II fore schools and private groups and not be con- fused by data in present day texts and glaring ad- During World II the Bui-eau found it necessary vertisements." The Bureau Research Associate plan to curtail activity on many programs as it devoted was explained to him whereby qualified scientists about 90 percent of its work to problems of de- may work at the Bureau on problems of public fense. It was declared a Restricted Area under the interest. The Bureau directs and supervises the supervision of the Departments of War and Navy. work and publishes the results. The sponsor pays The American Dental Association sent a prompt the salary of the associate. Dr. Weinstein accepted and patriotic request to the Bureau authorizing it the opportunity and supported the program for six to feel free to transfer its dental associates to Bu- years (1922-1928). reau military work as needed, salaries to be continued by the Association. Both, Bureau and 9. Organized Opposition Association scientists, filled assigned military positions with distinction. Dental research was The first comprehensive report [5, 6, 7] on the restored to its full-time activity at the close of the Weinstein Research Associate's achievements was war. given at the Dallas meeting of the American Den- 12. Rapprochements tal Association. It was quite evident at this meeting and in other happenmgs that a fission between Within 10 years after the previously mentioned vested interests and the research associates was oppositions were met, all dissenters had disap- developing. These interests wanted to crush the peared. The schools, societies, and manufacturers research. They appealed to the Secretary of Com- were joining in their commendations of the work. merce, insisting that the reports were creating New and and better materials and techniques have confusion among the dentists and in the schools been discovered. Perhaps the comment of one man- and requested that the work be stopped. The Bu- ufacturer can be used as a summary: "The re- reau was asked for a statement on the claims. The search and specifications are our only effective Director explained to the Assistant Secretary that protections against competitoi-s trying to exploit the research was a public operation and all findings inferior materials by glaring advertisements and were announced promptly. Furthermore, he stated personal endorsements". that the program was a health-saving research needed by every citizen of the United States. No 13. An Unfinished Research cease or desist order was issued. By some means the dentists across the Nation become aware of One project on the possible "Remineralization what was about to happen and were preparing for of Dentin", started by Schoonover and Souder, a fight. Had this developed, with the support of was dropped by reason of the more important war 90,000 dentists across the Nation, it would have needs. The illustrations in the preliminary report made the Bureau's later AD-X2 confrontation [9] show a deposit of some material in the hardness look like child's play. indentations; the length of the indentation mark is shortened after exposure to a fluoride solution. 10. American Dental Association The body's ability to repair a broken bone suggested the possibility of some such attempt to Meanwhile, the American Dental Association repair injury to dentin, under favorable condi- had expressed an interest in and a desire to co- tions. Dr. von Buest of the University of Louisville operate formally and actively in the program. The Dental School was the only one to give us en- details for this cooperation were completed in coiiragement on the idea. Quotes from recent re-

5 :

ports on the effects of fluoride are given below [2] Souder, Wilmer, Measurement and application of Bierman, M.D. [10] "the (cancerous) bone certain physical properties of dental amalgam. J. Dental Res. 7:173 (June becomes harder and more durable". Hoffman, 1927). [3] iSouder, Wilmer H., and Peters, Chauncey G., An NDRI [11] "topical fluoride can reverse the dental investigation of the physical properties of dental caries process and may even heal incipient caries". materials. Dental Cosmos 62:305 (March 1920). ". [4] Burgess, George K., Dental research by Department A research [12] . . into the molecular basis of Commerce. J. Am. Dental Assoc. 11:249 of disease" has been announced. Even a TV (March 1924). puppet declares that fluoride makes a tooth [5] Coleman, R. L., Physical properties of dental ma- stronger. Perhaps some budding researcher may terials (wrought gold alloys). J. Am. Dental decide to make an independent evaluation of these Assoc. 12:.520 (May 1925). [6] Coleman, R. L., Physical properties of dental ma- reports. terials (III) Progress Report of research on the 14. Fare Thee Well dental casting process. Dental Cosmos 68:743 (August 1926). With these scattered reminiscences my part in [7] Coleman, R. L., Physical properties of dental materials (gold alloys and accessory materials) this program must close. These are no feelings of BS J. Research 1, 867 (1928) RP32. resentment toward anyone. Rewards and gestures [8] Souder, Wilmer, and Paffenbarger, George C, Phys- of appreciation have been excessive. My response ical Properties of Dental Materials, NBS Circ. 433, 222 pages (1942), Foreword. is, Thank You and Best Wishes for the Future. [9] Souder, Wilmer, and Schoonover, Irl C, Experi- mental remineralization of dentin. J. Am. Dental 15. References Assoc. 31:1579 (December 1944). [10] J. Am. Med. Assoc. 208: (No 6) 953, May 12, 1969.

[1] Peters, C. G., and Souder, W. H., Some physical [11] Am. Dental Assoc. News Letter 22 : (No 7) March 31, properties of dental materials, Phys. Rev. NS 13, 1969. 302 (April 1919). [12] Physics Today (We Hear That) June 1969, p. 97.

6 ;

NATIONAL BUREAU OF STANBARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Dental Research at the National Bureau of Standards—History and Individuals

W. T. Sweeney*

Chief, Dental Research Section, NBS, 1953-1968

Dental research at NBS was initiated in 1919 by Dr. Wilmer Souder with work on dental amalgams. Dr. Souder built the Section on the principle of cooperation between revsearoli laboratories, manufacturers and the dental profession. In the 1920's research associates, first from the Weinstein Research Laboratory and later from the American Dental Association were added to the NBS dental research staff. Dr. N. O. Taylor and Dr. George C. Paffenbarger were the first ADA Research Associates. In 1945 while Dr. Irl C. Schoonover was Chief of the Section, the laboratory staff was further enlarged by the addition of Guest Workers from the Armed Services. Among the many important areas of research were precision casting, dental cements, polymers for dentures bases, composite I'estorative materials, high speed turbine handpieces, panoramic X-ray equipment, studies of natural tooth structure and development of standards and specifications.

Key words : Amalgam, dental ; American Dental Association casting, dental ; dental ;

research ; Dental Research Section, NBS guest workers Paffenbarger, George C. ; ; Schoonover, Irl C. Souder, Wilmer .specifications, dental. ; ;

1 . Introduction dental products purchased for the Federal services was referred to Dr. Souder for a routine re- 'The objective of this paper is to' give a brief and ply. Amalgam was the material of interest. Sou- concise report of the most outstanding events, per- der, as a young physicist fresh from his Ph.D. sonnel, and accomplishments of the dental re- studies at the University of Chicago, naturally search program at the National Bureau of Stand- looked at the problem scientifically. The more he ards in order to set the stage for the symposium examined the basis for selection of dental mate- honoring the 50th Anniversary of the program. rials, the more he was convinced that the science Only a few of the many persons and events can of measurement could make a contribution to the be discussed or even mentioned in the limited dental profession and the dental treatment of the time allowed. public. He, in short, visioned the substitution of The name of Dr. Wilmer Souder is the most quantitative measurements of properties for the outstanding of all the scientific staff because he is art of personal skill and personal recommenda- responsible for setting the character and ideals of tions. He had the zeal and forethought to advo- the program and was chief of the Section for the cate and demonstrate that physical science had first half of its life. The merit of bringing good a great many things to offer the health profes- scientific measurements into the health field as an sions. adjimct to better dental health care for the public A few people, notably G. V. Black, a dentist, was thereby established. He also realized that a had carried out some very basic investigations cooperative effort was essential between the three but the precision of a physicist was almost un- groups, dental profession, manufacturers, and known to the science of materials used by the general public as represented by the government. dentist. Souder first turned his attention to dental While this sounds trite, looking back today, it was amalgams and in 1919 published his report on not a very popular idea when he first started in the measurement of such properties as dimen- 1919, as there were very few who had the long- sional setting changes, strength, and flow. His re- range vision to see the advantages of such cooper- sults and methods are good today even after 50 ation. years and have led to much improvement of alloys and techniques for dental application. He 2. Early or Formative Period, 1919-22 held the theory that the best way to evaluate any material is to measure the properties that are im- This period was the most important as it is the portant for its application. He meant to obtain germ from which the Section grew. The original numerical evaluations by measurement and then request from the War Department to the National delineate his procedures so that any competent Bureau of Standards for assistance in evaluating scientist could repeat and check his results. Tliis was a unique point of view in the field of dental

Present address : School of Dentistry, University of Alabama, Birmingham, Alabama 35233. materials. He was told that it was impossible to

7 determine the elements in certain alloys, as the dental profession argued that it should know ex- analytical methods were too crude. As a matter of actly what it uses and the manufacturer thought fact this was correct, in some cases at least, but he lie needed protection because he spent a lot of answered this by saying that if we obtained a money to advance his product. The principle has scientifically trained staff, we could overcome this been to use trade names in publications if they are and he certainly proved it. necessary for understanding by the reader. Details Souder, realizing in this early period that a of methods of analysis were given in scientific re- working together oi the research laboratories, ports so anyone competent could evaluate proper- manufacturers, and profession was required if the ties versus composition. public were to receive maximum dental sei"vice, Verbal reports and many articles by the staff tried to interest the dental profession through of the Section during this period created such a their national organization to join in a coopera- demand by the dental profession for more scien- tive program. While certain individuals gave tific evaluations that by 1928 the profession was moral support to the idea, the orgaiiization was willing to support the research program and give not yet willing to put financial support into the official recognition to it by formal agreement be- research program. When this failed he was fortu- tween the American Dental Association and NBS. nate to secure financial assistance from Weinstein This arrangement made for a much broader base Research Laboratories, who provided sivbstantia^l of operation and the liaison has proven most prof- support from 1922 to 1928. itable to all concerned.

3. Expansion of the Program Via Research 4. Cooperative Program With American Associates, 1922-28 Dental Association, 1928—45

This period was most productive as it represents In April 1928 the coof)erative program between the first expansion of the dental work by the addi- the American Dental Association and the National tion of research associates to the group. The re- Bureau of Standards was initiated. The first ADA search associates were directed by NBS and sup- research associate was Dr. N. O. Taylor, a chem- ported by the Weinstein Research Laboratory. The ist, who was followed in 1929 by Dr. George C. first research associate, R. L. Coleman, an engi- Paffenbarger, the first dentist to work full time neer, was selected from the NBS Staff, a member in the Section. The period of the late 1920's and of the Weights and Measures Division. His work 1930's witnessed a great expansion of the research on measurements of properties of gold alloys and program through the addition of both NBS and precision casting techniques is dlassical in that it set ADA personnel. Basic research on the physical and the standard for precision casting, not only in the chemical reactions of many materials was under- dental field but later in industry, such as in the taken. This period was productive and an im- casting of blades for turbines. During this period proved understanding was obtained of the proper- the first dental specification for amalgam alloys ties of practically all the materials used in re- was promulgated, based on the original work of storative dentistry from amalgam alloys to denture Souder; and the properties of accessoi'y materials resins. The aj^proach was basically the same for used in dental casting, such as waxes, investments, all materials. First a study was made of the prop- and orthodontic wires were studied, and standard erties of the available materials and how they were measuring techniques were developed so that spec- affected by composition and by techniques used in ifications could be written later. practice. Typical examples were studies of the The chemical analysis of alloys of gold and the effect of heat treatment on gold alloys and how the platinum group metals by the joint efforts of mechanical properties were changed by cooling

Raleigh Gilchrist, an NBS chemist, and William rates in the dental laboratory ; how the method of Swanger, a research associate, was an example of mixing cements could drastically change the useful how the cooperative effort produced worthwhile properties of the material; and how exposure to scientific results. Their reports justify Souder's light affected the color stability of pink denture prediction that well trained chemists could deter- base resins. mine accurately the composition of dental alloys. Many special methods were developed for meas- After these reports, it was obvious that secret for- uring proi^erties of these different materials. The mulas would be secret only until someone was will- fact that dentistry uses small specimens in coming to spend the time and money to do' precision parison to those commonly employed in industry chemical analysis. This work created many con- makes it necessary to use unique test methods. Spe- troversial problems for NBS on the policy of pub- cial tests were developed to more closely simulate lication of analyses of trade brand alloys or mate- dental use. The fact that the oral cavity has a rials. These problems have not yet completely varied atmosphei'e with a variety of temperatures disappeared because it is a moot question as to how and humidity, makes it necessary that materials far a tax-supported institution should go in pub- be evaluated under conditions which are found in lishing trade secrets that individual companies clinical use. Biological requii-ements put many re- have spent much in developing. In general, the strictions on otherwise suitable materials.

8 The specification program was a major develop- for more technical information. This made it possi- ment as it made the results of research useful to the ble for a very valuable liaison to be initiated. general dentist and the public. The history of the Schoonover, with the help of Fischer (USAF) research program during this period followed a and Paffenbarger (USN), cooperated to set up consistent pattern of sufficient research on a type working arrangements with the Federal dental of material to understand the basic reactions and services that added great strength to the Section by properties of dental significance followed by a sur- increasing both personnel and funds. This made it vey of the existing materials to ascertain what possible for NBS to equip a clinical unit and cor- could be reasonably supplied and to determine the relate many laboratory findings with clinical prac- range of properties of the satisfactory materials. tices. These arrangements also made it possible for A specification was then written using the infor- personnel from the Federal dental services to work mation obtained to place numerical limits on the in the NBS Dental Research Section as Guest properties thought to be important. This was re- Workers for periods of one to several years. The viewed by a Specification Committee consisting guest workers included not only dentists with of representatives from the profession, the manu- much clinical experience but also scientifically facturers, and the government (representing the trained enlisted personnel (chemists, physicists,

public). Following the recommendation of this engineers, etc.) . In addition to the research accom- Committee the specification was adopted as an offi- plished, this program resulted in providing the cial specification of the American Dental Associa- services a number of dental officers trained in tion. Producers were asked to voluntarily guaran- materials research. Also to the benefit of NBS, tee their products to comply by signing a formal several of the enlisted guest workers remained document of agreement. Specimens were purchased after their tours of duty in service to become per- on the open market and tested for compliance at manent members of the NBS staff. NBS by the ADA research associates. Materials It has been the policy to draw no sharp distinc- which complied were listed in the ADA journal. tion between NBS staff members, ADA research Specifications developed for the Federal Gov- associates and guest workers in the operation of ernment and advice provided in writing purchase the Section. A spirit of cooperation has prevailed descriptions are among the many benefits of the with the result that many are the reports co- Section's program to the government. authored by representatives of each of the three The series of reports on dental casting and acces- groups. Guest workers have been encouraged to sory materials by Taylor, Paffenbarger, and take advanced study and frequently the research Sweeney made it routinely possible to make clini- carried out in connection with such studies in- cally acceptable castings. This is in contrast to volved collaboration with a Bureau senior scientist. an early investigation where only 2 of 25 leading This period was characterized by many sharp dentists were able to make an inlay casting that contrasts of opinions on the relative amount of would fit even reasonably well a standard die fur- effort that should be put on basic in contrast to nished them by NBS. applied research. It is my strong conviction that in The very excellent work on analytical methods an area such as dental materials and clinical den- for dental materials by Caul and Schoonover tistry the most return will be obtained by conduct- added new and precise information to the scien- ing, as we have at NBS, the two side by side. Each tific literature on such a wide variety of materials is an asset and a source of strength to the other. as chrome-cobalt alloy, mercury-containing alloys, In 1950 a very thorough review and bibliog- ^ resins, denture rubber, impression materials, etc. raphy was published by Schoonover and Souder and made it possible for manufacturers to improve to which the reader is referred for many details and better control their production. These methods about the Section's work up to that time. are used also to define composition limits of elements in many modern specifications such as those 6. Summary—Resume of History and for amalgam alloy, and gold alloys. Individuals

One cannot help but be impressed by reviewing, 5. Expansion of the Research Program to such as I have, the programs since my first contact Include Guest Workers From Federal on August 3, 1922. At that time the Section con- Dental Services and Foreign Countries, sisted of three scientists. Dr. Souder, Dr. Hidnert, 1945-69 and Mr. Coleman. This is in contrast with today's staff of thirty-five. About 175 individuals have been The period immediately following the end of associated with the Section during the past 50 World War II saw a major expansion in the pro- years. gram. As a result of the large scale procurement Space to even mention all the major projects is and utilization of dental materials by the Armed not available but items that come to my mind as Services, persons directly knowledgeable in dental being specially noteworthy will be recorded with materials research and clinical practice had first hand experiences which emphasized the great need 1 NBS Cire. 497, 14 (1950). 9 the individuals connected with them. In doing this properties of materials and the effect handling has it is realized that others may have different ideas on the clinical resiilts. but it is certain that these items have been an 5. The research reports and communications influence on dental practice. have been very productive in introducing new 1. The development of standards for dental materials and equipment ideas in the fields of both materials based on measurements of properties of clinical dentistry and dental research. The early dental significance. This is probably the most used work of Souder produced the dental interferom- and productive effort of the Section. It covered eter which has been adopted as standard around many types of materials, in fact, most of the ma- the world for measuring the setting changes in terials widely used in dentistry. The adoption of amalgam. The NBS standard MOD steel die is specifications and the publication of a list of cer- probably the widest used device for evaluating tified products by trade names added to the prac- precision casting. This was the result of the pre- tical usefulness of the program. The combination cision casting work of Coleman. The fused-quartz of professional approval and the scientific integ- tube method of measuring the thermal expansion rity of NBS combined to make a most reliable of solids, first developed by Hidnert and Sweeney buyers' guide for dentistry. Also, the specification in 1928, probably has been used more than any program served the manufacturers well, as it made other equipment for expansion measurement of available a standard to compare the quality of solids and especially for dental investments. products without any reference to selling price. The panoramic x-ray equipment now in uni- The success of this program has spread to many versal use in many clinics, was perfected by Hud- countries of the world and standards developed son, Kumpula, and Dickson and is now commer- here are used as models for others. cially available with continuing expanding uses in A survey a few years ago showed that only 0.1 dentistry for recording the condition of teeth and percent of the items on the published certified oral structures. This item has resulted in great lists did not in fact meet all the requirements. savings to the Federal dental services, more than This is proof that the production control by enough to repay the total cost of the whole Re- American manufacturers is very good. These researcli Section's expenditure of public funds. sults are based on tests of materials bought in the The turbine contra- angle high-speed handpiece open market. developed by Nelsen and Kmnpula is probably the 2. One item which is not usually emphasized most important advancement in dental equipment when discussing the Dental Research Section is in this century. It has inspired manufacturers to the effect the training of research personnel has use the principle to develop very sophisticated had on teaching of dental materials in the schools clinical turbines and has revolutionized the prac- and, in fact, on many techniques taught in restora- tice of operative dentistry. It not only makes the tive dentistry. The Federal dental services have work of cavity preparation and tooth reduction sent many of their best officers to the Section for faster and easier for the dentist, but more impor- training and many have received advanced de- tant it is much more comfortable to the patient, so grees using the research training in the Section much so that no modern dentist today can operate for credit. The cooperative program with George- without high-speed handpieces. The number of town University has resulted in 16 master's de- these instruments in use, both in this country and grees and two doctor's degrees. Eleven foreign foreign countries, is estimated to be in the hun- guest workers, for the most part representing den- dreds of thousands. tal schools, have been trained and returned to The many instruments modified, or new, devel- teaclaing and research. oped for use in specification tests are too numerous 3. The publication of scientific data and the to mention, but the method developed by Paffen- explanation of the physical-chemical reactions of barger for testing standard consistency of cements materials, for example, the setting mechanism of made it possible to compare on an equal basis the cements, the oral environmental effects on sur- properties of different brands of both zinc phos- faces, and the effects of particle shapes of alloys phate and silicate cements. The use by Sweeney of used for amalgam have lead to many material de- cross index marks or pin inserts on dentures as velopments and to superior techniques for using reference points for measurement of microscopic materials. The explanation of the cause of delayed dimensional changes, served to evaluate the expansion of zinc-containing alloys resulted in the accuracy and stability of a wide variety of denture elimination of the long used, undesirable, palming base materials from vulcanite to methyl metha- technique for mixing. crylate resins under clinical conditions. The tech- 4. In addition to publication, a much used nique has been used to study the effect of curing method of communication was the production of and repairing processes on stability or warpage of motion pictures on the properties and techniques dentures in service. for using many materials such as amalgam, ce- The early discovery of cristobalite for use as an ments, resins, gold alloys, etc. These pictures are investment material by Paffenbarger and Sweeney in constant use in schools for explaining the basic was rewarding even though the patent was finally

10 : — awarded to previous investigators using an entirely ing of the physical-chemical reactions that can different source and method of preparation. take place in the tooth. This will be of assistance in Schoonover developed methods for determining obtaining a better rationale for the cause of caries. the number average molecular weight of acrylic The following special items and the names of resin which helped exj)lain many problems in cur- those most responsible for them are given—^not in ing and construction of dentures. This was the first any order of relative importance report issued from NBS on effect of molecular (a) Physical methods of evaluating dental weight distribution on properties of a polymer.^ amalgam—Souder, Sweeney, Caul, The relationship between properties and molecular Burns, Dickson, and Oglesby weight has been given extensive attention by re- (b) Precision casting—Coleman search personnel of industry and government since (c) Chemical analysis of gold aJlloys—Gil- then, as it showed that chain length is as important christ and Swanger as elemental analysis in the area of polymeric (d) Analysis of dental cements Isaacs, materials. — Schoonover, Brauer, and Copeland The development of spherical-particle alloy for (e) Tests and evaluation of cements Paffen- amalgam by D. F. Taylor brought a new and — barger promising type of alloy into general use. The (f Packaging of cements for military uses control of properties by suitable size distribution ) Fischer and and the low packing pressures required have made Bums Chemistry of resin polymers Schoon- this one of the most significant innovations in (g) — over, Brauer and Caul amalgam research. Indications are that it may be (h) Physical properties, methods of resin eval- the most popular form of amalgam alloy in the uation— Sweeney future. The explanation of the setting mechanism (i) Gallium alloys as dental filling mate- of zinc oxide and eugenol cements by Copeland rials—^Waterstrat and D. L. Smith and Brauer and the development of a new cement, Turbine contra-angle high-speed hand- o-ethoxybenzoic acid (EBA), by Brauer are addi- (j) piece Nelsen and Kumpula tional landmarks of the program. The consistent — (k) Panoramic x-ray equipment Hudson, progress made by Bowen in the field of composite — Kumpula and Dickson filled resins has resulted in the marketing of several products that are bringing a new era in the (1) Standards and specifications for dental quality of esthetic tooth restorations for anterior materials—Souder, Taylor, Sweeney, teeth. The advancement toward adhesive materials Paffenbarger, Beall, Caul, Bums, is also very promising and points the way to a Brauer, Dickson and others major breakthrough in tooth restoration. (m) Promotion of national and international The development of methods for determining standards for dental materials—Paffen- the structure of natural teeth by Brown and barger, Sweeney and Dickson Moreno and physical properties of enamel and (n) Development of EBA cements—Brauer dentin by Stanford have advanced the understand- (o) Composite filling material—Bowen

2 J. Amer. Dental Assoc. 25, 1487 (1938).

452-520 0—72 2 11

II. Future Needs for Research in Dental Materials

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental ^Iatekials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

The Need for Basic Research in Dental Materials

Floyd A. Peyton

School of Dentistry, University of Michigan, Ann Arbor, Mich. 48104

The field of dental materials has benefited from basic studies of the silver-tin alloy system, of cobalt alloy systems, of polymers, and of many other areas. Future basic studies of the surface phenomena of wetting, spreading of liquid films, adhesion, diffusion into dental tissues, boundary interactions, and principles of viscoelasticity related to dental materials offer productive opportunities. Basic research is essential if advances are to be made in the improvement, modification, refinement, and development of servicable materials for the practice of dentistry. It is anticipated that such studies will increase in importance in the next quarter century and that basic scientists in many fields will be contributing to the improvement of dental service.

Key words : Basic research related to dental materials ; dental materials interdisciplinary ;

research ; training of dental researchers.

1. Introduction logically trained scientist working with the physi-

cal science investigator ; or the engineer, physicist, Those who have followed the growth and de- or chemist working to solve a problem of dental velopment of dental materials throughout the practice, all have made contributions to this field. world during the past century readily recognize The future holds even greater promise for interdis- the significant changes that have occurred in the ciplinary studies. As the subjects of bioengineer- research activities in this field. There is no evi- ing, biomaterials, and biomechanics are extended dence to indicate that further changes will not into medical science, it is satisfying to know that continue to occur in the years ahead. Certainly the a similar relationship has existed and promises to field is not dormant nor static at the present time. continue in the science of dentistry. The changes that have resulted since the last part of the 19th century represent the efforts of 2. Types of Research many persons who have conducted research studies, made laboratory and clinical tests and Perhaps no one should claim that one type of evaluations, and altered or modified materials, research is more important or desirable than an- techniques, and devices in order to serve the needs other. It can be recognized to advantage, however, of the dental profession to render more and im- that different forms of research exist, and through proved service to the patient. This has required all the combined efforts a contribution to society is forms of research studies, as noted by this sym- made. Some forms of research appeal to certain posium, including basic studies practical, applied, ; investigators more than others, resulting in a or clinical testing ; and developmental projects. It greater degree of concentration on one type of has required the effort, knowledge, and experience study than on another. of investigators throughout the world, located in Basic research in the physical sciences represents private or government supported institutes and those studies that explore the fundamental nature laboratories, universities, and dental schools, man- physical state, or con- ufacturer's laboratories, as well as dentists in pri- of a compound, a reaction, a vate or group practice. Not only have multiple dition, a mathematical relationship to a physical institutions contributed to the progress that has phenomenon, and describe the characteristics ob- been made, but also the training and background served. Such studies explore the fundamental of the individual investigators has been of a multi- physical forces and processes of nature. Basic sci- discipline character. ence studies, however, are not limited to physical At this time in history, when the multidiscipline phefiomena. Studies of a basic or fundamental approach to research is being recognized and en- nature are common in the biological, social, be- couraged, those in the dental materials field can havioral, political, and economic fields of en- take pride in the fact that for half a century inter- deavor. Perhaps all basic research studies attempt disciplinary studies have been common practice. to explore the fundamental nature of the subject, This practice has increased in recent years. The in contrast to the practical, applied studies which basic scientist working with the clinician ; the bio- translate into service the fundamental concepts.

15 Generally it is recognized that basic or funda- on various materials in the past. The National Bu- mental research studies involve investigations of a reau of Standards has contributed much to these new, different, and exploratory nature. The scien- basic studies. Frequently these have resulted in tific judgement of the investigator suggests that eventual applications to the solution of practical more knowledge in a particular area will con- dental pi'oblems. Only a few need be listed to illus- tribute to the welfare of mankind and aid in the trate the benefits from basic research studies in solution of present or future problems. The history dental materials. of scientific advances is filled with examples of One that is recognized by all who are associated such basic and fundamental studies. One needs only with restorative materials and the practice of den- to recall the advances resulting from basic studies tistry, is the metallurgical study of the silver-tin in the field of relativity, x-ray and other forms of alloy system and the reaction with mercury to pro- radiation, atomic structure, and many other basic duce dental amalgam. Many investigators have investigations to recognize that research in the contributed to the understanding of these alloys, fundamentals of science, is important as a founda- and the studies continue. The results have lead to tion for scientific endeavor. Productive basic re- a gradual improvement in the quality of dental search studies provide results that can open whole amalgam during the past 75 years, with better res- new approaches and solutions to practical prob- torations for the patient. lems. From the results of basic research studies, Another example of productive basic metallurgi- new concepts are established which advance the cal studies involves cobalt base alloys for use in technical and scientific skills in many related removable cast partial restorations. From these fields. studies, practical prosthesis are a reality. Closely Studies in areas of basic research are confronted, associated with this development, and of equal however, with numerous problems of varying mag- importance to the success of the small cast restora- nitude. Frequently such studies are time consum- tion, have been the basic studies of the precision ing, require the service of specialized and highly casting process. Dentistry has been accepted as a skilled personnel, or cannot be assured of success leader in this specialized interest of precision cast- and practical application, in advance of the project ing, with numerous basic studies reported, as well completion. The practical applications and poten- as practical applications described. tial significance of such basic studies frequently The development of the plastics industry within are the subject of debate and speculation among the past 50 years has stimulated many basic studies investigators having more practical interests. into the nature of plastics and the potential appli- Under these conditions, basic research studies fre- cation in restoring lost tooth tissue. Another group quently encounter difficulties in securing a source of materials that have special dental applications, of support funds, other than benevolent founda- made possible from basic chemical research studies, tions or some government agencies. Even from are the several types of elastic impression mate- these sources the basis of justification is critical rials presently in use in dental practice. These and requires well documented and carefully hydrocolloid and synthetic rubber compounds are planned statements of experimental design and examples of basic chemical studies, leading to procedure. This is proper, however, and only sug- practical applications which have resulted in al- gests that basic research studies are involved and tered methods and improved dental practice. complex, requiring the utmost in well coordinated Anyone who is familiar with the investigations efforts and skillful management for success. that have been conducted at the National Bureau The ultimate gratification comes usually to the of Standards, recognizes the contributions that investigator who is successful in his endeavors, and have been made to world dentistry through the de- ultimately sees his results applied to the solution velopment of physical test methods for various of practical problems. Perhaps there are relatively dental materials, and the establishment of stand- few investigators who delight in conducting basic ard test procedures and specification methods. To a research studies only for the sake of collecting such considerable degree, basic research studies were data. Rather, it is their hope to open a whole new involved in the development of this program. Sub- solution to practical problems, when such studies sequent assistance from the American Dental Asso- are undertaken by other investigators. It is im- ciation, the Federation Dentaire Internationale, portant to recognize, therefore, that normally fol- and now the United States of America Standards lowing a successful basic research investigation Institute, has carried forward the initial basic there will emerge numerous practical research undertaking, and developed practical means of studies of varying magnitude and nature. Often evaluation. Basic studies related to this program the dividing line is a thin one between basic, fun- undoubtedly will continue. damental research studies and practical research The list might be extended to include devices investigations. like hand pieces and x-ray machines, and the discussion expanded with documentation, but that is 3. Basic Research in Dental Materials unnecessary to indicate that basic research studies The field of dental materials has been fortunate in dental materials have been productive in the to have benefited from a variety of basic studies past, and have led to practical and applied studies

16 with benefits to dental practice. To extend this dis- wetting, spreading of liquid films, and adhesion to cussion would tread on the subjects of other speak- tooth tissue will be further studied basically and ers in this program. become better understood. Further studies with the electron microscope and electron probe should pro- 4. Current Basic Investigations vide a basic understanding of diffusion into dental With Promise tissues, and the interaction occuring within the material or at the boimdary surface. Future studies It might be well to list, without elaboration how- will show the true potential benefit of the laser ever, some current studies of a fundamental nature beam and the related process of holography. Fur- which promise significant benefits to dentistry in ther studies in the principles of viscoelasticity the years ahead. Some have been conducted here related to dental restorations, involving funda- at the National Bureau of Standards and others mental mathematical evaluations and interpreta- are being investigated elsewhere. Several will be tions to dental problems will continue. described in detail later in this program, so it is This list might be extended into many other sufficient to take note here only that unknown bene- areas of basic studies and investigation. Even fits to dentistry in the years ahead can be antici- though other persons might choose other examples pated from these efforts. of basic studies from the past or into the future One valuable area of study is that of composite of dental materials research, these suggested materials, particularly in the area of filled plastics. studies indicate the importance of basic research This subject is being explored by many investi- to the field of dental materials. gators. Another is the study of the basic nature of Since basic studies related to many disciplines the zinc oxide-eugenol compounds and the effects are involved in so much of the current dental ma- of chemical modifiers on the characteristic prop- tei'ials research and investigation, the education, erties of these materials. Practical results are be- training, and interests of the investigator have ginning to emerge from each of these long and been enlarged significantly in recent years. The difficult basic studies. training of the dentist and the dental teacher like- Studies on stress analysis in dental restorations wise has been influenced by this trend in materials and supporting structures, by photoelastic and research. A broad spectrum of physical science other methods of analysis have involved numerous training presently is necessary for the application basic problems. In like manner the studies that are of these principles to dentistry, whereas in the past being conducted on the friction and wear of ma- the biological sciences were predominant in the terials used in the mouth require basic evaluations training of dental investigators and practitioners. that are new to dentistry. It seems probable that this trend in training will In the same way studies on tissue reaction and continue in the years ahead. compatibility with restorative materials include Most research investigators have a definite con- even though many basic concepts in biological and physical cept of what constitutes basic research, it may differ to some degree from one person to sciences. The studies being conducted on clinical another. It is recognized by all, however, that basic evaluation and correlation with laboratory obser- research is essential and fundamental if advances vations, presently are basic in nature but can pro- are to be made in the improvement, modification, long duce range practical benefits for the dental refinement, and development of serviceable ma- profession. terials for the practice of dentistry. Basic research studies in dental materials have come to be recog- 5. Future Basic Investigations nized as being essential to the further advances in this area of dental science. It is anticipated that As one surveys the horizon of the years ahead, such studies will increase in importance in the next one sees the opportunity for basic research investi- quarter century. So long as the dental profession is gations that can lead to productive practical prob- concerned with improved dental service for the lems extending into most areas of dental mate- patient, it seems certain that the basic scientist in rials scientific endeavor. It can be anticipated that all fields will be offering his assistance to make that the fundamentals of the surface phenomena of service a reality.

— —

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Research Needed by the Federal Dental Services*

Henry I. Copeland

Air Force Systems Command, United States Air Force, Washington, D.C. 20331

The dental research laboratories of the Federal Government function within the

Departments of Commerce ; Health, Education and Welfare ; and Defense. These laboratories and the multidisciplined scientists therein along with those of the Veterans Administration provide a much needed national resource. The Federal Dental Services have problems in common with each other and with the profession at large, in addition to having problems unique to themselves. Each Federal Dental Service needs a scientific staff capability to solve its own immediate problems or to contract for the solutions.

Key words : Capabilities and needs of Federal Dental Services ; dental research laboratories,

Federal ; dental staff ; Federal Dental Services.

years to develop to its present capability. This 1 . Introduction resource is needed by the nation in its struggle to Half a century ago it was an innovation at the cope with the ever escalating i^roblem of assuring National Bureau of Standards to apply the scien- oral and dental health to a growing population tific approach and the science of metrology to solve and to assist the Federal Dental Services in dis- a specific dental problem generated by the Federal charging their responsibilities. Services—how to assure the procurement, storage, To examine dental research requirements of the and delivery of a dependably usable material to a Federal Dental Services it is necessary to study Federal dental clinic. the functions of the individual service and the en- This 50th Anniversary of the Dental Research viroimient in which it operates. The Federal Section, National Bureau of Standards, is an ex- Dental Services have problems in common with cellent time to take a look at dental research prog- each other and with the profession at large. In ress, dental research resources, and how these addition, because of special operational conditions, resources may be best applied in the future—par- each service and agency lias problems 'and research ticularly to problems of the Federal Dental capabilities which are unique to itself. Services. The Department of Health, Education and Wel- Since metrology broke the ice with dentistry, fare (HEW) is the major federal instrument for scientists of other disciplines have become in- dental research support. HEW is concerned with trigued with dental problems and have contributed all aspects of the dental and oral health of the immensely toward the establishment of an orga- entire population. Its National Institute of Dental nized body of knowledge relating specifically to Research (NIDR) conducts broad extramual and dentistry. Concurrently, there has been a prolifera- intramural research programs primarily of a basic tion of laboratories devoted to dental research and dental research nature that are designed to ad- development. As a result, there has been a tre- vance fundamental techniques and to establish a mendously significant improA^ement in the methods, broader base of knowledge for development ap- materials, and equipment used in dentistry. plication. HEW's Division of Dental Health has been assigned the task of researching ways and 2. Dental Materials Research in the means of increasing the productivity of dental Federal Government manpower; of upgrading the teaching of dental personnel; of improving the distribution, orga- At least three cabinet departments of the Federal nization, and financing of dental services; and of government have dental research laboratories developing better preventive, control, and treat- Commerce; Health, Education, and Welfare; and ment procedures. HEW also maintains a uni- Defense. The Veterans Administration has an formed dental service as a part of the United extremely active research program with a number States Public Health Service (USPHS) which of laboratories. These laboratories and the dedi- participates in clinical investigations related to the cated scientists of many disciplines who staff them overall objectives of the department. The USPHS are a true national resource that has taken many has a special interest in indigenous ethnic groups along with its concern with the health of the popu- The opinions expressed herein are those of the author and do not represent official USAF policy. lation as a whole.

19 The Veterans Administration (VA) gives spe- answers to particular problems of the uniformed cial emphasis to the study of normal aging, chronic services and requires tangible results from funds and degenerative processes associated with aging, expended. Therefore, within the DOD, emphasis the detection and localization of oral lesions, oral is on development, test, and evaluation. Common tissue metabolism, and problems dealing with tis- to all the dental services within the DOD—Air sue restoration of war incurred wovuids or losses Force, Army, Navy—is the problem of how to caused by surgical treatment for oral cancer. The most efficiently and effectively manage an over- VA research program is a continuing effort to en- whelming demand to treat oral and dental disease. hance the ability of this agency to maintain the In this objective, the Federal Dental Services have oral health of debilitated patients. No other common cause with all elements of dental re- agency is so directly concerned with tliis popula- search—universities, industry, and other govern- tion group. The VA program has made important ment agencies. However, this demand is made more contributions to the research community and the acute by the requirement that the dental services profession through its cooperative studies with of DOD must maintain the effectiveness of the man schools, other government agencies, and related in the field, the integrity of air crews, ship and medical areas. In addition, it provides opportuni- submarine crews, and the personnel of support ties for performing cooperative studies as the need elements essential to all three services. arises. Within DOD, the Air Force, Army, Navy each The Department of Commerce (National Bu- have problems peculiai- to their particular branch. reau of Standards) (NBS) with its Dental Ee- Each service requires a capability for rapid research Section, which we are saluting at this 50th sponse to requests from the field for effective Anniversary Symposium, occupies a particular answers to practical problems that may not have niche in the hearts of all those interested in the application to the profession as a whole. There is advancement of the science of dentistry. Begun in a need for a capability to adapt and design equip- 1919 in response to the needs of the uniformed ment and facilities that meet special requirements services, its value to the services, and to the proof the individual agency; and there is a require- fession as a whole, is fully recognized by the ment for a user test capability for the evaluation leadership of the dental profession. Its research of equipment, materials, techniques, and manage- program is currently carried on in cooperation ment concepts. For example, the Air Force has with the Council on Dental Research of the Amer- long been concerned with dental problems asso- ican Dental Association, The National Institute ciated with altitude, acceleration, and, more re- of Dental Research, the Army Institute for Dental cently, man in space. The Navy has had a special Research, the Dental Sciences Division of the Air interest in problems related to extremely cold Force School of Aerospace Medicine, and the Vet- weather and extended submarine operations. The erans Administration. Army has concentrated on solutions to the need to The unique capability of NBS for physical re- treat their large number of patients with traumatic search is not duplicated anywhere else in this injuries and those needing prostheses as a result. country. All dental health components having a To meet its requirements, both common and direct interest in this area contribute to this capa- special, each Federal Dental Service needs an bility through the activities of the Dental Mate- informed staff representing a sufficiently broad rials Group, a constituent of the International spectrum of professional, scientific, and techno- Association for Dental Research. Representatives logical disciplines. The staff must have a suitable of organized dentistry, the Federal Dental Serv- facility in which to work. The staff must maintain ices, and the dental industry unite with an inter- an awareness of current knowledge in dental and change of ideas in this activity because it is one related sciences. It must assure a close liaison with of a kind. Their affiliation with different levels ex- universities, industrial, and other government lab- tends also to the Federation Dentaire Interna- oratories. The research staff of a Federal Dental tionale and the International Organization for Service must know who is studying what and Standardization. All of this adds to the singular which laboratory has the expertise and apparatus effect of NBS on the dental research community. to address particular problems. The staff must Another unique feature of the cooperative pro- perform investigative studies pertinent to opera- gram is the fact that participating agencies have tional problems peculiar to its own service. To had the advantage of assigning guest workers to augment and complement this capability each Fed- the cooperative program and thus have gained eral Dental Service needs a provision and funds staff with capability members a knowledge and to negotiate contracts for special studies and for in research disciplines. cooperative efforts in areas of mutual interest. In the Department of Defense (DOD) context, research is thought of as research and develop- 3. Summary ment. The broader connotation encompasses all research, development, test and evaluation In summary, the Dental Research Section of the (RDT*&E) activities sponsored by DOD. DOD re- NBS triggered the impact of basic sciences on the search in the dental area is concerned with specific dental profession. This impact and the results of

20 it, have had a most beneficial effect on the profes- have a continuing need for research in areas ap- sion and the services it provides the nation and plying to the population as a whole and to specific humanity everywhere. Recognition of these facts problems relating to special operational condi- has led to a greater interest and support of dental tions. Finally, each Federal Dental Service needs research by the government. There has been a a scientific staff capability to solve its own im- minimum of duplication due to the special objec- mediate problems or to contract with imiversity, tives, interests, and requirements of the various industrial, or other government agency labora- agencies involved. The Federal Dental Services tories for their solution.

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Resbabch, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Research Needed by the Dental Industry

John F. Glenn

Central Research Laboratories, Dentsply International, Inc., York, Pa. 17404

Dental industry requires new research information wliich must come from nonindustrial research-oriented institutions. Two major requests of dental industry are for basic research and better clinical coiTelation with laboratory data. Many compilations of physical test data on a variety of existing materials or ill-deflned experimental ones are being reported, but the major need is for evaluation of the.se same materials in well controlled eUnical applications. More rapid access to results of investigations, especially government-supported grant. research, is needed.

Key words : Clinical and laboratory data correlation ; dental industry research needs ; dental

materials ; dental research, necessity for rapid communication of.

1. Introduction The profit motive, as a basis of operation, creates very profound differences in research personnel The research needs of the dental industry could and research eff orts in the dental industry com- be discussed from several viewpoints ; for example, pared to that of the so-called supported institu- the need for money, for equipment, for personnel, tions. The goals of our research efforts must be better management and administration, and so on. specific and directed towards getting answers All of these needs are internal within each orga- quickly to the multitude of problems at hand, not nization in the dental industry and are essentially just any answer, but one which will be bought by our own problems. the customers. Moreover, this answer must be ob- On the other hand, I have chosen here today to tained in the most economical way and its manner discuss the subject of the research needs of the of pursuit has to be contmually justified so that it dental industry from the aspect oi needs that can will be acceptable and will produce the desired be satisfied by external sources. I would like to profit. The attendance at meetings, the writing of consider how the laboratories supported by gov- papers, the delaying completion of the project for ernment, the schools, and others, whose only busi- several months or even a year to investigate a ness is research and whose only products are in- tangential but fascinating avenue of research, all formation and research papers, can assist us in play a very secondary role to the industrial scien- furthering the work of the dental industry and tist in his struggle to get materials out the front thereby help dentistry in the long run. door. I believe we, as a group, would like to be as As stated, this discussion is directed to the needs serendipic as one can be in the academic world, but of the dental industry and needs, by their very these pursuits have to be very materially tempered the economic press. nature, are something that we do not have or do by The dental industry is quite small as chemical not have enough of. The assessment of one's needs industries go. The sale of its products is limited to must always be based on a critical examination of the dentist population and the quantities involved what is now available. This is exactly what I have for sale to this limited group of customers are tiny. done. It is certainly intention not my to fault any Therefore, the industry is just not financially able group whose services may lie within the area of to support extensive research programs aimed at this discussion. On the other hand, it is intended developing fimdamentally new information about merely to review the situation with some sugges- matter, and must of necessity look elsewhere for tions that perhaps "something must be added." this kind of data.

2. Dental Industry Background 3. Research Needs of Dental Industry

By way of background I should point out, first, In the last few years a great deal of money and that the dental industry is in business to make and effort has gone into dental research. In a published market materials and equipment for use by the survey for 1966, PePlman said research grants were dentist, the dental laboratory technician, and ulti- allotted 10.2 million dollars, scholarships 1.4, and mately the public. It is supposed to be a profit- training grants another 5.1 million. There were making industry. reported to be 1,288 researchers in schools in that

23 year. Undoubtedly this expenditure has continued it seems to be in competition with the dental indus- and even increased in the last years since then, and try development laboratories which generally are the likelihood is that such support will continue better equipped for product development through into the future, even though prospects look some- their own orientation, close contact with the dental what gloomy right now. This will not last—we profession, incentive, and even the profit motive. just cannot afford to cut off even the majority of It seems to us from where we sit that there would ithe projects sponsored by NIDR, Public Health, be a greater contribution to the industry and to the Army, the Navy, and the Air Force, and the dentistry if the research-oriented institutions internally supported projects within some of the would establish the principles on which products dental schools. can be developed. In the past there has been some Perhaps, however, this is a time when the re- history of doing this pioneering, for example, search efforts that have been made in the past and Taylor and Demaree in the spherical alloys, Bowen are currently going on, should be examined with in his composite restorative materials, Brauer in regard to goals and their real value to dentistry the inauguration of the EBA's, Phillips putting and to the dental industry, which depend on and fluoride in filling materials, among several others. must use much of this information to provide bet- However, there seems to be a tendency of late ter dental care. With all the effort that is being through some pressure or other to develop complete put into this type of work, I feel it is very oppor- or more or less finished products. Perhaps the in- tune that the researcli group in the dental industry dustry is somewhat at fault in not taking up the is invited to indicate its needs. ideas coming forth fast enough and gives the ap- To make this discussion more meaningful, I have pearance of inadequately fulfilling the best needs taken the liberty to carry this invitation to the in- of dentistry. dustry personnel and have conducted a small sur- As an illustration of this need for basic work let vey of a number of dental-materials industrial us consider, for example, the program of the 1969 researchers asking them to* indicate their needs meeting of the Dental Materials Group—lADR. I from research oriented institutions supported out- have made a tabulation of the papers that were side of the dental industry itself. While I must presented and arbitrarily divided them into four admit that some of the comments and the frank- subject categories, determined from the title, the ness of those interrogated were surprising to me, abstract, and listening to some of them myself. there were certain dominant clearcut ideas almost Papers which could be considered as descriptions universally expressed which could be valuable as of tests, new or otherwise, constituted 15 percent of a guide in plaaining the nature of future research the presentations. New and more-or-less finished projects. materials not previously described were the subject of 14 percent. Papers which discussed some aspect 3.1. Need for Basic Research of existing materials accounted for 59 percent, of which 38 percent dealt with the subject of amalgam First, there is a crying requirement for more alloys, or amalgams, or handling amalgams. Basic fundamental or moi^e basic research in dentistry research papers concerning some phenomenon or and secondly, for more and better clinical corre- reporting facts which could be construed as con- lation of the laiboratory test data that is being pro- tributing to the general fund of dental informa- duced from many sources. In these two areas the tion, constituted only 12 percent of the offerings. dental industrj^ is just not geared to function, espe- These percentages more or less speak for them- cially, in basic work. In contrast, research insti- selves and would substantiate, I think, the need tutes and organizations are usually equipped with mentioned. There is an overabundance of testing, excellent instrumentation or at least instrumental retesting, and testing again, on current materials, availability, and in my opinion they have person- ad nauseum. nel who are well oriented to and are quite capable Be all this as it may, at least to industry per- in the basic sciences. They should be supplying sonnel, it appears that much of the data cannot be fundamental information, for example, on teeth, used and more attention could be given to acquir- their composition, the forces involved within the ing basic information. tooth, the nature of changes of oral tissue upon application of various dental procedures, or with 3.2 Need for Improved Clinical Correlation in time, the physical requirements which must be met Laboratory Testing for successful restoration of the function of oral tissues, just to mention a few. I am sure you are The second area expressed as being greatly well aware of the basics in dentistry. needed is more effort directed to better correlation In a number of instances research institutions of the quantities of physical data being accumu- have taken upon themselves or have been directed lated today, with clinical function. The clinical by someone to engage in activities which are inter- evaluation" of materials and processes, whether it preted as product development to fulfill some spe- be dental or medical is extremely difficult, prob- out. cific urgent need or other, which is current at the ably the most difficult type of research to carry moment. No matter how one looks at this endeavor. It requires very careful and detailed planning, the

24 institution of extensive controls, tlie development be made available to dentistry through the indus- of careful and uniform observation techniques try. That is what the industry is for. There is no which at best can be only empirical and subjective, need to bury research work in secrecy until the the selection of satisfactory material in sufficient project is completed or dropped, especially when numbers to be really significant, and some sort of it is supported by public funds. It has been sug- useful and understandable tabulation of the rested that this whole matter of communication be f results. All this is tedious, time consuming, expen- ooked into by the several grant-issuing agencies sive, and quite frequently disappointing in the and that they do some soul searching. If research clarity of answers that result from the experiment. institutions wish to convey the fact that it is their Also, more times than we would like to admit desire to aid the public and to produce better den- before the experiment can be completed, the ma- tal services ultimately through their work, then it terial being evaluated becomes obsolete and then would certainly seem that it would pay to team up the experiment is reduced to an exercise in futility. by better communication with the industrial re- There have been a few brave souls who have ven- search groups, who will have to produce the prod- tured forth into this area and are producing some uct eventually anyway. As it stands now the usual very notable contributions such as Ryge, Paffen- channels of publication and reporting are woefully barger, Myers, Phillips, and a few others. Unfor- inadequate and behind in keeping up with the tunately, the formidableness of this type of re- work that is being done. As suggested, the burden search has scared off, or so it seems, many who of the dissemination of information lies with the actually are equipped to contribute greatly. I fail sponsoring agency and much more would be ac- to understand why the materials departments of complished if these agencies would recognize this most dental schools, with their attendant clinical responsibility and attempt to issue informative facilities, do not take up this opportunity to con- reports on a regular, rapid, and broad basis. tribute to dental materials development in a useful These three points constitute the major areas manner. which were almost universally mentioned as The dental industry is not actually interested needed by dental industrial research personnel to in comparing one brand of material with another pursue their efforts in the adaptation of ideas and brand of the same type of material, although this findings of research laboratories and make them seems to be a very favorite endeavor. We are inter- practical in the form of dental products. ested, however, in learning what the generic types of dental materials do to the oral tissues and, of 3.4 Additional Dental Industry Research Needs course, what the oral tissues do to materials and what are the requirements from a clinical view- There are sevferal other areas which might be point for satisfactory replacements. said to be of lesser concern to industrial research personnel, but should nevertheless, feeil, The dental materials industrial researcher is I be constantly asked to produce and continually im- mentioned. prove materials and equipment and to design pro- One point is the request by many for the devel- realistic tests. cedures for replacement of tooth and other tissues, opment of more physical At the but few guidelines exist as to what strength, hard- present time, the testing procedures for dental ma- ness, fatigue, or other detailed characteristics are terials are more or less haphazard in their vailidity of basis of selection. tests actually required. It is somewhat akin to trying to The that are part of the various specifications are based a physical build a bridge without knowing what is going to on few properties of existing products the limitations go over it or how long it will be. Good clinical and study will assist materially in answering the ques- are those of the existing products. While we may tions with facts that are needed in the design of be getting by with this at the present time, we should be striving for well organized tests that will materials. Here then is our second major need, a simulate aspects clinical serVice the difficult one no doubt, therefore a real challenge, so of and define service attributes of a material specifically. let us pick up the challenge. more Such tests, coupled with actual clinical observa- 3.3 Need for Improved Communication tions, will establish what limits are really necessary from a functional standpoint, and the re- of Current Research Results quirements of the specifications will fall in line and be more meaningful in the judgment of The communication of the results of investiga- materials. tions by research institutions to others outside of It has been whispered that some of the physical their cwn agency, which sponsors the research, has test data are being developed by individuals who not been the best. In fact, it is quite poor. In many have a veiy limited background in the science of cases, the dental industry and other interested par- dental materials. They sometimes arrive at impli- ties have had to wait until the information was cations and conclusions supported only by a bit of published in a journal or in a report iss^ied by the statistical jugg'ling. There are many grants at pres- National Technical Information Service a year or ent, so called training grants, issued to individuals so later. When something that is worthy of note is who carry out a few physical tests on one product accomplished by a research institution, it should or another. Eventually the results are published.

25 Unintentionally, some misleading infonnation is Specification Committee and the addition of put into our journals, especially the popular ones. others, the tempo of specification writing and test The publication of this type of data and the procedures development has increased notably. manner in which it is developed and concluded, A.ny good specification for a material will have be- unfortunately for us, gives the impression right hind it a battery of tests which will have been or wrong that there is some incompleteness in developed through actual laboratory evaluation dental research. Perhaps this is part of the train- and the amassing of sufficient data to prove that ing grant situation and, if it is, then the procedure the tests can be carried out with reasonable uni- should be altered so that the trainee can still get formity in several laboratories by a number of this needed support but merely write a term paper different operators. Much of this work is going which will be buried in a library some place or on at the headquarters of the ADA, which is con- in somebody's desk drawer. This could be quite tributing very heavily and effectively to this effort. easily accomplished by the sponsoring agency A nurnber of schools have been cooperating and are changing its publication requirements, so train- spending time and money on test development. ing becomes indeed a training course and not a Likewise, industrial research laboratories are paper mill. doing the same and have contributed in a great Up to now I have probably sounded, although measure to the mass of data that backs up a num- unintentionally, somewhat critical of the present ber of these tests. state-of-the-art, so let us turn to a couple of farther All of this work costs money, considerable out thoughts for the future. money. Why not set up a series of grants to any A single central research control agency, possi- institution, including industrial laboratories, for bly in the ADA or NIDR or other governmental test development and data accumulation for speci- institution is suggested to consider the problems of fication work. Good physical facilities are avail- dentistry and parcel out the projects and investi- able in the industrial laboratory, as is wide testing gations on the basis of dental need and likelihood know-how. Use should be made of it, but at the of success, rather than fancy proposals. Perhaps same time it could be supported and the burden this is already established, but it is not evident spread. to us. 4. Summary There are several agencies at the present time which attempt to pass out research projects, but I have discussed with you briefly here the re- to industry people there appears to be a com- search needs of the dental industry—the needs petition between these agencies even though there that can come from outside the industry itself, probably is not. At any rate the job has the ap- from institutions that are in the research business pearance of not being done and there seems to be and supported by grant funds from one source or considerable duplication and needless effort. A central agency could control this effort more usefully. another. More extensive basic dental research, For many years there has been much effort to carefully controlled clinical studies to correlate develop specifications for existing materials. Re- physical laboratory data and establish realistic cently, with the formation of the USASI Com- limits, and better communication, are the funda- mittee from the older Dental Materials Group mental needs of the dental industry.

26 ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseabch, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Research Needed for Dental Education and Practice

Ralph W. PhiUips

School of Dentistry, Indiana University, Indianapolis, Ind. 46202

The current research effort in dentistry, and that which will occur during the next decade indicate that dental practice will be characterized by certain trends. These trends will require an acceleration of research on adhesive film forming systems, sealants for pit and fissures and means of improving anticariogenic characteristics of restorative materials. The interdisciplinary aspects of dental materials will be increasingly important. Greater focus will be on the biological properties of materials, materials and devices for oral rehabilitation, materials for implantation and the interaction of tissue and substances used to restore lost tooth structure. Further definition of the oral cavity can be expected to become a responsibility of the materials scientist. Increased emphasis can be expected in dental education on a more intimate basic science-clinical application orientation.

Key words : Dental education ; dental materials, research needs ; dental practice future

dental research ; materials, dental, research.

vide of dental service, the role of 1 . Introduction a broader type the dental auxiliaries will be expanded, as will We are now living, and shall continue to exist, their training. in a research oriented society. While too little Let me defend these predictions and attempt to attention appears to be focused on the quality of relate them to research in dental materials science. research, the sheer magnitude of the investigative effort in science is impressive. The health profes- 2. Future Trends in Dental Research sions, and especially dentistry, are no exceptions. It would be naive and dangerous to believe that In recent years, at least one-third of the dental this ever broadening base of knowledge will not research effort in the United States has been de- change decidedly the nature of dental care and the voted to ways for preventing dental caries while content of the dental school curriculum. It is upon another one-third has been concerned with elimi- that premise that the following remarks are based. nating diseases of the soft tissues. The response of It is indeed difficult to trace a profile of any the profession to the rapid escalation of research profession a decade hence, particularly in one such in these fields has been one of increasing alarm as dentistry where such a marked transition is that coincident with any substantial reduction in occurring, not only in the specific treatment of the incidence of dental caries and/or periodontal dental disease but also in socio-economic attitudes disease will be a reduced need for the dentist, a that have not as yet acquired a definite character. fear that he will eventually be phased out of the I am all too aware that even with much more health professions and will become somewhat of tangible data, predictability may sometimes go a historical curiosity. The outcry first occurred awry. when fluoridation became a recognized mode for However, it is becoming apparent that certain the arrestment of dental caries on a mass scale. trends are crystallizing in dentistry and I would To be sure, the goal of any health profession is characterize them as follows : ( 1 ) The present in- that research will lessen the demands that are roads that are now being made in reaching a total placed upon it. comprehension of the etiology of dental caries and Preventive measures such as fluoride therapy, designing the related therapy to arrest this disease which were the fruits of dental research, have al- will soon make certain dental restorative proce- ready altered the nature of dental practice. How- dures obsolete and the general nature of dental ever, they have certainly not led to a lessened need service may be quite different from that of the for the dentist or his auxiliaries. To the contrary, present. (2) With more emphasis upon the dental dental research has permitted the profession to ini- specialties, group practice will become more popu- tiate programs of total oral care rather than sim- lar. (3) Greater demands will be made upon reply playing catch-up with the overwhelming back- search and teaching programs that will extend log of dental treatment that has existed and which dental treatment to the entire population. In other will be required by an ever expanding population. words—community dentistry. (4) In order to pro- Furthermore, in being offered superior service, the

452-525 0—72 3 27 —)

public has become more dental conscious and mo- If one accepts as a basis for discussion that such tivated to seek dental care—^tlius actually increas- trends could be the natural outgrowth of current ing the demand for a lower dentist-patient ratio. programs of dental research, where does the den- Parallel to this apprehension of the young dental materials scientist fit into such a scheme ? First, tist of the impact upon his profession of research as for the dentist, the time allocated to certain in the biological sciences, one often hears depress- materials or technology, will be weighted entirely ing predictions coming from the dental materials different from allocations of today. Materials and teachers or researchers. Since the remaining one- concepts that at present may receive only a casual third of the bodies and monies now involved in glance will come to the fore while some of the clas- dental research are directly concerned with im- sical systems upon which we have lavished our proving dental materials and instrumentation, it energies will be of secondary interest. is predicted that this research in the physical sci- For example, although this is almost sacrilege ences will shortly spill forth an avalanche of su- to say, in time dental amalgam will no longer perior materials and that most of the former occupy its lofty seat as the material upon which problems associated with the clinical failure of the attention of dental materials researchers has materials and appliances will be conquered. In for so long been concentrated. For all of its merit turn Utopian technics will attain such a degree of as a system that has served dentistry so heroically perfection that further research will be unneces- and one that has offered so much scientific intrigue, sary. It is further rationalized that as dental dis- the need for research on dental amalgam will ease becomes better controlled through preventive diminish since it will be replaced by other mate- measures, it must therefore follow that the need rials or therapies. The same may be said for a for restorative materials and appliances will be- number of other materials, such as silicate cement come less important, as will the personnel in- and the current resins. They will be surplanted in volved in their development. time (and please don't ask me to suggest a target share this alarm. Conversely, the theme I do not date ! ) by adhesive restorative materials that will of this paper is centered on an opposing premise hopefully approach or duplicate the physical and namely, that although the teacher or researcher in chemical properties of the tooth itself. dental materials in the year 2,000 will likely be It would be naive if I were not to acknowledge concerned with entirely new problems and will that this learned audience is quite sensitive to the be attacking them in a vastly different way, his importance of research in the area of dental ad- relevance to the profession will be even more hesives. However, since the title of my assign- secure, as will the horizons for a meaningful ca- ment was to be centered upon the future needs of reer in research or education. dental materials research as they relate to prac- As I have suggested, any realistic evaluation tice and education, this subject affords an unusu- of the accomplishments that are now being made ally good example of the type of investigative in providing a means for irradicating dental caries effort with which we should and will be occupied. clearly indicates that the magnitude of this disease No other affords greater opportunities for a broad will diminish in a somewhat regular order. If a spectrum of dental applications, for the outlet of major breakthrough should occur, restorative den- an imaginative mind, or for the interplay between istry will then be concerned principally with re- disciplines that have historically been the trade- placement of the fractured tooth or the one lost mark of the science of dental materials, and of from periodontal disease. In any case, it is reason- the 50 year leadership given to us by the dental ably safe to predict that within the forseeable program at the National Bureau of Standards. future the carious lesion will no longer be the Adhesive systems afford opportunities to render with exist- focus of attention, or of time, of the profession. a type of dental care that is not possible they fit nicely into the The arrestment of dental caries will in turn trig- ing materials. Furthermore, of the future practice of dentistry as ger other widespread changes in the nature of framework I have sketched it. For example, the elimination dental practice. For example, the need for endo- of the microleakage phenomenon Avould alter much dontic treatment will become less. Since fewer of the basic biological and physical concepts that teeth will be lost as a result of caries, a lower now prevail in restorative procedures. The adhe- of prosthetic appliances will be required. number sive dental restoration would conceivably be less Thus it might be envisioned that the typical complicated to place, biologically superior and dental practice of the future will emphasize : ( 1 ]Dhysically improved. However, the true lure of preventive dentistry, (2) restorative measures that dental adhesive molecules lies in uses other than will be necessitated from the loss of tooth struc- for the restoration of a carious tooth or the ture due to reasons other than caries, (3) the attachment of orth,odontie brackets, important to maintenance of the health of the supporting tis- though these may be. Adhesive films applied inhibit the for- sues, and (4) the treatment of dental disorders tooth structure, if dura;ble, could that have previously been neglected by virtue mation of the dental plaque and thereby the caries of the overwhelming preoccupation with the rav- process or the deposition of calculus. Certainly ages of caries. adhesive phenomena, yet to be adequately explored

28 by the dental materials scientist, are involved in will slowly and methodically, and I would em- the retention of dentures, the bonding of ceramics phasize the latter, shift from studies such as the to metallic restorations, the retention of re- niceties of phase reactions in amalgam or the implanted teeth and of resin implants to tissue dimensional changes that occur in gypsum prod- and in the efficacy of maxillo-facial appliances. ucts. Merely to further accumulate such informa- Each of these uses is different in the sense of tion as a sort of capital investment will not be the structures that are involved, the respective insufficient. Instead the scientist will be explorin,g the terfaces and phases, the conditions to which the interaction of polymeric systems to both hard and bond will be subjected, and the physical and bio- soft tissues, will develop criteria for relating biological properties of the system. The scientist in logical characteristics of materials to their struc- dental materials is the most likely candidate to ture, will define the exact nature of tooth structure lead meaningful programs in the development of at interfaces, and will better characterize the stress such compounds, by virtue of his experience with patterns placed upon dental restoratives, appli- the physical behavior of materials in the oral ances or films. cavity, the highly specific test methods and specifi- I have used the example of adhesion because it cations he has developed for assessing materials does embody so well the changing complexion of and his obvious ability to synthesize various dis- dental materials research and its application as I ciplines into one that serves the multiple and com- see it, as well as the responsibilities which may plicated demands that dental treatment requires. thereby accrue. I have suggested, by using this There is yet another example that might be particular illustration, that the breadth of dental used to illustrate the need for adhesive systems materials research and education will range even since it is one which could be part of the routine further from very complex problems to those that type of therapy offered in the dental practice of are inordinately simple. For example, to develop a tomorrow. I am, of course, referring to a sealant system that will adhere to the formidable tooth for pit and fissures in the child patient. This is a structure in its aqueous environment will continue dental materials problem, even though it is essen- to require the attention of unusual research skills tially preventive dentistry. A number of persons and talents. On the other hand, the application Of in this audience are already deeply involved in that adhesive to a pit and fissure will be a very this area of research. An adhesive and durable simple treatment as compared to existing opera- restorative material that would prevent the prop- tive procedures. Thus one sees the need for so- agation of caries in these vulnerable areas has phisticated research done in great depth, to be great import to the treatment of mass popula- paralled by training programs for the dentist and tions. As I stated earlier in this paper, I am con- especially for auxiliaries who may be required to vinced that the profession will be forcefully utilize the end product of that research. moved, both by direct means and emotionally, to As I just stated, the role of adhesive molecules provide dental care to all segments of the in dental application is merely one of many that population and in all countries. Of immediate could be cited to indicate the future nature of application in mass care would be the arrestment dental materials research and its application. of existing caries or its prevention in the young These include the evolution of implant materials population. Adhesive sealants, particularly if for prosthetic appliances, of materials that will be anticariogenic, could become a major treatment more esthetic and superior in properties than the objective for such programs. Furthermore, it ones now used for maxillo-facial appliances, of would not be heresy to suggest that in certain methods for providing a more precise restoration instances this type of care might be administered of teeth that are lost from accident or diseases of by properly trained dental auxiliaries. the soft tissue, of ways for producing mineralized I am certain that most of you are in complete tissues or methods for protecting such tissues and sympathy with the concept that we must lend our specialized materials for all of the various dental assistance in every way possible to make dentistry specialties. As dental care becomes more specific, available to everyone and especially to those in greater emphasis will be placed upon diagnostic underdeveloped areas in this and other countries. aids and devices for maintaining a better control The mode for accomplishing this is indeed con- of dental treatment. The researcher and manufac- troversial and does not concern this group. What turer of dental materials long ago demonstrated does is the development of materials and the train- his ingenuity in that field and will no doubt be ing of people to cope with the magnitude of the called upon to develop technology for monitoring problem. Those materials and that type of train- physical and chemical changes in dental structure, mg may be considerably different from what now for minimizing the human variables in providing prevails. dental care and for ways to reduce the cost of the If one acknowledges that the broad area of dental operation. dental adhesives is a reasonable example of the Whether it be fundamental research in such type of research that fits into the context of "Den- divergent areas or other programs designed to tal Practice in the Year 2000", then it seems logical further the education of the dentist and his aids, that the interests of the dental materials scientist the point has been made that the field will remain

29 a most viable one, hopefully unfettered by previ- the biological characteristics of materials and the ous tradition as to its limits. Personally I have interaction of materials with their environment. every confidence that the scientists and educators The dental materials science course could very well now in this field do have the wisdom to make that encompass the physical structure and physical con- adjustment and rise to meet, not obstacles, but an stants of ail oral tissues. everchanging science landscape. Elaborate apparatus plays an important role in science today 4. Summary but we are too inclined to forget that the most important instrument in research must always be In summary, I am convinced that our future lies the mind of man. Kettering has said, "A problem in a greater interdisciplinary approach, not only is not solved in the laboratory. It is solved in some in the specifics of research on materials but in re- fellow's head. He only needs the laboratory ap- lating that research to the other physical, biologi- paratus to get his head turned around so he can cal, and clinical sciences. Furthermore, communi- see the right thing". A multitude of opportunities cation on an international basis at a level far for scientific enrichment in our discipline can be greater than that which now exists is essential in seen without much movement of the head. order to make best use of the talents now available, to prevent duplication of effort, and to concentrate 3. Future Trends in Dental Education a multitude of single isolated investigations into unified attacks on problems of mutual interest. I I would like to spend just a minute to relate the guess what I am saying is that the profession will foregoing sketchy remarks to dental education. If demand more from this science than any other dental education in the future assumes the same discipline in turns of adjustments to newer areas posture that has prevailed in the past, it by neces- of investigation and at a high level of competency. sity will reflect changes in dental practice. There- At the same time it will be charged with accepting fore, as emphasis is placed upon the dental special- a greater responsibility in the training of individ- ties, on community dentistry and upon caring for uals who will deliver dental service. The dentist all types of dental disorders, the dental school will request, and rightly so, materials and tech- curriculum will witness a more intimate relation- nics for handling a different type of dental prac- ship between the basic sciences and the clinical sub- tice than he is conducting today and he will ex- jects. No doubt this will be reflected in a more ver- pect educational programs for auxiliaries that will tical alignment of courses. This will be true for be heavily weighted in the subject matter pertinent each individual science, including dental materials. to usage of these materials. Coincident with this, Every effort will be extended by dental school ad- one may anticipate that the dental manufacturer ministrators to organize a proper atmosphere to will in turn become more specialized, skillfully fertilize this type of orientation. developing what at present might be called bizarre The dental materials course must eventually products. change to reflect the new duties of the dentist and I guess my major concern in this transition lies the changing nature of his practice. Certain main our capability and desire to attract future re- terials will become obsolete, others will receive searchers and teachers. We are now and will con- greater stress. In all likelihood the systems and the tinue to be competing with scientists and educators techniques that will eventually be paramount are in fields that now appear to have a greater allure as yet still under research. In the undergraduate and upon which the future practice of dentistry training of the dental student it seems logical that seems to be more directly based. I contend that the there will be less emphasis upon those materials science of dental materials will comprise a pillar and technics that will have become principally the for that foundation as strong as that of any other responsibility of the technician, assistant or hy- discipline—but only if it accepts a critical, reflec- gienist. However, as dental schools assume a tive attitude so that it can envision and react to greater role in the training of dental auxiliaries, the new challenges which a changing profession the content of the courses given to those groups will offer. If it does, then the values and rewards will be considerably strengthened in materials and are such that this science can compete on a scien- it will no doubt be presented at a higher level. tific and academic level equal or superior to that In addition, there will be greater emphasis upon with which we are now blessed.

30 III. Metals Research

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued Jime 1972).

Amalgams in Dentistry

Knud Dreyer j0rgenseii

Kebenhavns Tandlaegehejskole, Copenhagen, Denmark

Corrosion is a major factor in the failure of amalgam restorations through the deposition of corrosion products which may cause chronic periodontitis or through corrosion fracture of the margins which may cause secondary caries. Galvanic corrosion attacks only the tin- mercury 72 phase. Mercury released by this corrosion causes a mercuroscopic expansion resulting in a deflection of the amalgam from the cavity wall at the margin. Corrosion can be considerably reduced by minimizing the porosity of the amalgam restoration through overfilling the cavity, burnishing the margins, eliminating excess by carving and using zinc-containing amalgams.

Key words : Amalgam, dental ; amalgam, dental, effect of technique on corrosion ; clinical

failure of amalgam corrosion, dental amalgam ; dental materials marginal fracture, dental ; ; a malgam mercuroscopic expansion, dental porosity of dental amalgam. ; amalgam ;

of foils, particularly foil. those days 1 . Introduction metal gold In it was a very tedious task preparing a good filling The 50th anniversary of the Dental Eesearch of gold foil; it frequently took more than three Section of the National Bureau of Standards coin- hours. By comparison amalgam was advertised as cides with another noteworthy date: it is exactly a material of Which a filling could be made in just 150 years ago this year that dental amalgams first three minutes, painlessly and at no discomfort to made their appearance in the service of dentistry. the patient. Only some of the details of the amal- Countless teeth have been restored and preserved gam techniques of those days are known but there in the course of the past century and a half, thanks is no doubt that cavity preparation techniques, to the development of these materials. There is matrix techniques, and the actual amalgam tech- some uncertainty about who can claim to have in- nique were extremely primitive. It is scarcely sur- vented amalgams for the restoration of human prising therefore that amalgam fillings frequently teeth; it seems to have been the British chemist, caused very serious secondary injuries, and that Charles Bell, in 1819, who produced an amalgam many dentists reacted strongly against the appli- of silver filings and mercury; his product was cation of amalgam as a filling material. advertised under the name "Bell's putty". A few The debate on the use of amalgams in operative years later, in 1826, the French dentist, Onesiphore dentistry was confined almost exclusively to the Taveau, claimed to have invented an amalgam suit- United States, where the material had been intro- able for filling carious teeth; the material was duced by the Crawcour brothers in 1833 and where made by mixing mercury with filings cut from sil- the famous amalgam war raged imtil 1856. It was ver coins and went under the name pate d'argent, in this country too that almost all the advances took i.e., silver paste. place that have since led to the amalgam alloys We know that amalgam as made according to and amalgam techniques of today. The people who Bell and Taveau had several unfortunate pro^Der- headed this progress were men like Josiah Flagg, ties : the consistency of the freslily made mixes was founder of the constructive critical-analytic harsh or sandy, they hardened slowly and were method in operative dentistry generally, and amal- subject to excessive expansion. It was soon discov- technique in particular ; G. V. Black, who on ered, however, that the latter two drawbacks could gam the basis of extensive and detailed experiments de- be greatly reduced by adding tin filings to the silver filing or by cutting filings from an alloy of sil- fined the optimum composition of the amalgam ver, copper and tin, i.e., the metals which even alloy, and also systematized the technique and today are still the main components in dental amal- morphology of cavity preparation; Wilmer Sou- gam alloys. der, the first head of the Dental Section of the Na- The dental amalgams were received with great tional Bureau of Standards, recognized as the interest by the profession, which is not perhaps founder of the science of dental materials and surprising in view of the apparent simplicity of author of the first standard specifications for a den- amalgam technique; the filling materials which tal material, viz., amalgam alloy ; and many other amalgams could replace were primarily a variety authorities most of whom are present with us today

33 . —

and still active in the field of research into dental materials. It is important for me to stress that all the research carried out today in the amalgam field is based on the comprehensive and carefid studies of these authorities—and would not have been possible without their vital contributions. All organized research requires a specific objective in order that we may formulate and attack the appropriate problems. The study of dental amalgams is pure applied research, which means that it is justified only inasmuch as it is demonstrated to be oi use to operative dentistry. The debate in earlier decades was on whether amalgams were of any use whatever filling maiterial. Modem amalgam research on the other hand is given the task of examining (1) the extent to which amalgam fillings fail, (2) an analysis of Figure 1. Typical corroded Class II filling with the gingi- the physical and chemical processes leading to val and central area covered by a crust of dark products of corrosion while the occlusal portion remains clear failure, the influence of materials and of and (3) corrosion. technique on these processes, particularly with a view to discovering whether it is possible by alter- ing the composition and properties of the materials and by changing the technical working methods to halt or delay the processes.

2. Definition of Amalgam Failure

A failing amalgam filling can be defined as a filling that has been a contributory cause of secondary injuiy in the organ of the tooth, i.e., the tooth itself and its surrounding connective tissue. In the following discussion it is a condition that the amalgam fillings initially Avere morphologi- cally correct, which means among other things that the surfaces of the fillings were smooth and that a careful inspection revealed neither marginal excess nor marginal deficiency. It is moreover a condition that the fillings have been made of alloy and mercury, fulfilling the requirements of the relevant ADA specifications. The only forms of secondary damages of clinical significance are chronic gingivitis and secondary caries. experimental Chronic gingivitis occurs in this connection Figure 2. Enlarged view of the surface of an non-contaminated amalgam specimen showing small sur- when solid products of corrosion, promote which face Misters and solid corrosion products after four the retention of materia alba and plaque, are de- months exposure of the original smooth polished surface posited on the surface of amalgam fillings on the to artificial materia alba. Magnificationr-HO X peripheral surfaces of the tooth in close relation to the gingiva. Figure 1 shows a typically corroded corrosion. Although the accepted view of the Class II filling in which the gingival and central causal rdation between corrosion of fillings and one-third part are covered by a crust of dark prod- gingivitis is to some extent based upon presump- ucts of corrosion, the occlusal portion remaining tion, it seems to be probable enough to provide suf- clear of corrosion. Figure 2 is an enlarged view of ficient grounds for a study of the circumstances part of the surface of an experimental noncontami- influencing the process of corrosion. nated filling from a laboratory experiment; the filling had originally been smooth and polished 3. Corrosion of Amalgam but after four months' corrosion in artificial materia alba its surface was roughened by small Schoonover and Souder were the first to discover surface blisters and solid corrosion products. The galvanic corrosion attacks only the tin- that _ small hemispherical nodules are drops of mercury mercury 72 phase, and that only the tin dissolves covered by a thin crust of deposited products of the released metallic mercury remaining in the . :

amalgam; much of the tin is deposited locally, either within the corroded amalgam or on the amalgam surface. Many subsequent studies have served to confirm this view. It is typical of amalgam corrosion thalt in oral conditions it can occur only in the presence of a difference of oxygen concentration on the amalgam surface ; the area with the lowest concentration of oxygen forms the anode. Areas of this nature regu- larly occur in the most gingival parts of fillings on peripheral surfaces where the amalgam is covered by gingiva or plaque; the surfaces of the filling facing the cavity, and the walls in the pores of the amalgam also have a relatively lo w oxygen tension. Since it is only the 72 phase that is subject to galvanic corrosion, it is importanit that we become familiar with the morphology of this phase. If the individual 72 grains do not anastomose, only the surface grains in contact with the electrolyte can be dissolved by corrosion—which would have to cease once this dissolution had terminated. Figure 3 shows a section of an amalgam with a relatively high content of mercury. The amalgam was exposed for four months to corrosion in a 1 percent NaCl solution; the corroded surface of the amalgam—facing downward in the picture—^is partly covered by products of corrosion. The 72 phase is observed only in the upper part of the section as irregular, dendritic grains ; nearer the surface the phase has dissolved, and the amalgam has become severely porous. There is no doubt that in this case the surface 72 grains anastomosed with all FiGUBE 3. High mercury content amalgam after deeper lying 72 grains. Experiments in which the four months exposure to 1 'percent NaCI solution. 72 phase was dissolved with the aid of a 10 percent sodium citrate solution have revealed that even in Corroded surface (facing downward In picture) Is partly covered by corrosion products. The 72 pliase well-condensed amalgams all the 72 phase always (irregular dendritic grains) is present only in upper part of the section, having been dissolved in the forms a cohesive network of crystals. Corrosion can lower portion to form porosity. Magniflcation-approx. therefore continue into the depth of the amalgam 227 X until the whole of this phase has disappeared. As far as the structure of the 72 phase is concerned, difference in the chemical composition of the al- there is no essential difference between the various loys but is probably due to the fact that the zinc- types of amalgams; spherical and nonspherical, free alloys are not so easily wetted by mercury as with or without zinc, mercurized or non- those with zinc, which results in a slightly higher mercurized ; all show the same network ^ructiire porosity in the amalgams. It is a fact that zinc- of the 72 phase. As there is only a slight difference free amalgams characteristically have a slightly in the amount of 72 phase between an amalgam higher porosity than those with zinc, and that this with a high content of mercury and one with a lo w porosity is often localized as small nests in the content, the corrosive properties would for all amalgam. practical purposes be independent of the amal- Since otherwise the three main phases, 7, 71 and gam's content of mercury. 72, present in all standard amalgams can from an The intensity of corrosion of dental amalgams electrochemical point of view be considered iden- on the other hand depends very much on the po- tical, it is unlikely that amalgams produced from rosity of the amalgams. Porosity exposes a greater different brands should demonstrate different de- area of the 72 phase and thus increases the anode grees of corrosion. area. Experiments with a 1 percent solution of The means at our disposal to reduce the rate NaCl as electrolyte have shown that intensity of of corrosion in amalgams are therefore as follows corrosion triples when porosity of the amalgam in- (1) omission of zinc-free alloys, and (2) use of a creases from approximately 1 to 4 percent. It was filling technique that reduces porosity of the amal- also shown that amalgams containing zinc cor- gam as much as possible. Since porosity can vary rode less than amalgams without zinc. The pri- considerably according to the filling technique, mary reason for this does not appear to be the this latter factor is the most important. ,

4. Effect of Technique on Porosity larly in the bottom corners of the individual con- of Amalgam densed increments. Figure 5 illustrates porosity in the bottom corner of a well condensed, typical Detailed analysis of filling techniques have Class II filling made by conventional technique; shown that the consistency of the amalgam when the gingival part of the section is about % mm it is placed in the cavity is the factor of prime long. Figure 6 shows the same part of a filling importance as regards porosity of the condensed made by Avet technique; the much lower porosity amalgam. A relatively dry and more or less is remarkable. The diagram (fig. 7) shows poroK- crumbling amalgam will always result in greater ity and y-phase content in gingival corners in porosity than a soft, plastic amalgam from which Class II fillings condensed by means of dry tech- ( a considerable quantity of excess mercury can be nique with two different condensers Ap and Af ) expressed. by conventional technique (B) or by wet technique Curve P in the diagram (fig. 4) shows the (C). The diagram shows that the wet technique average porosity in cylindrical, mechanically is an efi^ective means of reducing the porosity of condensed amalgam specimens, dependent on the the amalgam, and that it does not lead to a higher content of mercury in the amalgam when conden- mercury content in the condensed amalgam than sation commenced. do the other forms of amalgam technique. It may The vibration technique, requiring as it does a be noted that G. V. Black many years ago stressed dry amalgam, will thus always produce a high the importance of using relatively plastic amalgam degree of porosity in the amalgam. The same ap- during condensation. plies in the case of hand-condensed fillings made With regard to the relation between plasticity by conventional technique in which the amalgam and porosity it is worth mentioning the rather is triturated with excess mercury, most of the ex- marked tendency of manufacturers in recent years cess being expressed before the amalgam is placed to mai'ket rapid-setting alloys with greatest possi- in the cavity. Fillings according to the so-called ble one-hour strength. The clinical value of such dry or Eames technique are also fairly porous. alloys is open to doubt. During condensation amal- The lowest porosity is achieved when a significant quantity of mercury is expressed in the cavity and when each new increment is condensed in the free mercury thus expressed—the mercury being removed only when it forms a layer of more than approximately ^ mm on top of the condensed amalgam. The latter method has been given the name "wet" teclinique, wet being a reference to the mercury content. Porosity in fillings made by vibration or by hand condensation with dry or with conventional techniques is concentrated at the bottom and particu-

P M S

P X o

46 50 54 58 62 7oHg

Figure 4. Relationship hetween mercury content of mix and porosity (P), mercury content (M) and strength FiGURE 5. Porosity in the bottom comer of a well (8) in cylindrical, mechanically condensed amalgam condensed, typical Class II filling made iy a specimens. conventional technique. Magnification^approx. To convert kp/mm" to MN/m^ multiply by 9.807. 115X.

36 :

Figure 7. Porosity and y-phase content in gingival comers in Class II fillings condensed by a dry technique with

two different condensers (Ap and Af) < by a conventional technique {B) and by wet technique (C).

fillings on peripheral dental surfaces—can lead to minimizing secondary damage to the gingiva we can draw up the following list 1. Avoid as far as possible use of zinc-free alloy. 2. Do not condense the amalgam for a longer period than that corresponding to a mercury absorption of 2 percent. 3. The amalgam should be condensed not with a vibrator but with a hand condenser. follow Figure 6. Reduced porosity (compare with fig. 5) 4. The condensation technique should in the 'bottom corner of a well condensed typical the main principles of the so-called "wet" Class II filling made by a wet technique. Mag- technique. nification^approx. 115 X- 5. The surfaces of the filling should be made as smooth as possible in order to reduce gams of this type quickly become dry and crumbly, the chances of retention of plaque and the and can therefore result in a significant increase in resulting increase in corrosion intensity. the porosity of fillings and a similar increase in 6. Patients should be instructed to maintain the the likelihood of corrosion. Increased porosity highest level of oral hygiene because this means at the same time a decrease in strength of brings about a reduction both of the anode the amalgam : a 1 percent increase in porosity has area and of the time a given part of the about the same efi'ect on the strength as a rise of 10 filling surface is covered with plaque and percent in the mercury content of the hardened can thus operate as an anode. amalgam. The rate at which a freshly triturated There is no doubt that certain details of cavity amalgam loses its plasticity can be registered in a preparation, matrix technique, and condensation simple manner, for example, by measuring at vari- technique also affect, though indirectly, the tend- ous times after amalgamation how much mercucy ency of amalgam to corrode. I do not however can be expressed from the amalgam. The difference intend to discuss these in detail here. in the percentage mercury content between an amalgam which has been condensed as early as 5. Marginal Inaccuracies of Amalgam possible and an amalgam which is condensed later Restorations can be called the mercury absorption of the amalgam. According to a standard method used for As mentioned in the introduction, secondary routine tests in Copenhagen, an amalgam which caries is the other form of damage that can result has a mercury absorption of 2 percent five minutes from the failure of amalgam fillings. Secondary after finishing the trituration, i.e., 41/2 minutes caries can be defined as a type of caries with a after the earlist possible condensation, is consid- causal relation to marginal inaccuracies of resto- ered optimum for medium-size fillings. The great rations. majority of amalgams however have a much higher There are many causes of marginal inaccuracies mercury absorption during this period. in amalgam fillings. A very frequent cause is mar-

To summarize the factors that—in amalgam ginal excess or deficiency ; but as these defects are

37 only very slightly connected with the material filling with typical, marginal corrosion fractures. properties of the amalgam, they will not be the An analysis of the processes leading to corrosion subject of detailed discussion at this time. Con- fracture is rather complicated, involving both the traction of the restoration during setting, with the materials science and technology, of amalgam. resulting gap between the filling and the cavity The processes are as follows: shortly after a wall, is unlikely ever to have been the direct cause filling has been made, saliva penetrates between of secondary caries ; if however the amalgam mar- the filling and the wall of the cavity ; because of gin is not supported by the enamel wall, then this the relatively low concentration of oxygen on the margin if located on a loaded tooth surface will surface of the filling facing the cavity wall this during mastication be deflected in toward the surface will form the anode in a galvanic corrosion cavity wall ; whether marginal fracture then occurs element. The mercury released by corrosion dif- will depend on the relation between the width of fuses into the amalgam and causes mercuroscopic the gap and the maximum flexibility of the margin. expansion. As this expansion involves only the Calculations have demonstrated that weak amal- surfaces facing the cavity wall, the margins of gam margins can fracture under a very moderate the filling are deflected away from the walls. Fig- load and even with such a slight deflection as 2 yu,m. ure 9 shows the state of the filling margins after This shows that contracting amalgams involve an mercuroscopic deflection. Mercuroscopic expansion increased risk of marginal fracture and that they is fairly considerable, even when only relatively should not therefore be accepted. small quantities of mercury are absorbed; the It is a recognized fact that delayed expansion amalgam expands about 5 percent linearly when can in certain circumstances produce a marginal it absorbs 1 percent by weight of mercury; the step of a size very likely to lead to secondary caries. expansion increases significantly with increasing Delayed expansion can also cause crevices in cases porosity of the amalgam. During mastication the where the side walls of the cavity diverge toward unsupported amalgam margin will be deflected in the surface of the tooth. The cause of this is that toward the wall of the cavity, and it will fracture the delayed expansion frequently occurs at consid- if the width of the corrosion gap exceeds the maxi- erably varying rates in different directions in the mum flexibility of the margin. Figure 10 shows a amalgam, and is greatest in the direction parallel typical corrosion fracture with the resulting V- to condensation. Delayed expansion can thus be a shaped marginal defect; it has been shown that contributory cause of marginal fractures in which the fracture surface stretches considerably under the margin of the cavity. It is not too complicated a process to avoid the causes of marginal inaccuracy already mentioned, and it is therefore unnecessary to go into further detail. However, there is one further cause of marginal defects, namely marginal corrosion, which very frequently and seriously predisposes the margin of the filling to fracture. Figure 8 shows an occlusal

> . • -i

FiouRE 8. Occlusal filling with typical, marginal corrosion fractures. Figure 9. Filling margin after mercuroscopic deflection.

38 —

degree of porosity increases to a great extent both the rate of corrosion and, independent of this, the mercuroscopic deflection of the margin. Increasing porosity reduces at the same time the margin's maximum flexibility to a marked degree. In order to reach a state in which it can fracture during mastication a porous amalgam margin requires less mercuroscopic deflection and it deflects more quickly than a margin with no or less porosity. All such factors which influence the porosity of the amalgam margin are thus of significance to corrosion fractures. The diagram (fig. 11) shows the result of experiments on the significance of some technical factors for marginal porosity; A represents the dry technique, B the conventional technique, and C the wet technique. 1 means just filling the cavity with neither excess nor deficiency and without any further technical procedures, 2 means overfilling the cavity and eliminating the excess by carving, while 3 means overfilling, burnishing the margins, and then removing the excess by carving. It is seen that technique no. 3 ensured

by far the lowest level of porosity ; when this technique was followed, it was unimportant whether technique A, B, or C was employed. It may be added that the content of mercury in the amalgam margins examined was independent of the technique used; only in technique 1 it was somewhat higher. It is probable but not directly proved that the

Figure 10. Typical corrosion fracture with resulting marginal porosity will be increased by the use of V-shaped marginal defect. nonzinc amalgams and rapid-setting amalgams. Figures 12 and 13 illustrate the porosity in the risk of caries is very small when the width of amalgam margins made by different techniques. the defect is less than 50 /^m; if this dimension Figure 12 shows a margin with a porosity of ap- however exceeds 50 /xm, the risk of secondary caries proximately 10 percent as a result of technique grows almost in proportion to its size. B-1, in other words, a conventional technique with- An understanding of the processes which result out overfilling or burnishing. Figure 13 is a margin in marginal corrosion fracture permits an analysis of the factors that influence these processes from the point of view that if it is possible to prevent or delay them, it will also be possible to prevent or delay corrosion fractures. For marginal corrosion to occur requires—as already mentioned—^that saliva penetrates the gap between filling and cavity wall. Two methods have been used to prevent this penetration, namely an improvement of the adaption of the amalgam to the cavity wall, and a varnishing of the cavity wall, primarily with copalite varnish; both of these methods reduce saliva penetration considerably but not enough to prevent corrosion or merely influence the rate of corrosion. The special structure of the phase, even in well-condensed standard amalgams containing a FiQTJSE 11. Effects of some technical fac- minimum of mercury, shows that alterations in the tors on marginal porosity. quantity of this phase by modifying the condensa- A-dry technique B-conventional technique tion technique is also unable to reduce or prevent C-wet technique corrosion. 1-cavlty not overfilled, margins not burnished 2-cavity overfilled and excess removed by In contrast, porosity of the amalgam margin carving 3-cavity overfilled, margins burnished and plays a decisive role in the rate of fracture. A high excess removed by carving.

39 Figure 12. Amalgam margin with about 10 Figure 13. Amalgam margin with about 1 percent percent porosity produced a conventional porosity after overfilling, burnishing and re- technique without overfilling or burnishing. moval of excess. Magnification^approx. 115X. Magnification^approx. 115 X.

amalgam-filling failures is correct, it is also a fact with about 1 percent porosity after overfilling, that the percentage of good quality fillings can burnishing, and removal of the excess. be increased considerably by observing the con- Another factor of great importance to the occur- clusions of the analysis of defects. rence of corrosion fractures is the size of the mar- The main recurring theme in this analysis of de- tilling. deflection results ginal angle of the The that fects has been the corrosion of amalgam and the from mercuroscopic expansion decreases as the necessity of observing various details in the tech- marginal angle increases ; when the marginal angle nical procedure ; there is no doubt that future re- is 90°, deflection is zero. The size of the marginal search into the properties of amalgams must be is mainly cavity preparation. angle decided by An centered upon these two factors. attempt must be made to achieve approximately The deep corrosion in current standard amal- a right angle between the cavity walls and the sur- gams depends j^rimarily on the network structure tooth. face of the This rule can often—but of the 72 phase; if this structure can be broken, not always—be obeyed without causing any the deep corrosion will simultaneously cease. This disadvantage. appears to be possible only by a radical modification in the composition of the alloy, in which all 6. Conclusion the tin or most of it is replaced by another metal. Copper is not applicable in this connection be- The foregoing analysis of the primary causes cause, among other things, it may form a rather of the failure of amalgam tillings provides the rapidly corroding cohesive copper amalgam phase conclusion that moderate changes in the properties in the other amalgam ; silver seems to be unsuitable of the alloy, the amalgam technique, and cavity too. On the other hand, palladium has certain preparation can produce vast differences in quality properties that are promising. of fillings. It is accepted that many amalgam fill- If the amalgam technique is to produce fillings ings demonstrate excellent restorative properties; of high quality it is essential that several details insofar as the analysis of the causes of many of the technique are executed with pedandic accu-

40 racy. For various reasons it is not always possible general conclusion : Vast improvements have been to meet this condition in clinical practice—which made in the quality of dental amalgams over the must be regarded as a fault in the technique. Fu- past 50 years, particularly due to the research and ture research into amalgams must certainly be standardization work carried out by the Dental concentrated on this point also. The spherical Research Section of the National Bureau of Stand- amalgams may have a promising future in this ards, to the benefit of the dental profession, par- respect; if their chemical composition were such ticularly, and humanity in general. I feel con- that, during setting, a coherent, corrosive phase vinced that you will continue in the future to make did not form, they might form the basis for a new significant contributions to the development of this and simplified filling technique very close to the important material. It is a great pleasure to be ideal. your guest here, and I thank you most warmly I want to close this paper with the following for your gracious invitation.

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Basic Metallurgy of Dental Amalgam

L. B. Johnson, Jr. and H. G. F. Wilsdorf

Department of Materials Science, School of Engineering and Applied Science, University of Virginia, Charlottesville, Va. 22901

An "equilibrium" mixture of components of dental amalgam would contain the phases

/3i, yi and shown by Gaylor. Because dental amalgam is formed clinically at relatively low temperature only yi and ya are generally found. The tensile strengths of the Ag-Hg and the Sn-Hg matrix phases of dental amalgam are considerably below the tensile strength of the amalgam while the tensile strength of AgaSn is considerably above that of the amalgam. Cleaning the alloy powder with 5 percent HCl promotes better bonding between matrix and AgaSn and significantly increases the tensile strength. Tlie y: phase of amalgam is the only phase subject to destructive corrosion. The y- phase has been successfully eliminated from amalgam by two methods: (1) by inclusion of powder consisting of the Ag-Cu eutectic in the alloy and (2) by substitution of 10 percent Au for Ag in the alloy.

Key words : Amalgam, dental amalgam, metallurgy amalgam, strength ; corrosion of ; ;

dental amalgam ; dental materials metallurgy of dental amalgam. ;

trated the determination the setting 1 . Introduction on of reaction or reactions and the constitution of the set amal- Probably the most successful prosthetic material gam. Although advances in knowledge have been ever devised by man is dental amalgam. This ac- almost continuous, probably no single bit of in- counts, no doubt, for the millions of man hours of formation could be considered to be a breakthrough labor which have been devoted to research on this for dentistry such as Black's original contribution. material and the volumes of literature devoted to Nevertheless, a number of significant advances rejDorting this research. have been made. No group has contributed as much toward the Probably of foremost importance were the excellence of this material as the research group accurate determinations made of the binary con- at the National Bureau of Standards. It is they stitution diagrams for Ag-Sn, Ag-Hg, and Sn-Hg who have set the high standards for the develop- and determinations of crystal structures for most ment of a fine product. of the important phases in these systems. Out- In any review of the literature covering dental standing among these investigations were those by amalgam, one can only hope to scratch the surface Murphy and his co-workers Almost no major if he is to keep inside a reasonable time limit. changes have been made in the constitution dia- Much good work must be summarized in a sentence grams published by this group. Generally accepted or perhaps not mentioned at all. This paper will phase diagrams for these binary systems are attempt to review what the authors consider to be shown in figures 1, 2, and 3. the most significant advances in the knowledge of For the Ag-Hg system, Murphy established two the basic science involved in the composition, peritectic reactions resulting in two phases of re- strength and chemical properties of dental stricted composition. The (3 phase was determined amalgam. to be hep with a = 2.98A and c/a = 1.62 and the 2. Constitution y phase complex bcc with a = lO.OA. Berman and Harcoui-t described two naturally occurring Although the first person to use an amalgam as [2] phases of the Ag-Hg system (a and and sug- a dental restorative is not known, the first scientific y) gested the name moschellands-bergite for the investigation of dental amalgam must surely be y phase. Its composition is best represented as credited to G. V. Black, beginning about 1896. It AgoHga and the unit cell contains 50 atoms. was he who showed that the most useable dental Nial, Almin, and Westgren determined the alloy was one containing near 75 atomic percent [3] structure of AgsSn to be slightly rhombically de- Ag and 25 percent Sn, with perhaps a slight sub- formed hep with four atoms per unit cell and stitution of for Cu Ag (about 5%) desirable. a = 2.995A and c/a = 1.596 at 25 atomic percent Since the time of Black's work only minor Sn. changes have been made in the basic composition of dental alloy. ^ Figures in brackets refer to the literature references at the Eesearch efforts have been concen- end of this paper.

452-525 O—72 4 43 :

V. Simpson [4] reported the 7 phase of Sn-Hg (/3i, 7i and 72), but disagreed with the steps to have a simple hexagonal structure with a = involved, in order to attempt to account for the 3.213A and c/a = 0.931 at 7.19 atomic percent Hg. nonequilibrium manner in which dental amalgam Although a number of researchers had studied was prepared. He proposed the following the ternary system of Ag-Sn-Hg between about equations 1910 and 1930, undoubtedly the most extensive 1. 52+Ag3Sn (unreacted) investigation was that by Gaylor [5] from about Ag3Sn+Hg^7i + 1933 to 1937. Her 70° C and 100° C isotherms for 2. 7i + 52+Ag3Sn—>7i+72->7i+72+ft the system are shown as figures 4 and 5. Prior to Gayler's studies, McBain and Joyner Troiano defined S2 as SuyHg which was later [6] had proposed the first "setting reaction" for shown by Fairhurst and Ryge [8] to be 7 phase dental amalgam: of the Sn-Hg system. He also defined 7: to be Ag3Hg4. Extensive investigations of the setting AgsSn + Hg Ag3Hg4 + Sn reactions for dental amalgam were made from 1952-60 by Moffett, Ryge and Barkow [9]. Gayler showed that did Sn not form as a prod- This group used a recording x-ray spectrometer uct of the reaction explained setting and the by to determine the phases formed and their chrono- complex reactions which probably not to do go logical appearance. They concluded that Hg completion. the following She proposed equations: reacts with AgsSn to form a matrix of Ag2Hg3 (71) and the hexagonal 7 phase of the Sn-Hg 1. AgaSn + Hg ^ /3i + 72 system (72). These phases could be detected within -> 10 minutes of beginning of trituration and 2. /3i + 72 /3i + 7i + 72 the no other new phases were identified at any time. Her nomenclature follows from figure 4. Schmitt [10] used selective etching techniques

Troiano [7] agreed with Gayler concerning to distinguish only three phases in set amalgam, the equilibrium phases expected to be present, the original Ag3Sn (7), 71 and 72. 44 Sn 10 20 30 40 50 60 70 80 90 Hg WEIGHT PER CENT MERCURY

Figure 3. Constitution diagram for the Sn-Hg system.

45 FiGUBE 6. Microstructure of dental amalgam. B indicates the Sn-Hg phase [12].

only after the set amalgam had been stored at elevated temperatures (38° and 60° C) for considera-

ble periods of time. He considered that the /3i phase formed by a diffusion controlled transformation of the yi phase, the transformation rate being very temperature sensitive.

More recently, Otani [14] has identified the fii phase in dental amalgam stored at room temperature for two years. Furthermore, Otani, Tsutsumi and Yamaga [15] have demonstrated the strong dependence of the phase relations on temperature. In dental amalgam allowed to set at temperatures above 80° C, in fact, no yi phase was found at any stage, only p-^ and y^. The same amalgam, however, when moved to room temperature for 48 hrs, showed the formation of yi phase at the expense of part of the ^1. Thus, the final amalgam prepared Figure 5. 100 °C isotherm for the Ag-Sv^Hg system [5]. in this manner satisfied Gayler's diagram. i It is interesting to note that the photomicro- I A great deal of impetus toward further research graphs of both Johnson and Otani (fig. 7 a & b) into the structure of set amalgam was provided showed the ySi phase as being continuous with the by the development of spherical-particle alloys by original y particles. An amalgam made entirely Demaree and Taylor [11]. Using this type alloy, of this "monophase" would be expected to have ex- ^ Wing and Ryge [12] identified the same three cellent mechanical and electrochemical properties. % phases in set amalgam, (fig. 6) (7, 71 and 72) and It appears at this point that there is no basic made the interesting observation that most of disagreement among the investigations reviewed the original alloy particles remained unreacted, above. It is apparent that Gayler is correct in her in even Targe excess of Hg (70 percent) . It appeared assessment of the phases present in an "equilib- that the reactions were "self limiting." rium" dental amalgam at, say, body temperature, None of these latter studies of the setting of den- i.e., i8i+7i + 72- It is equally apparent that a transi- tal amalgam were able to identify at any time dur- tion temperature exists near 80 °C, since Gayler's ing setting the presence of the pi phase called for 100 °C isotherm shows A +72+ liquid and Otani's by Gayler and Troiano. Johnson [13] showed the dental amalgams allowed to set at any temperature presence of the fix phase in dental amalgam, but above 80 °C showed and 72 as the only solid | 46 .

phases. Therefore, dental amalgam prepared above 3.1. Silver Amalgams 80 °C, then cooled to body temperature, will con- The largest volume of dental is oc- tain a near-equilibrium mixture of /3i-t-7i+72- amalgam Dental amalgam prepared under clinical condi- cupied by the Ag-Hg phase, present as an intermetallic compound closely represented the for- tions, however, is not at equilibrium and the work by mula AgaHgs. At present, there is no general reviewed here has shown clearly that the /3i phase theory explaining the mechanical properties of in- is not present at any stage in_ the setting of the termetallic compounds. amalgam—at least in a distinguishable quantity. In The fracture strength of Ag-Hg alloys was de- time, however, 01 phase forms at the expense of termined by Young and Wilsdorf 17]. Be- part of the ji phase—the higher the temperature [16, the more rapid the transformation. Therefore, den- cause of the large difference between the melting temperatures of the tal amalgam prepared at room or body tempera- two components, preparation ture will eventually contain a near-equilibrium of the cast alloys proved difficult. Suitable tensile mixture of 181+71+72 as required by Gayler's specimens, however, were prepared simply by diagram. The higher the temperature the sooner mixing Ag powder (99.99 percent purity) withHg will equilibrium be established. and compacting the mixture into a die at room The exact mechanism by which the setting of temperature. The Hg content was controlled by dental amalgam takes place is still not thoroughly varying the reaction time and the amount of Hg understood. It is greatly complicated by the fact available. Alloys were prepared in this manner that one begins with a problem of sintering in a containing 15-70 weight percent Hg. liquid medium and ends up with a problem in The tensile strength of these specimens is shown solid state diffusion. The one point that has be- in figure 8, ranging from more than 9,000 psi (62 come clear, however, is that every change under- MN/m 2) at low Hg contents to about 2,000 psi gone by the system is a striving for the attainment (14 MN/m ^) for alloys containing 50 to 70 weight of the state of equilibrium required by Gayler's isotherm. 3. Strength

Three intermetallic phases are the main constituents of dental amalgam, designated y, yi, and y2 as described previously. Since the alloy is produced by sintering in the presence of a liquid phase, many metallurgical factors influence the properties of the final material. Table 1 gives an indication of the complexity of the manufacture of this alloy and the effect on some properties. Before a discussion of the strength of dental amalgam is presented, the mechanical properties of the three phases, yi, ya, and y, will be discussed separately. Figure 7a. Alloy particle surrounded by /3i phase [13].

•' It _' *^

* *

f ' til

Figure 7b. ^1 phase in dental amalgam [l^] 47 Table 1. Metallurgical factors which influence dimensional stability, flow, and strength of dental amalgam

Metallurgical factors Mechanical properties affected

Silver-tin alloy: Increased tin content—^abnormal contraction Composition Increased silver content—>abnormal expansion Particle size Smollfir nnrfiplp <5i9ip —>stTPri0"t,ti fi.ttflinpH in sVinrt.pr f.imp Particle shape —^decrease in expansion Spherical powders are not so sensitive to packing pressures as cut alloys and therefore often yield more desirable over-all properties.

Alloy: mercury ratio Increasing the amount of mercury—>reduction in compressive strength —^undesirable increase in flow Too little mercury—^expansion of amalgam decreases unduly

Trituration Increasing the time of mixing—^increased strength

Condensation Increased pressure—^decrease in expansion —>decrease in flow

Contamination Through moisture (for alloys containing zinc) —^delayed and rather excessive expansion

percent Hg. In order to assure complete bonding, with certainty and since some ya grains could have specimens were pressed at 100° C for 3 days and a lower Hg content, tensile specimens for Sn amal- tensile tested. Although the strength was greatly gams containing from 2 to 20 weight percent Hg improved for alloy compositions between 15 and 60 have been tested. weight percent Hg (fig. 9), the strength for the Alloys were prepared by casting and also by a Ag-70 weight percent Hg specimen was compar- powder metallurgical technique [16]. able to the value shown in figure 8. Since the lat- The cast alloys were obtained by sealing instru- ter composition is near that of the Ag-Hg matrix ment grade Hg with chemically pure Sn into an phase of dental amalgam, it must be concluded evacuated quartz tube and holding it for 30 min- that its strength is considerably lower than the utes at 250° C. Frequent agitation assured homo- tensile strength of dental amalgam, the latter value geneity. The liquid alloy was quenched in liquid being 7,000-9,000 psi (48-62 MN/m nitrogen for solidification. After annealing for 7 days at room temperature, tensile specimens were 3.2. Tin Amalgams machined having a 1-in (25 mm) gage length and 0.250-in (6.35 mm) diam. Cylindrical compression The second constituent in the "matrix" of dental specimens were 0.375 in (9.52 mm) in height and amalgam is a Sn amalgam containing approximately 18 weight percent Hg. It has been identi- diameter. specimens were prepared by fied as SngHg and is generally referred to as the Powder-compacted is not mixing 325 mesh Sn powder with Hg in a plastic 72 phase. Since the exact composition known

3,000

6,000

30 40 50 WEIGHT PER CENT MERCURY

Figure 8. Tensile strength of cold-pressed Ag amalgams. (To convert psl to MN/m^ multiply by 6.895X10-3). 48 capsule in a Spex mixer mill. The plastic mass was to occur. The low strength above 18 weight per- immediately compacted in a die having the dimen- cent Hg is due to the presence of liquid Hg which sions of a flat standard tensile specimen. acts as crack nuclei in the alloy. The data obtained are shown on figure 10. Rising With the exception of the 20 weight percent Hg from the fracture stress of pure Sn at 3,600 psi (25 alloy, powder compacted specimens showed much MN/m^) to almost 10,000 psi (70 MN/m^) at 8 lower strength. Since the reaction between Sn pow- or 9 weight percent Hg, the tensile strength de- der and Hg is very fast, the compacting was vir- creased sharply as the Hg content was increased tually done with y particles. Consequently, the from 12 to 20 weight percent. It should be noted bpnding between the particles was poor. Porosity that the stress maximum coincided with the phase was not more than 4 percent. change from the a+y phase field to the y phase Compression tests on cast specimens showed of the system, which at room temperature may measurable ductility, particularly for those with contain approximately 9-18 weight percent Hg. low Hg content. A value of 10,000 psi (69 MN/m^) Beyond 18 weight percent Hg, free Hg coexiste was obtained for the 20 weight percent Hg with the solid y phase. specimen. The information from the phase diagram offers As indicated above, the y phase of the Sn-Hg a possible explanation for the variation of strength system is of importance in dental amalgam, the and ductility with Hg content. Between 2 and 9 most likely Sn-amalgam present containing about weight percent Hg, y is present in the a solid 18 weight percent Hg. The tensile strength of this solution as a second phase and thereby an effect alloy was found to be rather low, about 3,000 psi comparable to precipitation hardening is expected (21 MN/m^), and the compressive strength about 10,000 psi (69 MN/m^). Both values are considerably below the corresponding strengths of dental amalgam.

3.3. Dental Amalgam Alloy

"Dental amalgam alloy" is the name commonly used for the powder which, when mixed with liquid Hg, yields dental amalgam. Its chemical composition is AggSn. It is present in appreciable quantities as "unconsumed" alloy in the amalgam and consequently, must play a role in its mechanical properties. For the preparation of the alloy a vacuum technique was used [16] since alloys melted and cast in conventional ways were found to be extremely brittle. This latter was probably due to the intro- WEIGHT PER CENT MERCURY duction of oxygen since molten Ag is known to FiGUKE 10. Tensile strength of cast Sn amalgams. attract a substantial amount of oxygen which, upon (To convert psi to MN/m^ multiply by 6.895X10-^). solidification, precipitates at grain boundaries. 49 Dental amalgam alloy is the product of a peri- 3.4. Dental Amalgam tectic reaction and contains 26.85 weight percent The clinical importance of Ag-Sn amalgams in Sn. The times required to reach phase equilibrium dentistry has determined the nature of the research are extensive and appear to be prohibitive when that led to the excellent restorative we have today. manufacturing the alloy. In order to determine The restrictions in producing the amalgam under the sensitivity of the strength of dental amalgam clinical conditions are a challenge to the metal- alloy to compositional changes, two additional lurgist. First, the amalgam must be made easily alloys containing 23.85 weight percent Sn and and quickly in the dental office at room tempera- 29.85 weight percent Sn were prepared. The for- ture. Second, it must be capable of being molded mer, therefore, contained more of the /3 (Ag-Sn) fit the cavity in tooth so that mechanical forces phase while the latter contained more of the to a will keep it in place. Third, setting must occur at Sn-rich eutectic. with the reaching a high strength None of the specimens showed any ductility 37 °C, amalgam in a reasonably short period of time. beyond 1 percent strain, so that only the fracture of the most important clinical requirements strength for tensile loading and compression could Two should be mentioned at this point: a linear be determined. (1) of not more than 20 The compressive and tensile strengths of chill expansion of the restorative iu,m/cm and a high resistance to creep (in cast AgsSn were determined to be 75,000 psi (520 (2) usually referred to as "flow"). MN/m^) and 25,000 psi (170 MN/m-) respec- dental literature, Since dental amalgam is brittle, the compressive tively. Corresponding values for annealed dental strength been used as a convenient and impor- amalgam alloy were considerably lower. Young has tant criterion for assessing the effect of certain [16] found evidence for precipitated films in an- variables. nature of these varia- nealed specimens, which could be responsible for metallurgical The selection of publications. the reduced strength. It appears reasonable to as- bles will be indicated by a In 1949 Phillips investigated the time de- sume that chill-cast or slowly cooled alloy is repre- [18] the setting of amalgams. found sentative of dental amalgam alloy used in clinical pendence of He 10-hr strength about 80 percent of application. that the was the final, and that full strength was often attained The tensile strengths of chill-cast Ag-Sn alloys after setting times of 24 hr. Some alloys, however, with Sn contents 3 weight percent higher or lower required about 7 days to attain full strength, so than the y composition were found to be slightly from this time on compressive strengths were usu- increased (see table 2) though not to a marked ally given in terms of "7-day strength." degree. The microstructure, of course, was dif- A detailed study showing the effect of varying ferent, containing crystals in one alloy and 8 /3 condensation methods and time, specimen size, and other. far as dental amalgam crystals in the As deformation rate was reported by Taylor, Swee- this difference is hardly of is concerned, strength ney, Mahler and Dinger [19]. They also reported concern. It should be noted, however, that the elastic moduli ranging from 1.2 to 2.1X10^ psi difference in microstructure is most likely to influ- (8.3 to 14.5 X10« MN/m^) for five dental amal- ence the amalgamation reaction. gams. Studies of this nature, going into consider- Briefly, it has been shown that AggSn is con- able detail with respect to clinical factors, were siderably stronger than dental amalgam and, con- carried out through the 1950's and virtually consequently, must have a decisive influence on the cluded at the beginning of the last decade. At that strength of strength of the latter. time some concern was noted about the dental amalgam in tensile loading. Souder and Paffenbarger [20] reviewed the Table 2. Fracture strength of silver-tin alloys early work on that subject through 1942 and emphasized that the tensile: compressive strength Alloy containing ratio was extremely unfavorable, namely 1:8 or weight-percent Heat treatment Tensile Compressive Sn even poorer for some preparations. The renewed interest in tensile strength in the

psi psi 60's coincided with ( 1 ) the availability of spheri- 23. 85 Chill cast.. 30, 600 77, 700 cal dental alloy powders, used initially at the Na- 23. 85 Annealed *25, 000 65, 200 tional Bureau of Standards by Demaree and 26. 85 Slowly cooled 23, 000 26. 85 Chill cast. 25, 000 75, 000 Taylor [11], (2) the marked improvements in 26. 85 Annealed *17, 500 44, 800 metallographical techniques in handling dental 29. 85 Chill cast- . 500 200 26, 73, amalgam by Wing [21], and (3) the successful 29. 85 Annealed *14, 000 49, 300 application of the diametral test to this material by Burns and Sweeney [22]. Tensile specimens had a 1 in (25 mm) gage length and either a gage diameter of 0.140 in (3.56 mm) or 0.25 in Prior to the last mentioned investigation the (6.35 mm), the latter being indicated by *. Compression measurements were made on tensile specimens the specimens were 0.375 in (9.52 mm) long and 0.375 in shape of which had been conditioned by the brittle- (9.52 mm) in diameter. (To convert psi to MN/m^ multiply by 6.895X10-'.) ness and the manufacture of the material. In gen-

50 eral, the gage length was less than 0.5 in (13mm) 9,000 psi (62 MN/m^') (fig. 11), Modjeski and Nuckels psi and Nagai (Ward [23], Taylor [24], Kodriguez and Dick- 10,600 (73 MN/m^) [31] et al. psi son [25], and Mahler and Mitchem [26]). Addi- [32] 10,000 (69 MN/m^). tional studies were carried out in 1964 by Hollen- A comparison of tensile strengths obtained from back and Villanyi [27] and in 1968 by Nagai and dumbbell shaped specimens and from diametral Ohashi [28]. In general, these investigators tests indicates frequently a higher value for this found a 7-day tensile strength between 6,000 and property from the latter test, so that the ratio of 8,400 psi (41 and 58 MN/m"). The data on each tensile and compressive strengths is approximately material in each series of measurements were re- 1 :6. However Burns and Sweeney [22] obtained produced only over a wide range of values since the comparable values for the diametral and the axial fracture of the dumb-bell shaped specimens was loading methods. sensitive to surface flaws. Recently, Lautenschlager, and Harcourt [33] Following the work of Burns and Sweeney [22], pointed out that the diametral test is only applica- the diametral tensile test has been used to provide ble to homogeneous materials and that in speci- reliable data for dental amalgam. The advantage mens with pores and second phase particles the of this method lies in its insensitivity to surface actual values may be altered. Seen in this light, conditions of the specimen. The cylindrical speci- it is intei-esting to note that Nagai et al. [32] found men is placed between the plates of a testing ma- substantial strength differences for conventional chine and compressed transversely. A convenient and spherical alloys when specimens compacted specimen diameter is 4mm, the length of the cylin- at different pressures were subjected to the dia- tensile der is usually between 5 and 10 mm. According to a metral test. While conventional alloys showed a strength i-eduction of percent treatment by Timoshenko [29], the tensile stress, o-, about 50 is related to the compressive load, P, the diameter, for a lowering of the condensation pressure from 850 to 71 psi to 0.49 the strength and length, Z, of a right cylinder by the equation (5.9 MN/m=), difference for spherical-particle alloys was in general not more than 25 percent and often less. At this time it is not possible to arrive at an understanding of the mechanical properties of where

14,000 _

4,000 ALLOY

X "AS RECEIVED" lYOUNG AND WILSDORF) 2,000 "AS RECEIVED" (MODJESKI AND NUCKLES) "CLEANED" ALLOY (YOUNG AND WILSDORFl "CLEANED" ALLOY (MODJESKI AND NUCKLES)

42 46 50 WEIGHT PER CENT MERCURY

Jj'iGUBE 11. Tensile strength of dental amalgam [J7J. (To convert psi to MN/m^ inuUiply by 6.895x10-3).

51 dental amalgam is based on macroscopic tests. In order to separate different parameters, such as alloy particle size and shape, Hg content, trituration time, condensation pressure and time, specific surface area, and others, detailed studies of the microstructure are needed. This includes an understanding of the influence of pores on the shape of the phases, the effect of surface contaminants, and the atomic structure of phase and grain boundaries, to name a few topics only. Nevertheless, advances in the understanding of the strength have been made in recent years and these will be reviewed tofifether with new measurements on relevant properties. Strength measurements on dental amalgam normally show wide scatter, exceeding by far the experimental error. Anticipating that surface contaminations of the AgaSn particles could cause flaws in the amalgam and possibly would act as Figure 12. Schematic representation of micro-structure of dental amalgam. crack nuclei or, in any event, be detrimental to the bonding between phases, Young and Wilsdorf [17] prepared surface clean dental alloy. They going around the unconsumed AggSn particles. washed spherical alloy powder in 5 percent HCl Dental amalgam made from "cleaned" alloy par- and dried the particles with ethanol. It was ticles also fractured in an intergranular fashion observed that the reaction time for amalgamation with one remarkable difference—cracks were found was reduced by this process and that the strength not to avoid AggSn particles but proceeded of the resulting amalgam had increased from 9,000 through the particles by cleavage. This observation to 12,500 psi (62 to 86 MN/m^). Modjeski and gives a lead to an explanation of the increased Nuckels [31], using the same method, obtained strength of amalgams made from "cleaned" alloy. tensile strengths of 14,000 psi (96 MN/m^). An One could reason that the fracture of AgsSn par- explanation of the strength increase will be offered ticles is either due to a notch effect, which would in a subsequent paragraph. have weakened the relatively high strength AgsSn, The macroscopic failure of dental amalgam is or the bonding between the matrix and the par- classified as brittle fracture, i.e., the separation ticles was improved by the cleaning techniques. occurs in a catastropic fashion. Information on the Since the amalgam showed a definite increase in fracture mechanism is incomplete. Asgar and strength, it must be concluded that the bonding Sutfin [34] employed a microbend tester and used between matrix and AggSn was improved. light microscopy in order to study the subject on a At this stage of the development of the metal- microstructural basis. They observed that cracks lurgy of dental amalgam it is not possible to assess went through voids and propagated through the the influence of porosity on the mechanical proper- matrix phases (yi and yz are traditionally called ties in a quantitative manner. In the past, the fun- the "matrix") in an intercrystalline manner. These damental studies have tried to minimize porosity results were confirmed by Caron's [35] investiga- by using high condensation pressures (see for tion of fracture due to compressive loading. The example, Young and Wilsdorf [17] whose samples latter author also reported the fracturing of alloy had a porosity of less than three percent). How- particles in specimens produced at high packing ever, it can be expected that porosity in clinically pressures. prepared amalgam is higher and present tech- The application of electron fractography al- niques of investigation have to be improved in lowed additional observations at the submicro- order to make better quantitative measurements. It scopic level. A schematic representation of the is felt that the use of the scanning electron micro- microstructure of dental amalgam prepared from scope could be very helpful in this respect. spherical alloy particles is shown in figure 12. Another mechanical property of considerable Using replica techniques, Young and Wilsdorf importance for the clinical behavior of amalgam is [17] concluded from fracture surfaces that cracks creep. Extensive studies have been carried out, propagated in an intercrystalline manner for amal- but the actual mechanism in terms of advanced gams prepared from untreated alloy powder. They solid stat« theoi'y is still elusive. The intriguing noted, however, that the Sn-Hg phase showed a problems associated with the behavior of dental certain degree of ductility during failure, as evi- amalgams have attracted a larger group of solid denced by small dimples on the surface. The frac- state researchers, and a number of fundamental ture surface of Ag-Sn crystals showed primarily data which are needed for even a first understand- facets, indicating intergranular and/or interphase ing of the often unexpected properties of the alloy fracture, the latter of which was typical for cracks are beginning to appear in the literature. The work

52 on elastic constants of the y, yi, ya-phases and den- gam had lost most of its strength. This was attrib- tal amalgam by Grenobie and Katz [36] may be uted to the presence of concentration cells. quoted as an example. On the basis of their meas- Fusayama, Katayori, and Nomoto [38] dis- urements they attempted to develop a model of the agreed with the results of Schoonover and Souder. elastic behavior of dental amalgam which with For essentially the same conditions the Japanese further refinement could provide information on researchers reported no significant corrosion of the viscoelastic behavior and on the effect of poros- the amalgam surfaces except for a slight roughen- ity and mercury content on strength. ing around the region of contact with the Au. The amount of corrosion was not considered to be 4. Chemistry clinically significant. Swartz, Phillips, and El Tanir [39] observed The constitution and mechanical properties of the tarnishing of amalgams in solutions made up amalgams are primarily bulk properties of the of various concentrations of NajS, NaCl and material and can, in general, be considered to be H2O2, as well as in air, distilled water, and an unaffected by the environment, within practical artificial saliva. The degree of tarnish was deter- limits, at least for a period of a number of years. mined simply by visual observation. They con- Yet all metal or alloy surfaces react with their en- cluded that the highest degrees of tarnish occurred vironment to produce changes in the surfaces. in NasS solutions, the rate and severity of attack These changes depend on many variables, among increasing with sulfide concentration. Medium which are the particular constituents present in tarnish occurred in the artificial saliva. Experi- the environment, the composition of the metal it- ments with zinc and nonzinc amalgams indicated self, the physical character of the surface or inter- no measurable difference in the susceptibility of face with the environment, the flow of material the two types to tarnishing. past the surface, and the temperature. Wagner [40] was the first investigator to indi- The many variables involved make the study of cate that one specific phase of dental amalgam the chemistry of the system of dental amalgam was more susceptible to corrosion than the others. plus oral environment a difficult one. In vivo He pointed out that the ya phase was most subject studies are beset with the difficulties that no two to chemical attack and, further, that in amalgams human mouths offer the same environmental con- with a high Sn content, the yz phase could be stitutents, no two amalgam surfaces placed in continuous and corrosive attack could penetrate teeth are ever identical, and flow situations are deeply into the amalgam. He also, however, never the same for any two fillings. In vitro pointed out the importance of corrosion products studies, on the other hand, while allowing much as marginal "sealants." closer control of variables, have the decided dis- J0rgensen [41, 42] reported agreement with advantage of being very unlike clinical conditions. Wagner both on the continuity of the yz phase This not only makes factual correlations difficult in dental amalgam and the primary importance to make but also makes it difficult to convince of the corrosion of this phase. He found continu- clinicians that correlations exist even when they ous ya networks even when condensation reduced are apparent to the laboratory scientist. the Hg content as low as 40 weight percent. Nevertheless, advances in the understanding of Wagner's and J0rgensen's conclusions led Guth- the corrosion of dental amalgam have been made, row, Johnson, and Lawless [43] to an extensive especially in the last decade or so. study of the corrosion of the individual, isolated Prior to about 1940, although numerous studies phases of dental amalgam in Einger's solution related to the corrosion of dental amalgam were and in artificial saliva. Potentiostatic measure- made, the major emphasis was generally placed ments (figs. 13, 14, and 15) were complemented on the pathological effect on body tissues of gal- with electron micrographs of the surfaces (figs. vanic currents produced by dissimilar metals in 16, 17, and 18). This study supplied conclusive the mouth. Little effort was to note the effect made confirmation for the preferential and destructive on the metals themselves. corrosion of the ya phase. Only deposition (or In 1941 Schoonover and Souder [37] published tarnishing) occurred on the yi phase, while the the first really extensive study of the corrosion of phase was essentially dental amalgam. They reported rapid corrosion y neutral. Otani has summarized re- of dental amalgam in contact with Au in both [44] much of the 1-percent NaCl solution and in an artificial saliva, search in Japan on the corrosion of dental amal- but only gradual corrosion with loss of luster gam. In general, polarographic techniques have when it was not in contact with Au. Potential dif- been used rather extensively to attempt to deter- ference between the two metals varied between mine the effect of manipulative variables on corro- 446 and 548 mV. sion in various media. From their work, it appears Examination of 40 to 50 freshly extracted teeth that the more Hg an amalgam contains, the containing amalgam fillings showed that, where greater the corrosion ; the higher the condensation adaptation of the filling to the cavity wall was pressure, the less the corrosion spherical-particle ; poor, corrosion was often so severe that the amal- alloys provide amalgams with higher corrosion

53 1 — resistance; but, in general, dental amalgam is a electron microscope, and electron microprobe ex- highly corrosion-resistant material. aminations of the specimen surfaces. They inves- Mueller, Greener, and Crimmins [45], Mueller tigated a number of alloys and the effects of Sarkar, and Greener [46] and Mueller and clinical variables such as trituration time and Hg Greener [47] have made extensive electrochemical to alloy ratio and the effect of residual strain studies of amalgam corrosion both with open cir- energy on the corrosion rates. In general, practices cuit and anodic polarization techniques. These which resulted in the formation of more 72 phase measurements were complemented with optical. led to more corrosion. They also noted that the

0.9

0.7

0.6 Ag^Sn

dUJ 0.6 to 0.4

0.3

O.t

0.1

0 0

o - 0 1 o Ringer's solution -0.2 A Synthetic saliva

- 0.3

-0.4

J-

I 100 1000 10000 CURRENT DENSITY (^A/cm^)

Figure 13. Potential-current relations for AgaSn in corrosive media [J/S],

0.9 -

O.B -

- 0.7 AggHgj

0.6 -

0.5 - o 0.4 -

OS -

o - > 0.2

0 1 C

0.0 -

- 0.1 - o ° Ringer's solution

0.2 - 4 Synthetic saliva -0.3 -

- - 0.4

I 1 1 1 1 ' I -1 I I ' I 1 1—

10 CURRENT DENSITY (/xA/cm^)

Figure 14. Potential-current relations for AgzHgs in corrosive media [43].

54 I

- 0.3 o Ringer's solution

• 0.4 A Synthetic saliva

i I ' I ' ' ' i—

10 too 1000 10000 CURRENT DENSITY (/iA/cm?)

Figure 15. Potential-current relations for SnsHg in corrosive media US'i.

- -J

- *r

"* -

Figure 16. Electron photomicrograph of y phase after Figure 18. Electron photomicrograph of 72 p'lase after polarization in Ringer's solution for one hr at 0.08 V. polarization in Ringer's solution for one hour at —O.Slf (Orig. mag.XSOOO US]. V. (Orig. mag.XSOOO) US].

presence of Zn in the amalgam led to markedly different initial potentials but about the same as for nonzinc specimens after about 24 hr. Johnson and Lawless [48] reported that stress caused a marked increase in the anodic direction for the corrosion potential of the 72 phase, but only a slight change for dental amalgam or the y or yi phases. This effect was believed to be due to surface film rupture during plastic deformation of the specimens. They concluded that stress is probably not an important factor in the corrosion of dental amalgam under oral conditions. Mateer and Keitz [49] made a significant contribution toward the understanding of amalgam cor- 17. Electron photomicrograph of yi phase after Figure rosion in vivo. Ground and polished cross-sections polarization for 30 min in Ringer's solution at 0.08 V. extracted teeth containing amalgam {Orig. mag.XSOOO) US]. through 50 55 fillings were examined metallographically. The analyses of corrosion products scraped from authors reported a two-phase corrosion product amalgams in extracted teeth. They reported that deposited as layers in the marginal region between at least SuzSg and /SSnOz were present. the filling and the cavity walls, figure 19. These Mueller and Greener [47] usmg an electron mi- deposits tended to follow the phase and porous croprobe, 72 found chlorine present in the 72 areas areas in the amalgam, figi;re 20. Fine, branch-like of amalgams which had been polarized in a physio- penetrations of corrosion product deep into the logical saline solution and suggested that a tin amalgam were sometimes observed. It was con- chloride precipitate might be important in the cluded that the corrosive attack consumed the 72 passivation process. phase, and, indirectly, part of the y phase by for- It is clear from these studies that a wide variety mation of new phases due to reaction of Hg re- of corrosion products may form when dental leased in the original corrosion, with y phase. Recently, dental alloys containing small amounts of SnF2 have been placed on the market. This led Stoner, Senti, and Gileadi [50] to investigate the corrosion of amalgams with inclusions of such materials. They reported that the inclusion of SnFa in an amalgam enhanced its corrosion, as did also the addition of NaF into the corroding solution. The addition of SnF2 into the solution did not appear to affect the rate of corrosion of the amalgam. Many attempts have been made to identify the products of amalgam corrosion. Techniques involving chemical analysis, spectroscopy, electron microprobe, and electron and x-ray diffraction have been used on corroded surfaces, neighboring areas, and contacting solutions. Schoonover and Souder [37] and Hyselova, Zajicek, and Vahl [51] used chemical analyses and agreed that mainly Sn ions were in the corrosion products. Swartz, Phillips, and El Tanir [39] using x-ray diffraction and Guthrow, Johnson, and Lawless [43] with x-ray and electron diffraction, identified AgoS, HgS, and AgCl respectively. first group The Figure 19. Metallographic section through amalgam res- also concluded that a complex (Hg, Ag) Sx was toration in extracted tooth.

present while the latter found complex salts which Cavity wall In enamel (A) and dentin (B) Is at left. Light-gray corrosion product (C) has deposited against cavity wall. could not be identified by reference to ASTM files. Darker gray corrosion product (D) has deposited against Mateer and Reitz amalgam and has penetrated into regions which were originally [49] made x-ray diffraction interconnected (72 Sn-Hg and porosity) [49].

FiGUKE 20. Metallographic section showing fatigue crack and bulk corrosion of amalgam in margin region of extracted tooth. Amalgam fragments have chipped away from cavity wall (A), causing a gutter. Bullj deposition of the darlj gray corrosion product has taken place in 72 and porous regions within the amalgam, whereas cracking is visible at the exposed surface. The two types of attack appear to contribute mutually in destroying integrity of the restoration [49].

36 amalgam is subjected to an oral environment. Since the environments to which they are subjected vary widely, this is entirely reasonable. Au-CONTAINING AMALGAM In general, however, it appears that Sn is the main 14 DAYS OLD constituent which dissolves from the amalgam and is found in solution, while adhering corrosion products are probably high in Ag and Hg, are primarily sulfides, and contain some Sn. The most important conclusion to which nearly all corrosion studies have pointed is that the 72 phase of dental amalgam is the only phase subject Au-CONTAINING AMALGAM to destructive type corrosion or dissolution. This I Hr. OLD led J0rgensen [42] to suggest that research should be directed toward elimination, or at least reduction in amount, of the 72 phase, perhaps by replacing part of the Sn in the original alloy with Pd. Results of such efforts have not been reported as yet. Successful elimination of the 72 phase from dental amalgam has been reported for two recently developed alloys. reported studies In 1963 Innes and Youdelis [52] g on the dispersion strengthening of dental amalgams. They mixed various amounts of metal powder consisting of the Ag-Cu eutectic (71.9 65 60 55 50 45 40 35 30 percent Cu) with powdered AgaSn (78.2 percent DEGREES 29 Ag, 26.8 percent Sn). The combination was later marketed under the trade name Dispersalloy. FiGXJEE 21. X-ray diffraction scans of conventional and Mahler [53] studied amalgams made from the Au-containing amalgams showing absence of 72 phase in Au-containing amalgam at I4 days 157]. above alloy. He used an electron microprobe and reported (1) "the presence of a reaction phase surrounding the dispersant particles which was deter- in a "reaction ring" around the alloy particles, mined to have a composition intermediate between apparently similar to that for Dispersalloy. the intermetallic compounds of CugSn and Cug The question remains open as to how much cor- Sns," and (2) "the absence of the Sn-Hg phase rosion of amalgam is desirable. The sealing of (72) in this amalgam." margins and crevices by corrosion products has Johnson [54, 55], considering Ag-Sn-Hg alloys been demonstrated and surely may be desirable. as electron compounds, showed that it should be On the other hand, weakening of the amalgam structure by the destructive corrosion of the possible to dissolve more Sn in the 71 phase than 72 phase has been demonstrated and is also surely un- had previously been thought possible. Under non- desirable. It may be possible to eliminate destruc- equilibrium conditions, it appears that all of the tive corrosion effectively by forming an amalgam matrix Sn in dental amalgam could be incorpo- without the 72 phase, while at the same time allow- rated in a single phase with the same structure ing the formation of adherent films of oxides and as the 71 phase. sulfides to accomplish the marginal sealing. Only showed that the Grenoble and Katz [56] 72 long-term clinical testing of 72-free amalgams dental amalgams which phase disappeared from can give the final answer. had been subjected to extremely high pressures at or 750,000 room temperature (about 50 kilobars 5. Summary psi (5,200 MN/m^)). In order to bring about the elimination of the A brief review has been made of the significant 72 phase at body temperature, Johnson [57] then advances made over the past half century or so substituted 10 weight percent Au for 10 percent towards an understanding of the constitution, of the Ag in the original alloy. He reported (fig. strength, and chemistry of dental amalgam. 21) that the 71 and 72 phases formed as usual The conclusion has been reached that an "equi- upon trituration with Hg, but that after 14 days librium" mixture of components would contain the

at body temperature, x-ray diffraction patterns phases fii + ji + y-' as shown by Gayler's 60 °C showed only the presence of the 71 phase. This isotherm. Because dental amalgam is formed clini- material had a very high corrosion resistance. cally at a relatively low temperature, only the 71 More recent studies by Mahler [58], however, in- and 72 phases are generally found. Recent work dicate that part of the Sn may be tied up by Au has shown that ^1 does form, however slowly, thus

57 showing that the system is striving toward the [13 Johnson, L. B., J. Biomed. Mater. Res., 1, 285 equilibrium state—the higher the temperature the (1967), and 1, 415 (1967). [14 Otani, H., J. Osaka Univ. Dent. School, 10. 57 sooner equilibrium is established. (1970). Although numerous measurements have been [15 Otani, H., Tsutsumi, S., and Yamaga, R., J. Osaka made of mechanical properties of amalgams and Univ. Dent. School, 10, 69 (1970). Young, a much clearer picture of the microstructure has [16 F. A., Jr., D. Sc. Thesis, Univ. of Va., 1968. [17 Young, F. A., Jr., and Wilsdorf, H. G. F., Research evolved, it is still not possible to attain a complete in Dental and Medical Materials, p. 69 (Plenum understanding of the behavior of this material. Press, 1969), J. Biomed. Mater. Res. 2, 401 (1968). The major defect appears to be its multiphase [18 Phillips, R. W., J. Dent. Res. 28, 348 (1949). [19 Taylor, N. nature, which results in poor tensile properties. O., Sweeney, W. T., Mahler, D. B., and Dinger, E. J., J. Dent. Res., 28, 228 (1949). It does not appear to fit into any of the alloy cate- [20 Souder, W., and PafEenbarger, G. C, Physical Prop- gories from which the metallurgist can draw on erties of Dental Materials, NBS Circ. 433, 222 a theory of strengthening. Nevertheless, recent pages (1942). Wing, improvements in the study of microstructure, [21 G., Doct. Dissertation, Univ. of Sydney, Australia, 1961; Aust. Dent. J. 11, 105 (1966). modern advances in the application of electron [22 Burns, C, and Sweeney, W. T., I. A. D. R., 43d Gen. fractography, recent measurements of elastic con- Meeting Abstr., p. 122 (1965). stants of the individual phases, and the possibility [23 Ward, M. L., J. Am. Dental Assoc. 11, 487 (1924). of the elimination of one phase of the amalgam, [24 Taylor, N. O., J. Am. Dental Assoc. 17, 112 (1930). [25 Rodriguez, M. A., and Dickson, G., J. Dent. Res. all offer considerable hope for a better understand- 41, 480 (1961). ing and improvement of the mechanical properties. [26 Mahler, D. B., and Mitchem, J. C, J. Dent. Res. 43, A knowledge of the chemistry of dental amal- 121 (1964). gam is complicated not only by the multiphase [27 HoUenback, G. M., and Villanyi, A. A., J. South. Cal. State D. A. 32, 355, 379, 455, (1964) 33, nature of the material but even more so by the ; 423 (1965). extreme complexity of the environments in which [28 Nagai, K., and Ohashi, M., J. Nihon Univ. School of tooth fillings are located. Corrosion products and Dent. 10, 39 (1968). mechanisms have been difficult to identify. Never- [29 Timoshenko, S., Theory of Elasticity, p. 104 (McGraw-Hill Book Co., Inc. York, theless, one outstanding bit of information has New 1934). [30 Eden, G. T., and Waterstrat, R. M., J. Am. Dental evolved that the ya phase of amalgam is preferen- Assoc. 74, 1024 (1967). tially attacked in a detrimental manner. The pos- [31 Modjeski, P., and Nuckels, D. B., private communi- sibility of eliminating this phase, therefore, holds cation (1969). considerable promise toward improving the [32 Nagai, K., Ohashi, M., Habu, H., Nemura, M. Korenaga, F., Goto, N., Nagata, Y., and Fujimoto, lasting qualities of the material in an oral Y., J. Nihon Univ. School of Dent. 12, 9 (1970). environment. [33 Lautenschlager, E. P., and Harcourt, J. K., J. Dent. Res. 49, 175 (1970). [34 Asgar, K., and Sutlin, L., J. Dent. Res. 44, 977 We owe it to those colleagues, both past and (1965). present, of the Dental Research Section of the [35 Caron, J. C, Master's Thesis, Georgia Inst. Tech. (1967). National Bureau of Standards, in whose honor [36 Grenoble, D. E., and Katz, J. L., J. Biomed. Mater. this paper has been presented, that the search for Res. 5, 503 (1971) ; 5, 515 (1971). a full understanding of this most useful and in- [37 Schoonover, I. C, and Souder, W., J. Amer. Dental triguing alloy be continued. Assoc. 28, 1278 (1941). [38 Fusayama, T., Katayori, T., and Nomoto, S., J. Dent. Res. 42, 1183 (1963). 6. References [39 Swartz, M. L., Phillips, R. W., and El Tannir, M. D., J. Dent. Res. 37, 837 (1958). [1] Murphy, A. J., J. Inst. Met. 35, 107 (1926). Murphy [40 Wagner, E., Deutseh Zahnaerztl Z. 17, 99 (1962). and Preston, G. D. J. Inst. Met. 46, 507 (1931). [41 J0rgensen, K. D., Acta [2] Berman, H., and Harcourt, G. A., Amer. Min., 23, Odont. Scand. 23, 347 (1965). 761 (1938). [42 J0rgensen, K. D., and Saito, T., Tandlaegebladet, [3] Nial, O., Alniin, A. and Westgren, A., Z. Phys. 73, 55 (1969). Chem., 14, 81 (1931). [43 Guthrow, C. E., Johnson, L. B. and Lawless, K. R., [4] V. Simpson, C, Z. Physik. Chem., 109, 187 (1924). J. Dent. Res. 46, 1372 (1967). Gayler, M. L. V., Brit. Dent. J., 269 [5] 54, (1933) ; 56, [44 Otani, H., Ann. Dent. 27, 32 (1968).

605 (1934) ; 58, 145 (1935) 59, 245 ; (1935) ; 60, [45 Mueller, H. G., Greener, E. H., and Crimmins D. S., 605 (1936) 61, 11 (1936). ; J. Biomed. Mater. Res., 2, 195 (1968). [6] McBain, J. W., and Joyner, R. A., Dent. Cosmos, 54, [46 Mueller, H. G., Sarkar, N., and Greener, E. H., Elec- 641 (1912). trochemical properties of commercial dental [7] Troiano, A. R., J. Inst. Met, 247 63, (1938). amalgams. Annual lADR Meeting New York, [8] Fairhurst, C. W., and Ryge, G., Adv. X-ray Anal., N.Y. (1970). 5,64 (1961). [47 Mueller, H. G., and Greener, E. H., Electron micro- [9] Moffett, J. C, Ryge, G., and Barlfow, A. G., .J. Appl. surfaces. Phys., 23, 1188 (1952). probe analysis of dental amalgam Annual lADR Meeting, Houston, Texas (1969). [10] Schmitt, G., Deutseh. Zahnaerztl. Z., 15, 736 (1960). [11] Demaree, N. C, and Taylor, D. F., J. Dent. Res., 41, [48 Johnson, L. B., and Lawless, K. R., J. Biomed. 890 (1962). Mater. Res. 3, 569 (1969). [12] Wing, G., and Ryge, G., J. Dent. Res., 44, 1325 [49 Mateer, R. S., and Reitz, C. C, J. Dent. Res., 49, (1965). 399 (1970).

58 .

[50] Stoner, G. E., Senti, S. E., and Gileadi, E., The [54] Johnson, L. B., J. Biomed. Mater. Res., 4, 269

effect of NaF and SnFj on the rate of corrosion (1970) . of amalgams. J. Dent. Res., 50, 1647 dental [55] Johnson, L. B., Amount of Sn in the 71 phase of (1971). dental amalgam, J. Biomed. Mater. Res., 5, 239 [51] Kvselova, J., Zajicek, and Vahl, J., Deutsch (1971) . Zahnartztl A., 23, 631 (1968) [56] Grenoble, D. E., and Katz, J. L., J. Dent. Res. 50, [52] Inees, D. B. K., and Youdelis, W. V., J. Canadian 109 (1971). Dent. Assoc. 29, 587 (1963). [53] Mahler. D. B., Microprobe analysis of a dispersant [57] Johnson, L. B., A new dental alloy, lADR Program amalgam, lADR Program of Abstracts and Pa- of Abstracts and Papers, No. 23 (1971). pers, No. 14 (1971). [58] Mahler, D. B., personal communication.

452-525 0—72—5 59

: ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Mateeials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Casting Alloys in Dentistry

Kamal Asgar

School of Dentistry, University of Michigan, Ann Arbor, Mich. 48104

A major difference between the two techniques, hygroscopic and high heat or thermal expansion, used for casting gold alloys is that the wax pattern can offer some resistance to hygroscopic expansion of the mold while it does not resist thermal expansion. Researchers have condemned the hygroscopic technique because of nonuniform expansion of the wax pattern but have not given suflScient attention to other nonuniform dimensional changes which are present in both techniques and which may combine to affect favorably or unfavorably the accuracy and retention of the casting. Studies in this field should not be limited to simple measurements of thermal, setting and hygroscopic expansions of the investment but should be expanded to include strength of the investment along with its roughness, strength and behavior of waxes, and shrinkage patterns of molten metals. A need for new types of investments is presented by the recent development of many new nonprecious alloys for dental castings.

Key words : Castings, dental, accuracy ; castings, lost wax technique ; dental materials hygroscopic expansion gold alloys, dental ; high heat casting technique ; ; investment, dental thermal expansion technique wax, inlay casting. casting ;

1. Introduction to the profession by various researchers. As examples, Phillips [3], Hollenback [4], Sweeney Casting metals by the lost wax process has been [5], Coleman [6], Crawford [7], Shell [8], Ire- known in art and industry for many years. Since land [9], Smyd [10], Gustafsson and Hedegard the turn of the century, the idea of using the lost [11], J0rgensen [12], Coy [13], Asgar and Mahler wax procedure in preparing metallic substances [14], Fusayama [15], and Peyton [16], could be to replace the missing portion of tooth tissue has mentioned. These are modifications and offshoots been routine practice in dentistry. Yet, the re- of two main techniques, the high heat or thermal quired accuracy of the fit of the cast structure by expansion technique and the hygroscopic tech- the profession has not been completely resolved. nique. Theoretically, there is very little difference The early castings were grossly short of what is between these two main techniques. All of the in- considered today, an acceptable fit. This was vestments used today in dentistry have some set- mainly due to the inferior materials available at ting expansion, some hygroscopic expansion, and the time. The development of the cristobalite in- some thermal expansion. The investments are vestment by two independent research groups, formulated so that if they are properly handled, Sweeney and his co-workers at the National Bu- the sum of their setting, hygroscopic, and thermal reau of Standards, and Coleman and Weinstein expansion will be equal to the contraction of dental [1] ^ at the Weinstein Research Laboratories of gold alloys and wax. The difference between these New York, probably was the most important single two techniques boils down to the fact that in the achievement in the field of dental castings. Dental high heat technique, a major part of metal shrink- gold castings made by using the cristobalite invest- age is compensated by thermal expansion of the ment were first introduced to the dental profession mold. This is obtained by heating the ring up to at the annual meeting of the American Dental about 1200-1300 °F (650-700 °C) . The setting and Association in Buffalo, New York, September, hygroscopic expansions of such an investment are 1932, by Jesrich. At about the same time, Scheu rather low. In hygroscopic techniques the major [2] also presented his hygroscopic casting part of the total expansion is obtained hygro- technique. scopically, and the thermal expansion of this type of investment is low. 2. High Heat and Hygroscopic Although they are alike theoretically, practi- Techniques cally there is a great difference between these two techniques. Two of the most important differences Since 1932, over a dozen dental gold casting are procedures and techniques have been introduced 1. In the hygroscopic technique the wax pattern is present as one piece and solid while 1 Figures in brackets indicate the literature references at the end of this paper. hygroscopic expansion is taking place. Thus,

61 the wax pattern could offer some resistance, X A X whereas in the high heat technique the major part of expansion is taking place while the ring is placed inside an oven and the wax pattern no longer exists as solid or one piece. Thus, it could not offer any resistance to the expansion forces. 2. In the hygroscopic technique the mold temperature is 700-900 °F (370-480 °C). The mold temperature in the high heat technique is about 1200-1300 °F (650-700 °C) or about 400-500 °F (220-280 °C) higher than that of the hygroscopic technique. Figure 1. Distances A and B were measured precisely These two points could be considered as the real before investing the wax pattern and after expanding differences between these two techniques. If they the investment hygroscopically. were recognized completely and handled properly, many practical problems would be avoided. have shown that due to the setting and hygroscopic The classical woi'k of Docking and his associates expansion of an investment, the percent of expansion in cervical areas is more than that in the [17], reported in 1948 and 1949, was the first major research work to clarify the behavior of the pulpal floor area. Therefore, the hygroscopic tech- hygroscopic investment. Later, Landgreen and nique has been condemned. Yet the investigators have failed to go ahead and complete the casting Peyton [18], Delgado and Peyton [19], Asgar- procedure. fact is that the dimensions of den- zadeh and Mahler [20], Lyon and Dickson [21], The tal casting are also nonuniformly during Ryge and Fairhurst [22], Donnison and Docking changed solidifi.cation and shrinkage of cast molten metal. [23], Skinner and Degni [24], Mumford [25], These studies have clearly indicated that the wax Fusayama [26], Mahler and Addy [27], Eam- pattern expands nonuniformly under setting and shaw [28] , and many others reported their findings on hygroscopic expansion of investments. As a hygroscopic expansion forces of the investment. result of these findings, the hygroscopic expansion of an investment is no longer a mystery. 4. Effects of Nonuniform Shrinkage of the Alloy and Surface Roughness on the 3. Effects of High Heat and Hygroscopic Dimensions of the Casting Techniques on the Dimensions of the Although the nonuniform expansion of the wax Casting pattern has made researchers condemn the hygroscopic technique the nonuniform shrinkage during The effect of hygroscopic expansion on wax pat- solidification and cooling of cast alloys has been terns is somewhat different from that of thermal completely ignored. explain this point more expansion. The shape and size of the wax pattern To clearly, let consider types of patterns, a full makes some difference on hygroscopic expansion us two crown type and a Class I type. If a full crown type and the resultant cast piece. As an example, let us exact dimensions consider the behavior of two different patterns of casting were made having the such as, one surface inlay (Class I) and three sur- of its die, obviously such a casting would not fit. In order to have a full crown type of casting fit face inlays (MOD) . In the case of one surface inlay, the investment expands away from the wax onto a die, the casting should be slightly oversized if I, one pattern and it really does not matter to a great ex- (fig. 2). On the other hand, a Class or tent what type of wax is used. In the case of an surface inlay, had exactly the dimensions of its die, MOD pattern however, investment located be- such a casting also would not fit. The casting in tween the two axial walls during setting and hy- this case should be slightly smaller in order to fit groscopic expansion of the investment will be into the die. The question may arise concerning exerting some forces trying to enlarge the size of the fit of an MOD pattern. Mesial-distally, it is the wax pattern in the mesial-distal direction. The similar to a full crown, which means the casting wax pattern, in the meantime, will be offering some should be slightly larger than the dimensions of resistance to these expansion forces. Depending on the die, whereas in buccal-lingual direction, it be- the type of wax used, hard or soft, more or less haves like a Class I or one surface inlay which resistance forces can be offered by the pattern. Fur- means it should be slightly smaller than the dimen- thermore, on the pulpal floor area of the axial sions of the die. It should be recognized that the walls, the resistance is greater than on the cervical effective shrinkage of metal is also not uniform. In areas. a Class I, or one surface inlay, the alloy shrinks Some investigators have placed reference marks to its maximum amount. It could be visualized that on the cervical and pulpal floor areas of an MOD the cast structure, after solidification, may shrink type of pattern measuring exactly the distances according to its coefficient of thermal contraction. A and B (fig. 1). Results of these measurements That is to say, for every degree that it cools, it

62 Figure 2. Dimensions of a and b on full crown type castings should be larger than ai and bi. Figure 3. Roughness of casting causes Oi and bi to get Dimensions of c and d in inlay type castings smaller than a and b. Roughness of casting causes Ct should be smaller than Ci and di. and di to become larger than c and d. becomes somewhat smaller until it reaches room it will prevent the casting from complete seating, temperature. On the other hand, in the case of a as if the investment did not have sufficient full crown type pattern, the alloy shrinks its least expansion. amount. In this case also, after solidification, metal To summarize, one may say that in a one surface should shrink according to its coefficient of thermal inlay, the shrinkage of metal is at its maximum, expansion, but due to the shape of the pattern, it but roughness of the cast structure aids the ex- should compress the investment inside of the pansion. Whereas, in full crown type patterns, crown. Due to the fact that metals usually do not shrinkage of metal is at its minimum but rough- possess high strengths at temperatures close to ness of the investment prevents the seating of cast their fusion range, they cannot compress the in- crown or acts against the expansion of the invest- vestment inside of the crown to any extent, and ment. If an investment produces roughness to such their effective shrinkage at high temperatures is a degree that the shrinkage of cast metal at its rather low. As the alloy coojs, its strength increases maximum (Class I or one surface inlay), minus and eventually reaches a point where its strength the roughness value of the cast piece, is equal to is higher than that of the investment and can com- the shrinkage of cast metal at its minimum (full press the investment inside of the crown. From crown type) plus the roughness of the cast piece, such a temperature, down to room temperature, it , then such an investment is capable of producing may shrink according to its coefficient of thermal good MOD castings. contraction. Thus, if all other conditions remained the same, stronger investments would produce full Shrinkage of Metal — Roughness crown type castings which seem to have expanded (Class I) more. The strength of the investment has little = Shrinkage of Metal + Roughness effect on shrinkage of a one surface inlay. (Full Crown) Not only could the magnitude of shrinkage of the cast metal vary in these two types of patterns, Usually investments producing smooth casting the effect of roughness of the cast structure is surfaces could produce good full crown castings, also quite different. is When a wax pattern prop- but it would be difficult to produce good one sur- erly invested, the investment completely covers all face inlays. Similarly, investments producing surfaces of the wax pattern. It should be remem- somewhat rougher cast surfaces, may produce good bered that the wax pattern is invested in a slurry one surface inlays, but it will be difficult to make which is made by mixing some water with solid a well fitted full crown casting. particles of various sizes (fig. 3). To eliminate the wax pattern, a mold is placed in an oven and 5. Retention of the Casting heated to casting temperature. During this process, the excess water in the investment, as well as For dentists, retention of casting is another im- the water of crystallization of various solids, will portant point. It is preferred if the casting offers be eliminated. Thus, the smooth surface of the some retention in the last millimeter before its investment next to the wax pattern becomes some- complete seating. Questions may arise as to what is what rough. Roughness in the case of a Class I, responsible for such a retention. If all dimensions or one surface inlay, acts as an aid to expansion of the die were exactly reproduced on the cast, and its effect would be as if the investment ex- theoretically there should not be any retention up panded more. In the case of full crown roughness. to the last micrometer away from complete seating.

63 All of a sudden, all surfaces of the cast should contact all surfaces of the die. To analyze the retention, one should recognize what is responsible for it in different shape patterns. Such a study can be done by making castings, seating them on the die, embedding them in some type of bioplastic and cross-sectioning the die and the cast. The space between the cast and the die can be examined under a microscope. Such a study shows that dental castings are far from a perfect fit. In the case of full crown and MOD type of patterns only on marginal areas are the castings contacting the dies (fig. 4) . In the case of Class I and Class V, due to the fact that shrinkage of metal is at its maximum, the line A in the cast is slightly smaller than in the die. Therefore, in Class I some surfaces of the cast can contact those in the die, Figure 6. Class V pattern loith pins. whereas in the case of Class V, such a contact is not Castings usually have some retention. possible (fig. 5). By placing some pins in Class V preparation, one could get sufficient retention (fig. 6). It should be recognized that the pins during setting and hygroscopic expansion will be actually bending slightly and causing the retention. The marginal adaptation of Class V inlay with pin, in a cast condition, probably is not any better than the one without a pin. The fact that castings having some retention can easily be finished and better marginal adaptation can be obtained at the chair during the delivery, is subject of another discussion. It is more interesting to observe the retention of Class III inlays (fig. 7). Under a normal expansion of the investment, the inlay usually has

Figure 7. Class III type of inlay. The retention can be obtained by contacting the two inner surfaces or by the two outer surfaces.

some retention. By increasing the expansion of the investment, the inlay would lose its retention; by further increasing the expansion of the investment, the inlay regains its retention. The retention first was obtained probably because the inlays were very slightly small, and like Class I restorations, they touched the two inner surfaces of the die. When the dimensions of the cast inlay were enlarged by increasing the expansion of the invest- Figure 4. The casting contacts the die only on the ment, the cast no longer could contact the surfaces marginal ridge areas. of the die, thus it would lose its retention. By further expansion of the investment, the outer surfaces of the inlay could come in contact. The retention of many patterns is due to the setting and hygroscopic expansion of the investment which produce non-uniform expansion of the wax pattern of very minute amounts. Both Moore

Figure 5. In the Class I type of pattern, casting con- of the Ransom and Randolph Company, and Nie- tacts die in some areas. man of the Whip-Mix Corporation have produced j ' The distance A is responsible for some retention. In Class V investments having no setting and hygroscopic type of pattern, easting does not make any contact usually with the die. expansion. Thermal expansion of this investment

64 was sufficiently high to compensate for the shrink- Moore[36], Watts[37], Collins[38], and a few age of gold alloys. others, there are good investments available for Nieman has even obtained a U.S. Patent for his dental gold castings. During the last fifteen years, investment. Although such an investment sounded many new nonprecious alloys have been developed good at first, the castings had no retention and did in industry. The physical properties of some of not satisfy the needs of the dental profession. The these alloys promise their potential use in the field investment was discontinued. of crown and bridge. Yet the available investments are not suitable for their casting. It seems that 6. Measurement of the Accuracy there is a great need for some basic research in of Castings this field with the hope of developing a series of new investments for casting of these new alloys. It is relatively simple to measure thermal, set- It may be time to look for the possibility of find- ting, and hygroscopic expansion of an investment. ing a different binding material than the three It is however, more complicated to measure dimen- already being used. Perhaps a refractory material sions of cast pieces and compare those with that of different from those used today in dentistry should the original die. In this case not only are the var- be employed for casting of the newly developed ious expansions of the investment important, but alloys. also the dimensional changes of the wax pattern

and the shrinkage of the molten metal. If however, 1.4 the quality of the fit of casting to a die is to be objective in considered—the main of researchers 1.2- ^ this field, as well as those in practice—then the 1

roughness of the casting and strength of the in- 1.0- vestment also should be recognized.

In the past, relatively little research work has .8 been reported in which the fit of dental gold cast-

ings were measured and compared. The early work .6 of Volland and Paffenbarger [29], Hollenback and his group [30], Mahler and his associates [31], .4- are among few published articles in this country

concerned with the fit of castings and relating it .2 / 10 Microincries to some expansion values. Pomes and Slack [32], Suffert and Mahler [33], and Barone and Dick- son [34] have measured the roughness of dental 200 600 1000 gold castings. According to their values, roughness TEMPERATURE V of dental castings could be as high as 0.4 percent. Figure 8. The casting fits the Bureau Standards die. When this value is compared with the total shrink- of The age of the gold alloys, the role of roughness in the Investment had 1.35 percent thermal expansion at a casting temperature of 1,000°F. with a roughness of 10 microinches. fit of dental castings becomes clear. In figures 8 The setting and hygroscopic expansion of the Investment was Identical to the one used in figure 9. and 9, the thermal expansion of two investments are shown. Both investments have identical setting and hygroscopic expansion. Two castings of the Bureau of Standards full crown die were made employing identical techniques. One of the castings fit the die (fig. 8) and the other did not (fig. 9). Roughnesses produced by these two investments were different. In figure 8, the roughness measured by the profilometer was 10 microinches whereas the casting in figure 9 has roughness value of 25 microinches.

7. Conclusion

The time has come when studies in this field should not be limited only to measurement of thermal, setting, and hygroscopic expansions of the investment, but should be expanded to include strength of the investment along with its rough- TEMPERATURE °F ness, strength and behavior of waxes, and shrink- Figure 9. The casting does not the die. age pattern of molten metal. fit The investment had 1.45 percent thermal expansion at a casting It should also be mentioned that due to the ef- temperature of 1,000°F. Roughness of the Investment was 25 microinches. The setting and hygroscopic expansion forts, and of the hard research work of Nieman [35], Investment was identical to the one used in figure 8.

65 8. References [20] Asgarzadeh, K., Mahler, D. B., and Peyton, F. A., The behavior and measurement of hygroscopic expansion of dental casting [1] Coleman, R. L., and Weinstein, L. J., Investment, investment, J. Dental Res., 33:519 U.S. Patent 1,932,202. (1954). [21] Lyon, H. W., Dickson, G., [2] Scheu, C. H., A new precision casting technique, and Schoonover, I. C, The mechanism of J. Am. Dental Assoc., 19:630 (1932). hygroscopic expansion in dental casting investments, [3] Phillips, D. W., Present-day precision inlay invest- J. Dental Res., 34:44 (1955). ing and casting technique, J. Am. Dental Assoc. [22] Ryge, G., and Fairhurst, and Dental Ck)smos, 24:1470 (1937). C. W., Hygroscopic expansion, J. Dental Res., 35:499 [4] HoUenback, G. M., Simple technique for accurate (1956). [23] Donnison, J. A., Chong, M. P., and Docking, A. R., castings : New and original method of vacuum The effect of surface investing, J. Am. Dental Assoc., 36:391, 1948. area on the hygroscopic setting expansion and strength [5] Sweeney, W. T., Cristobalite for dental investment, of casting investment, J. Dental Res., 36:967 J. Am. Dental Assoc., 20:108 (1933). (1957). [24] Skinner, E. W., and Degni, F., Hygroscopic [6] Coleman, R. L., Physical properties of dental ma- expan- sion of dental investments, terials, BS J. Research 1, 867 (1928) RP 32. J. Am. Dental Assoc., 54:603 (19.57). [7] Cra\vford, W. H., Selection and use of investments, [25] Mumford, G., sprues, casting equipment and gold alloys in mak- and Phillips, R. W., Dimensional change in wax pattern during ing small castings, J. Am. Dental Assoc., 27:1459 setting of gypsum investments, (1940). J. Dental Res., 37:351 (1958). [26] Fusayama, T., Hosoda, H., and Kher, [8] Shell, J. S., Hodgen-Shell Dental Materials, chap- V. M., Influ- ences of clinical variables ter 8 (The C. V. Mosby Co., St. Louis, Mo., 1938). on a cristobalite investment, .T. Prosthetic Dentistry, [9] Ireland, J., Vacuum investing and its relation to 11:152 (1961). [27] Mahler, D. B., and Addy, A. B.. cast surfaces, Brit. Dental J., 86:111 (1949). An explanation for the hygroscopic setting expansion of [10] Smyd, E. S., Wax, refractory investments and re- dental gypsum products, J. Dental Res., lated subjects in dental technology, J. Prosthetic 39:578 (1960). [28] Eamshaw, Dentistry 5:514 (1955). R., The effect of restricted stress on the setting expansion of [11] Gustafsson, C. G., and Hedegard, B., Investing and gypsum bonded investments, Australian casting technique. Acta. Odontol. Scand. 12:233 Dental J., 9:169 (1964). [29] Holland, R. H., (1954). and Paflfenbarger, G. C, Cast inlay technique, J. Am. Dental A.ssoc, [12] J0rgensen, K. D., Investigations on dental precision 19:185 (1932). [30] Hollenback. G. casting technique, Odontol. Tidskr., 62:443 M., and Rhoads, J. E., Comparison of the expansion of the mold cavity with the (1954) . linear casting shrinkage [13] Coy, H. D., Casting procedures in dentistry. Intern. of gold. S. Calif. State Dental J., 28:73 (1960). Dental J. 5:173 (1955). [31] Mahler, D. B., and Addy, A. [14] Asgar, K., Mahler, D. B., and Peyton, F. A., Hygro- B., The effect of water bath in scopic technique for inlay casting using controlled hygroscopic casting technique, J. Pros- thetic Dentistry, 15:1115 water additions, J. Prosthetic Dentistry, 5:711 (1965). [32] Pomes, C. E., Slack, G. L., and Wise, M. W., Surface (1955) . roughness of dental castings, [15] Fusayama, T., Technical procedure of precision J. Am. Dental Assoc., 44:545 (1950). casting, J. Prosthetic Dentistry, 9:468 (1959) ; 9: 1037 (1959). [33] Sliffert, L. W., and Mahler, D. B., Reproducibility [16] Peyton, F. A., Mahler, D. B., and Asgar, K., Con- of gold casting made by present-day dental cast- trolled water-addition technique for hygroscopic ing technics, J. Am. Dental Assoc., 50:1 (1955). expansion of dental casting investment, J. Am. [34] Barone, J. F., Huff, R. L., and Dickson, G., Surface Dental Assoc., 52:155 (1956). roughness of gold castings. Dental Prog., 1:78 [17] Docking, A. R., Chong, M. P., and Donnison, J. A., (1961). The hygroscopic setting expansion of dental cast- [35] Neiman, R., Investment composition, U.S. Pat- ent ing investments, Australian Dental J. 52:6, 160, 2,247,395; 2,247,585 ; 2,247,586; 2,247,587;

320 (1948) ; 53:261 (1949). 2,247,588. [18] Landgren, N., and Peyton, F. A., Hygroscopic ex- [36] Moore, T. E., Dental investment material, U.S. pansion of some casting investments, J. Dental Patent, 1,924,874. Res., 29:469 (1950). [37] Watts, C. H., Jr., Investment material, U.S. Patent, [19] Delgado, V. P., and Peyton, F. A., The hygroscopic 2,675,322. setting expansion of a dental casting investment, [38] Collins, P. F., Dental investment composition and J. Prosthetic Dentistry, 3:423 (1953). process, U.S. Patent, 2,006,733; 1,953,075.

66 NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Basic Metallurgy of Dental Casting Alloys

John P. Nielsen

J. F. Jelenko Co., Inc., New York, N.Y. 10801

The complex dental casting alloys have arrived at their present useful state mostly by

trial and error. The three areas of technical significance for these alloys are : inertness in the oral environment, fabricability (including soldering), and alloy strengthening. Basic metallurgical studies are being made in this area, and as the gap between the art and basic metallurgy is closed, improvement in properties and the cutting down of precious metal content can be expected. There are prospects of new alloy systems and improvement in casting technique, particularly in automatic casting. Ceramic metal systems need further study. For the long range, investigations of the prosthetic material-human tissue reactions are necessary. To assist studies in this field a dental materials handbook on properties of dental alloys and related materials should be compiled and published.

Key words : Alloys, dental casting ; ceramic-metal systems ; chrome-cobalt alloys ; dental gold materials ; gold alloys ; metallurgy, dental alloy ; porcelain-gold materials ; solder, alloy.

1. Introduction laboratories and the dental profession by continually modifying alloy compositions, involving hun- Basic metallurgy of dental casting alloys refers dreds of different melts, until a reasonable to the established knowledge of interatomic forces, optimum was achieved for the various perform- crystal structures, phase equilibria, solid state dif- ance characteristics. No doubt many a night was fusion and reaction kinetics, plastic deformation, spent with the melter and perhaps a somewhat crystal strengthening mechanisms, and the like technically oriented gold alloy salesman seeking of such alloys. The assignment to discuss this topic the alloys that would be favorably accepted by the was given with the suggestion that the future de- dental profession. Naturally little of such findings velopments be kept in mind more so than a re- appeared in the literature not so much by default view of past accomplishments. In pondering over as by an attempt to preserve secrecy for each little this suggestion it became apparent that it was an real or imagined discovery. appropriate one in that there is not all that much The chrome-cobalt dental alloys however were established basic metallurgy of these alloys to dis- developed with the advantage of established cuss. Dental cast alloys are still in the technologi- metallurgical technology for chemical resistant cal stage in which the art is ahead of their reliance alloys, namely, the Stellites. However, here the on basic metallurgy. For example, the dental gold diversity of alloy types applicable to dentistry, at alloys are generally complexes of four or five alloy least for the present, is not so wide as in the gold additions involving copper, silver, platinum, pal- alloys. Nevertheless, some metallurgical art run- ladium, nickel, zinc, tin, indium, and perhaps one ning ahead of basic metallurgy does exist in the or two others, yet the function of these elements is case of the base metal alloys for baked porcelain not well documented in the technical literature. applications. Apparently the commercial complex alloys have Another interesting aspect of metallurgical art been arrived at over the years by trial and error. in dental alloys is the technique of metal joining Two particular cases in point are the palladium by so-called soldering. Actually some significant and zmc alloying additions. We recently found achievements exist in this field, almost completely the palladium addition as singularly effective in developed by technicians. Indeed soldering as the reducing the tarnisli propensity of silver-contain- metallurgical term is known does not apply, but ing gold alloys. On checking the silver-containing rather "hard soldering", "brazing", and perhaps dental gold alloys on the market, it was surprising sometimes unfortunately "fusion welding". The how many contained the required amount of this gold alloy solders must be noble and of the same alloy addition for tarnish resistance. Zinc appears color as the joined material. Hence the simplest to be a specific element that minimizes the tend- procedure for making a solder is to take the alloy ency for gold alloy casting porosity. Here again to be joined and lower its melting range about most dental golds already carry the required zinc 150° C. with zinc and tin additions. In the hands addition. Evidently some unsung heroes in the of a skilled technician, joining gold alloys with past responded to the feedback from the dental such a solder works quite well. For the dental

67 behavior golds fortttnately the metals to be joined are noble, ; those on the upper end of the list will 1. e., inherently "clean", and hence fluxing what behave anodic to those below. No one seems to have oxides occur is easily done. The joining of the non- made such a list of the different gold alloys for gold cast alloys is however a considerably more sulfide-producing environments. The rating of serious problem challenging both the metallurgist gold alloys in tarnish tests is the nearest to such a and the technician. series. However, such tarnish tests have not been Perhaps the most interesting area of metallurgi- standardized. cal art versus basic metallurgy is the strengthen- In the case of the chrome-cobalt alloys, their ing of the cast alloys. The research as reported in inertness in certain environments is attributed not the metallurgical (as distinct from dental re- to nobility but to the ability to passivate, i.e., to search) literature indicates the logical approach produce a protective coating automatically on the for studying the strengthening of alloys (chiefly surface. Here again no criterion seems to exist that starting with binai-y sys- would assist the dental by age hardening) ; by profession in classifying tems, then going on to the ternaries, etc. Research- such alloys as to their inertness in the oral ers reporting in the dental literature begin, for environment. practical reasons, with the complex alloys already Thus we have two important classes of cast being used by dentists. The complexity of the al- dental alloys; the noble and the passive, with no loys however limits the attainment of data that re- established evaluating procedure for true inert- veal the mechanism of alloy strengthening. Thus, ness in the oral environment. although the two approaches are heading toward each other, there is a technological gap. 2.2. Fabricability A brief review of some basic physical metal- significant lurgy of cast dental alloys as found mostly in the A characteristic of the dental casting alloys is that metallurgical literature will define this gap more they cast readily and give faithful replication of the detailed clearly. contours of the impression pattern. This requires convenient melting and casting temperature ranges. Also the high density 2. The Basic Metallurgy Limits of Cast of the gold alloys yields a low kinematic viscosity, Dental Alloys so that the molten metal readily flows and fills the mold in all parts. The chrome-cobalt alloys, 2.1. Alloy Inertness although of lesser density, also rely on the generally low kinematic viscosity of the metals. Meas- The primary nature of the gold alloys is the urements of the speed of mold filling indicate fill- retention of some of the nobility inherent in pure ing rates for small castings in fractions of a second. gold. There are two basic criteria for evaluating The solidification takes an additional few seconds. the nobility of a metallic element. In one the molar A part of the castability feature of dental alloys free energy of the formation of a compound from is that they react very little with the mold ma- the element is measured for standard conditions. terial, so that little dressing of the casting is neces- If this free energy value is negative, the compound sary, at least little enough not to disturb the fit of will tend to form spontaneously under the specified the casting as a dental restoration, which is a rather conditions. If the free energy is positive, the com- precise requirement. pound will tend to decompose back to its compo- The gold-base alloys are easily joined by solder- nent elements. The degree of this decomposing ing with carefully matched soldering alloys. The tendency is measured by the positive free energy matching is for color and for tarnish resistance. value. This degree of instability of the compound However the technology seems to be completely is a measure of the nobility of the metallic com- in the hands of the technicians and the dental alloy ponent. If the positiveness of the free energy is manufacturers. There appear to be no criteria as maintained at elevated temperatures, where reac- to surface tension, viscosity, spreading power, or tion rates may be very high, then this tendency for physical properties of the solder metal. The solder the compound to decompose at high temperature joint metallographically appears good, although is the fire-refining phenomenon characteristic of there is little quantitative data as to the physical the highly noble metals such as gold and platinum. properties. Cui-rently the chrome-cobalt alloys are A second criterion for determining nobility is not soldered in dental applications. The metallur- the measurement of emf values against a stand- gical art on this problem has as yet not been suc- ard electrode in a standard cell under certain speci- cessful. Here may be an opportunity for basic met- fied conditions. Here again positive values of the allurgy to assist in solving the prolDlem of joining emf indicate nobility relative to the standard elec- the base dental alloys. trode, with the degree of nobility being indicated by the positive emf value. 2.3. Alloy Strengthening Neither of these criteria serve too well for alloys. Sometimes a galvanic series is set up for a par- The essential theory for the strengthening of ticular electrolyte, and the metals, which may in- gold alloys, as reviewed by the metallurgists writ- clude alloys, are listed as to their relative anodic ing for the dental journals is that an order-dis-

68 order transformation takes place. This comes from rich. Such alloys are generally not ductile, nor are the discovery of Kurnakov and his associates back they amenable to age-hardening. However, the in 1916 that the stoichiometric compositions, AuCu high rigidity of these alloys, twice that of the gold and AuCug change from random substitutional alloys, plus the inherent hardness of tw^o-phase solid solution to a superlattice type, all in an FCC mixtures, makes these alloys directly suitable for structure, on cooling below about 400 °C. The stress-bearing applications in dentistry—for ex- superlattice is of coui'se the harder form. The com- ample, for partial restorations. However, as Asgar plex dental gold alloys appear to be one-phase has so ably shown, it is not completely necessary alloys after aging and so the order-disorder theory to accept all the loss of ductility usually found in seems to still apply. these alloys. By lowering the interstitial carbon In the high fusing dental gold alloys for baked and molybdenum, and making one or two other porcelain applications, no copper is added (it minor modifications, the customary chromium- cobalt alloys can be toughened considerably, with- stains the porcelain green) , but still age hardening occurs; accompanied, incidentally, by much loss out any loss of rigidity, or significant loss of yield of ductility. There appears to be a massive phase strength. Perhaps there is a sacrifice of stable carbides, but this is important transfoi-mation—what it is, is not clear. When Fe only for high tem- or Sn are added in small amounts, on the order of perature applications, and certainly of no value in one atomic percent, in the presence of Pt in the dentistry. In these chrome-cobalt alloys basic alloy, age hardening is substantially increased. Ap- metallurgy is being intelligently exploited for parently Pt^Fey and Pt«Sn„ compounds of some dental applications. kind play a role. It is interesting to point out that for dental ap- 3. Future Expectations for Dental plications the age-hardening of gold alloys is de- Alloy Technology signed to occur automatically by simply avoiding quenching of the casting, by slow cooling in the Recognizing that a gap exists between metal- mold after solidification. Dental laboratories, or lurgical ait and basic metallurgy in cast dental dentists doing their own metal casting, are not alloys, one might well ask what are the future ex- prone to set up heat treating furnaces scheduled pectations for dental alloy technology. It should be for appropriate age-hardening heat treatment. noted that this gap at present is already being Here is a good example of a kind of metallurgical closed in part. The dental manufacturers are staff- art developed between the alloy supplier and the ing their research departments with metallurgists, laboratory technician to achieve acceptable results. and cui'i'ently there is evidence of good metallurgi- On the other hand, in the metallurgical litera- cal reseach being published by them. Further- ture there is little evidence that the convenience more, there is some sponsorship by the National of the dental laboratory technician is considered. Institutes of Health at several universities on den- Hence automatic age-hardening by mold cooling- tal materials research, including cast dental alloys. is practically never mentioned. Strengthening of It will take a little time for these institutions to be- gold alloys in general however has been in part come productive, or at least stimulative, of useful systematically pursued. The order-disorder trans- research results, but within 10 years physical formation is well recognized for gold alloys metallurgists and dental alloy experts should be containing substantial amounts of copper. The age- speaking the same language. hardening due to other causes is however also rec- With this in mind, and knowing the motivations ognized. In the AuCuAg ternaries there is x-ray in dentistry, in dental materials manufacturing, evidence of phase decomposition hardening—with and in government support for dental research, the Cu-rich phase perhaps exhibiting a superlat- some technological expectations can be put forth. tice structure, and the Ag-rich phase adding sub- This is best done by examining the broad range of stantially to the hardening in certain compositions. past developments up to the present and attempt- The AuPtPd ternaries show solid state decomposi- ing to sense the momentums into the future. tion into two phases on aging, as do the AuPt The nature of evolving systems is that they alloys, whereas the AuPd alloys do not. increase in their complexity by a process of suc- By the time Pt, Pd, Cu, and Ag are added to cessive mutations that continually increases the gold, the hardening is quite significant. All sorts adaptability of the system to its surroundings. of solid state transformations occur, most of which Each successful mutation is a quantum jump in completely escape the examination with the optical complexity, or a new branch in the existing branch microscope. Apparently the gold alloy technology structure representing the system. Thus we can gap exists mostly between the evolution by trial trace the tree-like growth of an organization, or of and error of the complex gold alloys vised in den- a science, or of a specific teclmology. Having done tistry and the physical metallurgy understanding this we are in the position to make some estimates of alloy strengthening in these alloys. as to the likely course of future developments in As far as the chrome-cobalt alloys are concerned, that system. these are two-phase alloys at room temperature, In the case of dental alloy technology one can one phase being cobalt-rich and one, chromium- rather easily construct the evolutionary process of

69 :

say the last 50 years. Shortly after 1919, due mostly 2. Development of new alloys. to the work by the dental research group of the a. lower precious metal-content alloys National Bureau of Standards, it was established but with no sacrifice in dental that four types of cast gold alloys could serve the performance. needs for the overall range of cast dental alloys. b. new precious metal alloy systems. The ancient lost wax process had been applied to c. new non-precious metal alloy systems. dentistry many years earlier, and by gradual im- 3. Development of new casting equipment provement in casting equipment and in investment such that good porosity-free restorations materials the technique used with gold-base alloys are obtained on every casting, requiring became commonplace, available to all dentists. little or no dressing. In the thirties the development of the chrome- 4. Development of new soldering alloys and cobalt alloys that were chemically passive lead techniques, particularly for the high heat naturally to the mutation of using such alloys in alloys. dentistry in place of the costly noble metals where There are other possibilities that can be pro- ductility or burnishability was not important. jected. Acrylic jacket techniques already in exist- In the forties the physical properties of gold- ence might call for new types of alloys. Perhaps base alloys were sufficiently standardized so that completely new metal fabricating techniques, such most gold alloy manufacturers published physical as electroforming or powder metallurgy, might property charts of their alloys, and the larger com- come into existence in place of casting. panies adopted quality control measures to assure that every melt met the physical properties listed 4. Recommended Basic Metallurgy in their charts. Programs In the late fifties the porcelain jacket technique was perfected using a gold-base alloy. The de- Up to this point we have reviewed the dental mands on this alloy are exacting. Its thermal ex- casting alloys both as to their metallurgical art pansion coefficient must mat-ch that of the porce- and their metallurgical technology aspects. Based lain, it must have high sag resistance at porcelain on judgments concerning past alloy developments, baking temperatui^es as liigh as 1050° C, it must a projection of the new alloy developments are be free of porcelain-staining elements, and it must given. Now, the motivation for the new develop- contain elements that contribute to the shear ments exist. The question is whether the motivation strength of the porcelain-metal bond, which relying on the slow growth of the metallurgical apparently is chemical in nature. art, i.e., by trial and error, is sufficient to bring In the sixties the technique of producing fine- about these improvements in a suitably short time. grain castings was brought under control, raising The answer is probably no in most cases. The pur- the toughness of the alloys and reducing coring, pose of applying basic technology rather than thereby minimizing heterogeneity of alloy proper- simply the trial and error technique is to shorten ties. This period saw also the emergence of white the time lag between needs and fulfillment of these alloys for porcelain baking. needs. In order to bring this about, it would not be to What may one reasonably expect of the future ? wise assign dental alloy development projects The list below has been developed primarily on categorically to the metallurgists and anticipate good results in short order. young metallurgical the basis of three motivating factors : ( 1 ) There is A always the pressure to reduce the jewelry aspect turk would quickly come up with new alloys, but of precious metal alloys, i.e., to cut down metal the chances are that they would not be as good as costs by reducing gold, platinum, and palladium the ones currently used. Some important but subtle factors would not content wherever possible. (2) It is always de- be taken into account. For example, the sirable to improve the physical properties such as proposed alloys might be substantially stronger and have the gold content in what is be- strength, toughness, and tarnish resistance for the lieved to be the safe range, but after several hun- same metal cost. (3) It is always desirable to cut dred restorations had been inserted in patients' down on the need for technicians' skills in the mouths, it would be discovered that they discolor fabrication restorations, of and obtain results as badly. Another likely occurrence in situations as possible much automatically and mechanically. when a young technically trained man is assigned It would seem therefore that in the next few to make improvements in an old established tech- years the following will emerge nology, is that old improvements are rediscovered. 1. Improvement in dental performance of The first order of business is then to close the present-day alloys with minor modi- gap explicitly between metallurgical art and basic fications. technology in dental alloys. The young metallur- a. improved tarnish resistance of gold gist with modern sophisticated tools, such as the alloys. electron probe, the scanning electron miscroscope, b. improved mechanical properties in and the computer, should first learn why the pres- both the gold and the chrome-cobalt ent alloys work as well as they do. For example, alloys. one of the first j)rojects to assign to the metallur-

70 gist entering the dental alloy field is the determi- most welcome. Incidentally, the high heat type of nation of the mechanism of strengthening of the dental castings, particularly the base metal alloys dental gold alloys by the somewhat automatic age- are generally not soldered well by the technician. hardening of dental castings by mold cooling. A Here the metallurgist has a clear problem to work second project might be to determine the subtle on immediately. dental golds in resistance to discoloration of the There is need for the eventual development of a the oral cavity. It is in the course of such a pro- simple automatic casting machine, especially for where the metallurgist is forced to learn gram, the high heat alloys. There is no reason why a some of the complicated requirements of dental dental casting cannot be made completely sound, castings, that he would develop an appreciation of faithful to the mold details, with little or no finish- research the true objectives of his development ing required, coming directly from some kind of work. It is very probable that in the course of such automatic casting machine. This development is of questions would be raised whereby a program new course a little outside the domain of the metallur- sophisticated improvements could new and more gist. New heat sources for the melting and the conceived in dental technology. For example, be possibilities of vacuum casting have to be explored, the of crown be cast in some in- how can bulk a and perhaps the development of new investment strength metal, then be covered expensive high materials must come about to aid the metallurgist. with a soft burnishable material that is completely One particularly good area for the metallurgist, noble? together with the ceramist for closing the gap we Another good question for the biomedical metal- are discussing, is the study of the very interesting this IS would become, would lurgist, for what he and quite successful technique of porcelain bond- surface reactions metal-tissue contact. be the on ing to metal restorations. Here, once the metallur- new and interesting avenues would be Many gist and ceramist working as a team have caught opened up for exploration. An enormous amount up with the technique as it is now practiced, new will eventually be done on this whole of work improvements would surely be forthcoming. question of prosthetic materials-living tissue Finally some long range projects might be de- reactions. scribed. With the laboratory tools of today it It is while the biomedical metallurgist is rang- would not be difficult to document all the phase ing over these wide problems that he would also compositions that occur in the complex gold-base learn to keep his feet on the ground with such alloys. Quaternary and higher component phase boundary conditions as precision castability, join- diagrams are too difficult to construct. However ability, etc. There is little point in making a a coordination of metallography and rasters avail- superior alloy as to strength and tissue inertness, able with the electron microprobe (these show if it cannot be fabricated or soldered, or if it is area distribution of selected elements in a micro- nonburnishable. structure) would furnish the next most useful One interesting area that a metallurgist could information, short of knowing the phase diagrams. get into almost immediately is the development of Indeed, such data may actually be more usefiil a technique for producing sound castings. Dental than that obtained from phase diagrams. Micro- castings as a rule are reasonably sound. The alloys probe analysis can monitor nonequilibrium com- are sufficiently deoxidized to make them gas free. positions that occur in castings and on heat Porosity and surface defects, however, remain the treatment. most serious problems in dental castings. Here the Another worthwhile long range project would metallurgist can quickly bridge the art versus tech- be the determination of distortion, dimensional nology gap, because casting soundness is not one stability, and residual stresses in dental castings, of the successes of dental metallurgical art. The for both the gold and the base metal alloys. Along reason for this is that except for excessive porosity, these lines a study of the stress distribution in or surface defects, the laboratory technician is in porcelain-on-metal i-estorations could be impor- the dark as to the quality of the interior of a tant. A tempered porcelain, analogous to tem- casting. pered glass, where the tensile stresses are all Perhaps the largest gap between art and tech- confined to the interior might add to the already nology in dental alloys is the soldering or joining considerable usefulness of porcelain facings in technique. This technique has been developed dentistry. almost completely without metallurgy, between Perhaps the most important long range program the trial error the gold solder and of producer and would be the study of material-tissue reactions. of the skilled laboratory technician. The metal- This would naturally overlap medical materials. lurgist has a great deal to learn here. However, once the metallurgist has caught up with the 5. A Dental Materials Handbook and technician there is the strong possibility for the Dental Materials Publications development of techniques where manual skill is not required, skills being a manual commodity that The most useful reference sources we have to- is getting scarcer and scarcer. A good joining day on dental materials are the textbooks by Phil- technique of the furnace brazing type would be lips and Skinner, and by Peyton. However, being

71 :

textbooks, they cannot give space to numerous research on dental alloys. Indeed handbooks and phase diagrams, microstructures, and mechanical the like are an indispensable tool and no research properties for a large variety of alloys. The origi- laboratory functions well unless the researcher is nal sources as references are scattered too widely surrounded by the bric-a-brac of compilations, to be available to all dental materials research handbooks, wall charts, and data tables. workers. Therefore it seems iustifiable to have com- The appropriate group to issue such a handbook piled a dental materials reference handbook. The and renewing it on a five-year basis or so, is the coverage might be in the order of the following: ADA division at the National Bureau of Stand- ards. The first issue need be only a soft cover photo- 1. Physical Properties of the (appropriate) offset product. The natural demand that would de- Elements (Obtainable from the ASM velop will dictate its future development. Metals Handbook) To the dental materials researcher it is impor- 2. Phase Diagrams tant that he publish his work and that he follow This part should include about 50 commonly the work of other researchers. It will be a luxury used binary phase diagrams involving at least one soon to publish dental materials research papers in

of the following components : Au, Ag, Pd, Pt, Cr, general dental research publications. It is a waste Co, Ni, Fe, Sn, Hg, Cu, Si02 AI2O3, CaO, MgO, already today to have to subscribe to three or four NaoO, and KjO. Discussions of each phase dia- dental journals for the 5 percent or so of papers gram should include crystal structure of phases, that apply to dental materials. Furthermore, when and lattice parameters, where possible. Likewise papers are prepared for publication, the authors there should be about 30 ternary systems described may have to strain to connect their paper to den- in isothermal diagram form, involving at least two tistry. For example an excellent paper may be un- of the components of the above. The phase dia- fairly criticized for not having clinical data, when grams can be mostly obtained from Hansen]s and actually the information without clinical data is from Guertler's compilations. The nonmetallic sys- directly useful to dentistry, or may be a prelude tems are obtainable from "Phase Diagrams for to clinical work. Whether there are enough dental Ceramists," published by the American Ceramic materials papers to have special issues devoted to Society. Quaternary alloy diagrams are not too this subject is still uncertain. However, this can common or useful. "However, in due coarse there be anticipated in the next few years, particularly should be electron rasters of complex alloys in the since dental materials research is increasing equilibrium state. Wlien these apply to dental ma- significantly. should be included. terial systems, they 6. Summary 3. Mechanical Properties It appears that cast dental alloys arrived at their should be searched for assembling The literature present usefulness mostly by trial and error with mechanical proper- as completely as possible the coordination between the dental alloy producers polymeries. ties of pertinent alloys, ceramics, and and the dental laboratories and dentists. The three These should include data for cast, wrought, an- areas of technical significance for these alloys are nealed and heat-treated conditions. alloy inertness in the oral environment, fabrica-

bility (including soldering) , and alloy strengthen- 4. Corrosion Data ing. One can find some basic metallurgy in the lit- for electro- The literature should be searched erature in these areas, but nevertheless a gap exists literature chemical tarnish, and corrosion data. The between the art and the documented technology. on electrical contact technology should not be over- However, metallurgists are entering the field more looked in this connection. and more, and one can expect advances in the improvement in the properties of the present alloys 5. Miscellaneous Properties and in the cutting down of the precious metal con- and color data would be use- Heat conductivity tent while maintaining current properties. There is alloy and ceramic systems. ful for certain a prospect of new alloy systems, mostly in the non- 6. Miscellaneous Materials precious or low precious metal area. Improvements Standard information on waxes, resins, invest- in casting technique, mostly in the direction of au- depend- ment materials, and cements generally used in tomatic casting can be expected. Likewise, dental techniques should be included. ence on the necessary manual skills in soldering, One major advantage of such a handbook would both of the precious metal and the base metal type, works be the sparing of the dental materials textbook au- will be reduced. The trial and error method thors the chore of including tables of routine data too slowly to bring these improvements to the fore programs in their books. A textbook should not have to serve in the near future. Hence research research as a data reference book, but as an expounder of should be initiated ; first to acquaint the in the concepts so that the information in handbooks can metallurgist as to what is already good be intelligently used. However, the principal ad- present-day dental alloys and techniques, and sec- vantage would be to facilitate basic metallurgy ondly to open avenues as to where improvements

72 can come about. Ceramic metal systems need 7. Bibliography further study. Dimensional stability of castings is also important. A most interesting long range 1. Asgar, K., Techow, B. O., Jacobson, J. M., A new alloy for partial dentures, J. Prosthetic Dentistry project, and one that is important for medical 23: 36, February 19(0. is the study of prosthetic ma- materials in general, 2. ASM Handbook, 8th Ed. vol 1, pp. 1174-1196 (Ameri- terial-human tissue reactions. Finally, it seems that can Society for Metals, Metals Park, Ohio, 1961). certain research aids are in order to assist the new 3. O'Brien, W. J., Kring, J. E., and Ryge, G., Heat treatment of alloys for porcelain metallurgist working on basic metallurgy of den- to be used fused technique, J. Prosthetic Dent. 14 : 955 Sept-Oct, tal alloys. A dental materials handbook should be 1964. compiled and revised periodically. Dental ma- 4. Peyton, F. A., Restorative Dental Materials, 3rd Ed. terials papers should be given special issue status (The C. V. Mosby Co., St. Louis, Mo. 1968). 5. Shell, J. W., Hogden-Shell Dental Materials (The in the dental journals. C. V. Mosby Co. St. Louis, Mo. 1938). 6. Skinner, E. W., and Phillips, R. W., The Science of Dental Materials, 6th Ed. (W. B. Saunders Co., Philadelphia, Pa., 1967). Considerable assistance in the preparation of 7. Souder, W., and Paftenbarger, G. C, Physical Prop- this report came from Joseph Tuccillo of the Je- erties of Dental Materials, NBS Circ. 433, 34-80 lenko Co., Edmmid M. Wise (formerly of the In- (1942). 8. Vines, R. F., and Wise, E. M., Platinum Metals and ternational Nickel Co.), and Professor Ernest Their Alloys (International Nickel Co., 1941). Levine of New York University. Many references 9. Vines, R. F., and Wise, E. M., Age Hardening Pre- were used and it is difficult to make detailed cious Metal Alloys, pp. 190-230 (American Society for Metals, Metals Park, Ohio, 1940). credits. The following were the ten most frequently 10. Wise, E. M., Gold (D. Van Nostrand Co., New York used. 1964).

IV. New Development in Nonmetallic Restorative Materials

452-525 0—72 6

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Dental Porcelain

John W. McLean

Institute of Dental Surgery, Eastman Dental Hospital, London, W.C.T England

Current research on dental porcelain has been concentrated on methods of improving fracture resistance of porcelain restorations. Enamelling of metals or dispersion strengthening of glasses offer the greatest possibilities for this purpose. The fusion of porcelain to metal copings has proved very successful under clinical conditions, but the necessity for masking the metal substructure presents difficulties in obtaining the desired aesthetic characteristics. Replacement of the metal copings by a higher strength ceramic should improve aesthetics and reduce productive costs. Dispersion strengthening of glass with high strength alumina crystals has resulted in the production of a new range of aluminous porcelains for jacket crown and bridge pontic construction.

Key words : Aluminous porcelain ; ceramic materials, dental ; dental materials ; dental

porcelain ; dispersion strengthening of glass ; porcelain, dental ; porcelain-gold techniques and materials.

1. Introduction under clinical conditions. Providing a strong bond is achieved between the porcelain veneer and the The exacting aesthetic requirements of dentistry metal, there is little chance of leakage at the inter- have largely dictated the composition of dental face. In addition the porcelain enamel is reinforced porcelain. In order to produce highly translucent by the metal and is less likely to be placed under enamels, dental porcelain must contain a high pro- a tensile force which can cause brittle fracture. portion of glassy material. Regular dental porce- The attachment of porcelain to gold in dental lain has been developed from feldspathic glasses restorations has been variously ascribed to a com- modified by small amounts of crystalline materials bination of mechanical, wetting, or Van der Waals and coloring pigments, and suffers from the in- bonding. Vickery and Badinelli [1] ^ have exam- herent brittleness associated with all glasses. ined the interfacial effects between porcelain and It is hardly surprising that priority in research single crystal and polycrystalline gold surfaces has been given to methods of reinforcing dental and showed that, whereas mechanical and com- porcelain, since no other incident is more likely pressive forces played significant roles in attach- to frustrate the dentist in his work than the ment, Van der Waals forces probably do not make breakage of a porcelain jacket crown. The chip- a meaningful contribution. The only true bonding ping or complete fracture of porcelain pin teeth or effect derived from direct reaction of porcelain bridge pontics can also occur, involving both den- with interstitial compounds developed at grain tist and technician in time-consuming repair work. boundaries in the metal substrate. No evidence was found for a chemical bond between porcelain 2. Research Avenues to Improve Dental and gold per se. Industrial research on metal- Porcelain Strength ceramic bonding using modern techniques such as electron microscopy or electron probe micro- In order to improve the strength of dental porce- analysis has revealed that a glaze-body interface lain a number of avenues are open and future is always a region, sometimes very thin indeed, research would appear to center around five where diffusion of ions has been going on in both approaches. directions [2]. At the elevated temperatures em- 1. Enamelling of metals. ployed, the ions are quite mobile and will diffuse 2. Dispersion strengthening of glasses. from one material into the other. This might be 3. Enamelling of high strength crystalline termed chemical reaction but examination of the ceramics. interface revealed no definite compound formation. 4. Controlled crystallization of glasses. It therefore seems probable that the major con- 5. Production of compressed surface layers in tribution to effective bonding of dental enamels to dental porcelain via ion-exchange. gold is to be found in the compressive stresses developed through accurate adjusting of the thermal 2.1. Enamelling of Metals expansions of the gold and porcelain. The fusion of dental porcelain to metal copings 1 Figures in brackets indicate the Uterature references at the is now widely used and has proved very successful end of this paper.

77 Whatever the exact nature of this bond the most veneered with translucent enamels, produced much effective important fact is that it has proved very more life-like crowns since some light was still clinical conditions. Providing the technician being under diffused through the core procelain [4]. By takes care in degassing and cleaning his casting contrast the totally opaque calcined alumina core prior to applying the opaque porcelain the inci- porcelain caused much more light reflection from dence of porcelain actually shearing off the metal its surface, and crowns constructed from this mate- is very low. rial produced very similar effects to the opaque The somewhat doubtful aesthetics of the an- porcelain used in the fused porcelain to gold tech- terior fused porcelain to gold crown is perhaps of nique. It became clear that opaque porcelains must greater clinical significance. In order to ensure strike a fine balance between masking out cement that the appearance of the crown is acceptable, the linings or metal but at the same time not reflecting dentist is often required to remove more dentin too much light. It is doubtful whether this fine than is biologically desirable. This heavy destruc- balance will ever be struck with the fused porcelain tion of soimd dentin has increased the risk of pulp to gold restoration since it is essential to use very mortality and considerable effort is now being opaque materials in order to avoid any metallic made to reduce the thickness of metal, opaque and greyness showing through. For this reason it is enamel porcelain used in the standard type fused likely that any further improvements in aesthetics porcelain to gold crown. will be achieved through careful control of dentin In order to mask the metal substructure effec- translucencies in relation to a wide range of neu- tively, the opaque porcelains are heavily loaded tral opaque tones. Unfortunately the manufac- with opacifiers. These opaque layers will also pre- turers are placed in a dilemma in this situation sent problems since light reflection from their sur- since they could very easily produce dentin colors faces can often be severe. The manufacturers are with higher translucency if these colors were used therefore restricted in the amount of light trans- in consistent thickness. The variable thicknesses mission they can allow through their dentin porce- encountered in crowns made for clinical use will lains which are generally more opaque than the often rnean that a body dentin color of less than 0.5 regular porcelains used in jacket crown work. The mm thickness must be used. In this situation a anterior fused porcelain to gold crown will often commercially produced porcelain powder must lack depth of translucency, and due to the in- have a built-in safety factor, and light transmis- creased opacity, light reflection from the surface sion through the porcelain must be reduced to a is higher than in an all porcelain restoration fired level where the opaque porcelain does not have in vacuo. This property not only exacerbates any a dominating influence on color. If such an influ- metameric effects but can result in a deadness in ence occurs, the clinician will face serious meta- the mouth, reminiscent of some of the early air- meric problems once the patient experiences fired porcelain production teeth. Vacuum firing of varying artificial lighting conditions. In particu- the enamel and translucent porcelains can improve lar color changes of this type can often be severe the incisal translucency but the cervical two-thirds in subdued tungsten light. It is obvious that these of the tooth presents a formidable problem to the problems could be further reduced if the metal technician if the ultimate in aesthetics is to be copings were made thinner, and the search for achieved. higher strength metals which also have a high There is little doubt that the manufacturers have rigidity modulus is continuing with accelerated made considerable strides in solving some of these interest. The stellite-type casting alloys based on problems. Firstly by increasing the shade range of chromium, cobalt, nickel and molybdenum are an the opaque porcelains so that each opaque color is obvious choice but obtaining good casting accuracy closely matched in hue to the dentin colors and in thin section and a strong bond with dental secondly by using translucent overlay enamels to porcelain can often prove difficult under average give an illusion of depth to the crown. Built-in laboratory conditions. However the manufacturers concentrated colors can also improve appearance are devising improved techniques to and greys and blues can enhance the eifect of in- simplify these procedures cisal translucency. Surface staining will never and acceptable restorations can now achieve the same effect and if natural enamel is to be made. be simulated to perfection a ceramic crown must Successful bonding of porcelain to these non- have depth of translucency. precious metal alloys is dependent upon the pre- This problem was particularly noticeable dur- vention of a buildup of too thick an oxide layer ing the course of experiments on aluminuous por- at the porcelain-metal interface. A clean and celain core materials [3]. Porcelains containing homogeneous casting is the basic requirement for 40 percent very fine calcined alumina (specific sur- successful work. Sprues should be attached in such face area 6,000 cm^/g) were, for all practical a manner that a vertical or horizontal beam serves purposes, totally opaque whereas core porcelains as a reservoir with short 2 mm connections to the containing coarser grades of fused alumina (spe- casting itself. The casting must be thoroughly cific surface area 1,150 cm^/g) transmitted up to sand-blasted and cleaned with a solvent such as 20 percent light. The latter porcelains, when acetone or alcohol before applying the opaque.

78 .

After the opaque porcelain is applied, the car- Alumina has been used as an additive in regular dinal point to bear in mind is that vitrification dental porcelain for many years [8] but its use has must be rapid and complete before a thick oxide been mainly confined to making the so-called "hard separating layer can form and interfere with bond- core" porcelain which surrounds the pin anchor- ing. The use of metal firing trays to improve the age in a vacuum-fired tooth. In order to use alu- thermal conduction, and furnace muffles with even mina reinforced porcelain in greater section it heat distribution will assist the technician. Short became necessary to prepare specialized materials firing cycles of one to two minutes are also recom- in which much greater attention had to be paid to mended and opaque porcelains which can vitrify in the matching of thermal exjDansions and the sinter- this time are now available. Because of the strong ing characteristics of the glass matrix powders and oxide formation in metal areas which are not cov- alumina crystals. ered by porcelain it is good practice to extend the Research indicated that if the alumina crystals opaque layers into these transition areas and grind were dispersed in a borosilicate glass of matched away any excess porcelain after the final bake. expansion to the alumina, the strength resulting More clinical experience is required before any from this stress-free matrix was improved [6]. assessment of the merits of nonprecious metal al- Further work revealed that when fused alumina loys versus gold alloys can be made. Even in the crystals of 99.5 percent purity and with a size case of the precious metal alloys, the fine balance range of 20 to 30 /xm (specific surface area 1153 that needs to be struck in applying the various cm-/g) were incorporated in a specially prepared layers of porcelain taxes the skill of the most borosilicate glass containing a high combined experienced ceramist. With a world-wide shortage alumina content that "Aluminous Porcelains" in this skill it is hardly surprising that many could be produced with transverse strengths of fused porcelain to metal crowns do not do justice over 20,000 psi (140 MN/m^) and with a light to the full potential of these materials. transmission of up to 20 percent on 1 mm thick

However, despite these deficiencies, the fused discs [3] . This improvement in strength is approxi- porcelain to metal restoration has proved of mately double that of regular dental porcelain and enormous benefit to dentistry and has widened the clinical trials with these aluminous porcelains indi- aesthetic scope of both fixed and removable partial cated that they could be used in jacket crown prosthesis. manufacture [9]. It may be generally stated that the strength and 2.2. Dispersion Strengthening of Glasses opacity of an alumina reinforced porcelain is a function of its crystal or particle size. The finer The elimination of metal substructures in crown the crystal size the greater the strength and and bridge work is desirable, not only for aesthetic opacity. Research by the dental manufacturers has of reasons but also because of the high cost therefore been directed towards optimizing the production. crystal-glass particle size relationship to obtain The replacement of metal copings by a higher improved sintering characteristics, strength and strength ceramic seemed to offer possibilities and acceptable translucency. ceramics in- a survey of high strength industrial One such study reports the results with a special of dicated that increasing use Avas being made borosilicate fusion, high in alumina, in which the in in- ceramic oxide crystals as a reinforcing phase glass was tailored to have good sintering character- dustrial porcelains [3]. istics and high strength in the sintered state [10]. elasticity It is well known that the sti-ength and Addition of a closely sized alumina aggregate in physical interaction of glass can be increased by which the preferred alumina grains had fairly with an included phase of high elasticity [5, 6, Y] smooth surfaces improved the sintering behaviour prin- Dispersion strengthening of glass utilizes this of the alumina-glass composite, and high sintered cei-amic crystals of high strength ciple whereby densities were obtained. Modulus of rupture values and elasticity are fused in a glassy matrix to form in the range of 22,000 to 24,000 psi (160 to 165 composites crystal-glass composites. These form MN/m-) were obtained and in some cases ceramic a constant strain system and fracture has an equal bodies with strength values of over 30,000 psi (210 chance of starting in either phase. In the absence MN/m^) were produced. Relationships for sinter- of thermal expansion differences the strength and ing behaviour and strength were obtained that elasticity will inci-ease, approximately, in propor- were at least qualitatively predictable. These be- tion to the amount of the crystal phase. The choice haviour patterns closely followed Griffith's flow of ceramic crystals is fairly wide but for dental theory for strength and a viscous flow theory for purposes factors such as fusion temperature, bond- sintering. Lowered strengths resulted from solid ing with dental porcelain, color and aesthetic inclusions having unfavourably low thermal ex-

' values must be taken into account. Alumina pansions or by reactions or diffusion resulting in (AI2O3) was found to be the most suitable ceramic unfavourable grain boundary stress. Pore struc- for this purpose [3] and this material is re- ture was dependent upon particle size distribution ceiving increasing attention from the dental of the initial powder and upon the time-tempera- manufacturers. ture firing cycle.

79 The current alumina reinforced porcelains have might make it possible to produce lingual alumina benefited from this type of research and new types whisker-reinforced glass shells which could be of isotropic glasses containing high combined lightly ground to fit the jacket preparation. In alumina contents have been developed for veneer- this case the shells could be produced by vacuum ing the aluminous porcelain cores. It is likely that hot pressing. However it is doubtful whether this further improvements in the strength of these alu- method could rival the comparative simplicity of mina reinforced ceramics is possible but at present the fused porcelain to gold technique. the aluminous porcelain jacket crown cannot be The use of preformed "green" whisker-glass regarded as a replacement for the fused porcelain composites in tooth production should not be ig- to gold crown. The dentist must still be aware that nored since it may be possible to tailor-make the he is dealing with a brittle material and alumi- reinforcement to fit exactly in the tooth mould nous porcelain jackets must have adequate tooth prior to applying the overlay dentin and enamel support and at least 1 mm clearance on the lingual colors. surface if the alumina reinforcement is to make a significant contribution to strength [11]. 2.4. Enamelling of High Strength Crystalline Due to the semitranslucent nature of the alumi- Ceramics nous core jDorcelains and the high translucency of The the enamel veneere the anterior aluminous porce- bonding of aluminous porcelain to high strength recrystallized lain crown is superior in aesthetics to the fused alumina provides an alternative porcelain to gold crown. method of reinforcing the enamel veneers. Recrystallized Clinical experience has indicated that aluminous aluminas in excess of 85 percent purity porcelain should be used primarily as a replace- are one of the strongest groups of industrial ceramics. ment for and to extend the use of regular dental Recrystallized or "high alumina" is particularly porcelain. a attractivB material from the dental standpoint since it may be colored to dentin shades by the use of 2.3. Dispersion Strengthening of Glass With high-temperature-resistant Alumina Whiskers ])igments such as manganese-alumina pink, vana- dium-zircon blue or praesodymium-zircon yellow. Alumina whiskers are filamentary single crys- The resultant color backgrounds are very recep- tals having both high surface and crystalline per- tive to veneering with dental enamels and in addi- fection. Their strength at room temperature can tion a strong ionic bond is formed between the be as high as 2X10« psi (14X10^ MN/m^) with a alumina and aluminous porcelain veneer. This modulus of elasticity of 70X10« psi (48X10* chemical bonding is fairly tenacious and ion ex- MN/m==). change at the surface of the alumina is such that Dispersion of these whiskers in a glass matrix diffusion of ions from the porcelain into the alum- may be done in several ways. The following two ina can be denionstrated imder ultraviolet light methods were tried in jacket crown work. The [3] . High alumina ceramics also exhibit high ten- whiskers were intimately mixed with a finely sile strengths in excess of 17,000 psi (117 MN/m^), ground glass powder used for making aluminous and their modulus of rupture is in the region of porcelain and applied to a platinum matrix by 50,000 psi (345 MN/m^) [3]. standard technique. It was found that whisker More recently methods of firing high purity concentrations above 15 percent created difficulties alumina containing 0.2 percent magnesium oxide with regard to sintering behaviour, and high den- have been devised in which it is thought that a sities could not be obtained. It became clear that spinel is formed at the grain boundaries, slowing if sufficient whiskers were to be incorporated in the down grain growth, and allowing the diffusions glass powder to make a significant contribution of porosity along the grain boundaries. The re- to strength that alternative methods would have sultant materials are almost pore-free which in to be used. Vacuum hot pressing techniques allow turn high whisker concentrations to be used but such produces a highly translucent ceramic body. methods are not commercially viable in dentistry. A completely transparent recrystall ised alumina An alternative method was tried in which pre- has been developed by the General Electric Com- formed "green" whisker tapes were incorporated pany ajid marketed under the trade name of "Lu- in the jacket crown. These tapes were produced by calox". This type of material is particularly useful using the finely ground glass powder and highly for making thin shells for crown facings. oriented whiskers held in an organic binder [12]. High alumina can be manufactured in fairly

The tapes were incorporated in aluminous core intricate shapes to a high degree of tolerance ( ±2 porcelain on the lingual surface of the crown by percent linear) and preformed high alumina rein- a- lamination technique. Once again problems of forcements for crown and bridge construction sintering the alumina whisker-glass composite are now available. These reinforcements are sup- arose, since the layers of porcelain m between plied in the form of rods, sheets, tubes, and dove- the tapes tended to be very porous despite firing tail veneers and a number of types of construc- in vacuo. It was considered that future research tion are available to the technician [13].

80 Oval alumina tubes provide strong anchorage and the starting glass must be homogeneous with areas for custom-built or factory-made bridge qualities like optical glass. Spodumene is a suitable pontics and rival the fused porcelain-to-gold glass, compounded from the oxides of lithium, pontic in strength. Alumina tube pontics may be aluminum, and silicon and has been used exten- used with conventional golds and have the added sively to make "Pyroceram" cooking ware. Nor- advantage of replaceability. They can also ex- mally the ware is heated to a temperature where tend the use of porcelain in cases of long span nuclei are formed from the dissolved titanium bridges and markedly improve the aesthetics of dioxide. This happens at a temperature a little the anterior fixed bridge. Similar tubes can also above the annealing point where the glass shows be used to reinforce post crowns and very good the first signs of softening. After myriads of nuclei clinical results have been recorded in close bite have been formed in this way, the glass is slowly cases [14]. heated to higher temperatures where tiny spodu- Sheets of high alumina, 0.8 mm thick, may also mene crystals grow on the nuclei, converting the be incorporated in the lingual surfaces of alumi- transparent glass to an opaque white mass com- nous porcelain jacket crowns to reinforce the criti- posed chiefly of spodumene crystals. cal biting area. These crowns are more aesthetic The opacity of these glass-ceramics makes them than the metal reinforced restorations and clinical unsuitable for dental application but more re- trials have indicated a very comparable strength cently McCulloch [17] has reported on experi- performance [4]. ments with a glass-ceramic based on lithia-zinc All-ceramic fixed bridges can also be constructed oxide-silica and using metal phosphate nucleating using aluminous porcelain jacket crowns con- agents. This glass was transparent and amber in nected by high alumina rods. However these color in the glassy state but became translucent bridges are better confined to single tooth replace- and tooth-like after crystallization or ceraming ment since on long spans there is a risk of the abut- for 1 hr at 600 °C. At this stage modulus of rup- ment aluminous porcelain crown being placed in ture figures in excess of 18,000 psi (124 MN/m^) tension and fracturing down the midline. The were obtained. alumina reinforced fixed bridge will not replace This glass-ceramic was used to produce posterior fused porcelain to gold, but for the manufacture teeth by moulding in a die and counterdie. These of small span anterior bridges, alumina is a ma- production teeth were made only in a single shade terial of great aesthetic appeal. but further experiments were performed in which The construction of high alumina copings di- bars of the vitreous glass were made photosensi- rectly on to platinum matrices has also proved tive by using silver as a nucleating agent. On cool- possible [4], but the control of firing shrinkage is ing, the bars responded to ultraviolet light so that still a problem. Regular dental porcelain or alu- by differentially irradiating the surface, the glass, minous porcelain contains sufficient glass phase to on heating to the ceraming temperature, could be deform and slump over the platinum matrix. By made to crystallize at different rates, thus creating contrast, high alumina is extremely resistant to a polychromatic effect. It was shown that further pyroplastic flow and when shrinkage takes place characterisation might be accomplished by apply- during sintering, cracks or fissures will open up ing printed transfers, containing tooth pigments, in the alumina core powder. In some cases the to the surface. platinum will be permanently distorted, as the These experiments have not vet reached the stage alumina crushes it [4]. Experiments are continu- of commercialization and considerable problems in ing toward producing high content alumina-glass fabrication and color control must still present powders in order to increase pyroplasticity with- themselves to the tooth manufacturer. The current out seriously impairing strength. high standard of aesthetics in a vacuum fired anterior porcelain tooth has been created by building 2.5. Controlled Crystallization of Glasses in concentrated colors and multiblended veneers of dental porcelain by manual application. The crystallization glass developed Controlled of was fact that this process has not yet lent itself to auto- by Stookey in the United States and new and [15] mation is indicative of the tremendous problems unique properties were observed in these glass- encountered in duplicating both the depth of ceramics. Not only was the strength of these ma- translucency and color of human enamel and den- terials markedly improved but very high thermal tin. Dentistry still demands a high degree of shock resistance was imparted. artistic skill even in a mass produced porcelain Controlled crystallization of glass depends upon tooth. the fact that glass, at ordinary temperatures, is an undercooled liquid is in an unstable which 2.6. Production of Compressed Surface Layers state [16]. It can be made to crystallize by heating in Dental Porcelain Via Ion-Exchange to the proper temperature with crystal seed or nuclei present. The glass is then converted to a Chemical toughening of glass is widely used in dense mass of very tiny interlocking crystals. industry and these techniques may have some ap- Titanium dioxide is an effective nucleating agent plication in dentistry.

81 Isard and Lehman [18] have reported on experi- alumina reinforced bridgework or for direct bak- ments in which specimens of soda glass were heated mg of porcelain jacket crowns has reached a stage in a bath of fused potassium nitrate to effect ion- where commercial materials are now available. exchange. The compressed surface layer produced Vickei'y [22] has described a technique for using by this treatment imparted considerable strength a castalale, refractory, ceramic die composition in to the glass and it was claimed that the high which it is claimed that accurate reproduction of strength was not removed by abrasion. The treat- impressions taken in silicone or polysulphide elas- ment could be applied to the complicated shapes tomers is possible. When coated with a special with re-entrants such as jacket crowns and would polyphase refractory suspension the dies may be be particularly suitable for use with aluminosili- employed for the direct construction of porcelain cate glasses used in dental porcelain where ion- jacket crowns and inlays and butt-shoulder mobilities are relatively high. porcelain-on-gold restorations without the use of The application of this technique to clinical platmum foil. The die may then be removed di- practice presents certain problems since any oc- rectly from all of these preparations. No details clusal adjustments of the crown would remove of the composition of this refractory die material the compressed surface layer. However it might were given, and its accuracy against the platinum be possible to toughen crowns chemically after foil technique has yet to be established both in grinding in the mouth and this area of research de- the laboratory and clinic. serves further study. Chemical toughening of den- Miller Yardley [23] has published a clinical re- ture teeth does not appear to be a practical prop- port on the use of multiunit aluminous porcelain osition since these teeth are generally adjusted for bridgework constructed on a refractory die mate- occlusion after processing the denture. rial made specially for use with alumina reinforced ceramics. The ceramic refractory was used to in- 3. Additional Areas of Dental vest the platinum matrices prior to joining the Porcelain Research aluminous porcelain copings with high alumina I'ods. The fit of these bridges was clinically accept- 3.1. Silane Bonding of Acrylic Resin able and case reports were presented covering an to Porcelain 18 months' postoperative period.

Other areas of research on dental porcelain have 4. Summary been directed toward improving the bond between porcelain teeth and the acrylic denture base. It Current research on dental procelain has been was shown that when feldspathic porcelain teeth concentrated mainly on methods of improving the were treated with gamma-methacryloxypropyl- fracture resistance of porcelain i-estorations. trimethoxysilane that a bond was formed between Enamelling of metals or the porcelain and acrylic of such strength that dispersion strengthening of glasses offer the greatest possibilities when failure occurred, fracture took place in the at the present time and a review is given of the progress porcelain [19]. made in each field. Semmelman and Kulp [20] considered that this The significance of the role of bond was not adequate in heat-cured denture bases opaque porcelains is discussed in relation to their when measured from a clinically determined effect on the aesthetics of enamel veneers. standpoint and that the pin anchorage was still The use of nonprecious metal alloys as an alter- necessary. This evidence was in agreement with the native to gold is also receiving increasing attention work of Myerson [21] who considered that the and methods of improving the bond at the metal- unsatisfactory bond strength was produced by ex- ceramic interface are described. cessive shear stress during cooling. He showed that Dispersion strengthening of glass with high by reducing the temperature of processing, the strength alumina crystals has resulted in the pro- cooling shear stress was reduced and concluded duction of a new range of aluminous porcelains for that the self-cui'e systems produced the strongest jacket crown and bridge pontic construction. bond with the silane treated porcelain tooth. His Chemical toughening or nucleation of glasses final conclusion, which is probably most pertinent, may also provide otlier means of reinforcement but stated that the large differential in thermal expan- at the present time these methods are still in the sion represented the weak point in the silane experimental stage. bonded porcelain-acrylic system. This must be Other areas of research have been concentrated accounted for before the obvious advantage of sil- on impi'oving the bond between acrylic resin and ane bonding can be used. porcelain denture teeth by the use of silane 3.2. Refractory Die Materials coupling agents. The large differential in thermal expansion between acrylic and porcelain must be The development of new types of refractory accounted for before these methods become com- die materials for the construction of all-ceramic mercially viable.

82 .

5. References [11] McLean, J. W., The alumina reinforced porcelain jacket crown. J. Am. Dental Assoc. 75, 621 (1967) [12] Wakelin, R. J., Personal communication. [1] Vickery, R. C, and Badinelli, L. A., Nature of attachment forces in porcelain-gold systems, J. [13] McLean, J. W., High-alumina ceramics for bridge Dental Res. 47, 683 (1968). pontic construction. Brit. Dental J. 123, 571 [2] Batchelor, R. W., Harrison and Son, Stoke-upon- (1967). Trent, England. Personal communication. [14] McLean, J. W., The alumina tube post crown. Brit. [3] McLean, J. W., and Hughes, T. H., The reinforce- Dental J. 123, 87 (1967). ment of dental porcelain with ceramic oxides, [15] Stookey, S. D., Catalyzed crystallization of glass in Brit. Dental J. 119, 251, (1965). theory and practice. Ind. and Eng. Chem., 51, [4] McLean, J. W., The development of ceramic oxide «05 (1959), reinforced dental porcelains with an appraisal [16] Greene, C. H., Pyroceram., The New Scientist, p. of their physical and clinical properties, MDS 1708. (Dec. 1960). thesis. University of London, (1966). [17] MacCulloch, W. T., Advances in dental ceramics. [5] Batchelor, R. W., and Dinsdale, A., Some physical Brit. Dental J. 125, 361 (1968). properties of porcelain bodies containing corun- [18] Isard, J. O., and Lehman, M. L., Letter to Editor. Transactions the International dum, of 7th Ce- Brit. Dental J. 120, 56 (1966). ramics Congress, London, England, 31, (1960). p. [19] Paffenbarger, G. C, Sweeney, W. T., and Bowen, Binns, D. B., Some physical properties of two-phase [6] R. L., Bonding porcelain teeth to acrylic resin crystal-glass solids. Science of ceramics Vol. 1, denture bases. J. Am. Dental Assoc. 74, 1018 315-334 (Academic Press, London, pp. England, (1967). 1962). [20] Semmelman, J. O., and Kulp, P. R., Silane bonding [7] Hasselman, D. P. H., and Fulrath, R. M., Proposed porcelain teeth to acrylic. J. Am. Dental Assoc. fracture theory of a dispersion-strengthened glass 76,69 (1968). matrix, J. Amer. Ceram. Soc. 49, 68, (1966). [21] Myerson, R. L., Eftects of silane bonding of acrylic [8] Fonvielle, F. P., and Semmelman, J. O., Uses of resins to porcelain on porcelain structure. alumina in porcelains, lADR Program of Ab- J. Am. Dental Assoc. 113 stracts and Papers, No. 181 (1967). 78, (1969). [22] Vickery, R. Badinelli, L. A., [9] McLean, J. W., A higher strength porcelain for C, and Waltke, R. W., crown and bridge work, Brit. Dental J. 119, 268, The direct fabrication of restorations without (1965). foil on a refractory die. J. Prosthetic Dentistry, [10] Harman, C. G., and Wiener, I., Sintered glass-basis 21, 227 (1969). composites for iwrcelain prosthesis. lADR Pro- [23] Yardley, R. M., Multi-unit aluminous porcelain gram of Abstracts and Papers, No. 339 (1969). bridge work. Brit. Dental J. 126, 177 (1969).

; ; —

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseaech, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Dental Silicate Cements

A. D. Wilson

Laboratory of the Government Chemist, London S.E. 1, England

A systematic search for improvement of dental silicate cements requires detailed knowledge of their formation and structure. Experimental evidence shows the effective bonding medium in these cements is an amorphous aluminum phosphate. Physicochemical examination of a number of dental silicate powders showed them to be powdered fluorine- containing alumino-silicate glasses. The mechanism of the cement-forming reaction was studied for one cement following extraction of soluble ions by water at various time intervals after preparation. Infrared spectroscopy was used to record the changing absorption spectra of the setting cement. The microstructure of a number of fully hardened cements

was studied by a variety of techniques : optical and electron microscopy, electron probe microanalysis infrared spectroscopy and x-ray diffraction. Dissolution in acidic media of aluminum phosphate bonded silicate cements is an inherent defect stemming from the fundamental chemistry of the system.

Key words : Aluminum phosphate, amorphous ; dental materials ; dental silicate cements

electron probe microanalysis ; glass, alumino-silicate ; silica gel ; silicate cements, dental

silicate cements, mechanism of hardening, microstructure of, chemical nature ; silicate cements, susceptibility to acid attack.

1. Introduction is capable of significant improvement. This question can only be answered with certainty from In this paper the formation and structure of the knowledge of the fimdamental nature of the dental silicate cement is described in the light of setting mechanism and the microstructure. recent fundamental work much of which is as yet Now, although this material has been known for unpublished. This work was carried out in the a considerable time, little fundamental work has U.K. at Government research stations by B. E. been published. Undoubtedly the reason for this Kent, R. J. Mesley (Laboratory of the Govern- is that since the material is apparently amorphous ment Chemist), R. P. Miller, D. Clinton (National the principal structural determining technique Physical Laboratory), and K. E. Fletcher (Build- x-ray diffraction—is not applicable. ing Research Station) in collaboration with the The generally accepted view has been that the author. Results obtained enable the basic limita- dental silicate cement is bonded by a type of silica tions and future prospects of this cement to be dis- gel [3] . This is not an unreasonable hypothesis ; if cussed from a fundamental viewpoint. there is an interaction between an acid and a The dental silicate cement has a long history silicate, silica, often in the form of silica-gel [8] its successful introduction into dentistry followed must result, and indeed there are cements where from the Steinbeck patents of 1903^ [1] ^. It re- the essential bonding medium is known to be mains today a popular material despite the intro- silica-gel [9]. These are the industrial silicate duction of alternative resin systems [2, 3]. What is cements formed by reacting aqueous solutions of

less certain is whether it has a future ; for its draw- soluble silicates with acids, and, unlike dental backs—failure by erosion in the mouth, ability to silicate cements, they are acid-resistant—because irritate pulpal tissues etc.—are well Imown [3]. silica-gel is insoluble in acids. Although some research studies have been re- They are also weak when compared with dental ported [4, 5, 6] there has been as yet, no basic silicate cements. The dental silicate cement is one change in formulation of commercial materials of the strongest inorganic cements known and since 1907 when Schoenbeck [7] introduced the strength values in the region of 3,000 kg/cm^ (300 use of a fluoride flux in the preparation of the MN/m-) in compression have been recorded [10] cement powder. Following this innovation there which are much greater than those reported for the

have been some 60 years of practical development industrial silicate cements : 150-270 kg/cm^ (15-27 which, although it has led to improvements, has MN/m^) [9]. not appreciably changed the charactei- of this ma- There are therefore sound reasons for doubting terial. The question remains whether this cement that silica-gel is the essential bond in dental silicate cements, although, of course, it must be pres- 1 Figures In brackets Indicate the literature references at the end of this paper. ent in the cement.

8S Studies on the eflfect of acids on these materials (table 1) liave shown that these are all of the same have shown that disintegration is accompanied by basic type. The liquids are strong aqueous solu- loss of phosphates from the cements, an observations of phosphoric acid-containing metals. tion which is consistent with the presence of a cementing phosphate bond [11]. The electrical conductivity of the freshly prepared cements has been observed to drop sharply during setting [12] (fig. 1) which suggests that setting is the result E of a type of precipitation process. The increase in pH which has been observed as cements age is consistent with this view [13, 14]. These problems have been largely resolved by recent research where a number of methods of structural analysis were employed.

2. Physico-Chemical Nature of the Cement Materials

Part of the explanation of the formation of the 60 eo dental silicate cement follows from a considera- aqe o[ cement, min tion of the physico-chemical nature of the cement Figure 1. The changes in electrical conductivity a forming materials. Qualitative analysis of 16 mate- of dental silicate cement, 20 °C. (Laboratory of the Govern- rials and quantitative analysis of 4 materials ment Chemist).

Table 1.

Chemical Composition of Dental Silicate Cement Powders (given as percent w/w)

Silicate powder Chemical species A B C D

SiOj 41. 6 38. 8 34. 5 35. 9 AI2O3 28. 2 29. 1 28. 3 29. 0 CaO 8. 8 7. 7 8. 5 6. 1 NajO 7. 7 8. 2 11. 2 14. 5

ZnO . 3 2. 9 . 1 . 3 P2O5 3. 3 3. 0 3. 3 4. 4 F 13. 3 13. 8 18. 1 15. 2 H2Of550°C) 2. 2 1. 6 1. 7 1. 4

MgO . 1 . 1 . 1 . 1

SrO . 2

105. 7 105. 2 105. 7 106. 8 Less 0 for F _ 5. 6 5. 8 7. 6 6. 4

Total 100. 1 99. 4 98. 1 100. 4

Median particle size Cmicrometers) 8. 6 10. 6 11. 5 15. 5

Chemical Composition of the Liquids (given as percent w/w)

Chemical species A B C D

PO4'- 47. 3 47. 8 53. 8 63. 8 A15+ 1. 6 1. 9 2. 0 1. 8 Zn2+_. 6. 1 4. 2 9. 1 Mg2+ 1. 3

86 .

The powders are unique amongst cements in that I4O0 I2O0 lOOO 900 eOOcm'' 1 1 1 1 1 they are ground glassy bodies and not crystalline clinkers, a physical state which may account for the translucent nature of the final cement. The dental silicate glass is prepared by fusing a silica and alumina mixture in a fluoride flux containing minor amounts of phosphate. The fluoride flux is lowers / based on cryolite ; the addition of fluoride n/ \ its fusion point. *. c they are silicates, but J / \ The glasses are unusual ; unlike most, readily dissolve in mineral acids. \ I They are also opal glasses of extreme composition If n where the fluoride content is higher than that of any other opal glass. This opacity is caused by the phase separation of two types of minute droplets

1 X-^'i 1 1 1 1 approximately 4000 A and 250 A in diameter 7 8 9 lO U 12 microns which can be observed under the electron micro- Figure 2. Changes in scope. This phase separation is to be expected as the infrared spectrums of a dental silicate cement as it sets. {Laboratory the Govern- glasses the of large amounts of fluorine in weaken ment Chemist). continuity of the 3-dimensional network. Curve I. Freshly prepared cement. The other feature of interest is that these glasses Curve II. Set cement. are aluminosilicates with a mole ratio of Si : Al Absorption bands (a) slUca-alumlnophosphate gel unity. structure of this type of (b) sUica gel approximating The (c) HsPOi Ion aluminosilicate plays an important role in cement formation and requires further discussion. the aqueous phase of the cement, and be rendered A simple silicate is a macromolecule based on a continuous series of Si-0 bonds, and structurally water soluble. consists of chains or 3-dimensional networks of The course of this process is illustrated by results that have been obtained simple wet- linked [SiOJ tetrahedra. Because of its atomic from dimensions aluminum can replace up to half of the chemical experiments. Cements of various ages were silicon atoms in a network which for valency rea- crushed and the concentration of water sons acquires a negative charge. These are the soluble ions determined by leaching with water. so-called aluminosilicates. In dental silicate glasses Cements were prepared using a powder-liquid ratio of 4g/ml. The chemical composition of the where the Si : Al approximates to 1 : 1 there are many -[Si04]-[A104] -links; the negative charge powder was : 18.7 percent Si, 15.0 percent Al, 6.3 on the network being balanced by the positive percent Ca, 5.7 percent Na, 13.3 percent F, 1.0 per- charge on the glass cations, calcium and sodium. cent P, 40.0 percent O, and that of the liquid 49 percent H3PO4, 1.25 percent Al, 6.5 percent Zn. Results are shown as a plot of the concentration of 3. Cement Formation leachable ions against the age of the cement (fig. 3) The dental silicate glass can be regarded as a Now it is evident from these data that decomposi- negatively charged polymer surrounded by a cloud tion and liberation of glass ions (AP^^, Ca^"^, Na"^, of cations. This type of structure is susceptible to F") from the powder is rapid. By the time mixing attack by acids, since the positively charged pro- is completed (1 min) considerable amounts of tons can easily penetrate into the negatively sodium, calcium, and fluoride have been rendered charged glass network, a type of acid-base reac- water soluble, that is they have been liberated from tion. The principal sites of attack will be at the the powder as a result of its decomposition. aluminum atoms which attract most of the nega- Indeed at this point, these levels are at a maxi- tive charge; when there are a large number of mum for calcium and fluoride because subsequently Al-O-Si links, complete disruption of the network these ions are removed by precipitation and inter- will result. This is exactly the situation that oc- pretation of results is obscured. For this reason the curs when the dental silicate powder is mixed with initial decomposition process is best illustrated by the cement liquid, when infrared spectroscopy the curve for water extractable sodium as sodium shows (fig. 2) that the alumino-silicate network is forms no water insoluble product. By the time rapidly degraded to a silica or aluminosilica gel, mixing has been completed some 10 percent of the and that HoPOl ions are formed as protons are sodium contained in the powder has been liber- lost from the liquid ated the increasing to a maximum of 20 [15]. ; amount Now this process of decomposition will be accom- percent in the following 30 min or so. If this figure panied by the liberation of ions locked previously corresponds to the amount of powder decomposed in the glass lattice. As a result these ions would be then the powder is in five-fold excess over the expected to migrate into the cement liquid, that is liquid.

87 unlikely that it will be remedied at all for the present type of dental silicate cement. Even when the cement is fully hardened some

soluble material remains ; the salts of sodium with dihydrogen phosphate and aluminofluorides. This is soluble material that is eluted from cements in the specification tests for solubility and distinte- gration. What relationship it bears to the clinical durability of a silicate restoration is uncertain, for clinical durability will be related to the resistance of the insoluble part of the matrix towards oral fluids. It should be noted that the setting mechanism of the dental silicate cement is totally different from that of the Portland cement. In the latter, interaction between powder and liquid is slow and is accompanied by formation of the cementing matrix. The situation for the dental silicate cement is totally different since setting and hardening are controlled by increases in pH. The interaction between powder and liquid is very rapid and most of it occurs in the initial stages, during and shortly after mixing. This process is accompanied by a relatively small change in the nominal pH of the system which only increases from an initial figure aqe of cement hours of 0.8 to 1.7 at the initial set [14]. Subsequent the next 48 is caused a Figure 3. Variation of the concentration of water soluble hardening during hr by ions contained in a dental silicate cement as it set and much larger change in pH, an increase to 5.0-5.5, hardened, illustrating the liberation of ions from the which completes the precipitation process. Little glass matrix and their subsequent precipitations. (Lab- further interaction between powder and liquid is oratory of the Government Chemist). required to produce this pH change ; its magnitude The accumulation of ions in the aqueous phase is so small that it cannot be detected by present methods. This behavior follows from a considera- of the cement paste is accompanied by a decrease in tion of the nature of the neutralization curve of its acidity, for this is an acid-base reaction. As a

phosphoric acid ; this system has a strong buffering result AP% Ca-''*, Zn^^ F", and the H2PO4- progres- capacity in the region of its pKa's but sively form insoluble substances causing the cement (2.1, 7.2) little in the intermediate regions. to set. This is a type of precipitation process and its correlation with setting is shown in figure 3. It is best illustrated by the curves for zinc and 4. Microstructure phosphate, since if the small amount of phosphate The set dental silicate cement has unusual and in the powder is neglected, these are present interesting features. The best examples are initially in a soluble ionic form. There is no prior amongst the strongest inorganic cements known extraction process to obscure interpretation of and are unique in being substantially amorphous. results. Inspection of this data showed that the X-ray diffraction of seven examples showed that sets in 6 min when 50 percent of both of cement they contained only minor amounts of crystallites, subsequent these species have precipitated. The which were however well-defined. These were iden- post-set hardening is due to the continuation of tified as augelite, AIPO4 A1(0H)3 and fluorite, this process and is complete in 48 hr, for the system CaF2. described. The appearance of polished surfaces of dental This mechanism explains why dental silicate silicate cements under incident light is that of a cements are adversely affected if exposed to mois- poorly connected mosaic of highly reflecting angu- ture shortly after set and before fully hardened. lar grains, as a micrograph of a typical example Ions which would have contributed to the forma- shows (fig. 4). The grains, which appear un- tion of the matrix are removed by leaching [10] attacked, vary in size over a broad range, 1-100 and optimum strength can never be obtained. Any fim in diameter. Porosity shows as black non- in cement which is dependent on a precipitation set reflecting areas, which are more numerous will be sensitive to moisture in this way. Unfor- weaker cements. Under cross-polarizers and re- tunately, most rapidly setting cements, and these flected light a network of birefringence is apparent are required for dentistry, are based on this type in cements with strengths in compression in excess of mechanism. So the problem of the elimination of 1,750 kg/cm2 (170 MN/m^) (fig. 5). This evi- of this drawback will not be solved easily. It is dence taken together with the amorphous nature 88 . . . .

Figure 4. Micrograph of a polished surface of a dental Figure 6. Scanning electron micrograph of a strong silicate cement, taken by transmitted light, showing cement. (National Physical Laboratory) glass grains and porosity. (National Physical Labora- tory).

Figure 7. Scanning electron micrograph of a weak cement. (National Physical Laboratory). Figure 5. Network of birefringence shown by reflected light analysed by crossed polarizers. (National Physical Laboratory) the bonds between particles and matrix are com- paratively weak. Attempts have been to improve the strong- of the bulk of the c«ment matrix and the isotropic made est properties of cements by the inclusion of fibers nature of the crystallites indicates that this bire- and other forms of inert aggregate. In our labora- fringence is the result of orientation due to inter- tories it has been found that these are ineffective. nal stresses. A micrograph of a fiber-containing cement shows The scanning electron microscope with its high that the matrix has shrunk from the fiber with magnification coupled to a considerable depth of crack initiation (fig. 8) focus is a powerful tool for determining the physical microstructure of materials. Fractured surfaces 5. Microchemical Nature of the Matrix are normally chosen for examination. A micrograph of a typical strong cement is shown (fig. 6). Electron-probe microanalysis is an important The dominant feature is a glass particle about 10 technique for determining the chemical nature of fim across, which by its angular appearance is the microstructure of amorphous materials. In this apparently unattacked. Inhomogeneities on its technique a narrow electron beam, some 2 nm surface are indicative of local attack, these also across, scans a flat polished section and records the serving to distinguish it from the surrounding chemical composition of each point on the cement matrix. The matrix which appears particulate in surface. This enables the distribution of elements strong cements is plate-like in weaker ones (fig. 7) to be mapped as white points on a black The appearance of the fractured surfaces suggests background.

89 . ;

The association of the various cations and anions in the matrix has, as yet, not been resolved completely. The detection of minor amounts of crystallites, augelite, A1P04A1(0H)3, and fluorite, CaFa, in the matrix may be indicative of the association of ions in the predominant amorphous phase of the matrix. Infrared spectroscopic data are consistent with the presence of amorphous aluminum phosphate which must be considered as the major constituent of the matrix. Aluminum phosphate is known in ceramic science as a binding agent [16] and must be regarded as the effective bonding medium. The exact nature of the polymer remains a subject for speculation. Results from other studies indicate that the function of the powder may only be to supply ions needed for the reaction at an

FiGTJEE 8. Scanning electron micrograph of a cement appropriate rate. Cement strength is far more containing glass fliers. {National Physical Laboratory) dependent on the phosphoric acid concentration of the liquid than the chemical composition of the powder (fig. There are therefore good reasons In the case of dental silicate cements the element 10). for supposing that structure the phosphoric distribution can be compared with the physical the of liquid is important, for it is dependent acid positions of the particles in the matrix, as shown acid on concentration is that by the back scattering of electrons, thus enabling [17]. It known aluminmn a in acid solution the chemical composition of both particles and forms complex with phosphate as proportion of increases, matrix to be established. and that the aluminum dimers and polymers are formed where both hy- An electron-image photograph and a set of ele- the ment-distribution photographs of the same area of droxyl and phosphate act as bridges between aluminum atoms . Therefore there are grounds study for a dental silicate cement are shown in [18] for supposing that the setting of the cement pro- figure 9. The cement used was similar to that em- ceeds the progressive formation of an alumi- ployed for the reaction studies, metals being with num-oxygen-phosphorous polymer as aluminum omitted from the liquid to avoid confusing experi- enters the liquid from the powder. It is plausible mental data. From an examination of these it is to suggest that a pH-dependent polymerization apparent that phosphorus is almost entirely in occurs in the cement to give a water- insoluble poly- the matrix. Silicon is associated with the particles, nuclear species, which is the effective binding and there is none in the matrix. medium, and is based on phosphate and hydroxjrl Calcium and aluminum are present in both regions. Since originally both were present solely in between aluminum ions. The structure of the dental silicate cement is the glassy particles, it is clear that there has been simple than might have been supposed from some migration into the aqueous or matrix phase. more combinations possible of elements No data are shown on the diagram for sodium and the many present. For spatial reasons a combination of sili- fluorine because of experimental limitations ; how- is not possible. Moreover, ever more recent experimental results using im- con and phosphorus if it were, a combination is unlikely; proved techniques indicate that some sodium and even such the rare Si-O-P bond is not of a strength com- fluoride also migrate from the powder grains into parable with the Si-O-Si, Al-O-Si and Al-O-P the liquid. bond, and stable structures do not result [1^]. Detailed electron scans show that in fact the Another possibility is the metal alumino phos- glass cations are displaced from the surface region phate of Barrer and Marshall [18] (M20)n_m of the individual particles, leaving a central un- ( ALOa ) „ ( P2O5 ) m • xH^O. However, such bodies, at attacked core in the case of the larger particles. least in crystalline form, have not been identified This displacement of cations from the glassy possibly the affinity of calcium for fluoride in- particles is a result of the penetration of protons hibits their formation. The indications are that the from the liquid which means that water will have matrix is a simple mixture of a basic aluminum entered the surface layer of each particle. Since phosphate and calcium fluoride, a little of which silicon remains in this degraded surface layer it has crystallized. may be inferred that it is present as a type of This structure of the dental silicate cement can silica gel. Infrared data confirm this deduction. It be used to predict the clinical behavior of a res- is the retention of this siliceous framework which toration. The effective binding medium, basic preserves the geometry of the glassy particles aluminum phosphate, is insoluble in neutral solu- thus giving them the appearance of remaining tion, but dissolves in acidic and in neutral citrate unattacked. media. A silicate restoration would, therefore, be

90 Figure 9. Electron probe microanalysis showing physical microstructure and element distribution {white spots ona Mack background). (Building Research Station).

(a) Electron image micrograph, (b) Silicon distribution, (c ) Phosphorus distribution, (d) Calcium distribution, (e) Aluminum distribution.

452-525 O—72 7 91 .

undergo a type of precipitation process as the pH increases and the cement sets. The effective binding agent is amorphous aluminum phosphate; the matrix also contains calcium fluoride. There is some crystallization of these compounds. Since these substances are acid-soluble this type of cement can never be designed to withstand adverse oral conditions. The hardened cement con- • tains some soluble material, the sodium salts of dihydrogen phosphate and alumino fluorides. The specification test only determines these soluble constituents and therefore does not necessarily give an indication of the durability of the water- insoluble phase of the matrix under oral conditions.

The author thanks Mr. J. K. Foreman, Super- intendent, Research Division, for his constructive comments and continuing support and Dr. D. T. 30 'O SO 60 Lewis, C.B., the Government Chemist, for per- concentration of H^PQ, % mission to publish.

FiGUEE 10. The effect of acid concentration on Crown Copyright. Reproduced by permission the compressive strength of a cement. of the Controller of Her Britannic Majesty's To convert psi (pounds per square inch) to N/m- Stationery Office. (newtons per square meter) multiply by 6,895.

7. References expected to give good service under favorable oral conditions. Saliva itself acts as a near neutral [1] Steenbock, P., German Pat., 174557 (1903) buffer, and so a silicate restoration washed by Brit. Pat. 15,176; 15,181 (1904). [2] Buonocore, M. G., Int. Dental J. 18, 406 (1968). saliva should have a long life. However if sited in [3] Skinner, E. W., and Phillips, R. W., The Science of regions where local acidic condition can develop Dental Materials, 6th Edition, p. 60 (W. B. or if the saliva contains appreciable citrate, early Saunders, Philadelphia and London 1967). failure is to be expected. This argument may be [4] Wright, J. W., J. Dental Res. 1, 35 (1919). [5] Manly, R. S., Baker, C. F., Miller, P. N., and Welch, extended to any dental silicate cement as at present F. F., J. Dental Res. 30, 145 (1951). proportions of formulated. Whatever variation of [6] Rockett, T. J., I.A.D.R. Abstracts #433 (1968). the basic constituents of both powder and liquid, [7] Schoenbeck, F., U.S. Pat. 897,160 (1907). it is difficult to see how the present type of dental [8] Murata, K. J., Bull. U.S. Geol. Surv., 950, 25 (1946). Vail, J. G., Soluble Silicates Their Properties and silicate cement can be further developed, for its [9] Uses, Vol. 2 (Reinhold Publishing Corporation, faults are inherent. It would seem therefore that New York, 1952). future research for an improved translucent [10] Wilson, A. D., and Batchelor, R. F., J. Dental Res. cement will have to follow radically different lines. 46, 1425 (1967). Wilson, A. D., and Batchelor, R. F., J. Dental Rea. If the cement is to be of the precipitation-set type [11] 47, 115 (1968). for rapid setting then it will have to be based on [12] Wilson, A. D., and Kent, B. E., J. Dental Res. 47, a more acid resistant binder than aluminum 463 (1968). phosphate. [13] Norman, R. D., Swartz, M. L., and Phillips, R. W., J. Dental Res. 45, 136 (1966). 6. Summary [14] Kent, B. E., and Wilson, A. D., J. Dental Res. 48, 412 (1969). On mixing the dental silicate powder and liquid, [15] Wilson, A. D., and Mesley, R. J., J. Dental Res. 47, protons from the liquid penetrate the surface 644 (1968). Kingery, W. D., J. Amer. Ceramic Soc. 33, 242 layer of the glassy particles. The negatively [16] (1950). silicate is charged alumino network disrupted and [17] Van Wazer, J. R., Phosphorus and its Compounds, glass ions migrate into the aqueous phase. Silicon Vol. 1, Chemistry (Interscience Publishers, New remains in the surface as a type of silica gel thus York-London, 1958). preserving the initial shape of each particle. With [18] Salmon, J. E., and Wall, J. G. L., J. Chem. Soc. 1128 (1958). the exception of Na^, ions which have accumulated [19] Barrer, R. M., and Marshall, D. J., J. Chem. Soc, in the aqueous phase (AP"", Ca^"", F", H2P04') 6616 (1965).

92 NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Composite Restorative Materials ^

R. L. Bowen,* J. A. Barton, Jr.,** and A. L. Mullineaux***

Dental Research Section, Institute for Materials Research, National Bureau of Standards, Washington, D.C. 20234

This article describes the development of dental composite materials, which have quickly established a place in the practice of dentistry because of the shortcomings of alternative esthetic direct filling materials. The goal-directed basic research in which the composite materials were developed was initiated at the National Bureau of Standards in 1956. Late in 1957, a monomer was synthesized that is suitable for use with reinforcing fillers. Commercial dental composite materials available in 1969 use this resin together particulate fillers they have lower polymerization shrinkage with and various additives ; and coefiicient of thermal expansion, and higher compressive strength and stiffness relative to unreinforced resins. Compared to silicate cements, the composites have lower solubility, higher tensile strength, and comparable compressive strength. Composite restorative materials, when they have been fully developed, should provide the dental profession and the public with greatly improved restorations for anterior teeth.

Key words : Biomaterials composites dentistry ; fillers glass methacrylates ; monomers ; ; ; ; ;

prostheses ; resins ; restorative material.

1. Historical Introduction from the cavity walls [5], allowed the exchange of oral fluids around their margins [6] and were Although composite dental restorative materials associated with pulpal inflammation [7-9] and are yet in their infancy, the roots of their develop- a high incidence of recurrent decay [10-11]. ment can be traced back through the scientific lit- Furthermore, discrimination between decalcified erature for many years. In 1843 a German chemist, dentin and the unreinforced direct filling resin Joseph Redtenbacher, reported a new acid to was difficult because they were both radiolucent. which he gave the name acrylic acid [1].^ In the For these and other reasons, a great many practi- last part of the century, methacrylate esters and tioners returned to the use of silicate cements [3]. their polymers were discovered [1]. Probably be- The shortcomings as well as the merits of sili- fore 1940, Schnebel had found that tertiary amines cate cements are well known. After nearly a hun- (such as Michler's ketone) would activate benzoyl dred years [12] of development and use, these es- peroxide to give self-hardening resins [2]. So far thetic materials with good thermal and anticario- as can be determined, the first self-curing methyl genic properties still have distinct disadvantages methacrylate resin formulation for filling teeth [3]. Most obvious is the susceptibility to disinte- was introduced in the United States in 1948. gration in a decay-like manner [13-15], typically There soon appeared on the market a number after about four years in the mouth [3, 14]. Fur- of direct filling resins, and many believed that ther description is given by Wilson elsewhere in these materials would be the answer to the short- this publictaion. comings of the decades-old silicate cement. How- Because of the persistent need for a permanent, ever, this idea was short-lived [3]. These resin esthetic direct filling material, there were contin- resins. restorations discolored [4], tended to shrink away ued efforts to improve the methacrylate An aluminum silicate filler (about 15 percent) was

1 This research activit.v was supported in part by research part of one methyl methacrylate monomer-poly- grants to the American Dental Association from the Na- tional Institute of Dental Research and is part of the dental mer system [16]. The use of fillers with the proper research program conducted by the National Bureau of standards, In cooperation with the Council on Dental index of refraction and a low thermal expansion, Research of the American Dental Association ; the Dental and the development of adhesion in the resin were Research Division of the United States Army Medical Research and Development Command ; the Dental Sciences advocated [17]. Indeed, when inert fillers were in- Division of the School of Aerospace Medicine, USAF ; the National Institute of (methyl methacrylate) , there Dental Research ; and the Veterans corporated into poly Administration. Associate Director, American Dental Association Research was a reduction in the coefficient of thermal expan- Unit at the National Bureau of Standards. sion and in the water sorption in proportion to the **Guest Worker, Dental Research Section, National Bureau of Standards, Lieutenant Colonel, U.S. Air Force. concentration of fillers [18]. However, there were •Research Associate, American Dental Association Research Unit at the National Bureau of Standards. limited adhesive characteristics in these and cer- 2 Figures in brackets indicate the literature references at the end of this paper. tain other resins investigated [19].

93 Epoxy resins [20], developed early in the pres- annual production rate of ent century, had an Ha C=CC0CH2 -C-CHs o/(^\c-/()VoCH2 -C-CH2 OCC^CHs about 20 million pounds by 1954. Certain formula- H "^c^^^ H CH3 tions of these resins had intriguing properties in- H3 cluding adhesive characteristics and the ability to harden at moderate temperatures with little BIS-GMA shrinkage. These led numerous investigators to study epoxy resins as a binder for inorganic fillers 0 [19, 21]. Primarily with indirect techniques, com- HsC^CCOCHs posite restorations were prepared using powdered CH3 fused silica and porcelain, bonded together with MMA minor quantities of (heat-cured) epoxy resins. The good esthetics and other favorable properties of FiGUKE 1. Structural formula of BIS-GMA. these composites encouraged the further investiga- The asterisks indicate asymmetric carbon atoms that give rise tion of this approach to dental restorative to a number of steroisomers. Methyl methacrylate (MMA is shown for comparison [37]. materials [21]. However, attempts to develop composite materials using epoxy resins with various hardening corporated into the resin could do this [18]. Ex- agents for use in direct techniques met with nu- perimental composites prepared in 1953 utilized merous difficulties. Although epoxy resins were particles of fused silicon dioxide as a reinforcing initially adhesive to hard tooth tissues, the filler because of the extremely low coefficient of strengths of bonds to enamel or dentin after pro- thermal expansion of vitreous silica [21]. With longed exposure to water were disappointing the epoxy resin system, there was relatively good [19, 21, 22]. adhesion between the resin and the surfaces of these particles. 2. Early Composite Developments Later in the investigation, after the more adhesive epoxy resins had been replaced by the less The goal-directed research in which the com- adhesive dimethacrylate resins (BIS-GMA) [26], posite dental restorative materials were developed it became apparent that special means for attain- was initiated at the National Bureau of Standards ing adhesive bonding between the resin matrix in 1956. After unsuccessful attempts to use various and the filler particle surfaces were required. At epoxy resins and hardening systems, a compromise that time vinyl silane coupling agents appeared to between epoxy and methacrylate resins was con- be the most suitable means of improving this bond- ceived late in 1957 [23, 24]. The reaction sites ing [30, 31]. The more reactive y-methacryloxy- (oxirane rings) of the epoxy molecule were re- propyl silane coupling agents [32, 33] had not then placed by methacrylate groups. This gave a hybrid become commercially available. A vinyl silane molecule that polymerized through methacrylate coupling agent applied to reinforcing filler gave groups [25]. It was suitable for use as a binder for composites that had about four times the tensile reinforcing fillers because it was nonvolatile, had strength of composites utilizing the same filler a relatively low polymerization shrinkage, and without its being treated with the vinyl silane hardened rapidly under oral conditions when suit- [34]. When the methacryloxypropyl silanes be- ably formulated with an appropriate initiator came available, their ability to promote adhesive system. bonding between the resin and glass was compared This dimethacrylate monomer (BIS-GMA) was with that of vinyl silanes ; the methacryloxypropyl bisphenol synthesized by the reaction of A and silanes gave significantly stronger and hydrolyti- glycidyl methacrylate The same product is [26]. cally more-stable bonding than did the vinyl also produced by the reaction of the diglycidyl silanes [35-37]. ether of bisphenol A and methacrylic acid [25]. Since the strength and durability of the com- Its structural formula [27] is shown in figure 1. posites depended on the quality of the bonding be- Although BIS-GMA has been referred to as an tween the organic matrix and the reinforcing filler epoxy resin, this is not the case ; the original epoxy (oxirane) groups disappear during the synthesis particles, it was of utmost importance that the best and are, in effect, replaced by methacrylate groups. coupling agents be used in the optimal manner. Unfortunately, this dimethacrylate monomer has The coupling agent most widely used was y-metha- also been called a polyester. The well-established cryloxypropyltrimethoxy silane, also known as 3- term "polyester" should be restricted to the poly- (trimethoxysilyl) propyl methacrylate. This or- condensation products essentially of dicarboxylic ganofunctional silane was hydrolyzed and con- acids with dihydroxy alcohols [28]. densed on the surfaces of the filler particles. It Since the coefficients of thermal expansion of made the particles water re]Dellent, reducing the organic polymers are many times higher than that water sorption of the composites; it also formed of the dental hard tooth tissues [6, 29], means for chemical bonds at the surface of the particles, con- lowering this value were needed. Fillers [17] in- necting the organic polymer and the inorganic

94 filler [38], converting inert fillers to reinforcing erator than DMPT when compared on an equi- fillers [39]. molar basis [43]. Although fused quartz or fused silica received In a composite material developed in Great Bri- a great deal of attention in experimental work [14, tain [44], a long-chain mercaptan, together with 21, 23, 24, 26, 34], many other filler materials were an aliphatic peroxide, served as the hardening also investigated [16, 18, 19, 21, 31-34]. These in- initiators. cluded the synthetic mineral y8-eucryptite and Although other peroxides have been used [44], closely related materials [35, 40, 41], microcrystal- benzoyl peroxide has been the one most used in line glasses [42], aluminosilicate glasses, bariuni- composite restorative materials. It has been dis- titanium glass, crystalline quartz and pyrogenic persed either in the reinforcing filler with the aid silica [36]. It has been widely held (although not of a solvent that is removed by evaporation, or is universally accepted) that condensation between dissolved in one of the monomeric liquids of the SiOH groups on the surfaces of these silica-con- composite formulation. taining fillers and SiOH groups of the hydrolyzed Sulfinic acids or their derivatives were used in silane accounted for the bonding between the cou- the initiator systems for experimental composite pling agent of the filler and the filler particle it- materials [14]. self. Copolymerization of the methacrylate groups of the silane with the methacrylate groups of the 3. Some Properties of Experimental and monomers comprising the organic continuous Commercial Composites phase has been thought to complete the chemical linkage between the phases of the composites [38]. Before proceeding with recent and current de- The BIS-GMA monomer was too viscous to be velopments, perhaps it would be well to define readily mixed with these fillers without first being "composite." For this discussion, the term "com- thinned with some suitable monomer having a posite restorative material" refers to a man-made, lower viscosity. Among others, methyl methacry- three dimensional combination of at least two late and ethylene and tetraethylene glycol di- chemically different materials with a distinct in- been used for this purpose methacrylate, have terface separating the components [39], proper- [23, 24]. ties are thus obtained for the restoration of the As with antecedent methacrylates [16], stabiliz- form and function of defective teeth which could ers such as hydroquinone were first used to inhibit not be achieved by any of the components acting premature polymerization and give the necessary alone. Thus, the use of the term "composite" dis- storage stability (shelf life) to the BIS-GMA tinguishes the combination of inorganic aggregates formulations [23, 26]. On polymerization, less dis- bonded together with organic polymers from un- coloration occurred with the use of the mono- reinforced direct filling resins and from silicate methyl ether of hydroquinone. Later, hydroqui- cements. none was replaced by BHT (butylated hydroxytol- It is primarily the high proportion of reinforc- uene; 2,6-di-i5er^-butyl-4-methyiphenol), a stabil- ing filler in composite restorative materials that izer that appeared to contribute no discoloring yields properties different from those of direct effects [43]. filling resins. Most of the physical properties of Other additives that contributed to color stabil- composite restorative materials have been im- ity of the restorations were ultraviolet-absorbing proved by the incorporation of a maximum amount compounds. While these may have contributed to of reinforcing filler together with a minimum the stability of the monomers during storage, their amount of a binder of cross-linking polymerizable primary purpose was to reduce the amount of organic resin. discoloration in the resin of the final composite The "first generation" of commercial dental material. Typical UV absorbei-s were 2-hydroxy- composite materials used BIS-GMA resin binder, 4-methoxy-benzophenone [43] and 2- (2'-hydroxy- described previously, together with major portions 5'-methylphenyl ) benzotriazole. of various particulate fillers and minor portions of Since most of the experimental work on compos- various additives. Data that have been reported ite materials utilized an amine-peroxide initiat- in the literature [45-60] and more recently ob- ing system, another important component has been tained [61, 62] show that composites have lower the amine accelerator. The most commonly used polymerization shrinkages and coefficients of ther- of these was N,N-dimethyl-;?-toluidine (DMPT). mal expansion, and higher compressive strengths This compound interacted with peroxides to pro- and stiffnesses relative to unreinforced resins. duce free radicals which brought about the poly- Compared to silicate cements, the composites have merization of the methacrylate monomer groups lower solubilities and higher tensile strengths. and thus the hardening of the resin binder of the More detailed data are given in tables 1 and 2, composite. Investigations of factors that produce including comparisons with dentin and enamel discoloration in these materials led to the introduc- [24, 63-68]. The quantitative information pre- tion of N,N-dimethyl-sym-m-xylidine (DMSX; sented in the tables is incomplete since the meth- also known as DMDA) , which not only gave less ods and conditions of measurement are not de- discoloration but also was a more effective accel- scribed; however, these data will serve to illus-

95 trate the ranges of values and order of magnitude generation" of composite materials. For example, to be expected with these materials. the BIS-GMA is not completely color stable and The smoothness of the finished surface is com- is too viscous for use without being thinned with parable with that of silicate cements [36, 57, 58] a monomer of lower viscosity. Furthermore, it can- but inferior to that of unreinforced resins [52] not be purified by distillation or by crystallization and most of the other restorative materials. The since it is inherently a mixture of non-volatile opti- abrasion resistance of composite materials is su- cal isomers. To overcome these problems, three perior to direct filling resins [36, 47, 48, 51, 56, 58]. dimethacrylate monomers were synthesized that Similar to some other dental restorative ma- can be prepared and purified separately by re- terials, problems encountered in packaging and crystallization [27]. These are the condensation storage include volatility of some ingredients and products of 2-hydroxyethyl methacrylate with premature gelation of resin components [36]. the acid chlorides of phthalic, isophthalic and terephthalic acids, respectively. Upon combining 4. Current Research these three crystalline products, a ternary eutectic liquid mixture is formed that has a viscosity low Current research has the goal of making im- enough to be used without the addition of any provements on this "first generation" of composite volatile ingredients. Preliminary tests show it to materials now available to the dental practitioner. be non-toxic and non-irritating [27]. For example, there is a degree of pulpal irritation Numerous other monomers [71-75] are cur- caused by some of these materials [69, 70], and it rently being synthesized and evaluated for use in has not yet been established which of the various dental composite materials. ingredients contribute most to the inflammatory The research effort to improve polymerization reactions of the dental pulp tissues. Other prob- accelerators [76] has led to the successful syn- lems that have been encountered are incomplete thesis of aromatic amines having unusually high color stability [62] and resistance to staining, molecular weights [77] so as to minimize their difficulties encountered in finishing and polishing, solubilities in tissue fluids and thus their freedom and lack of x-ray opacity. to diffuse into the pulp or other tissues. The struc- The recent investigations at the National Bu- ture of these amines is based partly on the fact reau of Standards have led to new dimethacrylate that the color stabilities of the composites have monomers, reinforcing fillers, accelerators, stabiliz- correlated better with the nature of the ring sub- ers, and adhesion-promoting coupling agents stituents than with the kind of nitrogen substitu- that will, hopefully, lead to an improved "second ents in the amine accelerators [76, 77].

Table 1. Some physical properties of esthetic direct-filling materials

Hardening time Polymerization Water sorption Solubility and shrinkage disintegration

minutes % (by volume) % (by weight) % (by weight) 4 [14] 3-8 [24, 26] Experimental composites 5-7 [34] 2 [23] 0. 9 [26] 0. 00-0. 08 [23] 7-10 [41] 2. 7-2. 8 [26] 0. 30-0. 35 [62] 0. 04 [26] 5 [61] 0. 15-0. 22 [61] 3. 5-5 [62]

3-4 [50] Commercial composites 1. 5-4. 5 [52] 0. 78-2. 46 [53] 1-1. 5 [55] 1. 2-1. 5 [52] 1. 60-3. 68 [56] 0. 01-0. 08 [53] 2. 5-5 [59] 0. 8-1. 6 [57] 0. 12-0. 16 [621 3-3. 5 [61] 1. 12 [62]

2 [5] Silicate cements 4-6 [14] 3. 3 [26] 0. 4-1. 1 [14] 4 [34] 1. 3 [60] 0. 4-1. 3 [26]

6 [14] 6 [34] 6-8 [5] b Unreinforced resin 3-3. 5 [50] 6. 2 [26] » (1. 0-1. 5 [5]) (0. 10 [26]) 4 [52] 5. 2 [52] (2. 1 [26]) 0. 1 [60] 2-4 [59] 5-6 [60]

" Volumetric expansion Self-curing denture base resins.

96 1 < ,'

u Tt* T-H (N (N 00 CD CO u > a O «0 (M o o . III I I tn IM 00 CO CO O to lO to O CO-* 00 i> to 00 00 res ve o a O) CI o d

tati 00 ,nd ^ IM CO 03 .CO, lO CD a 03 CO .CD 0) d CO CD l> O CD to to O CO lO d T3 05 «0 ^ IM iM I I ^ 1— d I I J. I lO 00 o o lO o O lO CO lO to 00 lO >o >o to to lo (M 1-1 05 CO

c3 s CO IB CO to f-l 05 (M CO to 05 (M t> (M IM i-H IM (M CO .-^.tP to lO »0 lO lO lO to to T—I lO to OOOQOOO OOOUOO O oo OO O ^OOOOOOO o o o o o o o o o o o o to o o o O 00 05 00 05 >0 O O lO O lO 00 O lO oI ^ to >o OO Tt< to lO lO CO 00

I I I I I I I I I I I I I J i I o o CO O lO rH lO O O O o o »o S"CO CO CO IM

I— TjH r- d to rt* (M to 05 O to I »C O >0 to O (M 00 IM (M '3 I I I I I I I I 04 t> o o ^ 1-1 CO IM Ol IM (M CO CO Oo 00

o.— o CO to to to to to to— to CO += IM IC 1-1 '— I ir—itN- to to IM

I I I I I * to IM 00 lO CO 1-1 rt* (N CO to 00 00 4l IM 1-H 00 I> . . IM CO (M (M

T-H I— T-I CO (N' CO . . . . iM' im' (M i C> o o O OOOO

TjH lO --I IM (M CO l> OOIM to lO to Tt< to to lO lO lO to 1—1 CO to to tOtOCD, d CO o o o o o o o o o o o o o o o to o o o o o o o o o o o o o o o o IM |> to IM O O o 1 lO 05 t--""co'"oriM"lM'~Co"tO TtT^jTco^io" co" cTi-T orco'"co'~ o (M IM rH 1-1 (M IM 1-1 M 1-1 1-1 (N 01 IM

,d i-H -& 1—I lO IM O IM Ol CO TfH to d £1 O OOOO o o o oO ^ r—it:^ i> o 05 05 ^to _ _ . _ lO (M T*( |> to CO lO .to.co. £i 21

I I I 03 oI o oI oI oI oI oI o o O O o o o o o a oooooo o o o CO o o o o o o o o O 00 o * 03 00 CO 1-1

co"'*"cf io'cd'co'" IM 1 I IM co^co'co"

o a o 6 a o s CO u o a> u d

ID d (a u 0} t-l s

Color stability is also profoundly affected by One of the coupling agents, NPG-GMA (the the characteristics of the stabilizers that are used. reaction product of N-phenylglycine and glycidyl Significant advances are being made in the syn- methacrylate) , has two functions in the same mole- thesis and utilization of stabilizers that are non- cule. One function is that of a chelate group that toxic and that do not cause discolorization of can form multiple bonds with calcium in the tooth composites. [43, 76, 77]. surface. The other function is that of a methacry- In the instance Avhere there has been evidence late group, which can copolymerize with the of greater surface staining of composite restora- hardening organic matrix of the composite mate- tions relative to the adjacent tooth enamel, it has rial [83]. A solution of the coupling agent placed not yet been determined to what extent this accre- on the prepared tooth surface with the excess tion of discolored material is due to inadequate solution being wiped away leaves an invisible film. hydrolytic stability of the adhesive bond between Following this, the freshly mixed (but not yet individual particles and the resin matrix, the hardened) composite material is placed on top of chemical nature of the reinforcing filler materials, this treated surface and the subsequent polymeri- the chemical nature and degree of polymerization zation apparently includes the methacrylate of the resin formulations, the roughness of the groups of the coupling agent. The coupling agent surface after the restoration is finished [78], or has presumably also formed ionic and coordinate factors some combination of these factors. These bonds with the tooth surface. Laboratory data on studied. are currently being tensile adhesion tests clearly demonstrate that the recently, there has been little or no effort Until use of the coupling agent significantly improved to prepare special materials to serve as reinforcing the adhesion between composite materials and fillers for dental composite materials, rather, ma- tooth surfaces [83, 84]. Clinical evaluations are terials already available were selected for this ap- under plication. However, the unique demands on this way [79]. kind of composite material warrant the develop- There is a rapidly-expanding wealth of experiment and use of the most suitable fillers as well ence and information regarding composite dental as other ingredients. materials, not only of the kind described here, but One experimental composite [79] contained a also of a wide diversity of materials and tech- mixture of two reinforcing fillers : spherical par- niques [85]. Composite restorative materials, when ticles of fused silica made up about two-thirds of they they have been developed to their best poten- the filler and smaller particles of an x-ray opaque tialities, should eventually provide the dental pro- glass made up about one-third of the filler. The fession and the public with greatly improved special bar- smaller particles were composed of a restorations for anterior teeth. ium fluoride-containing glass that was developed this purpose Radiopacity is desirable so for [80]. 5. References that dentists can distinguish between the filling material and any carious dentin that underlies the [1] Luskin, L. S., Milestones in the History of Acrylic filling. Products, Or-Chem Topics No. 23, Summer 1967, The most recent investigations are involved with Pliiladelphia, Rohm & Haas Co. Schnebel, Dentistische Reforme 4 and 1942 the preparation and evaluation of a single glass [2] 5, Blumenthal, L., Recent German developments in filler material containing silica and only enough the field of dental resins, fiat final report no. 1185, barium to give the desired refractive index and Field Information Agency, Technical, Office of Military Government for Germany (US) (27 May x-ray opacity Preliminary indications are [81]. 1947). that it will be suitable for converting into spheri- [3] FafCenbarger, G. 0., Dental cements, direct filling cal particles utilizing conventional techniques resins, composite and adhesive restorative mate-

rials : A Resume, Proceedings of the Engineering while retaining the appropriate refractive index, Foundation Research Conference on Engineering and will have a low coefficient of thermal expan- in Medicine-Bioceramics, J. Biomed. Mater. Res. sion. Chemically, the glass is formulated to have, (Special Issue 1972). [4] Caul, H. J., and Schoonover, I. C, Color stability at least in theory, hydrolytic stability in its bond- of direct filling resins, J. Am. Dental Assoc. 47, ing with the silane coupling agents. 448-452 (Oct 1953). Since these composite materials are not intrinsi- [5] Smith, D. L., and Schoonover, I. C, Direct filling resins : Dimensional changes resulting from po- cally adhesive to the prepared tooth surface, spe- lymerization shrinkage and water sorption, J. Am. cial coupling agents have been developed which Dental Assoc. 46, 540-544 (May 1953). Nelsen, R. J.; Wolcott, R. B., and Paffenbarger, improve bonding between such composite filler [6] G. C, Fluid exchange at the margins of dental materials and dentin and enamel. The chemical restorations, J. Am. Dental Assoc. 44, 288-295 design of these special coupling agents was based (March 1952). I. H., McLean, J. W., The response on empirical observations of the types of groups [7] Kramer, R. and of the human pulp to self-polymerizing acrylic that could displace water from the surfaces of restorations, Brit. Dental J. 92, 255-261, 281-287, powdered human enamel and dentin, thereby mak- 311-315 (May-June 1952). L. I., Pulp reaction to the insertion of ing the surface organophilic rather than hydro- [8] Grossman, self-curing acrylic resin filling materials, J. Am. philic [82]. Dental Assoc. 46, 265-269 (March 1953).

98 [9] Kramer, I. R. H., The relationship between pain and [31] Vanderbilt, B. M., and Jaruzelski, J. J., The bond- changes in the dental pulp following the insertion ing of fillers to thermosetting resins, Ind. and of fillings, Brit. Dental J. 96, 9-13 (Jan 1954). Eng. Chem., Prod. Res. and Dev. 1, 188-194 [10] Hedegard, B., Cold-polymerizing resins as restora- (Sept 1962). tive materials, Odontol. Tidskr. 65, 169-212 [32] Sterman, S., and Marsden, J. G., Filler-silane inter- (June 1957). actions. Modern Plastics 41, 254-266 (Oct 1963). [11] Hedegard, B., Co-Report, Synthetic Plastics, Intern. [33] Sterman, S., and Marsden, J. G., Silane coupling Dental J. 8, 249-250 (June 1958). agents as integral blends in resin-flUer systems. Modem Plastics 40, 125-177 (July 1963). [12] Fletcher, T., British Patent No. 3028, 1878, Fletcher, Bowen, R. L., Effect of particle size dis- T., Silicate of lime and alumina, Brit. Dental J. [34] shape and tribution in a reinforced polymer, J. Am. Dental Science 22, 74, (1879). Assoc. 69, 481-495 (Oct 1964). [13] Henschel, C. J., Observations concerning in vivo [35] Bowen, R. L., Development of an Adhesive Restora- disintegration of silicate cement restorations, J. tive Material, Adhesive Restorative Dental Ma- Dental Res. 28, 528-529 (Oct 1949). terials—II (Public Health Service Publication L., Paffenbarger, G. C, and Mullineaux, [14] Bowen, R. No. 1494, p. 225, Washington, D.C., 1966). of A. L., A laboratory and clinical comi>arison [36] Bowen, R. L., Unpublished data. direct filling resin, A silicate cements and a [37] Holliday, L. (ed), Composite Materials, p. 153 Progress Report, J. Prosthetic Dentistry 20, (Elsevier Publishing Co., New York 1966). 426-437 (Nov 1968). [38] Johannson, O. K., Stark, F. O., Vogel, G. E., and [15] Norman, R. D., Swartz, M. L., Phillips, R. W., and Fleischmann, R. M., Evidence for chemical bond- Virmani, R., A comparison of the intraoral dis- ing formation at silane coupling agent inter- integration of three dental cements, J. Am. faces, J. Comp. Materials 1, 278-292 (Jan 1967). Dental Assoc. 78, 777-782 (April 1969). [39] Broutman, L. J., and Krock, R. H. (eds.), Modern [16] Knock, F. E., and Glenn, J. F., Dental Material Composite Materials, pp. 501, 7 (Addison-Wesley and Method, U.S. Patent No. 2,558,139 (June 26, Publishing Co., Reading, Massachusetts, 1967). 1951). [40] Hatch, R. A., Phase equilibrium in the system

[17] Paffenbarger, G. C, Nelsen, R. J., and Sweeney, LisO-AlaOa—-SiOs , The American Minerologist 28, 471^96 (Sept-Oct 1943). W. T., Direct and indirect filling resins : A review of some physical and chemical properties, [41] Boyd, R. N., Colin. L., and Kaufman, E. G., Dental J. Am. Dental Assoc. 47, 516-524 (Nov 1953). Filling Composition of a Coefficient of Thermal [18] Rose, B. E., Lai, J., Green, R., and Cornell, J., Expansion Approximating that of Natural Tooth Enamel, U.S. (March Direct resin filling materials : Coefficient of Patent 3,503,128 31, 1970). thermal expansion and water sorption of poly- [42] Smoke, E. J., Ceramic compositions having nega- methyl methacrylate, J. Dental Res. 34, 589-596 tive linear thermal expansion, J. Amer. Ceramic (Aug 1955). Soc. 34, 87-90 (March 1951). [19] Rose, E. E., Lai, J., Williams, N. B., and Falcetti, [43] Bowen, R. L., and Argentar, H., Diminishing dis- J. P., The screening of materials for adhesion coloration in methacrylate accelerator systems, to human tooth structure, J. Dental Res. 34, J. Am. Dental Assoc. 75, 918-923 (Oct 1967). 577-588 (Aug 1955). [44] McLean, J. W., and Short, I. G., Composite anterior [20] Lee, H., and Neville, K., Handbook of Epoxy filling materials, Brit Dental J. 127, 9-18 (July Resins (McGraw-Hill Book Co., New York, 1969) 1967). [45] Hirasawa, T., Self-curing resins containing inor- [21] Bowen, R. L., Use of epoxy resins in restorative ganic fillers for dental restorations. II. Effect of materials, J. Dental Res. 35, 360-369 (June surface treatment of glass fillers on mechanical 1956). properties, Shika Zairyo Kenkyusho Hokoku 2, [22] Bowen, R. L., and Mullineaux, A. L., Adhesive 629-643, 644-655 (March 1965). restorative materials, Dental Abstracts 14, 80-82 [46] Going, R. E., and Sawinski, V. J., Microleakage of (Feb 1969). a new restorative material, J. Am. Dental Assoc. [23] Bowen, R. L., Dental Filling Material Comprising 73, 107-115 (July 1966). Vinyl Silane Treated Fused Silica and a Binder [47] Buonocore, M. G., Matsui, A., and Yamaki, M., Consisting of the Reaction Product of Bisphenol Abrasion of restorative materials, N.Y. State and Glycidyl Acrylate, U.S. Patent No. 3,066,112 Dental J. 32, 395-400 (Nov 1966). (Nov 1962). [48] Peterson, E. A., Phillips, R. W., and Swartz, M. L., [24] Bowen, B. L., and Rodriguez, M. S., Tensile strength A comparison of the physical properties of four and modulus of elasticity of tooth structure and restorative resins, J. Am. Dental Assoc. 73, 1324- several restorative materials, J. Am. Dental 1336 (Dec 1966). Assoc. 378-387 (March 1962). 64, [49] Bowen, R. L., Adhesive bonding of various mate- [25] Bowen, R. L., A Method of Preparing a Monomer rials to hard tooth tissues. VI. Forces developing having Phenoxy and Methacrylate Groups Linked in direct-filling materials during hardening, by Hydroxy Glyceryl Groups, U.S. Patent No. J. Am. Dental Assoc. 74, 439-445 (Feb 1967). 3,179,623 (April 1965). [50] Matsui, A., Buonocore, M., and Yamaki, M., Heat [26] Bowen, R. L., Properties of a silica-reinforced poly- of polymerization of certain new and conven- mer for dental restorations, J. Am. Dental Assoc. tional restorative materials, J. Dental Res. 46, 66, 57-64 (Jan 1963). 1106 (Sept-Oct 1967). [27] Bowen, R. L., Crystalline dimethacrylate monomers, [51] Gotfredsen, C, Addentd) Investigations of a plastic J. Dental Res. 49, 810-815 (July-Aug 1970). filling material, Tandlaegebladet 72, 407-429 [28] Bjorksten, J., Tovey, H., Harker, B., and Hen- (May 1968). ning, J., Polyesters and Their Applications, p. 11 [52] Macchi, R. L., and Craig, R. G., Physical and me- (Reinhold Publishing Corp., New York 1956). chanical properties of composite restorative ma- [29] Guide to Dental Materials and Devices (5th ed), terials, J. Am. Dental Assoc. 78, 328-334 (Feb pp 51, 98 (Am. Dental Assoc., Chicago, 1970-71). 1969). [30] Vanderbilt, B. M., and Simko, J. P., Jr., Silane [53] Freeman, F. H., Composite restorative materials. coupling agents in glass-reinforced plastics, Mod- Presented before the Dental Materials Group, ern Plastics 38, 135-217 (Dec 1960). lADR, Mar 21, 1969, Houston, Texas. 99 .

[54] Tani, Y., and Buonocore, M., Marginal leakage and [70] Langeland. L. K., Guttuso, J., Jerome, D. R., and penetration of basic fuchsin dye in anterior res- Langeland, K., Histologic and clinical compari- torative materials, J. Am. Dental Assoc. 78, 542- son of Addent with silicate cements and cold- 548 (March 1969). curing materials, J. Am. Dental Assoc. 72, 373- [55] Lee, H. L., Swartz, M. L., and Smith, F. F., Epoxy 385 (Feb 1966). resins in dentistry, presented at the ACS Na- [71] French Patent No. 2,008,541 (Jan 23, 1970). tional Meeting, April 1969, Minneapolis, Minne- [72] Mihailov, M., and Boudevska H., Synthesis and sota (in press). polymerization of polyestermethacrylates and [56] Gotfredsen, C, Physical properties of a plastic fill- terephthallic and furane-2, 5-diearboxylic acid, ing material (Addent®), Acta Odontol. Scand. Compt. Rend. Acad. Bulgare Sci. 18, 31-34 27, No. 6, 595-615 (1969). (1965). [57] Lee, H. L., Swartz, M. L., and Smith, F. F., Physical [73] French Patent No. 2,010,896. properties of four thermosetting dental restora- [74] French Patent No. 2,010,905. tive resins, J. Dental Res. 48, 526-535 (July-Aug [75] Atsuta, M., Nakabayashi, N., and Masuhara, E., 1969). Hard methacrylic polymers, II., copolymers of [58] Phillips, R. W., Swartz, M. L., and Norman, R. D., methyl methacrylate and 2,2-Di(4-methacryloxy- Materials for the Practicing Dentist, chapter 59, phenyl) propane, J. Biomed. Mat. Res. 5, 183-196 pp. 182-191 (The C. V. Mosby Co., St. Louis, Mo., (May 1971). 1969). [76] Bowen, R. L., and Argentar, H., Amine accelerators [59] Docking, A. R., Modem materials in dental prac- for methacrylate resin systems, J. Dental Res. tice, Australian Dental J. 15, 303-309 (Aug 1970). 50, 923-928 (July-Aug 1971). [60] Schouboe, P. J., Paffenbarger, G. C, and Sweeney, [77] Bowen, R. L., and Argentar, H., Tertiary aromatic W. T., Resin cements and posterior-type direct amine accelerators with molecular weights above filling resins, J. Am. Dental Assoc. 52, 584-600 400, J. Dental Res. 51, 473-482 (Mar-Apr. (May 1956). 1972). [61] Barton, J. A., Jr., Burns, C. L., Chandler, H. H., [78] Chandler, H. H., Bowen, R. L., and Paffenbarger, and Bowen, R. L., An experimental, intermediate- G. C, A method for finishing comjxxsite restora- restorative composite material (To be published tive materials, J. Am. Dental Assoc. 83, 344-348

in J. Dental Res. ) (Aug 1971). H., [62] Barton, J. A., Jr. et al., Unpublished data. [79] Chandler, H. Bowen, R. L., Paffenbarger, G. C, and MuUineaux, A. L., Clinical investigation of [63] Craig, R. G., and Peyton, F. A., Elastic and mechan- a radiopaque composite ical properties of human dentin, J. Dental Res. material, J. Am. Dental Assoc. 37, 710-718 (Aug 1958). 81, 935-940 (Oct 1970). [80] Bowen, R. L., and Cleek, G. W., X-ray-opaque re- [64] Stanford, J. W., Paffenbarger, G. C, Kumpula, inforcing fillers for composite materials, J. Dental J. W., and Sweeney, W. T., Determination of some Res. 48, 79-82 (Jan-Feb 1969). compressive properties of human enamel and [81] Bowen, R. L., and Cleek, G. W., A new series of dentin, J. Am. Dental Assoc. 57, 487-495 (Oct X-ray-opaque reinforcing fillers for composite 1958). materials, J. Dental Res. 51, 171-182 (Jan-Feb Stanford, J. W., Weigel, K. V., Paffenbarger, G. C, [65] 1972). and Sweeney, W. T., Compressive properties of [82] Bowen, R. L., Investigation of the surfaces of hard hard tooth tissues and some restorative mate- tooth tissues by a surface activity test. Pro- rials, 746-756 (June J. Am. Dental Assoc. 60, ceedings of the Workshop on Adhesive Restora- 1960). tive Dental Materials at Indiana University, [66] Craig, R. G., Peyton, F. A., and Johnson, D. W., Owen Litho Service, Spencer, Indiana-Publishers Compressive properties of enamel, dental Sept 1961 ; National Technical Information Serv- cements, and gold, J. Dental Res. 40, 936-945 ice, U.S. Dept. Commerce, Springfield, Va. 22151, (Oct 1961). PB 173009, Phillips and Ryge (ed.). [67] Souder, W., and Paffenbarger, G. C, Physical [83] Bowen, R. L., Surface-Active Comonomer and Properties of Dental Materials, NBS Circ. 433, Method of Preparation, U.S. Patent No. 3,200,142 222 pages (1942). (Aug 10, 1965). [68] Sweeney, W. T., Sheehan, W. D., and Yost, E. L., [84] Bowen, R. L., Adhesive bonding of various mate- Mechanical properties of direct filling resins, rials to hard tooth tissues (Parts I-V), J. Dental J. Am. Dental Assoc. 49, 513-521 (Nov 1954). Res. 44, 690-695, 895-902, 903-905, 906-911, 1369- [69] Stanley, H. R., Swerdlow, H., and Buonocore, M. 1373 (July-Dee 1965). G., Pulp reactions to anterior restorative mate- [85] Craig, G. G., The placement of composite resin rials, J. Am. Dental Assoc. 75, 132-141 (July restorations, Australian Dental J. 15:277-280 1967). (Aug 1970).

100 ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Cements Containing o-Ethoxybenzoic Acid (EBA)

Gerhard M. Brauer

Dental Research Section, Institute for Materials Research, National Bureau of Standards, Washington, D.C. 20234

Cements containing o-ethoxybenzoic acid (EBA) are reviewed. Studies of the mechanism of hardening of zinc oxide-eugenol cements indicated the potential usefulness of other chelating agents in dental cements. Products with greatly enhanced physical and mechanical properties are obtained by the partial substitution of EBA for eugenol. Similar to zinc oxide-eugenol cements, these materials are well tolerated by the tissues, but they also stimulate the formation of reparative dentin. Physical properties of the EBA-containing cements approach those of the biologically and physiologically less desirable zinc phosphate cements. The EBA cements have become commercially available and have been well accepted as luting agents for fixed restorations and as insulating bases. They appear to be the materials of choice for indirect pulp capping. Resin modified EBA restoratives show good stress-bearing characteristics and should find applications as an intermediate restorative. Possible studies to further improve these versatile materials are discussed.

Key words : Crown and bridge cements ; dental cements ; EBA cements ; EBA sedative

bases ; intermediate restorative ; o-ethoxybenzoic acid cements ; pulp capping materials zinc oxide-EBA-eugenol cements.

1. Introduction chelate with any excess eugenol being sorbed by both zinc eugenolate and the unreacted zinc oxide Cements containing o-ethoxybenzoic acid [10-11]. More detailed studies of the infrared (EBA) are an outgrowth of studies to improve spectra have confirmed the 6isdioxachelate struc- zine oxide-eugenol (ZOE) cements which were ture of zinc eugenolate [12]. first reported in the dental literature about one CH3 hundred years ago [1-3 J. ^ Over the years ZOE cements have found a wide range of applications -CH2— CH=CH! in dentistry: as temporary restoratives, sedative /V bases, cementing media, for crown Zn and bridge CHj=CH-ChA work, in pulp capping, as soft tissue packs in oral / \ \/ ^0 o surgery and periodontics, as root canal sealers in ins endodontics and with modifying agents as impression pastes. The ZOE cements possess much better The crystal structiu"e of this compound has been compatibility than most dental materials with investigated by Douglas [13] and Cartz [14]. X-ray both the hard and soft tissues of the mouth [4-7]. diffraction of commercial ZOE products or ex- They have excellent sealing properties [8] and perimentally prepared formulations which also their sterilizing effectiveness [9] has been dem- have a large excess of zinc oxide yield extremely onstrated. ZOE cement also acts as a palliative or small values for the relative intensity ratios of the anodyne and as a mild non-irritant antiseptic. principal peaks of zinc eugenolate to ZnO. This Their low strength which may not be large enough suggests that in these cements the zinc eugenolate is below the level to resist forces of mastication and their lack of reaction product formed 2% resolvable in standard preparations [15-16]. High resistance to wear and disintegration deter more resolution microscopy also indicates that the zinc extensive use in temporary restorations. eugenolate product in commercial formulations is small and is confined to the reaction interface [17]. 2. Setting Mechanism of ZOE Cements These formulations also contain about 5% free eugenol after hardening [18]. A number of studies have shown that the set The setting reaction is speeded up by the mass resulting from mixes containing the proper presence of an accelerator such as zinc acetate. proportion of zinc oxide and eugenol consists of Presence of water or a decrease in the pH of the zinc oxide embedded in a matrix of zinc eugenolate reaction mixture also reduces setting time. Water reacts with zinc oxide to form the hydroxide ' Figures In brackets Indicate the literature references at the end of this paper. which in turn reacts with eugenol [19].

101 (1) ZnO + n H2O > n Zn(0H)2, cm^, 600 psi) and a solubility and disintegration value of 0.008 g/cm^ [21]. A clinical evaluation of (2) n Zn(0H)2 + 2n eugenol > this cement as an intermediate restorative material

n Zn (eugenol) 2 + 2n H2O. at intervals of 60 days for a period of 12 months showed this formulation to be superior to three The formation of the hydroxide is slow, but acid other experimental cements with regard to the speeds up the reaction. maintenance of anatomical form and marginal in-

( tegrity [22]. Products incorporating poly (methyl 3) n ZnO + 2n CH3COOH > methacrylate) in the powder have become com- nZn(CH3COO)2 + u H2O, mercially available both as luting agents and intermediate restoratives. (4) nZn(CH3COO)2+2?iH20 The knowledge gained from the characterization of the reaction products Zn(OH)2+2?i CH3COOH. led to extensive investigations with the ultimate aim of obtaining improved n ZnO+?i HsO-^n Zn(0H)2 cements by substitution of other complexing agents for eugenol. Eugenol isomers capable of Since the zinc hydroxide reacts with eugenol, forming chelates were synthesized [23] and the ef- reaction goes to the right. Obviously, (4) zinc fect of the position of substituents on the behavior acetate can be substituted for acetic acid, but of the isomers as evidenced by their ionization con- to start the reaction a trace of water must be stants and reactivity with zinc oxide was deter- present. Commercial formulations generally con- mined [24]. In the reaction of eugenol isomers tain zinc acetate dihydrate as accelerator which with zinc oxide, the 1,2,3-trisubstituted isomers do furnishes the water to initiate the reaction. not react readily compared to the unsymmetrically 1,2,4- or 1,2,5-trisubstituted ones, indicating that 3. Modified ZOE and EBA Cements the chelation reaction is greatly influenced by steric hindrance of the bulky neighboring allyl Low strength is unquestionably the main groups. The chelation reaction may also depend to disadvantage of ZOE cements. A number of some degree on the ionization constants since the studies have been undertaken during the last more acidic chavibetol reacts somewhat faster than decade to upgrade the properties of the cements. eugenol. The following approaches have been taken: Zinc oxide will react with many chelate-forming 1. Replacement of zinc oxide by other metal compounds to yield coherent products [25]. Mixes oxides. containing o-ethoxybenzoic acid (EBA) salicylal- 2. Incorporation of fillers, reinforcing or modi- , dehyde, acetylacetone, o-ethoxyacetic acid or lactic fying agents. acid form hard products within one hour at room 3. Substitution of eugenol by other chelating temperature. Some of these products disintegrate agents. in water. Only limited studies have been made to substitute Cements obtained from o-ethoxybenzoic acid other metal oxides for zinc oxide [11]. Cupric (EBA) and metal oxides of group II of the eugenolate can be prepared by a procedure similar periodic table or lead oxide have been studied in to that used in the synthesis of zinc eugenolate. considerable detail Products formed from A slight excess of a methanolic solution of cupric [26]. zinc oxide and EBA-eugenol solutions in the ab- acetate monohydrate is reacted with eugenol at sence of an accelerator harden more rapidly and 60° C for one hour {16]. Employing the same have higher strength and lower solubility and dis- reaction conditions, that is, refluxing methanolic solutions of eugenol with mercuric-, lead-, nick- integration vakies than those made with EBA elous- or calcium-acetate did not yield the respec- alone (figs. 1 and 2) . Most useful properties are ob- tive metal eugenolates. Attempts to synthesize tained with liquids containing between 50 to 70 the cupric or nickel isoeugenolates also were percent EBA. However, these cements have a high unsuccessful. water solubility. Substitution of 2-propoxy-5- methylbenzoic acid, a higher, more hydrophobic Addition of rosin or hydrogenated rosin im- homologue for EBA, yields cements with unex- proves the- working properties of the mixes. Hy- pectedly high water solubility and disintegration drogenated rosin is commonly used since it is stable 27]. On incorporation of rosin or hydroge- to oxidation and yields cements with good color [26, nated rosin the water solubility is greatly reduced lability. Incorporation of up to 20 percent poly- [26]. Addition of a reinforcing agent such as mer dissolved in eugenol to improve the physical monocalcium phosphate, heat-treated fused quartz properties was first suggested by Curtis [20]. An with compres- experimental cement containing surface-treated or alluminum oxide gives products zinc oxide (80%) and poly (methyl methacrylate) sive, shear and tensile strengths that are three to four times those of conventional ZOE formula- (20% ) powder and a liquid consisting of 99 percent eugenol and 1 percent acetic acid had a com- tions. Thus, these EBA containing products have pressive strength of 64.0 MN/m^ (550 kg/cm^ mechanical properties similar to those of zinc

7,820 psi) , a tensile strength of 4.1 MN/m^ (42 kg/ phosphate cements.

102 % EBA

iqo 80 60 40 20 0

0 20 40 60 80

0 20 40 60 80 100 % EUGENOL % EUGENOL Figure 2. Effect of composition of the liquid on compressive strength and solubility and disintegration Figure 1. Effect of composition of the liquid on setting of ZnO-EBA-eugenol cements. time and standard consistency powder-liquid ratio of ZnO-EBA-cugcnol mixes [25]. X X solubility and disintegration, o o compressive strengtn [25].

The properties of cements, based on zinc oxide, adhesion values than ZOE products. Similar for- hydrogenated rosin, EBA, and eugenol have been mulations have become commercially available and studied in detail [28] . The strength of the cements have been well accepted as luting agents in crown is nearly independent of the particle size of zinc and bridge cementation. On incorporation of more oxide and fused quartz. The carbon dioxide and j)owder into the mix, excellent base materials can water content of the zinc oxide also have little be obtained. Especially desirable is their high ten- effect on the physical properties of the hardened minute compressive strength of 46.1 MN/m^ (470 product [28-30]. Alumina reinforced EBA ce- kg/cm% 6,680 psi) which can easily withstand the ments have physical properties superior to those forces encountered in condensing an amalgam. of cements reinforced with fused quartz [31]. The The brittleness of these luting agents limits their preferred composition contained 64 percent zinc use for temporary restorations of multiple surface oxide, 30 percent tabular ALOs and 6 percent hy- carious lesions in areas subject to heavy masticatory drogenated rosin. A slurry prepared from 1.7 g forces [32]. Stress bearing characteristics of powder per 0.2 ml of liquid can be mixed easily of EBA cements can and will harden in less than 10 minutes. The result- be improved through the incorpora- ing product has a compressive strength of 93 MN/ tion of powdered polymers of relatively low elastic m^ (950 kg/cm% 13,500 psi) and a film thickness moduli [30]. The most suitable resins are meth- of 26 jjm (table 1). With one-surface inlays, these acrylate copolymers, although vinyl copolymers cements adhere at least as well as commercial zinc may also be potentially useful. Other resins, be- phosphate cements and give much higher tensile cause of their resilient nature are difficult to obtain

Table 1. Physical properties of dental cements ^

Powder-liquid Setting Solubility quantities time Tensile strength^ Compressive strength ^ and disintegration

gm/ml min MN/m^ kg/ cm 2 kg/cm^ Percent

Zinc oxide-eugenol (ZOE) 1. 0-2. 2/. 4 3-8 1. 4-2. 5 14-25 16. 7-38. 3 170-390 0. 02-0. 20

Reinforced ZOE . 6-1. 1/. 3 3-8 1. 5-6. 9 15-70 39. 2-75. 5 400-770 . 05- . 80

EBA (AI3O3 reinforced) 1. 6-2. 0/. 2 7-9 3. 9-7. 4 40-75 58. 9-93. 1 600-950 . 05- . 13

EBA (plastic-modified) 1. 1-1. 3/. 2 7-8 6. 6-9. 8 66-99 50. 0-78. 5 510-800 . 13- . 94

Zinc phosphate cement _ 1. 4/. 5 7-8 3. 2-4. 6 33-46 68. 7-147. 1 700-1, 500 . 10- . 20

' Some of the data given in this table are taken from the results of the collaborative test program—Zinc oxide— Eugenol Dental Cements ISO Technical Committee 106/ WGl—Filhng Materials. 2 One day.

103 in powdered form. A cement made from powder 4. Mechanism of Setting of EBA Cements containing 58.2 percent ZnO, 27.3 percent AI2O3, 5.4 percent rosin and 9.1 percent methyl meth- The products formed on hardening of com- acrylate copolymer and liquid containing 62.5 per- mercial EBA cements have not been completely cent EBA and 37.5 percent eugenol had, after one characterized [16]. Cements made up of zinc oxide week, tensile and compressive strengths of 11.5 and a liquid consisting of either EBA or 62.5 per- MN/m^ (117 kg/cm% 1,660 psi) and 65.1 MN/m== cent EBA-37.5 % eugenol were pulverized. The (664 kg/cm% 9,430 psi) respectively. Thus, the powder was extracted by shaking with 50 ml of addition of the polymer greatly increases the ten- methanol for 6 hr. Aftei- centrifuging, the me- sile strength of the hardened cement. These mark- thanol layer was decanted and the remaining edly higher tensile strength values are probably powder was dried at 110 °C and weighed. From more important than the somewhat lower compres- cements containing EBA, the original liquid was sive strength; the clinical results described later removed quantitatively (table 2). Cements con- bear out this point. taining EBA and eugenol lost 60 to 70 percent of Cements containing rosin have somewhat higher their liquid component on extraction. Probably tensile strength, but are more soluble and set more all the EBA was removed by the methanol extrac- slowly than those to which hydrogenated rosin has tion. The results were confirmed by heating the been added [30]. Improvement in the physical dried unextracted solid residue in a crucible to re- properties of a formulation containing 58.2 per- move any remaining organic matter. There was no cent ZnO, 27.3 percent AI2O3, 5.4 percent hydro- loss in weight on heating the residue of the cement genated rosin and 9.1 percent methyl methacry- prepared with EBA liquid. The EBA-eugenol late copolymer in the' powder and 62.5 percent containing cement showed a 30 to 40 percent loss in EBA-37.5 percent eugenol in the liquid was not weight which accounted for all the unextracted obtained on increasing the (1) hydrogenated rosin liquid. Thus, eugenol is much less readily ex- content, (2) percentage of eugenol, and (3) ratio tracted, and hence more firmly bound than EBA in of zinc oxide to aluminum oxide reinforcing agent. the hardened cementa Addition of 0.5 percent aluminum sulfate decreases very slightly the solubility and disintegra- Table 2. Weight loss of cements on solvent extraction tion of cements containing rosin or hydrogenated rosin. A vinyl chloride-vinyl acetate copolymer- Cement Extrac- Loss of liquid Extrac- containing cement had one-week tensile and com- containing tion on heating tion zinc oxide with CH3OH insolu- with pressive strength of 9.8 MN/m^ (100 kg/cmS 1,420 and CH3OH ble residue CHCI3 psi) and 91.5 MN/m^ (933 kg/cmS 13,300 psi), respectively. Incorporation of an acrylonitrile-buta- diene-styrene terpolymer, a polyacetal resin, or Percent Percent Percent EugenoL 10-12 80-90 ~80 various of commercial polycarbonate mold- grades EBA 100 0 "119 ing powders produced materials with physical EBA and properties somewhat lower than those containing eugenol 60-70 30-40 "102 acrylic copolymer. A summary of the physical properties of conventional and reinforced ZOE "Including some solid extracted with CHCls. cements, alumina-reinforced, plastic-modified EBA cements, and zinc phosphate cements is To study the reaction product, zinc oxide (2 g, given in table 1. 0.0246 mole) and o-ethoxybenzoic acid (8 g, 0.048 A comparison of the first commercially avail- mole) were stirred together and set aside. After able EBA crown and bridge cements has been two months at room temperature, the mixture consisted of a soft layer on layer. made by Phillips and coworkers [33]. Proper- top of a hard X-ray diffraction did not give peaks. ties investigated were compressive and tensile any The product was partially soluble in hot water, insol- strength, film thickness, solubility and disintegra- uble in methanol, ethanol, chloroform, or dimethyl- tion in both water and acid and retentive charac- formamide. teristics as determined by the amount of tensile The reaction products were added to warm force required to remove one-surface inlays from acetone and filtered. On evaporation of the solvent prepared cavities following cementation with the an amorphous material separated from the filtrate. various agents. Physical properties of an experi- The solid residue, insoluble in warm acetone, could mental alumina-reinforced EBA cement and com- be dissolved by boiling in acetone for 10 min. A mercial EBA cements of unknown composition solid (mp 92-100 °C) crystallized on cooling to have also been studied by Custer and Anderson 0 °C. After drying in a vacuum, analysis for car- [34]. The properties of typical commercial EBA bon and hydrogen indicated that the compound cements have been measured in a collaborative test was zinc o-ethoxybenzoate (Anal: Calcd. for program which had as its aim the development of CisHisOeZn : C, 54.63, H, 4.58. Found: C, 54.8; H, a specification for ZOE and EBA type cements 4.6) Absence of an infrared absorption peak [35]. around 1,750 cm-^ indicated that no unreacted

104 : ,

COOH is present. The broad absorption band of greater the tendency for metals to combine with a o-ethoxybenzoic acid around 1,230 cm-^ which can given chelating agent, the greater the drop in pH. be attributed to the ethoxy group [36] is present OH —Me- in the zinc derivative as a sharp band at 1,240 cm-^. OCH3 The spectrum of o-ethoxybenzoic acid has absorp- OCII3 CllaO-/^ + Me++ + 2H+ tions at 1,745 cm-^ and also carbonyl absorptions at 1,594 and 1,609 cm"\ For the primarily ionic- CH2-CH=CH2 CHj-CH=CH2 CH2=CH-CH bonded zinc

II tions follow Beer's law. They show an absorption peak around 292 nm, whereas ZnSO^ does not Possible formation of a chelate was studied [16] absorb at this wave length. Job's method of by titration of an aqueous solution of EBA with continuous variation was applied to EBA-ZnSO^ base in the presence of zinc ions and Job's spectro- solutions having a total concentration of 4X10~* photometric method of continuous variation [38]. M. No maximum was observed by plotting optical The first procedure depends on the fact that density versus composition of the solution. This most metal chelates may be considered as formed result does not necessarily indicate that no chelate by the displacement of a weak acidic portion of the formation takes place since some chelates have chelating agent by a metal ion [39]. Thus, the absorption spectra that do not differ significantly addition of metal ion causes a drop in pH and the from those of the chelating agent. McKenzie and

105 coworkers [40] have shown that ionic chelates teeth restored with cyanoacrylate-containing re- have nearly the same absorption spectra as their storative material. The layer of reparative dentin chelating agent, but that the absorption spectra formed in response to the filling materials was for covalent bonded chelates contained strong ab- proportional to the degree of odontoblastic dis- sorption bands characteristic of the chelate, and ruption and the inflammatory infiltrate. It was hence of the metal-donor bond. Since EBA is a much more pronounced in the teeth that were re- moderately strong acid, its anion should be a stored with cyanoacrylate and EBA than in teeth fairly strong conjugate base. Formation of an filled with ZOE cements. Thus, in cases in which ionic complex with a divalent metal ion such as the therapeutic aim is to require a thick layer of Zn^"*" is likely to occur, especially since oxygen reparative dentin, a cavity base with either the donors favor ionic bonds. EBA or cyanoacrylate is preferable. Eugenol forms a five-membered chelate whereas It would be of interest to study if the reparative EBA may give a six-membered ring. Measurement and secondary dentin formation produced by EBA stabiUty constants K as well as cements is as rapid as that formed after the use of the chelate _ the formation constants ki and Atj for the reaction of calcium hydroxide. Coleman and Kirk [42] filled cavities scheduled for extraction for ortho- Me+++Chel-'^MChel+ dontic reasons with ZOE and EBA cements. The teeth were extracted after periods ranging from MChel+ + Chel- ^^MChel2 24 hr to 3 weeks. The teeith were fixed in 10 percent formol saline, decalcified, and examined his- [MChella tologically. Little or no pulpal reaction could be "[Me++][Chel-i]2 attributed to either the ZOE or the EBA materials. The odontoblast layer was usually intact. would yield valuable information since these con- Although some vacuolation was present in this stants would show the relative stabihty of the region, this was also seen in the control teeth where two ring systems. no cavities had been prepared. There has long been a deep interest in an im- 5. Clinical Studies proved ZOE cement that would be suitable for permanent cementation of cast restorations. The The main purpose of a temporary restoration modified ZOE or reinforced EBA cements have is sedation and protection of the tooth from irri- been employed successfully as crown and bridge tants and decay. The ZOE cement is essentially cements and appear to be well suited for this pur- neutral and, therefore, offers unusual pulpal pro- pose. The absence of irritation on the dentine-pulp tection. The EBA-containing cements have the complex and the resulting freedom from post- same mild tissue reaction of unmodified ZOE cementation symptoms gives them a big advantage cements, which are much superior in this respect over zinc phosphate cements. Since their compres- to zinc phosphate cements [41^3]. sive strength is much higher than that of un- Unset cements caused some necrosis and a mod- modified ZOE cements, their retentive properties erately severe inflammatory reaction when they are improved and approximate those of zinc were first inserted into rat muscle [42]. Granula- phosphate cements [33, 44, 45]. A clinical study tion tissue formed in the region and healing pro- using 186 full cast crown bridge retainers and 205 ceeded rapidly, producing a fibrous capsule to the full cast crowns has been reported by Horn [46]. implant. When set materials were used as im- The span of the bridges cemented with this mate- plants, the EBA-containing cement was always less rial was limited to a maximum of two consecutive irritating than zinc oxide-eugenol cements. Bhas- pontics of bicuspid width. Nineteen of the full cast kar and coworkers [43] investigated the pulpal crown bridge retainers were of the cantilever type. restorative materials. response of four types of Resin veneers were protected from the excess EBA Class cavities were prepared in 78 teeth of three V cement by coating them with silicone grease. No miniature swine. The animals were killed after cavity liners or medicaments were applied. The 1, 2, and 3 weeks and the teeth were examined crowns were not completely filled with cement, microscopically. A powder containing 57 percent but a coating was applied to the internal surfaces ZnO, 28 percent aluminum oxide, 9.5 percent poly (methyl methacrylate) copolymer and 5.5 percent and a small amount was allowed to flow into any rosin and liquid made up from 66.7 percent EBA crevices or depressions on the teeth prior to inser- and 33.3 percent eugenol was used. The restorative tions. Dryness was not maintained after the res- material appeared to be biologically acceptable to torations had been seated by applying pressure the dental pulp. No irreversible pulp damage was for two or three minutes. After cementation, abut- observed. Odontoblastic disruj^ti.on and inflam- ment teeth were not painful, and the marginal matory infiltrate were not severe. They were least relationships were not unduly distorted. In pronounced in the ZOE restoration, very slightly selected cases, tests with ice-cold water seemed to more marked in a commercial, resin-modified ZOE indicate excellent insulation. This effect may be and the EBA formulation, and most prominent in caused by the ability of the cement to obtund pain.

106 A number of crown and bridge cements incor- EBA cements, employed as cement bases, usually porating EBA have become commercially avail- utilize a higher powder-liquid ratio than when able in the United States and in Europe. Many employed as lutmg agents. It is good practice to products have film thickness of less than 25 /xm. incorporate the maximum amount of powder into They may be used for the final cementation the liquid consistent with a usable consistency so of metal crowns and bridges retained by metal that the powder will be in large excess in the crown, porcelain or plastic jackets, porcelain-over- hardened cement. EBA cements, because of their metal restorations, and gold-veneered-with-resin greater strength, seem to be well suited as a one- restorations. step base in deep cavities for gold, silicate cement

No clinical data are available regarding the re- and amalgam fillings [32] . When this cement was tention of orthodontic bands cemented with EBA ])laced in 32 vital but symptomatic teeth, symp- cements. Bands cemented with an experimental toms siibsided within two days. The cavity prepa- cement containing 90 percent ZnO and 10 percent rations were then completed and permanent resto- hydrogenated rosin powder and 62.5 percent EBA rations placed, leaving part of the EBA cement as and 37.5 percent eugenol required three times as a base. The teeth remained vital and non- much force to accomplish removal as bands ce- symptomatic. mented with ZOE, but the retention values were An alumina reinforced EBA cement was placed considerably lower than those obtained for as a base imder a series of amalgam restorations phosphate or silicate cements [47]. that were packed with a calibrated spring plugger Zinc phosphate cement has been the preferred at a pressure of 140 kg/cm^. When the teeth were material for use as an intermediary base beneath sectioned after 48 hr, the base was still intact metallic restorations. Zinc phosphate is preferred (fig. 3), whereas ZOE bases fractured at the be- over conventional ZOE and calcium hydroxide pulpal-proximal line angle [31]. cause of its superior strength, despite its inferior ZOE-type cements have been the preferred ma- biological and slightly poorer thermal diffusion terial for use over recently injured pulps caused by characteristics. When ZOE or calcium hydroxide deep and extensive operating procedures. This is bases are used, it is often recommended that they especially true in the teeth of children in whom be overlaid with the stronger zinc phosphate cement [48-50]. secondai-y dentin has not yet fonned a protecting The clinical significance of the compressive barrier within the pulp chamber. The materials are strength of a base material has not been defined. radiopaque [54], seal a cavity better than other It is obvious, nevertheless, that when amalgam is restorative materials [8, 55] and thus prevent condensed into the cavity preparation, the base organisms or moisture from the oral cavity gain- must have sufficient strength to support the forces ing access to the cavity floor. This may be the rea- of condensation. If the base does not have sufficient son for the higher percentage of negative cidtures strength, fracture or displacement of the base obtained when lining with ZOE instead of calcium could permit the amalgam to contact the underlying tooth structure and thus negate the thermal protection afforded by the base. Furthermore, in deep cavity preparation, the amalgam could be forced through microscopic exposures in the floor of the cavity and into the pulp [51-52]. Chang, Swartz and Phillips [53], in experiments conducted under laboratoi-y conditions, showed that with ZOE materials a minimum strength capable of suppoi-ting amalgam condensation ranges from 0.7 to 1.2 MN/m^ (7 to 12 kg/ cm2, 100tol70 psi). Interest has increased in recent years in using zinc oxide-eiigenol type cements as bases under amalgam and inlay restorations. These bases are nonirritating to the pulp and thus eliminate the need to protect the pulp Avith subbase materials when the zinc phosphate cements are used. The procedure is simplified, time is saved and the danger of pulp irritation is further controlled. The now available reinforced ZOE cements and especially the EBA cements, because of their increased early strength, are capable of withstanding the Figure 3. Section through an amalgam restoration con- forces developed during condensation of amalgam densed under HO kg/cnv^ (13.8 MN/ni^) packing pressure against an AhOa reinforced EBA iase with a and those forces which may subsequently be trans- ten-minute compressive strength of 470 kg/cm' (46.2 mitted through the restoration. MN/m") [31].

452-525 0—72 8 107 ,

hydroxide. Any residual organism probably re- tion, (4) resistance to abrasion and attrition for mains in a latent form under sound restorations, an extended time, (5) ease of manipulation and because the environment has been altered and placement, and (6) a longer service life than con- conditions for growth have become unfavorable. ventional ZOE temporary fillings. These organisms could become reactivated if saliva Polymer reinforced ZOE cements because of were to gain access to them through a leak in the their greatly improved physical properties [21], restoration. Such leakage is much less likely if a show a high degree of clinical acceptibility after

ZOE or EBA lining is used. Thus, EBA cement, a 12-month observation period [22] . On the other because of its excellent sealing characteristics [42] hand, fused quartz [32] or titanium dioxide [22] will assist in decreasing the number of organisms reinforced EBA cements, despite their improved remaining in the dentin and, furthermore, will strength, proved unsuitable as temporary restor- promote remiiaeralization of decalcified dentine at atives. Contrary to their low solubility and dis- the base of the cavity [43] . These properties should integration in water and dilute acids, the fused make EBA cements the material of choice, espe- quartz reinforced EBA restoratives disintegrated cially in indirect pulp capping procedures since rapidly under oral conditions [59]. Thus, a low in calcium hydroxide does not possess the excellent vitro solubility value may be useful in comparing sealing characteristics, whereas ZOE or modified the relative solubility of products of similar com- ZOE cement does not stimulate the formation of position, but is not necessarily an indication of the reparative dentin to any appreciable extent. success of such restorations in the mouth. Compari- Human pulps which had been exposed in the sons have been made of the in vivo intraoral dis- course o'f normal operative procedures in five non- integration of cements using specially designed symptomatic teeth were capped using an EBA con- partial dentures so that cements are exposed to the taining cement and the cavities were filled [32]. oral environment [60]. Results of these tests Within a week part of the material was removed showed considerable patient variation in the rate and permanent restorations placed, leaving part and amount of cement lost. Abrasion played an of the previous cement filling as a base. None of important role in the loss of material with the the patients reported any symptomatology. None greatest loss invariably occurring in those regions of the teeth showed any radiographic changes and most subject to abrasion by the tongue. Thus, the all responded normally to vitality tests within the in vivo disintegration appears to be an effect of two to ten months observation per'iod. the interaction of solubility and abrasion. Studies of the possible use of EBA cements in The resin-modified alumina- reinforced EBA root canal therapy and for gingival dressing are cement, because of its demonstrated mechanical not available. Requirements for an improved root and palliative properties, especially its much canal sealer would include (1) a suitable con- higher tensile strength, appears to be very desir- sistency at the time of insertion into the canal, (2) able for use as a long-duration temporary good dimensional stability to avoid fissures resiilt- restorative. ing from shrinkage, (3) good adhesion, and (4) In a limited clinical study [30], approximately high degree of insolubility to body fluids [56]. 50 restorations, including complex restorations ZOE yields a fairly satisfactory hermetic seal, but subject to heavy occlusal stresses, were placed using is only slightly adherent to the cavity walls. Fur- one formulation. Its powder component contained thermore, according to Erausquin and Munizabal, 58.2 percent ZnO, 27.3 percent ALOs, 5.4 percent ZOE can be irritating to the periapical tissues rosin and 9.1 percent methyl methacrylate [57], although the response is reduced by addition copolymer. of acrylic polymer spherules [58]. A powder-liquid ratio of 1.2 g powder per 0.2 In many patients it is desirable to treat dental ml of liquid was used. The material was usually restorative problems for extended periods of time mixed on a glass slab, but could also be prepared with a long-term temporary or intermediate restor- by mixmg in a capsule employing a mechanical ative material. Such occasion arises in teeth in mixer. The unreacted eugenol was removed by highly carious mouths, particularly those of chil- blotting or by squeezing the mixed mass in an dren where immediate excavation of all caries is amalgam squeeze cloth. The material presented indicated. This treatment arrests the caries and good manipulative properties and could be readily favorably alters the oral flora. Temporary restora- adapted to cavity walls and margins. Patients were tions are then placed and the permanent restora- recalled and observed periodically during the nine- tions are inserted as scheduling permits. The month observation jDeriod. The restorative did not Armed Forces also have unique dental require- dissolve or disintegrate in the oral fluids. All resto- ments such as emergencies at remote sites or in rations remained serviceable and showed only min- combat zones that preclude the insertion erf perma- imal signs of wear over the nine-month observa- nent restorations. Criteria for an acceptable "in- tion period. All teeth restored with this cement termediate" restorative include [22] : (1) a satis- remained asymptomatic for the entire period of factory seal between the cavity preparation and observation. the material, (2) biologic compatibility with the The formulation selected had a relatively high pulp, (3) easy removal from the cavity prepara- tensile strength, but the in vitro solubility and dis-

108 integration values were considerably higher (one Besides the enhancement of properties of EBA week solubility and disintegration 0.94 percent) cements resiilting from successful research efforts, than those of other promising mixes. Since all it is anticii^ated that a major advance in the near restorations remained serviceable over the nine- future will be the availability of many new com- month observation period, the in vivo solubility mercial products of this type, tailor-made to the did not appear to be of any significance. Unreacted wide spectrum of specific applications needed by liquid in the clinical mixes was removed by blot- the practicing dentist. ting. It is conceivable that lower in vitro solubility The product of the future will combine conven- and disintegration values would have been ob- ience with versatility. It will possess excellent tained if this procedure had been followed in the manipulative properties, will mix with ease to a preparation of laboratory test specimens. predetermined, carefully controlled powder-liquid ratio, will have a setting time that will be rela- 6. Direction of Future Work tively unaffected by environmental parameters such as temperature and humidity, will have a low The EBA cements, largely as a result of the ex- film thickness when required, and if desired, can tensive studies conducted during the last few years, be color coded to indicate the stage of treatment of have passed from their infancy to a state of incipi- the tooth. ent maturity. Further enhancement in physical The findings obtained on pulp capping and on and mechanical properties of EBA and other using EBA cements as sedative restorations and chelate-type cements for use as semipermanent "in- bases point, however, to avenues for further clin- termediate" restoratives would be desirable to ical research. Such studies should establish more make use of their excellent biological properties. clearly any potential advantages of EBA cements However, progress in improving these materials as bases under silicate cement or metallic fillings, may not be as rapid as one has become accustomed as pulp capping oi' cementing media and as inter- to during the last few years. mediate restoratives. Basic investigations that would lead the way to the development of improved products should in- 7. Summary clude pinpointing the exact mechanism of the setting of EBA cements. Determination of the chelate During the last few years considerable interest stability constants of potentially useful chelates as has been generated in improving zinc oxide- well as measurement of their hydrolytic stability eugenol cements. A better understanding of the would be most useful. Synthesis of prospective setting mechanism of these cements has become chelating agents and evaluating of the resulting available which has led to the development of cements would be desirable. products Avith enhanced physical and mechanical The existing fundamental knowledge makes it properties usually containing o-ethoxybenzoic possible to conduct development of better EBA ma- acid (EBA). These materials show the mild reac- terials along many lines. Modest improvements tions to the tissues including the dental pulp simi- may be obtainable by judicious selection of rein- lar to ZOE cements, but also stimulate the forma- forcing agents such as treated or spherical glass tion of reparatiA^e dentin. Physical properties of powders or the partial replacement of zinc oxide the EBA-containing cements approach those of the by another reactive metal oxide, such as mercuric biological and physiological less desirable zinc oxide. Upgrading of formulations by whisker re- phosphate cements. The EBA cements have been inforcement does not look promising. Incorpora- well accepted as luting agents for fixed restoration of slightly soluble fluorides that leach from the tions and as insulating bases. They appear to be the material of choice for indirect pulp capping. EBA cements at a controlled rate should be stud- Resin modified restoratives show good ied with the aim of reducing the solubility of the EBA stress-bearing characteristics and should find ap- components of tooth structure and thereby bestow- plications as an "intermediate" restorative. ing anticariogenic properties to these cements. Further studies of novel chelating agents to par- 8. References tially or wholly replace EBA in cementitious compositions, especially those that yield a strong bond [1] Molnar, E. J., Cloves, oil of cloves and eugenol. at the tooth-restorative interface, are a potentially The medico-dental history. Dent. Items of Interest 971 (June-Oct. 1942). fertile field for developing a greatly superior prod- 64, 521, 663, 745, 876, [2] Brauer, G. M., A review of zinc-oxide eugenol type is realized, uct. Even if such a breakthrough filling materials. Rev. Belg. Med. Dent. 20, No. 3, clinical application of such compositions would be 323 (1965). entirely dependent on the results of histological, [3] Brauer, G. M., New developments in zinc oxide- eugenol cement. Ann. Dent. 26, No. 2, 44 (1967). pathological and clinical studies in order to prove [4] Mitchell, D. F., The irrational qualities of dental beyond a doubt that such a product has the desir- materials. J. Am. Dental Assoc. 59, 954 (1959). able biological properties that have made ZOE [5] Manley, E. B., A review of pulp reactions to chemical irritations. Intern. Dental. J. 1, 36 (1950). and cements so desirable as dental restorative EBA [6] Stanley, H. R., Swerdlow, H., and Buonocore, M. G. materials. • J. Am. Dental Assoc. 75, No. 1, 132 (July 1967).

109 [7] Massler, M., Effects of filling materials on the pulp. [29] Norman, R. D., Phillips, R. W., Swartz, M. L., and N.Y. J. Dent. 26, 183 (1956). Frankiewicz, T., The effect of particle size on the [8] Massler, M. and Ostrovsky, A., Sealing qualities of physical properties of zinc oxide-eugenol mix- various filling materials. J. Dent. Children 21, tures. J. Dental Res. 43, 252 (1964). 228 (1954). [30] Brauer, G. M., Huget, E. F., and Termini, D. J., [9] Turkheim, H. J., In vitro experiments on the bac- Plastic modified EBA cements as temporary re- tericidal effects of zinc-oxide eugenol cement on storative materials, J. Dental Res. 49, 1487 bacteria containing dentin. J. Dental Res. 34, (1970). 295 (1955). [31] Brauer, G. M., McLaughlin, R., and Huget, E. F., [10] Copeland Jr., H. I., Brauer, G. M., Sweeney, W. T., Aluminum oxide as a reinforcing agent for zinc and Forziati, A. F., Setting reaction of zinc oxide oxide-eugenol-o-ethoxybenzoic acid cements. J. and eugenol. J. Research NBS 55, 133 (1955) Dental Res. 47, 622 (1968). RP 2611. [32] Civjan, S., and Brauer, G. M., Clinical behavior of [11] Vieillefosse, R., Vayson de Pradenne, H., and o-ethoxybenzoic acid-eugenol-zinc oxide cements. Zumbrunn, J. P., A study of combinations of J. Dental Res. 44, 80 (1965) types of oxides of zinc-eugenol cements and [33] Phillips, R. W., Swartz, M. L., Norman, R. D., 488 phenol plastics. Rev. Franc. Odontostomat 5, Schell, R. S., and Niblack, B. F., Zinc oxide and (1958). eugenol cements for permanent cementation. J. A., Ryabina, L. V., [12] Gerner, M. M., Zaidorozhnyi, B. Prosthetic Dentistry 19, No. 2, 144 (1968). Batovskii, V. N., and Sharchilev, "V. I., Infrared rg^, Custer, F., and Anderson, R. A., A comparison of spectra of eugenol zinc eugenolate. Russ. J. and o-ethoxybenzoic acid cements. Fortn. Rev. Chicago Phys. Chem. 40, No. 1, 122 (1966). Dent. Soc. 56, No. 3, 9 (1968). [13] Douglas, W. H., Studies of crystal structure of ISO/TC 106 1. (Filling materials) . Draft speci- zinc and magnesium eugenolates. (Abstract) J. ^ WG fication for dental zinc oxide-eugenol cements. Dental Res., 42, 1,108 (1963). [14] Natarajan, S., Gopalakrishna, and Kartz, L., Pro- [36] Bellamy, L. J., The Infrared Spectra of Complex gram and Abstracts of Papers, 50th General Ses- Molecules. 2nd ed. (Wiley and Son, New York, sion, International Association for Dental Re- 1958). search, No. 386 (1972). [37] Sievers, R. E., and Baylor, John C. Jr., Some metal [15] Norling, B. K., and Greener, E. H., X-ray diffrac- chelates of ethylenediamine tetraacetic acid, tion studies of the ZnO-eugenol reaction. Inter- diethylenetriaminepentaacetic acid, and tri- national Association for Dental Research, 46th ethylenetetraminehexaacetic acid. Inorg. Chem. General Meeting. Abstracts of papers. No. 430, 1, 174 (1962). San Francisco (March 1968). [38] Job. P., Formation and stability of inorganic com- [16] Brauer, G. M., and Wiedeman, W. H., Unpublished plexes in solutions. Ann. Chim. [10] 9, 113 results. (1928) ; Concerning hydrochloric acid and hydro- [17] Grant, A. A., Greener, E. H., and Meshii, M., High bromic acid solutions of salts of cobalt, copper resolution microscopy of dental cements. Aus- and bivalent nickel, ibid [11] 6, 97 (1936).

tralian Dental J. 13, No. 4, 295 (Aug. 1968) . [39] Martell, A. E., and Calvin, M., Chemistry of Metal [18] Molnar, E. J., Residual eugenol from zinc oxide- Chelate Compounds, p. 39 (Prentice-Hall, Inc. eugenol compounds. J. Dental. Res. 46, 645 New York, 1952). (1967). [40] McKenzie, H. A., Mellor, D. P., Mills, J. E., and [19] Vieillefosse, R., Hanegraaf, Ch., and Chastagner, N., Short, L. N., Light absorption and magnetic Rev. Franc. Odontostomat. 15, 467 (1968). properties of nickel complexes. J. Proc. Roy. Soc. [20] Curtis, D., U.S. Patent 2,413,294 (Dec. 1946). N. S. Wales 78, 70 (1944). [21] Jendresen, M. D., Phillips, R. W., Swartz, M. L., [41] Phillips, R. W., and Love, D. R., The effect of certain and Norman, R. D., A comparative study of four additive agents on the physical properties of zinc zinc oxide and eugenol formulations as restora- oxide-eugenol mixtures. J. Dental Res. 40, 294 ,tive materials. Part I. J. Prosthetic Dentistry 21, (1961). No. 2, 176 (1969). [42] Coleman, J. M., and Kirk, E. E. J., An assessment of [22] Jendresen, M. D., and Phillips, R. W., A compara- a modified zinc oxide-eugenol cement. Brit. tive study of four zinc oxide and eugenol formu- Dental J. 118, 482 (1965). lations as restorative materials. Part II. J. [43] Bhaskar, S. N., Cutright, D. E., Beasley, J. D., and Prosthetic Dentistry 21, No. 3,300 (1969). Boyers, R. C, Pulpal response to four restocative [23] Brauer, G. M., Morris, R. W. and Howe, W. B. materials. Oral Surg. Oral Med. and Oral Path. Synthesis of isomers of eugenol. J. Res. Nat. Bur. 28, No. 1, 126 (1969).

Stand. (U.S.), 67A (3 ) 253-259 (1963). [44] J0rgensen, K. D., and Hoist, K., The relationship [24] Brauer, G. M., Argentar, H., and Durany, G., Ioni- between the retention of cemented veneer crowns zation constants and reactivity of isomers of and the crushing strengths of cements. Acta eugenol. J. Res. Nat. Bur. Stand. (U.S.), 68A(6) Odont. Scand. 25, No. 4, 355 (1967). 619-625 (1964). [45] Oldham, D. F., Swartz, M. L., and Phillips, R. W., [25] Brauer, G. M., White, E. E. Jr., and Mashonas, M. Retentive properties of dental cements. J. Pros- G., The reaction of metal oxides with o-ethoxy- thetic Dentistry 14, 760 (1964). benzoic acid and other chelating agents. J. Dental [46] Horn, H. R., The cementation of crowns and fixed Res. 37,547 (1958). partial dentures. The Dental Clinics of North B. Co., Phila- [26] Brauer, G. M., Simon, L., and Sangermano, L., Im- America, pp. 65-81, (W. Saunders proved zinc oxide-eugenol type cements. J. Dental delphia, Pa., 1965). Res. 41,1096 (1962). [47] Williams, J. D., Swartz, M. L., and Phillips, R. W., orthodontic as influenced by [27] Brauer, G. M., and Simon, L., Synthesis of 2- Retention of bands Angle Orthodontia 35, No. 4, propoxy-5-methylbenzoic acid. J. Res. Nat. Bur. cementing media. Stand. (U.S.) 66A, (4) 313^317 (1962). 278 (1965). Mosteller, J. evaluation of intermediate base [28] Civjan, S., and Brauer, G. M., Physical properties [^^^ H. An 571 of cements based on zinc-oxide, hydrogenated materials. J. Am. Dental Assoc. 43, (1951). rosin, o-ethoxybenzoic acid and eugenol. J. Dental [49] Messing, J. J., Linings and their manipulation. Res. 43, 281 (1964). Dent. Pract. 8, 336 (1958).

110 [50] Hoppenstand, D. C, and McConnell, D., Mechanical [56] Erausquin, J., and Muruzabal, M., Root canal fill- failure of amalgam restorations with zinc phos- ings with zinc oxide-eugenol cement in the rat phate and zinc oxide-eugenol cement hases. J. molar. Oral. Surg. 24, No. 4, 547 (1967). Dental Res. 39, 899 (1960). [57] Erausquin, J., and Muruzabal, M., Tissue reaction [51] Lyell, J. S., Base forming materials for restora- to root canal cements in the rat molar. Oral Surg.- tions of silver amalgam. Australian Dental J. 5, 26, No. 2, 360 (1968). 132 (1960). [58] Erausquin, J., and Muruzabal, M., Periapical tissue [52] Phillips, R. W. Cavity varnishes and base mate- response to root canal cements with the addition rials. The Dental Clinics of North America, p. of acrylic polymer spherules. Oral Surg. 26, No. 4, 159, (1958). 523 (1968). [53] Chong, W. F., Swartz, M. K and Phillips, R. W., [59] Brauer, G. M., The relationship between laboratory Displacement of cement bases by amalgam con- tests on solubility of zinc oxide-eugenol type ce- densation. J. Am. Dental Assoc. 74, No. 1, 97 ment and their behavior in the mouth. N.Y. J. (1967). Dent. 37, No. 4, 146 (1967). [54] Sahs, E., Radiopacity of pulp capping materials. [60] Norman, R. D., Swartz, M. L., Phillips, R. W., and Northwest Dent. 46, No. 2, 113 (1967). Virmani, R., A comparison of the intraoral dis- [55] Nichols, E., Endodontics (John Wright and Sons integration of three dental cements. J. Am. Den- Ltd. Bristol, England, 1967). tal Assoc. 78, No. 4, 777 (1969).

Ill

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Organic Adhesives

Harvey Alter and Abraham Fookson

Gillette Research Institute, Rockville, Md. 20850

The materials and means for improving joint strengths in a wet environment between dentin or enamel and restoration or cavity liner are reviewed. Critical surface tension, 7c, a descriptive property of surfaces, offers a basis to evaluate capacity of adhesives to wet tooth surfaces. Used as a pretreatment, an adhesive may function by displacing water. Attachment to tooth surface through hydrogen bonding may also occur. Adhesion may be improved by use of liners, coupling agents, fillers and tooth surface treatments such as etching processes. Dental adhesive materials include acrylics, cyanoacrylates, epoxy resins and polyurethanes, the latter having apparently good promise as a dental adhesive of the future.

Key words_: Critical surface tension ; dental adhesion improvement through liners, coupling

agents, fillers, and surface treatment ; dental adhesives to include acrylates, cyanoacrylates,

epoxy resins and polyurethanes ; dental materials polyurethane. ;

itself) , in a wet environment, no matter if the ma- 1 . Introduction terial is to be applied as a restoration, adliesive, or Perhaps this review should have been entitled cavity liner. "Organic Structural Adhesives," or yet more confining, "Synthetic Organic Structural Dental Ad- 2. Some Surface Chemical Considerations hesives." A more confining title would not only better indicate the scope of the discussion to fol- The function of an organic dental adhesive is low, but also would help deliver the important obviously to bond to the tooth substance, whether message that dental adhesives are something spe- the tooth surface be mineral, protein, or other- cial; the requirements for their application and wise in nature. So before discussing adhesives, performance probably impose greater constraints per se, it is worth considering some aspects of the than for any other adhesive application. Certainly, chemistry of the tooth adherend surface. in no other technology would the designer of an Like other surfaces, tooth—either dentin or adhesive joint want to be restricted to a medically enamel—may be described in terms of its critical acceptable, low temperature and fast setting, high surface tension, i.e., the maximum surface tension strength adhesive, which at the time of applica- of a liquid which will wet or form a zero contact tion and over the long term, must be resistant to angle. The now familiar critical surface tension, water. Indeed, some adhesive experts, for example or yc, concept has been developed by Zisman and in the coatings or aerospace fields, may well doubt co-workers [1] ^ and has been related to adhesion that such requirements could ever be met. and abhesion [2]. The yc value of an adherend But certainly many of the requirements have solid then teaches what the surface tension of an been met. Organic polymeric restoratives are now adhesive must be to wet and spread. Also, the yc used in clinical practice, and many truly adhesive values are useful for ranking solids as high or low dental polymeric materials are currently under energy surfaces. For example, table 1 following, investigation. This review is intended to cover from Zisman [1], ranks some polymer surfaces, some of the more promising approaches of research a low value of yc indicating a low energy surface. in new adhesive systems, and to discuss means for It is generally believed that metals, minerals, enhancing joint strength with these materials. Ee- and certain other inorganic solids have high energy gretfiilly, the scope of this review must be limited surfaces, i.e., most liquids will spread on them. and cannot include all of the pertinent published Apparently, this is not true for tooth surfaces; work, the recent information on biological adhe- Uy and Chang [3] measured yc of dentin and 100 percent relative hu- sives (such as from the barnacle) , nor delve into enamel surfaces at 50 and differences among dental adhesives, adhesive midity (RH). Their results are given in table 2. restoratives, and cavity liners. This review is re- On comparing table 2 with table 1, the yc of stricted to materials and means for improving the tooth surfaces at 23 °C and 50 percent RH is joint strength of bonds between dentin or enamel 1 Figures In brackets Indicate the literature references at the and another member (which may be the adhesive end of this paper.

113 Table 1. Critical surface tensions of various polymeric tooth siurfaces. Thus, a nonmiscible, nonhydrogen solids [1] bonding adhesive (or adhesive solution) applied to tooth substance musit wet and spread on a low Polymeric solid Dynes/cm. energy surface of condensed water, as measured at 20 °C. by Uy and Chang [3] , and as might be expected to occur in the high humidity oral environment. Polymethacrylic ester of <^'-octanol 10. 6 Alternatively, the adliesive may displace the Polyhexafluoropropylene 16. 2 water from the surface. Such displacement is ac- Polytetrafluoroethylene 18. 5 complished largely by surface-chemical Polytrifluor oethylene 22 action, Poly(vinylidene fluoride) 25 rather than merely d.issolving away the water Poly(vinyl fluoride) 28 layer. The mechanism of water displacement by an Polyethylene 31 organic liquid has recently been analyzed by Zis- Polytrifluorochloroetbylene 31 Polystyrene 33 man and co-workers [7 J . It may be summarized as Poly (vinyl alcohol) 37 follows. Poly (methyl methacrylate)-.. 39 The "initial spreading coefficient," Sia, of liquid Poly (vinyl chloride) 39 b displacing liquid a, is related to the surface and Poly(vinylidene chloride) 40

interfacial : Poly (ethylene terephthalate) 43 tensions y by Poly(hexamethylene adipamide) 46 Sba = ya~ (ys + yab)

Generailly, when liquid a is water, for any homolo- Table 2. Critical surface tensions, dynes/cm \S] gous series of h liquids, Sm increases as the solubility in water increases and as the boiling point Surface 37 °C. (100 23 °C. (50 decreases. A high value of Si,a is desirable. This is percent R.H.) percent R.H.) summarized in figure 1, taken from reference 7. Note the high values, corresponding to effective Human dentin 39. 5 39-49 water displacement for some familiar liquids such enamel 31. 5 38. 5-40 Human as ethyl ether and ethyl acetate. Bovine dentin 28-31 38-40. 5 The displacement of water from a surface is Bovine enamel _ 24-49 38-42. 5 also facilitated by having the liquid partially mis- cible with water. For example, in figure 2 (also taken from reference 7), high values of initial about the value of polyvinyl chloride. Looking at spreading coefficient are associated with the solubil- the lower numbers for at 37 °C and 100 percent yc ity in water for a series of alcohols. The area of RH, the tooth surface appears to be like that of effective water displacement is clearly marked and polyethylene or lower surface energy polymer. At corresponds to compounds such as 1-butanol, 2- the time of publication, such conclusions were diffi- metliyl-l-propanol, 1-pentanol, and 2-methyl-2- cult to understand. Although Uy and Chang may butanol. be criticized for using many H-bonding liquids Tlius, in order to wet the tooth surface, the sur- in making the measurements, the final yc values are nonetheless an experimental result which today, face tension of the adhesive must be lower than yc, because of more recent publications by others can which may be as low as 24 dynes/cm, the yc of a be interpreted. condensed water layer. This is true only if the Other workers have examined supposedly high 60 energy surfaces at high humidities. Shafrin and • n-ALCOHOLS Zisman [4] found that the effect of water on the o CELLOSOLVES

• spreading of organic liquids on glass was to cause 5 50 A KETONES ACETATES an apparent change in the critical surface tension o ACETOACETATES from about 1 c; 46 dynes/cm at percent EH to 30 40 ETHERS dynes/cm at 95 percent for the same glass surface. Essentially similar results were found by Bemett 30 and Zisman [6] for borosilioate glass, quartz, and S sapphire surfaces (again all supposedly high energy surfaces) and by Alter and Cook [6] for 20 keratin fiber surfaces. In the latter study it was found that the yc of native and oxidized hair fibers ranged from 34 to 25 dynes/cm from 1 percent to 95 percent R.H, respectively. The higher yc is about what may be expected for a protein surface [6]. i 40 80 120 160 200 240 280 320 The lower values, 25 to 30 dynes/cm, are inter- BOILING POINT (°C) preted as measures of the critical surface tension for wetting a condensed water layer, whether the Figure 1. Initial spreading coefficient versus toiling point water layer is on glass, sapphire, keratin, or now, for various aliphatic %cater-displacing compounds [7].

114 1 1

1 1 Although the conventional aerylate dental mate-

1 1 rials do not adhere well to tooth substance, they 1 1 . 1

1 1 io illustrate a general means of applying adhesive, 1 1 1+ T viz., permitting the adhesive to 1 polymerize in place

1 to give monomer and polymer an opportunity to * absorb, and thus produce higher bond strengths. — 1 EFFECTIVE WATER

> 1 — DISPLACEMENT The importance of polymerizing the adhesive in {

1 place has been demonstrated at least for amine ^ 1 POOR WATER 1

/ ' DISPUCEMENT cured epoxy adhesives [14] ; the importance of 1 /o NONE 1 adsorption as an early and necessary step in the j

1 lo 1 • NORMAL ALCOHOLS adhesion process is generally accepted. 1 1

1 o BRANCHED ALCOHOLS An improved dental restorative material based

1 1 acrylic 1 on an copolymer has been one of the major 1 1 1 contributions from the dental research at the Na- 1 1

' 1 tional 1 Bureau of Standards. Bowen [15-18] has 100 0.01 0.1 1.0 10 described a composite restorative system based on SOLUBILITY IN WATER AT 20 °C (WEIGHT PER CENT) acrylic copolymers and their further improvement by attention to tooth surface preparation, mode of Figure 2. Relation of ivator soluMlity to initial spreading filling, and adhesion of the restorative to the filler coefficient and loater displacing aMlity of simple niono- hydric alcohols. particles. The resultant materials appear to be im- The vertical lines through points in the region of strong displace- proved restoratives, rather than new adliesives ment are proportional in length to the areas of a 2-mm water film displaced by the respective alcohols [T], [19]. 3.2. Cyanoacrylates adhesive is not hydrogen bonding. Alternatively, A relatively new class of acrylates, the alkyl the adhesive (as a pretreatment for the surface) a-cyano-acrylates, is being investigated as a pos- may displace the water layer. A third alternative sible restorative [20]. This class of adhesives is of is that the adhesive is hydrogen bonding, such as an epoxy or polyurethane. These types of adhe- interest because the monomers polymerize readily sives are discussed in later sections. with the aid of a weak base or water as an initiator. Interestingly, a general precaution in applying 3. Dental Adhesive Materials the cyanoacrylates as adhesives is to dry the adherend surfaces, relying on only trace water vapor 3.1. Acrylics as an initiator, and to avoid acidic surfaces. Long exposure to water, or water vapor, can have a Acrylic polymers, such as poly (methyl meth- deleterious effect on the bond strength [21]. acrylate), have long been used as restoratives [8] The polymerization of the alkyl a-cyanoacry- and much careful study has been devoted to these lates is highly exothermic, resulting in a peak materials, particularly at the National Bureau of polymerization temperature which may be uncom- Standards and by the group this symposium fortable in vivo. This peak temperature, and the honors [9, 10]. The attributes and deficiencies of ability of the monomers to displace blood (and this class of materials are well known and it is

probably water) , as well as certain of the proper- generally recognized that for all their merit in ties of the polymerized mass, such as mechanical clinical practice, they do not adhere to tooth sub- strength, rate of hydrolysis, and histotoxicity, stance. Some improvement in their adhesion can depend on the length of the alkyl side chain be obtained, at least qualitatively, by appropriate [20, 23]. The chemistry of this class of adhesive copolymerization [11]. 22, has been discussed in the context of use as tissue The acrylics are usually applied as a mixture of adhesive in surgical procedures and for hemostasis polymer, monomer, and polymerization catalyst. [22, 23]. Use in dentistry has been described by One reason for the widespread use of acrylics is Bhaskar and Frisch [24] and by Collito [25]. that such mixtures can polymerize in place quickly, particularly when an accelerator is included [12]. 3.3, Epoxy Adhesives The polymerization of such mixtures to polymer is usually accomplished by considerable The use of epoxy resins as dental restoratives shrinkage, 6 percent or more by volume [13], an was described early by Bowen [26]. Later, Lee intolerable level for adhesive bonding under any [27] reported additional experimental work di- circumstances. Means of avoiding the shrinkage rected at an epoxy material for clinical dental use. are the common pracitices of incorporating partic- Also, Lee and co-workers [28, 29] have synthesized ulate filler, dissolving polymer in the monomer new resins with the aim of achieving high bond before polymerization, and using the "brush" strengths. They found that a bulk restorative resin technique. could be made from the triglycidyl ether of tri-

115 hydroxybiphenyl cured with cyclohexane bis- work was dropped [32]. In short, epoxy resins, (methylamine) and that the cured material had a for all of their attributes in adhesives technology, good balance of properties for a restorative mate- do not appear to be broadly suited to dental rial. The resin sets moderately slowly and requires practice. mixing on a hot plate prior to insertion to achieve 3.4. Polyurethanes clinically feasible set times. Lee and co-workers [28] have synthesized and As used for dental adhesives, the term poly- investigated many types of epoxy resins as well urethane refers generally to the condensation as several different types of curing systems. None polymer between a polyol (derived e.g. from a seem feasible at this time for various reasons. polyester or polyether), and a polyfunctional Promising results were obtained by blending epoxy isocyanate. Polyurethane formation may be sum-

resins with polyurethanes ; the latter are reviewed marized as in figure 3. in the next section. The condensation reaction can be rapid at room The most promising of the epoxy restorative temperature and may be catalyzed to take place resins, one based on the triglycidyl ether of tri- in a few minutes by weak bases such as tin ootoate hydroxybiphenyl cured with cyclohexane bis- or tertiary amines. The isocyanates are applied

(methylamine) , was used in limited clinical trials as adhesives similar to acrylics or epoxies in that [29]. This resin system, filled with alumina, main- mixture of the monomers or prepolymers and the tained adhesion to human dentin in vivo when cross-linking agent (TDI in the example in figure placed in restorations prepared without under- 3) are applied to the joint and permitted to poly- cuts for at least three months [29]. merize in place. The clinical results were obtained with Epoxy- An interesting aspect of the chemistry is that lite resin NIH-27. If this material must be applied organic isocyanates will react with water to gen- in a mamier similar to Epoxylite resin NIH-7, erate CO2. (This is the basis of some urethane theia prior to inserting the restoration the epoxy foam production.) The reaction between tolylene resin and its hardener must be mixed on a tem- diisocyanate (TDI) and water, is shown in figure perature-controlled hot plate for 60-90 s, and this 4. This reaction scheme provides for excess iso- mixing must be well controlled. Then, the finish- cyanate in a urethane adhesive formulation acting ing should not be attempted for at least 30 mm as a Avater scavenger for water on the tooth sur- after mixing and preferably on a subsequent visit face. In this way, the polyurethane adhesive may by the patient to the dentist [29]. No adhesion bond to a dry (or temporarily dry) surface. The test data are given trying to relate the laboratory CO2 generated should be able to escape from thin procedures with the clinical trials [28, 29]. sections, such as a cavity liner. The water scaveng- Epoxy resins have also been used for bonding ing action and any effect it may have on dental orthodontic attachments to teeth [30]. The bond adhesions are speculative at this point. strengths of commercial epoxy adhesives under CHj laboratory conditions were reported to be in excess of the forces judged to be exerted on the orthodon- HO-R-OH+OCNi -NCO HO-R-O-C-N + Q + NCO o H tic attachment during treatment, and this finding- polyol \y was verified in limited clinical trials. This report [30] is brief and qualitative; it is impossible to tell from it exactly what was done. polyol HO - *- R-O-C-N+OfNCO HO-R-O-C-N+Ol-N-C-O-R-OH^-^ Although epoxy resins are useful structural ad- II 1 I I I II O H O H hesives for many applications, and despite the TDI large amount of research performed relative to their use in dentistry, such use seems severely limited. The reason for this is that generally, high ocN-^O to polymer bond strengths are obtained only after high tem- Jl'-|r"°~''~°"lr~';'HO OH {Of perature bakes and/or mixing of resin and hardener at elevated temperature. Typical curing Figure 3. Polyurethane formation. agents are polyfunctional amines, of questionable utility in contact 'with tissue of any sort [31]. CHj CH3 Even the so-called room temperature curing form- ^"aO ulations of epoxy adhesives require several hours |0)-NCO > 0CN-(0)-N-C-0H to develop their maximum strength, well beyond TDI substituted carbamic acid reasonable clinical practice times. In our laboratory we experimented briefly with epoxy dental C H 1 C H 1 C H3 adhesives which would cure hard in a few minutes at room temperature, even in the presence of water. X/ H O H \/ These formulations were catalyzed by BF3 and substituted urea its complexes, compounds not to come in contact with epidermis, let alone other tissue, and the Figure 4. Behavior of TDI in the presence of water.

116 . 1

The first description of polyurethane adhesives tions, and prevented water from entering the for dental applications was by Galligan, Schwartz restoration margin [33]. and Minor [33] who investigated adhesive liners The adhesion of the polyurethane can be im- based on equal parts by weight of TDI and a proved by other choices of polyol and isocyanate. poly (propylene glycol) of 2,000 nominal molecu- At the present time in our laboratory, we gen- lar weight. This composition contained a tenfold erally obtain higher bond strengths with the excess of isocyanate groups; 0.01 percent tripropyl- polyurethane from castor oil (triglyceride of amine was in some instances added as a catalyst. ricinoleic acid) cross-linked with PAPI - (a poly- The composition was tested as an adhesive liner phenylene isocyanate from Upjolin) than with for conventional restorations in freshly extracted, the polyurethane reported by Galligan et al. [33]. sound human premolars in conical cavities. (The However, the joint strength can be mcreased even cavities were "dried" with a blast of air for 5 s more by choice of filler, other additives, coupling before the liner was applied.) A summary of the agents, and means of preparing the tooth surface, data is given in tables 3 and 4. (The reader is than by making small changes in the choice of referred to reference 33 for the experimental reactants in the adhesive. Such means to increase methods.) In every case, including when the ure- adhesion are discussed in the next section. thane was used as a restoration, an improvement in properties was obtained with the test composi- 4. Means to Improve Adhesion tion compared to the results for the conventional restorative materials. Noteworthy, the joint Several means have been proposed to improve strengths of the acrylic restorations were increased dental adhesives by bonding the adhesive chemi- above zero, the value for the acrylic alone. The cally to the mineral or protein portion of teeth. liners were adhesive to teeth, and to the restora- For example, Buonocore [34] investigated the inclusion of glycerophosphoric acid dimethacrylate in an acrylic restorative with generally encourag- Table 3. Adhesion of restorations to teeth [33] ing results and obtained evidence to suggest that there may be some chemical reaction between the Com- Thermal components of the adhesive and the organic matter Restoration Strength pres- Shock of dentin. Similarly, Lee et al. [28] suggested sion cross-linking the collagen portion of dentin with zirconium acetate to render it hydrophobic and lb N Cycles Cycles thus improve bond strengths. Brauer et al. have Acrylic _ - . 0. 0 0 0 1 suggested using chelating agents [35] between the Urethane-lined* acrylic... 8. 3 37 >400 >40 adhesive and mineral portion of the tooth as well Amalgam 0. 3 1 0 1 as the eerie ion induced grafting of synthetic Urethane-lined* amalgam. 8. 6 38 240 >40 Silicate.. 5. 5 24 >400 15 polymers to the protein portion [36]. Other sug- Zinc phosphate 5. 1 23 220 11 gestions have been the use of rubbery liners, cou- Urethane* . .. ._ .. 3, 0 13 >400 >40 pling agents (such as silanes) , proper choice of fillers, and acid and enzyme debridement of the pre- *The urethane was the product obtained by curing a 1 : pared tooth surface to improve adhesion or bond mixture of tolylene diisocyanate and polypropylene glycol" strengtli. These latter suggestions are reviewed in some detail below. Table 4. Tensile strength of urethane-lined acrylic restorations [33] 4.1. Rubbery Liners

The adliesive failure of a joint the Urethane liner Tensile means that strength joint strength was exceeded by the applied stress. In practice, this stress may be tensile, compressive, or shear, and usually is a complex mode which Iso- often involves cya- Polyglycol a shear component. (See for ex- nate* ample, reference 37.) Dental adhesive joints are usually tested in tension [38] and there have been lb N no reports of a stress analysis of any dental ad- TDI Polypropylene glycol ("2010") 8. 3 37 TDI Ethylenediamine-based hesive test. Patrick et al. [39, 40] have suggested polyethylene-propylene that tests of dental adhesive joints include a shear glycol . - 9. 1 40 Polypropylene glycol ("1010") 4. 3 19 PAPI 'PAPI is: PAPI Polypropylene glycol ("2010") 13. 4 60 NCO NCO PAPI Polyethylene-propylene glycol . 6. 6 29 PAPI Trimethylolpropane based glycol.. _ . .. 8. 9 40 polypropylene Q)--C-H.2-

*TDI, tolylene diisocyanate; PAPI, polymethylene polyphenylisocyanate where n is approximately 1.

117 )

component, as well, and have devised a punch test The extension of coupling agent technology to for this purpose. Also, they pointed out that a dental practice is a logical one, and many workers joint between high modulus dental tissue and a have done so (e.g., references per- 27, 39, 40, 42) ; I'elatively brittle (although having a 20 times haps the earliest use was by Bowen [16, 43]. There lower modulus) resin produces a mechanical situ- is little question that coupling agents, particularly ation which enhances joint failure close to the in- silanes, improve the adhesion of joints with dental terface. Therefore, they proposed using a rubbery materials. Two examples are worth citing. liner between dental tissue and restorative, to pro- Patrick et al. [39], using the shear punch test vide a zone to absorb stress. Obviously, the rubbery described earlier, demonstrated the value of a si- liner must be adherent to the restorative and to the lane coupling agent with the rubbery liner. Some tissue. of their data, reproduced in figure 5, illustrate the Patrick et al. [39, 40] used a rubbery liner of higher bond strengths and greater resistance to partially saponified poly(acrylonitrire-co-butyl water immersion (in this case, synthetic saliva at acrylate) and tested composite joints of bovine 37 °C) foT joints prepared with silane coupling dentin (or einamel) -rubbery liner-epoxy resin by agent, compared to those without. Each point on their shear-punch test. These composite joints had this figure represents the average test result of an average test strength of 3415 ± 198 psi (23.6 ± from 15 to 120 specimens. 1.4 MN/m^) without the rubbei-y liner and 4300 ± In our laboratory, we have demonstrated that 254 psi (29.7 ± 1.8 MN/m^) with the liner. In ad- improved bond strengths are obtained (tested after dition, tlae streng'th of the joint Avith liner was less 3 days immersion in water) for acrylic restorations affected by thermocycling (4 °C, 10s; 65 °C, 10s; bonded to either bovine enamel or human dentin for each cycle) than similar joints without liner when a selected coupling agent is incorporated in [40]. either the acrylic or the polyurethane liner used

in this work . selected results are shown 4.2. Coupling Agents [44] Some in table 5. The term coupling agent usually refers to a com- The joint strengths were measured by the butt pound which can form a bond between the adhe- test of Lee et al. [38]. The polyurethane liner was based on and casitor oil and filled with sive and adherend. Frequently, such agents are PAPI titania. The (poly (methyl methacrylate) substituted silanes and are applied to the adherend PMMA was applied as preformed rod wetted with mono- surface before applying the adliesive. Such coup- mer, to simulate an acrylic restorative. The ration- ling agents are frequently used in glass-polymer ale behind the choice of coupling agent was to composites; one of their principal functions is to chose compounds with reactive moieties which will reduce the moisture sensitivity of the composite incorporate with both the acrylic and urethane [41]. Presumably, they will have the same func- polymers on polymerization. The compounds were tion for dental restoratives. Noteworthy, coupling mixed in the polyurethane before applying and agents need not be silanes, although, silanes are curing; their effect was to near double the joint used most often. strength of the cured system. (The formic acid

Shear value (psi) 5, 000

4,000

/\ Silane + liner

Liner

^3 Silane

o No treatment Expoxide restorations with hydroxylapatite filler ^^^^^

I 100 150 200 Consecutive hours

Figure 5. Effect of exposure time in 37 °C saliva on shear values as determined from model bovine restorations [39]. To convert shear value In psi to MN/m» multiply by 0.006895.

118 . . ; ;

Table 5. Effect of coupling agents on adhesion of PMMA to formic acid treated bovine enamel [44]

Amount coupling agent, phr* Joint strength with coupHng agents

a b c d

psi psi psi MN/m^ psi MN/m 2 0 700 4. 8 700 4. 8 700 4. 8 700 4. 8 2 1270 8. 8 1070 7. 4 1160 8. 0 1190 8. 2 5 1125 7. 8 1380 9. 5 1000 6. 9 1340 9. 2

a.2-hydroxyethyl methacrylate; b.2-hydroxypropyl methacrylate c . 2-t-butylaminoethyl methacrylate d.l, 3-di(allyloxy)-2-hydroxymethyl propane. *Parts per hundred parts of resin.

treatment is explained in a later section.) Similar which are known chelating agents for Ca ion. results have been obtained with other coupling Since then, several other workers have reported agents, including silanes, and will be repoi-ted the use of acids, sequestering agents, and later [M]. enzymes for the etching or debridement of 4.3. Fillers tooth surfaces prior to applying an adhesive or restorative (for example, references 28, 34, Particulate fillers are usually incorporated in 36, 42, 44). This appears to be a highly effec- an adhesive for several reasons, such as to better tive general means for improving the adhe- match the thermal expansion coefficients of ad- sion of dental adhesives. Also, the recent wide- herend and adhesive and thus improve bond per- spread use of the scanning electron microscope formance. Bowen has shown that the strength of (SEM) for examination of the treated dental sur- dental restoratives can be improved by proper at- faces has provided a clear understanding of what tention to filler particle size and distribution and, these debridement treatments are achieving. This is interestingly, the importance of using a silane illustrated below with examples from our own coupling agent to improve the bond between filler work. The use of the SEM in studying dental and resin [15]. Kelated to this, Alter has anatomy was recently reported [49]. shown [45] that thermoplastic polymers and rub- Figure 6 shows a series of SEM photographs bers are reinforced by certain particulate mineral tracing the appearance of bovine enamel ground fillers, depending very much on their particle size. wet with successively finer grit polishing paper. Reinforced thermoplastic polymers, especially using fillers pretreated with silane coupling, are Tapt.-r 6. Increase in adhesive joint strength of PAPI-castor articles of commerce [46] oil polyurethane due to addition of fillers [42] The effect of fillers on the adhesion of polyurethanes to dental tissue was demonstrated by Average Filler Amount force to Fookson and Ellison [42] . Table 6 shows some of break their data for the large increase (as much as four-fold) in the force to remove the filled polyurethane restoratives from a conical cavity in phr* lbs N human dentin after 3 days water immersion. Ad- None _ . _ . 7 31 ditional data illustrating the importance of proper ZnO, USP 50 12 53 choice and amount of filler in polyurethane dental ZnO, USP 100 19 84 materials will be reported later [44] Ti02, reagent grade 50 18 80 TiOj, reagent grade - 100 25 111 4.4. Surface Preparation

Super Floss " _ _ _ 50 18 80 In 1955 Buonocore [47] reported that a phosphoric acid and phosphomolybdate-oxalic acid Snow Floss ' - . 37 27 120 treatment of dental enamel increased the adhesion Adsorption alumina _ . . 50 31 138 of acrylic filling materials. More recently, New-

man and Sharpe [48] showed that one effect of Zr02, electronic grade - 100 10 44 such a treatment is to make the tooth surface Zr02, electronic grade - - 150 9 40 more wettable. Asbestos, crude - - 12K 14 62 In 1956, Bowen reported [26] an improvement Asbestos, crude _ _ 25 15 67 in the adhesive strength and moisture resistance of epoxy resins to dental tissue by washing the » A calcined diatomaceous earth—Johns-ManviUe Company, Incorporated. cut surface of tooth with a solution of ammonium i> An uncalcined diatomaceous earth—Johns-Manvllle Company, Incor- porated. triacetic acid or ethylenediamine tetra-acetic add, •Parts per hundred parts of resin.

119 120 and finally with y-alumina (Buhler), a very fine treatment with dilute acids and other agents, some particle size polishing compound. Although this undoubtedly physiologically acceptable. successive finer and finer polishing removes much of the debris from cutting, and although the speci- 5. Some Directions for the Future men in the final photo appeared to be highly polished to the eye, the final specimen can still Some of the state of the art reviewed here leads be considered rough and the enamel anatomy is to possible future directions for research in organic not visible. Figure 7 shows the effect of treating dental adhesives. So, we may gaze into a very the polished enamel with dilute acids for a brief time. The final polishing debris is removed and the rod structure of the enamel is clearly evident. Figure 8 shows a similar effect on dentin where the sample polished with y-alumina shows a rough surface and it is difficult to distinguish the farailar tubular structure of the dentin. Figure 9 shows that a brief etch with dilute acid again removed the debris and the tubular structure is clearly evident. Similar results have been obtained with human tooth tissue. It is possible to etch too far, as shown in figure 10. Here, a good deal of the surface tooth tissue was removed, exposing the deli- cate tubular labyrinth structure. Suoh a surface, as an adherend, could have la weak boundary layer (there is insufficient mineral to support the bond) and the joint shows a low strength. Surfaces of the types shown in the SEM photographs were used to make adhesive joints with a polyurethane adhesive and were tested by the butt joint method. Some selected results are given in tables 7 and 8. Work currently in progress in our laboratory extends these observations. There is every indication that bond strengths of polyurethane adhesives Figures. Bovine, dentin polished with y-alumina (Magni- to dental tissue can be greatly increased by short fication: 3000X).

Figure 7. Bovine enamel etched with 50 percent aqueous Figure 9. Bovine dentin treated with 50 percent aqueous formic acid for one minute {Magnification: 3000X). formic acid for one minute {Magnification: 3000X).

121 maybe an enzyme or sequestrant. The etched surface will be dried, maybe by surface-chemical displacement, and then a coupling agent, most likely a silane, applied. Alternatively, the coupling agent will be incorporated in the adhesive. Today, it appears that the most likely adhesive to be used will be a polyurethane. The reason for this is that the polyurethanes have the right bal- a,nce of properties: low toxicity (judged from the

literature) , controlled and short cure times within clinical practice, and apparently adequate mechanical and adhesive strength. The polyurethane of the future will probably be one based on PAPI, which is the isocyanate of least toxicity [50], so far, and the polyol will be derived from a polyether, because polyester derived ones have poor long-term hydrolytic stability in other applications [51]. The polyurethanes we have been using can be cured in clinically feasible time®, controlled from 5 to 20 min. The polyurethane will be filled and the filler may well be pretreated with a silane coupling agent. In addition, the particle size and size distribution of the filler will be carefully controlled. filled Figure 10. Excessively etched 'bovine dentin {Magnifi- The polyurethane may well -be the resto- cation: 3000X). rative itself, although, this seems unnecessary. There now exist many good restorative materials; it is just that they are not adhesive. We have found Table 7. Adhesion filled PAPI-castor oil resin to of that polyurethanes are adhesive not only to dental untreated and formic acid etched bovine tooth tissue tissue, but also to acrylics (especially with the right coupling agent) and, in preliminary work, Surface Treatment Joint strength to dental amalgam. Much research has yet to be done before this psi MNjnfi type of dental restoration can be clinically tested. Dentin.. None .. 173 1. 2 Formic acid " 228 1. 6 Indeed, much of the above speculation is based on Enamel. None - . . 373 2. 6 the assumption that in vitro adhesive j,oint testing Formic acid °. >> 705 4. 9 is translatable to the in vivo situation. This assumption will be tested soon in our laboratory in " 50 percent acid for one minute, then wash. animals. Hopefully, other laboratories will do Cohesive failure. similar work in the near future.

Table 8. Adhesion of filled PAPI-castor oil resin to citric acid treated bovine enamel The work reviewed here, the new results reported from our own research, and indeed the Joint strength Treatment time speculation as to the future, are all in large part possible because of the pioneering and active den- 25% aq. acid 50% aq. acid tal research program carried out these past 50 years at the National Bureau of Standards. Our minutes psi psi thanks to this group, for making it pos'sible, and 0 410 2. 8 410 2. 8 best wishes for continued contributions to dental 940 6. 5 1420 9. 8 1 1230 8. 5 1360 9. 4 research. 5 970 6. 7 1410 9. 7 6. References

[1] Zisman, W. A., Advan. Chem. Ser., 43, 1 (1964). [2] Zisman, W. A., NRL Report 5699, November 29, cloudy crystal ball and speculate what an ideal (1961). adhesive system for clinical practice might be some [3] Uy, K. C, and Chang, R., Adhesive Restorative years from now. Materials—II, Proc. 2nd Workshop, Bio-materials Advisory Committee, Nat. Inst. Dental Res., In preparing the restoration, no undercutting Res. (1965). will be necessary, the freshly cut tooth isurface will [4] Shafrin, E. G., and Zisman, W. A., J. Am. Ceramic be etched with a dilute acid (such as citric) and/or Soc, 50, 478 (1967).

122 . —

Lee, [5] Bernett, M. K. and Zisman, W. A., J. Colloid and [29 H. L., Cupples, A. L., and Swartz, M. L., Com- Interf. Sci., 29, 413 (1969). prehensive Report No. 67-112 to Nat. Inst. Dental [6] Alter H., and Cook, H., J. Colloid and Interf. Sci., Res., June (1967). 29.439 (1969). [30 Retief, O. H., and Dreyer, D. J., J. Dent. Assoc. R., [7] Baker, H. R., Leach, P. B., Singleterry, C. and South Africa, 22, 338 (1967). Zisman, W. A., Ind. and Eng. Chem., 59 (No. 6), [31 Skeist, I., Ch. 25 in Handbook of Adhesives, I. 29 (1967). Skeist, ed., (Reinhold Pub. Co., New York, 1962). [8] Skinner, E. W. and Phillips, R. W., The Science of [32 Alter, H., Unpublished. Dental Materials, Chps. 11 and 14 (W. B. Saun- [33 Galligan, J. D., Schwartz, A. M., and Minor, F. W., ders Co., Philadelphia, 6th ed., 1967). J. Dental Res., 47, 629 (1968). [9] PafEenbarger, G. C, Nelson, R. J., and Sweeney, [34 Buonoeore, M. J., Adhesive Restorative Dental Ma- W. T., J. Am. Dental Assoc. 47, 516 (1953). terials— I, Proc. 1st Workshop, Bio-Materials Res. [10] Coy, H. D., J. Am. Dental Assoc. 47, 532 (1953). Advisory Committee, Nat. Inst. Dental Res. 1961. [11] Schwartz, A. M., and Galligan, J. D., Adhesive [35 Driessens, F. C. M., Brauer, G. M., and Termini, Restorative Materials—II, Proc. 2nd Workshop D. J., lADR Program of Abstracts and Papers, of the Bio-materials Research Advisory Commit- No. 137 (1969). tee, Nat. Inst. Dental Res., (1965). [36 Brauer, G. M., and Termini, D. J., lADR Program of [12] Brauer, G. M., Davenport, R. M., and Hansen, Abstracts and Papers, No. 404 (1970).

W. C, Modern Plastics, 34 (No. 8), 153 (1956) ; [37 Patrick, R. L., ed., Treatise on Adhesion and Ad- Brauer, G. M. and Burns, F. R., J. Poly. Sci., 19, hesives, Vol. 1 (M. Dekker, New York, 1967). 311 (1956). [38 Lee, H. L., Swartz, M. L., and Culp, G., J. Dental [13] Brauer, G. M., J. Am. Dental Assoc. 72, 1151 (1966) Res., 48, 211 (1969). See also reference 8, p. 225. [39 Patrick, R. L., Kaplan, C. M. and Beaver, E. R., J. [14] Alter, H. and SoUer, W., Ind. Eng. Chem., 50, 922 Dental Res., 47, 12 (1968). (1958). [40 Patrick, R. L., and Kaplan, C. M., Adhesive Restora- [15] Bowen, R. L., J. Am. Dental Assoc. 69, 481 (1964). tive Materials—II, Proc. 2nd Workshop, Bio- [16] Bowen, R. L., J. Am. Dental Assoc. 74, 439 (1967). materials Res. Advisory Committee, Nat. Inst. [17] Bowen, R. L., J. Dental Res., 44, 895, 903, 906 Dental Res., (1965). (1965). [41 Brelant, S., in Treatise on Adhesion and Adhesives, [18] Bowen, R. L., U.S. Patents 3,194,783 and 3,194,784 Vol. 2, R. L. Patrick, ed., (M. Dekker, New York, (1965). 1969). [19] Lee, H. L. and Swartz, M. L., J. Dental Res., 49, [42 Fookson, A., and Ellison, A. H., lADR Program of 149 (1970). Abstracts and Papers No. 136 (1969). [20] Civjan, S., Margetis, P. M., and Reddick, R. L., [43 Bowen, R. L., Adhesive Restorative Dental Mate- J. Dental Res., 48, 536 (1969). rials—I, Proc. 1st Workshop, Bio-materials Res. [21] Coover, H. W., Jr., Ch. 31 in Handbook of Adhesives, Advisory Committee, Nat. Inst. Dental Res., I. Skeist, ed., (Reinhold Pub. Co., New York, (1961). 1962). [44 Ellison, A. H., and Fookson, A., Annual Comprehen- [22] Lehman, R. A., Hayes, G. J. and Leonard, F., Arch. sive Report to Nat. Inst. Dental Res. for 1968- Surg., 93, 441 (1966). 1969. [23] Woodward, S. C, Herrmann, J. B., Cameron, J. L., [45 Alter, H., J. Applied Poly. Sci., 9, 1525 (1965). Brandes G., Pulaski, E. J., and Leonard, F., Ann. [46 Sterman, S., and Marsden, J. G., Modern Plastics, Surg. 162, 113 (1965). 43, (No. 7), 133 (July, 1966). [24] Bhaskar, S. N. and Frisch, J., J. Am. Dental Assoc. [47 Buonoeore, M. G., J. Dental Res., 34, 849 (1955). 77, 831 (1968). [48 Newman, G. V., and Sharpe, L. H., J. New Jersey [25] CoUito, M. B., U.S. Patent 8,250,002 (1966). State Dental Soc, 37, 289 (1966). [26] Bowen, R. L., J. Dental Res., 35, 360 (1956). [49 Hoffman, S., McEwan, W. S., and Drew, C. M., J. [27] Lee, H. L., Adhesive Restorative Materials—II, Dental Res., 48, 242 (1969). Proceedings 2nd Workshop, Bio-materials Res. [50 Anon., Tech. Bulletin 105, Engineering, Mechanical Advisory Committee, Nat. Inst. Dental Res. Control and Toxicological Considerations (1965). Urethanes, The Upjohn Co., Kalamazoo, Mich. [28] Lee, H. L., Cupples, A. L. and Swartz, M. L., Annual (July, 1968). Comprehensive Report No. 68-140 to Nat. Inst. [51] Gahimer, F. H. and Nieski, F. W., Insulation, p. 89, Dental Res., June (1968). August, (1968).

452-525 0—72 9 123 I V. Mechanical Behavior 1 ; :

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseiarch, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Viscoelastic Behavior

Philip L. Oglesby

Dental Research Section, Institute for Materials Research, National Bureau of Standards, Washington, D.C. 20234

Since the mechanical responses of many dental materials are functions of time as well as of applied stress, viscoelastic theory, and experimental methods should be used in describing and characterizing these materials. Both static and dynamic methods may be used. Viscoelastic theory useful for characterization of dental materials and for interrelation of responses under different types of static and dynamic tests includes the Boltzmann superposition principle for linear materials and as modified and extended for non-linear materials, the time-temperature superposition principle and approximation methods for calculation of relaxation and retardation spectra. Methods which have been applied to dental materials such as amalgam, polymeric restorative materials, elastic impression materials and natural tooth structure include creep, stress relaxation, indentation and forced vibration methods.

Key words : Amalgam, dental creep ; dental materials ; denture reins ; indentation ; base

test method ; mechanical properties, dental materials rocking beam oscillator ; stress ;

relaxation ; torsion pendulum ; viscoelastic methods.

tion in a quantitative appropriate 1 . Introduction manner by transformation equations; (5) it enables one to Since many restorative as well as natural dental select a test or group of mechanical tests that will materials have deformation characteristics which fully measure the various mechanical phenomena are time-dependent as well as stress- dependent, occurring in the material and evaluate the me- viscoelastic theory and methods olfer the most chanical parameters of each; and (6) it isolates advantageous means for description and charac- individual mechanical phenomena and their pa- terization of their mechanical behavior. However, rameters at the macroscale level and together even though well established procedures for visco- with microviscoelastic theory, permits better de- elastic characterization are available, their appli- scription of the relation of mechanical response cation to dental materials has not been extensive. to microstructure. To encourage further work in this area, theory and Both static and dynamic test methods have been methods described in various publications (refer- used to investigate the viscoelastic properties of ences below) are brought together in this paper dental materials. Commonly employed static tests and discussed with particular attention to their are classified as : (1) a constant stress test, such as potential utilization in dental research. Several a creep test, (2) a constant strain test, such as a examples of the application of such methods to stress relaxation test, or, (3) a test where the dental materials are described. stress and strain are both varied slowly and in ^ The viscoelastic description has numerous ad- some cases cycled ; an example is the classical vantages: (1) it enables researchers to describe stress-strain test. All of the above types of static analytically and to predict the mechanical be- testing have been employed in dental research and havior of these materials both in laboratory ex- testing. Dynamic test methods may be generally periments and in the mouth; (2) it enables one to classified as follows : ( 1 ) free vibrational methods, separate and quantitatively describe the relative (2) forced vibrational methods, both resonance contribution of the various time-dependent and and nonresonance, and (3) propagation methods non-time-dependent mechanical responses occur- using either pulses or continuous waves. In many ring simultaneously in material it cases, the same dynamic apparatus may be used a ; (3) furnishes a unified theory whereby the mechanical behavior to determine the viscoelastic properties of a mate- of materials with time-dependent responses can rial by two or more of the above methods. Whether be compared with the behavior of materials with the method be static or dynamic, the following fac- non-time-dependent responses used for the same tors should be considered when testing a material purpose, such as the different types of anterior (1) The specific mechanical characteristics restorative materials; (4) it enables one to com- of the material to be investigated pare mechanical response of a material under one 1 It becomes difficult to make a distinction between static test condition to that under another test condi- stress-strain testing and dynamic testing when cycling occurs.

127 ; :

(2) hoAv this information on the material represented by a spring of compliance Jo in)- may be interrelated to that obtained stantaneous behavior) and a dashpot having a

for the material using other methods coefficient of viscosity r? (viscous behavior) in and series with one or more Voigt elements (retarded (3) how the mechanical phenomena and elastic strain). The graphical behavior of each of their parameters obtained for the the three types of strain and their combination, material may be directly or indirectly along with its accompanying model and corre- related to the microstructure of the sponding analytical strain function, may be seen material. in figure 1 where the strain is assumed to be a linear function of the stress a. When the strain is 2. Creep of Linear Viscoelastic Materials a linear function of the stress, the data obtained at different stresses may be reduced to a single The creep test as a method of investigation of the creep curve by plotting the creep compliance viscoelasticity of a material has the advantage of [J{t) = (.{t)la^ as a function of time or logarithmic simple instrumentation, and is preferred for the time. The creep compliance J{t) of the combina- long testing time required for those materials hav- tion of the three types of phenomena versus time ing retardation times that extend over a long time and the logarithm of time may be seen in figure 2. scale, but has the disadvantage of insensitivity to When the retarded elastic creep compliance has a the retardation behavior of the material in the continuous distribution of retardation times Tr, the initial short portion of the experimental time scale. combination creep compliance J{t)=Ji-{-J^-\-jR the following analytical A creep test is normally conducted on the material may be described by in the form of a specimen having a uniform cross equation sectional area. A constant load is applied either in tension, compression, or shear, and the defonnation Jit) = Jo+t/v+ r J{T)[l-e-"^]dT (1) is measured in the direction of load application as a function of time. The deformation, detected by such devices as a cathetometer, strain gage, or as shown graphically in figure 2a. Substitution of differential transformer, may be measured as a L{t)/t for J{t) in eq (1), gives: function of time for periods of less than one second to many years, if necessary. Creep curves for the material are obtained for different stresses, and then deformation behavior of the material as (2a) a function of time and stress is extracted from the family of creep curves. The creep curve of a strain-hardened specimen consists commonly of one or more of three phenomena, each of which may or may not be a linear function of the applied stress; while as a function of time, one is independent, one linear, and one nonlinear. The three deformation or strain phenomena are: Time t

(1) Instantaneous elastic strain «/, described (a) Instantaneous Strain by analogy to a spring having a compliance f/o, or a modulus Gq, where Jq=\IGq,^ (2) viscous strain which may be described as analogous to a dashpot when linear,

having a coefficient of viscosity r? equal to the applied stress divided by the strain rate, and (3) retarded elastic strain usually described by analogy to the so-called Voigt element or series of Voigt elements, the components of which consist of spring of Time t ^ a compliance Jr, t = Time of or modulus Gr in parallel \vith a dash- " Stress Removal pot having a coefficient of viscosity (c) Retarded Elastic Strain (d) Combination of Three Types of Strain E = Ej + + = J^o + (oe)/n+ oJjd-e"''"') f]R, where the retardation time r of e o£ J„ (1-e "^1) " E = J o + (ot)/n+ oS J„ (1 - e"'''i) the Voigt element is defined as r 1=1 Bi o 1=1 Ri

r=r)iiJR. " J„(r)(l-e ''^l)dT E = J o + (ct)/n+aS'° J-(t)(1 - e'''''')dT L O K o Ok If a material exhibits a linear combination of all three types of strain, €=ej+€r+er, it may be FiGUKE 1.

128 :

approximation methods as well as giving a better visual picture of the distribution of retardation times Tr for the material, and (2) in application of

the time-temperature superposition principle [1, 2, 3]^ to a linear viscoelastic material to obtain its creep compliance behavior at a given temperature that othermse would require experimental data over many decades at that temperature. There- fore, a creep curve may be obtained at a specific temperature T for times outside the range of practical observation at this temperature by obtaining creep curves at higher and lower temperatures within the time scale of the creep experiment, and then shifting the higher and lower temperature creep curves for the linear material along the axis of the logarithmic time scale until they all join into a continuous master curve for the temperature T. Before the experimental creep curves can be shifted to make the master curve at temperature T, the comphance values should in theory have a density (or specific volume) cor- rection, but, in practice, the density changes with temperature are often small enough to be neglected. The amount the curves are shifted along the In time axis is described by the equation

(In ^— In t/to=ln Ln t to)=\n At, (b)

where the shift is said to be positive when the FiGUKB 2. curve is shifted to shorter times (to the left on the In time scale) in forming the master curve. The time-temperature superposition principle is appli- The subtraction of tj-q from both sides of eq cable only to creep data, also to stress (2a) results in not but relaxation data and to dynamic ^ mechanical testing data. The At values obtained from the J{t)-tly)=Jo+ L{r)[\-e-"^]d hi T shift on the In time scale, or In l/w (reciprocal frequency) scale in the case of dynamic measure- (2b) ment, for these different methods of testing are where equivalent. The theory and application of the time- J{t) is the creep compUance temperature superposition principle to a linear Jo is the instantaneous elastic compliance viscoelastic material, as well as the theoretical tjr) is the viscous response where t is the time after significance of At for the material, will be dis- application of the stress, a, and -q is the coefficient cussed in a later part of this paper. of viscosity

3. Stress Relaxation of Linear Visco- elastic Materials Jo

Stress relaxation behavior is observed in vis are analytical forms of the retarded elastic re- coelastic materials; that is, the stress in the sponse with J{t) and Z(r) being forms of the material relaxes or decreases with time when the retardation spectrum of the material. material is deformed quickly and the deformation The graphical representation of eqs (2a) and is held constant. Stress-relaxation tests require (2b) is seen in figure 2b. It might be mentioned slightly more complex instrumentation than creep that J(t) — t/i] represents the elastic portion (in- tests in that, in addition to requiring a device for stantaneous plus retarded) of the creep compliance detecting deformation, a load measuring device is curve in stress also a constant experiment and required in order to follow the force change with represents, after appropriate transformation, the time. Like the creep test, the stress-relaxation recovery portion of the curve in figure 2 a when test is the preferred method for the long test time the stress Plots creep has been removed. of com- required for some materials, but has the dis- pliance J{t) {J{t) — versus In t rather than or t/n) advantage of the lack of sensitivity to relaxation versus t for a linear viscoelastic material, have the advantages of more direct utilization— (1) in cal- 2 Figures in bracl^ets indicate the literature references at the end of this culating the retardation spectrum Z(r) by various paper.

129 behavior exhibited by the material in the initial HAXHELL MODEL ' -"MW 3}"^) G short portion of the time scale. Stress-relaxation o -t/T data are generally more directly interpreted in o(t) = o e terms of viscoelastic theory than are creep data. -t/T G(t) = o(t)/e = G A tensile stress-relaxation device often consists simply of two clamps between which the specimen is attached ; the upper clamp is usually attached to a load detecting cell which is rigidly attached to a frame; the lower clamp can be adjusted up or down in respect to the fixed upper clamp to obtain various deformations in the specimen. Once the lower clamp has been adjusted to obtain the desired deformation in the specimen, the clamp is held fixed in respect to the frame. The amount of deformation in the specimen can be detected by means of a strain gage, differential transformer or cathetometer. The stress-relaxation curves are obtained at different deformation levels, thus the stress-relaxation as a function of time and deformation is obtained for the material from the family of stress- relaxation curves at different deformations. The stress is plotted as a function of linear time or of logarithmic time. If stress is plotted as function of strain for a common time value from the family of curves, the stress-strain curves for the fixed time values will yield the functional relationship between the stress and the strain. With (c) materials exhibiting a linear stress-strain relation- Figure 3. ship, dividing the stress values for each curve by the corresponding fixed strain value gives relaxation modulus G{t) = a{t)/€o and the family of stress- previously mentioned, there is the lack of initial relaxation curves reduces to a single master sensitivity to stress-relaxation behavior of certain modulus relaxation curve. If the material is materials as a consequence of the finite time linear in its stress-strain behavior at a fixed time, required for application of the fixed strain (rather the usual curves plotted are relaxation modulus than idealized instantaneous application of the curves either as a function of linear time or strain). This lack of initial stress-relaxation logarithmic time. sensitivity can be discussed in terms of a material The simplest analog to describe stress-relaxa- which can be represented by a single Maxwell tion behavior is the Maxwell model which is a element. If such a material has a short relaxation series combination of a spring of modulus Go and time, T, or expressed another way, the stress- a linear dashpot having a coefficient of viscosity relaxation time scale is short compared with the 17 which has the differential equation of motion: time required for application of the strain, the instantaneous maximum stress and corresponding de (7 1 d(T limiting modulus are indeterminate. The (3) Go=(Jojt Tt^v G'o dt' result of this can be seen for a simple Maxwell material represented by the dotted line in figure In stress test as the normal relaxation mentioned 3(a). previously, is the strain held constant after initial Most materials, including dental materials, cannot be characterized by a single Maxwell model rapid deformation. Hence, becomes zero and ^ having a single relaxation time. These materials have more than one relaxation time and, in many the solution is: cases, so numerous are the relaxation times that

Got they can be treated as continuous in their distri- (3a) bution, the continuous function H{t) being called the relaxation spectrum. When the relaxation mod- A material that can be described by a Maxwell ulus G{t) for a linear viscoelastic material is plotted model is said to have a single relaxation time against logarithmic time (over an extended time defined as r=?j/(?o. Such a material is shown range), most often a decreasing sigmoidal shape graphically by a plot of stress as a function of curve is obtained, as illustrated in figure 4. In linear and logarithmic time and also by a plot figure 4, the limiting short time value of the relax- of relaxation modulus (stiffness) G{t) = a{t)/e ation modulus is called the glass modulus Gq, while as a function of logarithmic time in figure 3. As the value of G{t) at infinite times approaches an

130 : —:

and the creep compliance function J{t) are reciprocally related only at the limiting values, and, therefore, their curves on a logarithmic time scale are not true mirror images for a given linear viscoelastic material. It has been shown by

Gross [4], and more specifically by Leaderman [5], that the two functions G{t) and J{t) are related by a reciprocal relationship between their respective Laplace transforms

pL[J{t)]=l/pL[G{t)] (7)

where the Laplace transform is defined as follows

L[f{t)]= j\-^'Mdt equilibrium modulus value or zero, depending on the microstructure of the material. The differ- While equation (7) relates the two functions ence between the glass modulus Go and the equili- J{t) and G{t), there is difficulty in obtaining one brium modulus Ge is defined as the decay modulus function from the other by means of equation (7) Gr= Gq— Ge. For example, in the case of amorphous due to the obstacles involved in the inversion of polymer, the value of Ge would depend on whether the Laplace transform. More often the functions the polymer was crosslinked or not. The cross- G{t) or J{t) are given as empuically determined linked polymer would exhibit a value G^ at infinite data and numerical inversion is required. One of time, but the noncrosslinked would approach zero the most severe drawbacks in using the above at infinite time. Commonly, even though Ge=0 equation is that usually one function has been for noncrosslinked polymers, an intermediate pla- determined only over a limited time scale and the teau is observed, the length of which is molecular entire time scale is requked to make the inversion. weight dependent. Since most materials must be The logarithmic time plot for relaxation modulus described by a finite number of relaxation times G{t) illustrated in figure 4 has the same advantages or a continuous relaxation spectrum, the modulus- of direct use as in the case of creep compliance relaxation behavior shown in figure 4 may be de- versus In t: (1) in determining the relaxation scribed by one of the following equations. In the spectrum iJ(r) by approximation methods and case of a finite number of relaxation times, the giving a qualitative visual picture of the distri- relaxation modulus equation takes the form bution of relaxation times r^, and (2) in applying the time temperature superposition principle to the relaxation modulus data for a linear material. G{t) = ^G,e-'^+Ge- (4) Again, as in the case of creep compliance data, if the relaxation modulus G{t) data are obtained at various temperatures above and below In the case of continuous distribution of relaxation some reference temperature T, these higher lower times, the relaxation modulus equation takes the and form for a linear material temperature modulus curves may be shifted along the In time axis until they form a master curve at the reference temperature T. The amount of G{t)=j^^ GiT)e-"^dT+Ge (5) shift again being described by (In t — In tg) =ln tlto=\n At, where the values obtained by the shifts of the relaxation data are equivalent or, upon substituting G{t)^H(t)/t, the equation to those obtained from the shift of the creep becomes compliance data on the same linear material^ this result is encompassed in the theory of the G{t)=p°' H{T)e-"^d In r+Ge- (6) time-temperature superposition principle.

4. Superposition Prin- The monotonic decrease of G{t) from Go to Ge, Boltzmann as shown in figure 4, can be described by eq (6). ciple—Relation to Stress- Strain It is noted that the sigmoidal relaxation modulus Behavior G{t) curve in figure 4 roughly approximates the mirror image of the creep compliance J{t) plot According to Boltzmann's superposition prin- shown in figure 2, both plots being on a logarithmic ciple, if a series of stresses have been applied to a time scale. As recalled from earlier discussion, the specimen at various times, the deformation at glass modulus and glass compliances are related any subsequent time is simply the summation as G0=1 /Jo- Also, the equilibrium modulus of the deformations which would have been and steady-state compliance Je are reciprocally observed at that time if each of the stresses had related. However, the relaxation modulus G{t) been applied independently.

131 :

Recalling the definition of creep compliance, or by the use of the chain rule and where the entire J{t) — 6(t)/cr discussed earlier, it has been noted stress history from e^—co is considered, the that in a linear viscoelastic material, when a above equation takes the following form: single stress

Thus, the above finite relationship relates the a{t)=i:,G{t-e,)e, (15) CO strain e(i) at time t to all the previous stress history by way of the creep compliance J. The where G is the relaxation modulus defined earlier, above equations are finite forms of the Boltzmann thus relating the stress to prior strain history. superposition principle. These equations are useful When the strain increments are introduced in a in step-function experiments. For ex- describing continuous manner, the above equation becomes described ample, consider the creep experiment by a stress integral as follows: figure 2(a) where the stress has been introduced and removed in a stepwise manner and the stress a was maintained over a sufficient time for a

e{t)=a^Jo+J^+^-^^'j-

tinuous spectrum form of the creep function given Now, substituting e(0) for (t(Jo+«/r+W'7) by eq (2 a),

e{0)-e{t)=a[J{t)-tM (12) J{t)=^^'" L{r)[l-e-''^]d In r + tln+J, which is the transformation mentioned earlier for the unloaded portion of the curve. differentiation results in If the stress increments are not introduced in in eq (14) followed by steps, but in a continuous manner with time, the strain function becomes a strain integral in accord- ^=i^=J(0. (17) ance with the superposition principle taking the ad (J at following form: Thus, the slope of the strain-stress curve in a constant stress rate experiment is the creep function is given A

132 : : :

stant strain rate case for a stress-strain experiment, 6. Dynamic Test Methods by substituting eq (6) into the superposition eq Often a type of test is used where the system in which the material is a component is excited (16), and assuming e (a constant) , one ob- de into vibration. The stress and strain in the tains the result material vary sinusoidally with time as opposed to the creep or stress relaxation tests which are essentially step-function experiments. Such -{£: TH{T){l-e-'i')d hi r+Get](18) vibratory tests measure the dynamic response of a material. As in the earlier discussion on creep and relaxation, this discussion will be restricted (j{t) the relaxa- which relates the stress integral to to linear viscoelastic materials where the strain tion spectrum H{r) and thus to the relaxation amplitude is proportional to the stress amplitude. modulus function of the relaxation experiment in a The vibrational character of a system composed simple manner since for the constant stress rate of a linear viscoelastic material may be described <=e/e; thus stress versus strain is experiment, by a linear time-invariant differential equation equivalent to a stress-time plot. of the form " d^T 5. The Spherical Indentation Test (19)

The problem of the contact of a smooth rigid where represents the output and x the input. sphere of radius R against an initially plane sur- y Such systems described by the above differential face of a linear viscoelastic material was analyzed equation may also be described by m^ans of their by Lee and Radok [6]. Their analysis has been the weighting or impulse functions [10]. The impulse basis of further study and generalizations by other response function g{t-T) = W{t-T) associated with workers [7]. The solution developed by Lee and the above differential equation has the form Radok is

16 g{t — r) = '£^ Ctgiit-r) (20) 1 = 1 X de where giit-r) is a set of basic solutions to the where homogenous differential equation. The relation the total force as a function of time or of the input x to the output y for the above the loading history system is expressed by the superposition principle J{t) the shear creep compliance for the linear as follows material y(t)- T)x{T)dl R the radius of the indenter (21) m the radius of indentation contact a{t) the depth of penetration of the sphere The ease of determining the weighting function into the plane surface. depends on the form of the differential equation If the total force is applied in the form of instan- of the system. A third method of description of taneous constant force Po, then P{t) = PoH{t) the linear viscoelastic systems is by means of a where H{t) is the Heaviside step function [H{t) = 0 frequency function [4, 11]. This frequency function bears a close relation in properties to the for t <^ 0, Hit) = 1 for i > 0] and the above solution takes the form in. terms of the shear system's Aveighting function and is related to the compliance as follows: weighting function by the follo\ving integral pair: 16 [mf 16ffl^^[a(0F Jit) = _ (22) 3RPoH{t) 3PoH{t)

Using this equation J{t) in. theory could be deter- W{t-T)=^ H{ii,)e"'^'-^^dw (23) mined by measuring the central indentation depth J^" a{t) as a function of time. Practically, however, this equation has application only when the period The relation of the system's input x to the sys- of load application is short compared with the tem's output y in terms of the frequency function experimental time scale needed to evaluate J{t) is given by the following completely. The first group to recognize the application of y{t)=^ f^" H{io:)Xiio,)e'"'dw (24) he spherical indenter as a viscoelastic test for various dental materials was W. T. Sweeney and co-workers [8, 9]. where X{iw) is the Fourier transform of the input x.

133 : :

The difRculty in determing H{iui) from the differ- force, that is \vhen/(0=O, the resulting homog- ential equation of the system is similar to that of enous differential equation is the weighting function determination. The terms such as dynamic modulus and com- Ajy+B,y+Ky=0 (27) pliance, mechanical impedance, logarithmic decrement, and phase angle most often determined The displacement yc takes the following form when for a vibrating linear, viscoelastic system may be the system is underdamped arrived at by examining the resulting differential equation of motion or a system with a single {B,'-AAiK<0): degree of freedom having an elastic element as y, (28) well as viscous character such that the sum of the forces are zero in accordance with d'Al t's ember where a is the attenuation factor and the principle as given damped oscillation frequency. The attenuation factor a is related to the logarithmic decrement Ajy + B,y + Ky^ = /(O (25) 5 and the damping frequency and to the co- _ /inertiaX /viscousX / elastic\ / exciting\ efficients of the differential equation by: \ force / \ force / \ force / \ force /

(29) In the above system, the viscous and elastic forces are usually exerted by the viscoelastic material The logarithmic is Avhile the inertia force may be due to the material decrement defined as the natural logarithm of the ratio of sample or may be due to another part of the two successive maxima and system, depending on the system design. The y^ y^+x which occur in the period Ta=2irlwa and is given by: complete solution of the above differential equation is the sum of the complementary solution 5=ln {y„-,/yn)=aT, (30) yc and the particular solution yp-.

Since the antilog of eq (30) is y^Vc+Vv (26)

e^=yi/y2=y2/ys= • =yn/yn+i The complementary solution yc is the solution then to the above differential equation when the exciting force is zero or what is called the homogenous {e^Y={yilyi+i){yi+ilyi+2) . (yi+n-i/yi+n)=e"'' differential equation. The complementary solution or yc to the homogenous portion of the differential equation describes the displacement behavior of 5=-ln (yi/yi+n) (31) the system upon removal of the exciting force or what is called the free vibrational character of the Therefore the logarithmic decrement can be system. An important term evaluated for a system obtained from two displacement maxima separated containing a viscoelastic material in free vibra- b}^ n periods by use of eq (31). The elastic energy tion when the system is in the underdamped stored by the system described by eq (27) when condition is the logarithmic decrement. This term the displacement i/ is at a maximum during the will be discussed in more detail later. The second free oscillation is represented as follows for each part of the solution to the above differential successive maximum: equation called the particular solution is a specific solution of the equation when/(i) is not zero, that is when the exciting force f{t) is acting on the V,=lKy,^V,=lKy^ . . . V„=l Ky.' system. It is a specific solution in that it describes (32) the displacement behavior of the system under steady-state conditions after the transient be- The energy loss between two successive maxima havior (which is described by yc) has vanished as Vi and V2 divided by the original energy is a component of the displacement y. The terms commonly evaluated for a system containing a viscoelastic material, under steady-state vibratory conditions, are the mechanical impedance V, KyJ and phase angle for the and, or the mate- system (25)2 (25)3 rial, as a function of exciting frequency. Also, the (33) 2! 3! mechanical Q factor is determined from the frequency response curve of the mechanical im- When 5 is small, the energy loss per cycle is pedance. All of these will be discussed later in approximately more detail. First considering the free vibration of the i2b (33a) viscoelastic system after removal of the exciting V

134 : : :

Next, looking at the linear system under steady- It is noted that the absolute values and their state conditions when force vibrated, the motion respective storage and loss values of moduli and

of which is governed by eq (25) , when the exciting compliances are related as: force is harmonic, the steady-state displacement is a harmonic function. Hence, let the exciting yp \G*\ = ^lG'^+G"^ (40a) force be represented by the harmonic function.

f(^t)=Foe*"' (external excitation) (40b)

Now the displacement would take the form The ratio of either G"/G' from equations (37 a and b) y=Yoe'"'. Substitution of these two functions or J"/J' from (39 a and b) is: into (25) leads to:

G"/G'=J"/J'=t&n e (41) (34) y j-o where 6 is the phase angle between input and output and tan 6 is called the loss tangent. If one uses where is defined as the complex modulus or G* eq (35a) for the harmonic input f=FQe^"', the stiffness. Therefore, a system like that described output y of the system for steady-state motion is by eq (25) has a steady-state modulus G* or compliance J*=l/G*. From eq (34) we note that: 1. (42) f=G*y (35a) The output y lags the input / by the phase angle 6 (35b) y=J*f Now, returning to eq (34), it is seen that for a system in steady-state motion governed by eq (25), complex quantity G* can be written in The the expressions for \G*\, \J*\ and tan 6 are as rectangular form and polar form as follows: foUows

Fo 1 gi6 = e-\-i sin e) = \G*\e'' G*=G'+iG" \G*\icos Yo J*

Fo (36) (43a) \G*\ = is the absolute value the where Fo of tan d=wB,/iK-AjO)^) (43b) stiffness G* and 9 is the phase angle between the input and the output / y. The two terms above are usually measured as a Also: function of exciting frequency w. The phase angle Fo 6 is measured between the exciting input function G' = \G*\ cosd- cos 6 (37 a) Yo and the output function for a given exciting fre- and quency. Usually, the reciprocal of is measured at each driving frequency and plotted as a func- G" = \G*\ sine-- sin 6 (37b) tion of frequency. This reciprocal the magnification factor of the system, is determined by The moduli G' and G" are defined as the storage measuring the amplitude of the output in respect and loss moduli of the system. The complex to the input amplitude at a given exciting fre- quantity, G* represents simultaneously the elastic quency. The magnification factor is: and damping properties of the system. Since the 1 complex modulus G* and the complex compliance 1/|6'*| = - J* are reciprocally related as defined by (35a) and (44) (35b) then from eq (36) Kyln- where

J*=l/G* -id J*| {cos d-ismB)= J' -iJ" Fc (38) =resonance frequency of an undamped system 90° where J' and J" are the storage and loss com- Resonance occurs when the phase angle 6 is and w=co„. Substituting u=u„ in eq (44) gives: pliances :

Yo 1^ T' — \7* J*\ cos e-- cos d (39a) [V\G*\]res = (45) Fo or Yo K_ J"^\J*\ sin e-- sin 6 (39b) i^[l/|6^*|]..,= (45 a) Fo

135 — . : —

The quantity given by equation (45a) is called due to a viscoelastic sample as described above and the mechanical Q factor of the system described the external excitation is of the form Foe*"* (a by eq (25). The Q factor of the system is usually harmonic function) then the differential eq (25) obtained by taking the frequency difference for can be expressed in terms of the viscoelastic two frequencies on each side of the resonance material moduli G' and G'' as follows maximum [12] where Ajy + {G:-\-iG'r:)y=Foe*''* ^^2 K or 2 BnWn Ajy+Gly=Foe*-' (50)

In the absence of external excitation, /(<)— in called the half power points; thus Q is defined as 0, the case of the system described by eq (50), K some investigators [13] have attempted to solve Q- (46) the free vibration by trying to solve the homog- CO2 — Wi enous differential eq (50).

If the system contains a viscoelastic material Ajy+{G',,+iG'r:)y=0 sample where the damping and elastic responses (51) of the system are entirely due to the sample and But it must be pointed out that the form of G'm the inertia of the sample is negligible compared and G'J, in occurred only because the form with that of the system so that when the phase (50) of the external exciting function fit) was known angle is measured between the input at the sample and the solution under steady-state conditions is and the output displacement, the expression the same form as that of the excitation source. (43b) becomes But there is no such analogous situation in the CO case of free vibrations. Assuming the solution is tan B„ (47) K of the form expressed by eq (28), then

the which shows that the phase angle between G:r. = A,W-Cc')/9( (52a) force on the sample and its displacement is directly proportional to the driving frequency co G'^=2c,,aAr/ 9( (52b) and is equal to given by eq (46) when the driving where frequency is the resonance frequency o)„ of the total system. It is further noted that the definitions c

It is seen that G" is directly proportional to 9{ = shape factor the damping constant and the exciting frequency w. Now again examining the case where If damping is small enough so that B^^

(53b) (49a) G':^<^nB,l9(

One sees that G'J^ in (53b) is of the same form as (49b) in (49b) and will be equal when the exciting 9i G'rl^ frequency in (49b) is equal to Wn- Further, it is noted that if where G^ and G'J, are the storage and loss moduli of the material when corrected with the B,'«4.Aj appropriate shape factor £/^. It is further noted then that the loss modulus of the material G'^ is the GL-=Kl9{ (54) same as that of the system since all the loss is in the material. Now if the system described by eq which is the same as the G^ in the steady-state (25) has its damping and elastic forces completely case of (49a)

136 . : :

Dynamic methods in which the inertia of the the logarithmic decrement 5 as expressed by eq system is large compared with that of the specimen (29), substitution into (59a) and (59b) leads to of viscoelastic material being characterized are the result: sometimes called dynamic methods ^vith added inertia. Two examples of this type of method are (60a) the torsion pendulum which is usually used (1) or in free vibration and is very useful for measurement of energy loss in linear systems through the G'^=Ij^\^-^-h')ldi logarithmic decrement, and (2) the rocking-beam oscillator. Both of these methods are low-frequency (60b) methods but may be used in free or forced oscilla- TT tion. The torsion pendulum essentially consists or of shaft composed of the viscoelastic material a Gl=^irIU5l9i to which a disk of moment of inertia / is attached. The torsion pendulum as a dynamic test method where /„ is the resonance frequency of the un- and the theory of its free vibration is discussed damped vibration of the system recalling that in detail in an article by Nielsen [14]. The differ- w„ 27r/„. the ratio ential equation of motion for the torsion pendulum = Now of G'^ to G'm using eqs (60a) and (60b) would be the following for would be that given by eq (25) where Ai=I, the a viscoelastic material moment of inertia, y=B, the angular displacement for an external torque j{t) GL' 4ir6 (61) ie+B,e+Ke=j{t) (55) G' '47r2-52

Equation (55) holds for small angular displace- which for small damping takes the form ment. Since the inertia of the viscoelastic specimen shaft is negligible compared to the moment of /, Gm /G!n=d/ir (62) inertia of the disk, the equation can be written in accordance with eq (50a). The shape factor ^ takes the following forms [14] for (1) rectangular cross-section specimens /H«?;+^•6^;')5=/oe-' and (2) circular cross-section specimens of the or viscoelastic material: (56)

(1) 9(=CD'M/16L In free vibration /o6*'''=0 and eq (56) takes the form of eq (51). (2) 9( =TrRy2L

ie+{G:+iG':)e=Q (57) where or C= width of the specimen ie+Gie=o Z>= thickness of the specimen i?= radius of the specimen where the complex modulus G* in this case is the Z= length of the specimen complex shear modulus of the viscoelasitc material M—Si shape factor obtained from tables with G' and G" being the storage and loss com- The rocking beam method [15], like the torsion ponents. The solution to would of eq (57) be the pendulum, has the restoring force and damping form of eq (28) force provided by the sample of viscoelastic 0=0oe(i"d-«)' (58) material. The excited motion of the rocking beam like that of the torsion pendulum may also be expressing the angular rotation 6, as a function of described by eq (25) , but the displacement of the time for free vibration. Since the above solution is sample in the case of the rocking beam is linear, of the form expressed by eq (28) then G' and G" not angular, the inertial term Ai is the effective are related to the attenuation factor according to mass (me) of the S3^stem, and the external exciting eqs (52a) and (52b) as follows: source is a force /(i), not torque in the equation as shown G'„=IW-o^)l9{ (59a) m,y+B,y+Ky=f{t) (63)

(59b) The rocking beam consists of a beam of large mass compared to that of the sample, with the where S(. is again the shape factor. Now assuming beam resting on a knife edge and having adjustable damping is small enough so that a)d=aj„ and re- weights. The specimen is positioned in the vertical calling that the attenuation factor a is related to direction vnth one end clamped to the horizontal

137 : : : : : :

beam out from the knife edge and the other end The effective mass of the system is determined to a fixed position. The specimen undergoes motion by substituting a spring of negligible mass and only in the y coordinate upon excitation of the determining free vibrational frequency a)i=27r/i of beam for small oscillation. Returning to the equa- the system with the spring in place of the vis- tion of motion of the rocking beam, it is noted coelastic specimen. Now if the natural frequency that since the mass of the viscoelastic specimen a)o = 27r/s of the spring is determined using a known is small compared to the effective mass of the mass mo in free vibration or if the spring constant beam, eq (63) can be written using eq (50 a and Ks is determined by adding weights to the spring b) in the following form and using Hooke's law, F=~KsX the effective mass nie is given as follows m,y + {E:+iE':)y=U"^' (64) or Ks o:s' fs' mey + E*y=foe'^' me=—|=mo -^=mo (70) where E*^ is the complex Young's modulus with E' and E" being the respective storage and loss The shape factor 3^ for the rocking beam is as follows components. Now in free vibration again /o^'"' = 0 and eq (64) becomes: 9(=A/l

m,y-\-E:y=0 (65) where I is the stretched length of the viscoelastic or specimen and A is the corresponding cross-sectional m,y+iE:+iE':)y=0 area at such length I. The cross-sectional area Ao of the undistorted specimen is related to A as The solution to eq (65) is again of the form of follows, assuming constant volume of sample eq (28) y = yQe''''^d-0L)t (gg-) IqAq— Ia or expressing the linear displacement as a function of time for free vibration for an underdamped Ao=AI/Io=a(i+^^ system; therefore, E^ and E^' are related to the attentuation factor a according to eqs (52a) and where Iq is the undistorted length and Al the (52b) as shown change in length upon stretching.

E:=m,W-a')l9( (67a) 7. Theory of the Time-Temperature ¥;' = 2mewW^ (67b) Superposition Principle

Earlier in this discussion it was stated that the If damping is small enough so that Wd=co„ and time-temperature superposition principle is appli- using eq (29) which relates the attenuation factor cable to various mechanical test data, creep, a to the logarithmic decrement, upon subsititution stress relaxation, and various dynamic mechani- in eq (67a) and (67b) the moduli are related to cal tests, and that Leaderman [1] and others the logarithmic decrement 5 as follows; [16-20] have demonstrated that a composite curve of the mechanical response can be constructed by shifting of data obtained at different temperatures along a logarithmic time or recip- or rocal frequency axis. This has also been shown K=mefn'{^T'-8')/9( (68a) to be applicable to dielectric or magnetic relaxation phenomena. If this principle is applicable to the data obtained on a material under different IT test methods, a micromechanism occurring in the 0 material may be confirmed and examined by each E'^=^Tmefn'8/ 9( (68b) of these methods; for example, transitions in an amorphous polymer may be explored by j where w„ = 27r/„ with being the undamped examining the distinct dispersion regions by i vibrational frequency of the system. Now the mechanical means. For certain materials, in order \ ratio of E'„! to Em is the following: to obtain the composite curve for the reference 5 of measurements at temperature Tq by reduction {

K7S;=4r6/(47r^-6^) (69) various temperatures, the modulus function must .1 be multiplied by a factor Topo/Tp where To is which if damping is small reduces to the reference temperature and po is the density of the material at that temperature, while p is E"/E'm=5/ir the density at T, the temperature of data meas-

138 :

urement. This factor times the modulus function ments. Upon analysis, they concluded that the true value of AHa was obtained from the tangent the is plotted against In J- or In (coAt) to give loss (internal friction measurements). They explained the discrepancy in the AHa values from composite curve; recalling that In (In /—In to) At= creep or stress relaxation measurements on the expresses the amount the curves at temperature basis of the temperature dependence of the shifted along the In time axis to form the T are relaxed and unrelaxed compliances or moduli composite curve at the reference temperature Tq. which are the limiting values. As pointed out in This application of the factor ToPo/Tp to the the earlier discussion on creep and stress relations, mechanical response of a material is in accordance the unrelaxed creep compliances and moduli are with the method of reduced variables which the instantaneous values, while the relaxed com- involves two assumptions. The first is that the pliance is the instantaneous compliance plus the modulus function for the material is proportional limiting or asymptotic value of the retarded to the absolute temperature T and density. elastic compliance, and the relaxed modulus is Therefore, just the equilibrium modulus Gg. The inclusion of (71) the temperature dependence of the limiting moduli and compliances into the time-temperature superposition principle led McCrum and Morris where and are the modulus function at Mo M _ [23] to the following relations which define the temperatures To and T. The second assumption McCrum-Morris superposition principle is that the molecular mobilities of the material at any temperature T all have the same frequency- (74a) temperature dependence which can be described A. as a single activation energy of flow AHq which is related to the shift factor At and temperature by JRT Jul GuT—GRT Ct dn (74b) the Arrhenius equation. J, 'J uTo ""^0 ^ RTo

R In An Jut J^î'q AH. (72) _ _ (74c) \l/T-l/To, 'T T " uTn ûTp where R = the gas constant. JRT ^Rô The temperature dependence of the activation (74d) energy of flow AHa for the viscoelastic response of JRTn GRT amorphous polymers at and above the glass transition temperature Tg of the polymer has where J^t and Jrt are the unrelaxed and relaxed been described by the WLF equation, a semi- compliances for any temperatm-e T while J^ra empirical relation developed by Williams, Landel, and Jrt^ are values at some reference temjierature and Ferry [21] and given as: To- 111 the above equation Gut, Grt, and Gu^,, and Grto are the respective unrelaxed and relaxed moduli at an}- temperature T and reference tem- (73) {51. Q+T-TgY perature To- From eqs (74 a, b, and c) it can be shown that the This equation shows AHa to increase with Tg and retardation spectrum L-r (In t) and the relaxation to fall off above Tg. Since AH. measures the spectrum Hj, (In r) at temperature T, for a solid, change in molecular motion with temperature, obeying the Boltzmann superposition principle, AHa is a maximum at the glass-transition tem- can be related to their respective spectra at a perature of an amorphous polymer and is low at reference temperature Tq hy the following relation: temperatures far above and below the transition. Using tensile creep data, Bueche [22] calculated Lr (In T) = bTLTQ (In r/A^) (75a) the experimental AHa versus T for polymethyl- acrylate in the vicinity and above the glass b Ht (\nT)=-^Hr, (Inr/A^) (75b) transition temperature Tg. Bueche gives an extensive discussion of the molecular theory of such transitions and their relation to AHa. From these results, McCrum and Morris developed McCrum and Morris [23] examined the experi- the following general relations for the creep com- mental values of AHa for various polymers in- pliance and the relaxation modulus at tempera- cluding poly(methyl methacrylate) and polytetra- tures T and T.: fluoroethylene as obtained by creep or stress relaxation experiments and compared them to J^^(t/Ar)=~ JT{t)+JuTo (76a) those obtained by internal friction measurements. They observed that the AHa values from creep and stress relaxation measurements were larger GromT)=-^6r{t)+GRT,(^^'^) (76b) than those obtained by internal friction measure-

139 452-525 0—72 10 : : :

The above equations are very general and it can dG{t) appropriate for relaxation experiment be shown that with assumptions, such c^ln t as bT—dr and Cr= 1 that WLF relations are special 80(a) cases. Glass state transitions may be evaluated by means of the above equation. For example, such for dynamic experiment transitions have been usually described by the din cj following (80b) J^^it/Ar)=JT{t) (77a) Schwarzl and co-workers [5, 26] developed approximations of the spectra using higher (77b) derivatives of the experimental response function J{t), J'{o:), J"(w), G{t), G'io:), and G'^ico), but, which can be obtained from (76a and b) by assum- while these relations are better approximations ing 6r=Cr=dr=1.0. But McCrum and Morris to the spectra, the observed response functions assumed for such transitions that br=C7.= d7-5^1 are not accurate enough to get second derivatives which leads to the following from the experimental data in most cases. The approximation relations are expressed as follows:

(78a) Retardation spectrum

for creep data (Sla) 6roit/Ar) = CrGT{t) (78b) ^('/2'=f^-rf^

The above equations, when applied to creep data Z'VlM=?[/,(„)-,^{;^,] for dynamic and stress relaxation data in the vicinity of a glass compliance data transition [23] gave values of AHa which agreed with the values obtained by internal friction meas- (81b) urements when applied to data for various where transitions. m=J{t) or J{t)-th m=J"i.^)-^l^-n 8. Approximate for Calcula- Methods Relaxation spectrum tion of the Relaxation and Retarda- dG{t) d'G{t) tion Spectra of a Material and the , for relaxation data Interrelation of Viscoelastic Response dint' d{\n ty Functions (82a)

Alfrey [24] developed first order approximations 77/(1/0;)=^ [^^"'(co)-^^^^] for dynamic data for the calculation of the relaxation spectrum and (82b) the retardation spectrum of a material. The retardation spectrum using Alfrey's method is An interrelation between two different visco- simply found by taking the slope of the creep elastic responses was given earlier in the case of compliance (J{t)-t/ri) versus the logarithm of creep and stress relaxation function by eq (7) . The time (t), or, in the case of a dynamic experiment relation between the dynamic compliance and the the slope of the storage compliance versus the retardation spectrum of a linear viscoelastic mate- logarithm of frequency (w) as expressed by the rial can be shown to be represented [5] by the following equations [5, 25] following

d[J(t)-th] J'(co)=e/,+ -^^^^^Inr (83a) for creep experiment d\n t (79a) ^"(-)=r"fx^2^1nr-h- (83b) diJ'ic)] for dynamic experiment d\n CO The relations between the dynamic moduli (79b) (storage and loss) and the relaxation spectra are as follows To calculate the relaxation spectrum using this method one takes the slope of the relaxation G'{.^)= f^" ^^_;^f;' ^ln T+G. (84a) modulus G{t) versus the logarithm of time (t), or, for the dynamic experiment, the slope of the + " iJ(T)cOT storage modulus G'{w) versus the logarithm of (84b) ^"(a>)=J_ dXvL frequency (co) given by the following equations

140 : .

and static responses are where . or . Since the dynamic JnitMM ....<„) J„{t,Ti,T2, . . Tn) interrelated through the spectra for the material, are characteristic weighting functions of the one can be calculated from the other using the excitation-response system; where above equations and eqs (2a and b) and (6) Jn{t,ti,t2 ti>t 9. Nonlinear Viscoelastic Behavior Jn{t}Ti,T2 rn)=0 Ti<0

Leaderman [1] found in his earlier investigations This general relationship can be applied to of the creep of fibers that the creep and recovery static as well as the dynamic response of a behavior were nonlinearly related to the stress, nonlinear viscoelastic material. It will be discussed and he proposed a modification of the Boltzmann later in this paper in respect to the superposition principle containing an empirical nonlinear behavior of dental amalgam on which creep data were function y[o-(T)]. However, Leaderman's modifica- ob- tained. If the response of the system is tion did not explain nonlinearities later observed stationary, the above equation takes the following in the mechanical behavior of polymers. In 1960, form Nakada [27, 28] (see also Greene and Rivlin [29]) proposed a theory of nonlinear responses where e(t) = j^yit-TOMTr)dT, the excitation o-(0 and the resulting response are a series of normalized orthogonal /»-|_co related through J-f-co functions cr{t — Ti)a{t — T2)J2{Ti,T2)dTidT2 - CO J — CO (90) cr(i) = Sai0i(O (85) i = l a{t-TMt-r2Mt-T3) -co J — CO J — CO (86)

i = l J2{Ti,T2,Tz)dTidT2dT3-\- . where the coefficients aj and 6< are given by where Jn{ri,r2 T2)=0 Ti

For example, if we assume a dynamic excitation of (T{t)i{t)dt (87) the form o-(<) = cro cos wt, then the resulting re-

sponse, e {t) , from the above equation is b^= e{t)^iit)dt (88) e{t)=GQ[Ai cos ut-\-A2 sin wt]

-\-(Tq^[Bi cos 2co<+52 sin 2uit-\-B'^ ^i{t) being the complex conjugate of 4>iit). If the excitation always precedes the response +o-o^[Ci cos 3wi+C'2 sin 3cof and if each of the coefficients, 6/s, of the response function can be expanded by means of Taylor + (73 cos w^+C4 sin . . . (91) series in the coefficients, a/s, of the excitation where function (x{t), then the response function e{t) can yli=J/(co), ^2=J/'(co), be represented by e{t) = j^y{t,)Jrit,U)dtr B2

+ f f

C2 =1 Jz" -co J — CO J —

Mt,t:,t2,t3)dt4t2dt3+ (89) — lxi), (73 =^ J3 {03,W, or For step function excitation or creep where <*) = 0, the above integral takes the form c{t — Ti)

J3{t,Ti,T2,T^dTidT2dTz-\- . response of dental amalgam [30].

141 , .

10. Yiscoelastic Methods Applied To hour period during which the mechanical char- Dental Materials acteristics of the specimen are changing to a considerable extent. There are numerous repoi'ts in the literature con- Stress-strain measurements on dental amalgam taining data showing viscoelastic characteristics of in compression were carried out by Smith, Caul, dental materials. However, in only a few of these and Sweeney [56] prior to 1956, while tensile has there been an attempt to analyze the data in stress-strain measurements on amalgam were con- terms of rheological theory. A few examples are ducted by Rodriquez and Dickson [57] 1960-61. cited below. Because these stress-strain measurements resulted The primary polymeric component of denture in a nonlinear relation w^iich depended on the strain base resins is usually poly (methyl methacrylate) rate, a study of tensile viscoelastic behavior one of the dental material components to which of dental amalgam was undertaken using static viscoelastic methods have been applied. Both static creep tests on tensile specimens. These creep studies and dynamic methods have been used to study its by Oglesby, Dickson, Rodriquez, Daven- port, mechanical behavior and the relation to its chemi- and Sweeney [30] showed that dental amal- cal structure. Creep studies [31-36] and stress re- gam is nonlinear in its viscoelastic behavior. Strain hardened in tension laxation studies [37], as well as dynamic testing amalgam was found to exhibit [38^2], have been reported. Denture base resins three types of mechanical behavior: (1) instan- were studied mechanically by the use of stress- taneous elastic strain, (2) retarded elastic strain (transient strain tests conducted by Barber [43] in 1934 and creep), and (3) viscous strain (steady state creep). by Sweeney and Schoonover [44] in 1936. Their Viscoelastic theory was applied to time dependent characteristics were pointed out by the creep data obtained for dental amalgam to calculate Sweeney, Paffenbarger and Beall [45] in 1942. the tensile stress-strain behavior. Since the amalgam creep strain was a nonlinear function Later Myerson [46, 47] using the dynamic torsion pendulum method, studied the change in stiffness of stress, the usual Boltzmann superposition prin- and damping capacity as a function of tempera- ciple could not be used so the nonlinear generali- ture and cross-linking of various methacrylates zation of the superposition principle developed by used in denture base resins and examined the tran- Nakada [27] Avas applied. The experimental ten- sitions occurring in various methacrylates at differ- sile stress-strain data were found to be in good ent temperatures. Application of the d.ynamic tor- agreement with that obtained by transformation of sion pendulum was later extended to the study of the creep data. Since the creep data indicated that denture base viscoelasticity by Braden and Staf- the retarded elastic behavior could not be clearly ford [48] who compared the denture base polymers separated from instantaneous elastic behav- of vinyl acrylate with polycarbonate. They ex- ior due to short retardation times, it was neces- to amined the shear modulus and loss tangent sary employ dynamic tests to obtain the in- (damping capacity) of these polymers as a func- stantaneous elastic modulus. tion of temperature. Static viscoelastic testing of The first method used for this purpose was the ultrasonic methyl methacrylate direct filling resin and arti- pulse-echo technique. Young's modulus as well as shear and bulk for ficial teeth was carried out by W. T. Sweeny and modulus amalgam were determined by this technique by Dickson and co-workers [8, 9] using the indentation testing procedure mentioned previously. Oglesby [58]. Later testing on amalgam has been carried out by Larson using Viscoelastic investigations of impression mate- [59] a forced resonance technique (50 kHz range), advantageous in rials were conducted by Braden using a cylinder measuring internal friction. Larson also measured viscometer [49] on alginates and both a cone and Young's modulus obtaining values in the range of plate, and a parallel plate viscometer [50] on rub- those reported by Dickson and ber base impression materials. Oglesby by the ultrasonic pulse-echo technique. the Much information has been published on the Recently using torsion pendulum in free vibration Barton and flow of dental waxes [51-53], but the emphasis has Dickson measured the shear modulus as well been on determining temperatures at which large [60] as the damping capacity of amalgam over the changes in flow rates take place, rather than in temperature range 23-51 °C. They found the shear making comprehensive analyses of the viscoelastic modulus to be about 3 X 10" psi, and the internal properties of the materials. friction was found to increase with temperature in The viscoelastic nature of dental amalgam has the range covered. Mahler investigated the sec- been recognized for many years. Data on the creep ondary or nonrecoverable creep of amalgam of dental amalgams for a period of up to 8 days [61] and found it to be related to clinical performance were shown by S6uder and Peters in 1919 [54], and, in fact, a flow or creep test was included in the in the mechanical function of amalgam in the Federal Specification for amalgam in 1926 [55]. mouth. This secondary creep behavior of amal- This test would be difficult to analyze precisely gam was studied as a function of temperature by since it is used as a measure of the setting time or Dickson, Oglesby, and Davenport [62]. The creep setting rate of dental amalgam. Thus, it provides and stress-relaxation responses have been used to a summation of the creep of a specimen over a 21- study the setting time of amalgam by Fuse [63]

142 .

Compressive stress-strain tests were carried out [8] Sweeney, W. T., Sheehan, W. D., and Yost, E. L., Mechanical on dentin, by Peyton, Mahler, and Herslienov [64] properties of direct tilling resins, J. Am. Dent. Assoc. 49:513 (Nov. 1954). and again later by Craig and Peyton [65]. In this [9] Sweeney, W. T., Yost, E. L., and Fee, J. G., Physical work Craig and Peyton also conducted compressive properties of plastic teeth, J. Am. Dent. Assoc. creep tests on human dentin. Dentin was shown to 56:833 (June 1958). be a viscoelastic material with a yield point above [10] Merriam, C. W. Ill, Mathematics for linear systems. Chapter 4, in Control Systems Engineering, which it appears to exhibit (1) instantaneous elas- Seifert, W. W., and Steeg, C. W., Ed. (McGraw- tic strain, (2) retarded elastic strain, and (3) vis- Hill, New York, 1960). cous strain, but below the yield point only the first [11] Roesler, F. C, Some applications of Fourier series two phenomena are observed. AVork on the creep in the numerical treatment of linear behavior, behavior of dentin should be extended in order to Proc. Phys. Soc. B68:89 (1955). [12] Kolsky, H., Stress Waves in Solids, Chapter 5 analytically describe the viscoelastic behavior in (Oxford University Press. London, 1953). compression and then the trans- using appropriate [13] Buchdahl, R., The rheology of organic glasses, formations, the correspondence between the com- Chapter 4, Rheology, Vol. 2, Eirich, F. R., Ed. pressive stress-strain and creep behavior shovdd be (Academic Press, Inc., New York, 1958). demonstrated. These data could then be compared [14] Nielsen, L. E., A recording torsion pendulum for the with that obtained by stress relaxation experiments measurement of the dynamic mechanical properties of plastics and rubbers. Rev. Sci. Inst. 22:690 experiments accounting aniso- and dynamic for (Sept. 1951). tropy effects. In Lugassy and Korostoff 1968, [66, [15] Nolle, A. W., Methods for measuring dynamic 67] rejwi'ted an experiment using sti'ess relaxation mechanical properties of rubber-like materials, techniques in dentin in whicli they studied the vis- J. Appl. Phys. 19:753 (1948). coelastic properties in relation to structural ani- [16] Tobolsky, A. V., and McLaughlin, J. R., Elasto- sotropy and compared the residts to polymer viscous properties of polyisobutylene, V. The transition region, J. Polymer Sci. 8:543 (1952). viscoelasticity. Recently Barton and Dickson [60] [17] Ferry, J. D., Mechanical properties of substances of have used the torsion pendulum to determine the high molecular weight, VI. Dispersion in con- moduli and the damping capacity of dentin at centrated polymer solutions and its dependence on temperature and J. mouth temperature as well as at room temperature. concentration, Am. Chem. Soc. 72:3746 (1950). Little work has been done in the relation of the [18] Fox, T. G., and Flory, P. J., Viscosity-molecular viscoelastic behavior of dental materials and tis- weight and viscosity-temperature relationships sues in relation to their atomic, molecular, and for polystyrene and polyisobutylene, J. Am. Chem. Soc. 70:2384 microstructural properties and processes yet there (1948). ; [19] Tobolsky, A. V., and Andrews, R. D., Systems is a large amount of theory and experimental tech- manifesting superposed elastic and viscous be- nology that could be broug'ht to bear. Two Avorks havior, J. Chem. Phys. 13:3 (1945). in the dental field in i-elation to microproperties [20] Andrews, R. D., Hofman-Bang, N., and Tobolsky, A. v.. Elastoviscous properties of polyisobutylene, are Myerson's work [46, 47] on methacrylate cross- I. Relaxation stress in whole polymer at different linking ill respect to mechanical properties and molecular weights at elevated temperature, J. the work of Dickson and Oglesby [62] on the Polymer Sci. 3:669 (1948). steady-state ci'eep of amalgam and the detei'mina- [21] Williams, M. L., Landel, R. F., and Ferry, J. D., The temperature dependence of relaxation tion of the activation energy of the process. mechanisms in amorphous polymers and other glass-forming liquids, J. Am. Chem. Soc. 77:3701

1 1 . References (1955). [22] Bueclie, F., Physical Properties of Polymers, (Wiley [1] Leaderman, H., Elastic and Creep Properties of Interscience, New York, 1962). Filamentous Materials and Other Higli Polymers, [23] McCrum, N. G., and Morris, E. L., On the measure- (The Textile Foundation, Washington, D.C., ;ment of the activation energies for creep and 1943). stress relaxation, Proc. Roy. Soc. A281:258 [2] Catsiff, E., and Tobolsky, A. V., Stress-relaxation (1964). of polyisobutylene in the transition region (1, 2), [24] Alfrey, T., Mechanical Behavior of High Polymers, J. Colloid Sei. 10:375 (1955). p. 551 (Interscience, New York, 1948). [3] Ferry, J. D., Viscoelastic Properties of Polymers [25] Nielsen, L. E., Mechanical Properties of Polymers (John Wiley and Sons, Inc., New York, 1961). ( Beinhold. New York, 1962 ) [4] Gross, B., Mathematical Structure of the Theories [26] Schwarzl, F., and Staverman, A. J., Higher approxi- of Viscoelasticity (Hermann, Paris, 1953). mation methods for the relaxation spectrum [5] Leaderman, H., Viscoelasticity phenomena in from static and dynamic measurements of visco- amorphous high polymeric systems. Chapter 1 in elastic materials, Appl. Sci. Res. A4:127 (1953). Rheology, Vol. 2, Eirich, F. R., Ed. (Academic Press, Inc., New York, 1958). [27] Nakada, O., Theoiw of non-linear responses, J. Phys. Soc. [6] Lee, E. H., and Radok, J. R. M., The contact problem Japan 15:2280 (1960). for viscoelastic bodies, J. Appl. Mech. 27:438 [28] Leaderman, H., McCrackin, F., and Nakada, O., (Sept. 1960). Large longitudinal elastic deformation uf rubber- [7] Nolt, I. G., and Meier, J. A., On the spherical in- like network polymers, II. Application of a gen- denter as a means for determining viscoelastic eral formulation of non-linear response. Trans. material functions. Proceedings of the Fourth Soc. Rheo. 7:111 (1963). International Congress on Rheology. Part 2, Lee [29] Greene, A. E., and Rivlin, R. S., The mechanics of E. H., Ed., pages 167-181 ( Interscience Pub- nonlinear materials with memory. Arch. Rational lishers, New York, 1965). Mech. Anal. 1:1 (1957).

143 [30] Oglesby, P. L., Dickson, G., Rodriguez, M. L., [48] Braden, M., and Stafford, G. D., Viscoelastic prop- M., T., Visco- Davenport, R. and Sweeney, W. erties of some denture base materials, J. Dent. elastic behavior of dental amalgam. J. Res. Nat. Res. 47:519 (1968). Bur. Stand. (U.S.), 72C (Eng. and Instr.) No. 3, [49] Fish, S. F., and Braden, M., Characterization of the 20^-213 (July-Sept. 1968). setting process in alginate impression materials, A. A., plastic structures for [31] MacLeod, Design of J. Dent. Res. 43:107 (1964). complex static stress systems, Ind. Eng. Chem. [50] Braden, M., Viscosity and consistency of impres- 47:1319 (1955). sion rubbers, J. Dent. Res. 46:429 (1967). [32] Weber, C. H., Robertson, E. N., and Bartoe, W. F., [51] Craig, R. G., Eick, J. D., and Peyton, Time- and temperature-dependent modulus con- F. A., Proper- ties of natural waxes used in dentistry, cept for plastics, Ind. Eng. Chem. 47:1311 (1955). J. Dent. Res. 44:1308 (1965). [33] Fukada, E., On the relation between creep and [52] Ohashi, M., and Paffenbarger, vlibrational loss of polymethylmetacrylate, J. G. C, Melting, flow and thermal expansion Phys. Soc. Japan 6:254 (1951). characteristics of some dental and commercial waxes, [34] McCrackin, F. T., and Bersch, C. F., Creep behavior J. Am. Dent. Assoc. 72:1141 of transparent plastics at elevated temperatures, (1966). [53] Ohashi, M., Soc. Plastics Eng. J. 15:791 (1959). and Paffenbarger, G. C, Some flow characteristics [35] Thompson, E. V., Secondary processes of poly- at 37 °C of ternary wax mixtures (methyl methacrylate) and their activation that may have possible dental uses, J. Nihon energies as determined by shear and tensile creep Univ. Sch. Dent. 11:109 (1969). compliance measurements, J. Polymer Sci. Part [54] Souder, W., and Peters, G. C, Investigation of the A-2 6:433 (1968). physical properties of dental materials. Dental [36] Bueche, F., Viscoelasticity of polymethacrylates. J. Cosmos 62:305 (1920). Appl. Phys. 26:738 (1955). [55] Souder, W., Measurement and application of cer- [37] McLaughlin, J. R., and Tobolsky, A. V., The visco- tain physical properties of dental materials, J. elastic behavior of polymethyl methacrylate, J. Dent. Res. 7:173 (1927). Colloid Sci. 7:555 (1952). [56] Smith, D. L., Caul, H. J., and Sweeney, W. T, [38] Deutsch, K, Hoff, E. A., and Reddish, W., Relation Some physical properties of gallium-copper-tin between the structure of polymers and their alloys, J. Am. Dent. Assoc. 53:677 (1956). dynamic mechanical and electrical properties, [57] Rodriguez, M. S., and Dickson, G., Some tensile Part I. Some alpha-substituted acrylic ester poly- properties of amalgam, J. Dent. Res. 41:840 mers, J. Polymer Sci. 13:565 (1954). (1962). [39] Heyboer, J., Dekking, P., and Staverman, A. J., [58] Dickson, G., and Oglesby, P. L., Elastic constants The secondary maximum in the mechanical damp- of dental amalgam, J. Dent. Res. 46:1475 (1967).

of : of ing polymethyl methacrylate Influence [59] Larson, R. V., Internal Friction Damping in Dental temperature and chemical modiflcaltions, Chapter Amalgam. Thesis, Dept. of Mechanical Engineer- 14. Proceedings of the Second International ing, University of Utah, August 1967. Congress on Rheology. Harrison, V. G. W., Ed. [60] Barton, J. A. Jr., Dickson, G., and Oglesby, P. L., ( Butterworths Scientific Pub., London 1954). Determination of properties of dental materials [40] Koppelmann, J., Tiber den dynamischen Elastizi- 'by means of a torsion pendulum, lADR Program tiitsmodul von Polymethacrylsauremethylester and Abstracts of Papers, p. 172 (March 1968). bei sehr tiefen Frequenzen. Kolloid Z. 164:31 [61] Mahler, D. B., and Von Eysden, J., Dynamic creep (1959) . of dental amalgam, J. Dent. Res. 48:50 (1969). [41] Nielsen, L. W., Dynamic mechanical properties of [62] Dickson, G., Oglesby, P. L., and Davenport, R. M., high polymers, Soc. Plastics Eng. J. 16:525 The steady state creep behavior of dental amal- (1960) . gam, J. Res. Nat. Bur. Stand. (U.S.), 72C (Eng. [42] Schmieder, K., and Wolf, K., Uber die Temperatur- and Instr.), No. 3, 215-229 (July-Sept. 1968). und Frequenzabhangigkeit des mechanischen Fuse, N., Flow, stress relaxation and compressive Verhaltens einiger hochpolymerer Stoffe. Kolloid [63] strength of dental amalgam. Bull. Tokyo Med. and Z. 127:65 (1952). Dent. Univ. 16:17 (1969). [43] Barber, Ronald, Preliminary tests of some of the A., B., B., newer denture materials, J. Am. Dent. Assoc. [64] Peyton, F. Mahler, D. and Hershenov, 21:1969 (1934). Physical properties of dentin, J. Dent. Res. 31 :366 (1952). [44] Sweeney, W. T., and Schoonover, I. C, A progress report on denture base material, J. Am. Dent. [65] Craig, R. G., and Peyton, F. A., Elastic and mechani- Assoc. 23:1498 (1936). cal properties of human dentin, J. Dent. Res. [45] Sweeney, W. T., Paffenbarger, G. C, and Beall, J. 37:710 (1958). R., Acrylic resins for dentures, J. Am. Dent. Assoc. [66] Lugassy, A. A., and Korostoff, E., Comparative 29:7 (1942). viscoelastic properties of dentin and bone, lADR [46] Myerson, R. L., The Internal Friction of Linear and Program and Abstracts, 46th General Meeting, Crosslinked Phosphate Glasses and Methacrylate p. 99 (March 1968). Polymers. M.I.T. Master's Thesis. May 1961. [67] Lugassy, A. A., Mechanical and Viscoelastic Prop- [47] Myerson, R. L., Effect of crosslinking on mechanical erties of Cow Bone and Sperm Whale Dentin damping and stiffness of methacrylates, lADR Studied under Compression. Thesis, Graduate Abstracts, Fortieth General Meeting, p. 89, School of Arts and Sciences, University of Penn- (1962). sylvania (1968).

144 NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseakch, Proceedings of the oOth Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Low Frequency Determination of Mechanical Properties

Richard L. Myerson

Myerson Tooth Corporation, Cambridge, Mass. 02139

The torsion pendulum is a valuable tool for use in determining the mechanical properties and molecular structure of dental materials. With the torsion pendulum the

modulus of a material may be determined from the frequency of oscillation ; the internal friction can be obtained from the rate of damping. Change in internal friction with change in frequency provides insight into type of structure and relaxation mechanisms. Activation energy of processes can be determined from shift in peak energy absorption temperature with change in frequency. A principal advantage of low frequency testing is in the resolution of dispersions or of internal friction peaks. The torsion pendulum method has been applied to various polymer materials in studies of the relationship of composition and transitions, effects of cross-linking, comparison of cast poly (methyl methacrylate) and powder-liquid molded poly (methyl methacrylate) and the relationship of impiact resistance to peak energy absorption temperatures. Studies have been made of the relationships between peak temperatures and composition and cross-linking in sodium phosphate glasses. A review of equipment is presented.

Key words : Activation energy ; dental materials glass, sodium phosphate ; internal ;

friction ; mechanical properties ; polymers, dental ; torsion pendulum.

1. Introduction crystallites in the case of a crystalline or semi- crystalline material. The focus of this report is to be on the use of the Braden and Stafford [2] reviewed the viscoelas- low frequency mechanical properties as a tool for tic properties of some denture base materials using analyzing the molecular structure of the materials a torsion pendulum at less than one cycle per sec- being considered. The emphasis will be on the tor- ond. They determined glass transition tempera- sion pendulum. With the torsion pendulum one tures which are important in setting a limit of can determine the modulus of a material by ob- temperature to which a material can be safely serving the frequency of oscillation. The internal exposed in cleaning, repair, etc. They determined friction of a material is obtained from the rate of the shear modulus with good agreement to calcula- damping. By noting the change in values of inter- tions made from the tensile modulus. Finally they nal friction with changes in frequencies, one can determined the internal friction spectrum from gain insight into the type of structure and the + 20 °C to +105 °C. They concluded that some relaxation mechanisms. The activation energy of of the materials of high impact resistance gained the processes involved can be determined from the their impact resistance by sacrifice of the glass shift in peak temperature with change in transition temperature, which limited their higher frequency. temperature usefulness. One impact resistant ma- If the internal friction of a material is plotted terial also had a higher internal friction. The au- versus temperature for a fixed frequency of oscilla- thors stated that this might mean less elastic tion, there is usually a general rise in the value behavior. with a rapid increase in the area of softening. There I suggest that the value of tan 8 (internal fric- will often be maxima at various lower tempera- tion) as an indication of specific end use properties ture points. For silicate ceramics the lowest tem- might be misleading. First of all, the experimen- perature peak has been usually ascribed to an al- strains operating kali ion diffusion under stress, an intermediate tation is done with very small peak to oxygen ion diffusion, and finally the high in the linear viscoelastic range. Most stresses that temperature peak to visco-elastic motion. In poly- are damaging will, of course, involve larger strains mers these peaks at the lowest temperature have and will be nonlinear. Secondly, the loss com- been ascribed to side chain motion and/or very ponent in plastics may be due to side chain motion local motion on the main backbone chain [1].^ The rather than main chain motion, and this would higher temperature peaks are usually caused by probably reflect itself in retarded elasticity rather softening of the material or by the melting of than in creep. A significant use of the internal friction spectrum is as a means of characterizing 1 Figures In brackets Indicate the literature references at the end of this paper. a material. It will provide some indication of the

145 : likelihood of a build-up of heat during a cyclic 3. Theoretical Aspects and Interpre- loading. It provides a signal that there may or tation of Torsion Pendulum Data may not be creep under long term steady state loading by identification of relaxation processes The theoretical aspect of an oscillating stress as and the time-temperature effects. applied for a torsion pendulum has been well The energy absorption patterns versus fre- covered in the literature, and suitable references quency and temperature will yield information are included with this paper. Because of limitations that, in part at least, is attributable to molecular of space, only the conclusions, with a minimum structure specifics. By controlled experiments—as of the background, will be presented here. examples, varying the composition of a copolymer In the Voigt solid (fig. 1), a model of a vis- or the alkali ion I'atio in ceramics—and by use of coelastic solid is presented. The relationship of other information such as diffusion data, dielectric stress and strain for such a solid is as shown in losses, etc., the peaks of energy absorption can be eq 1. identified. Then the effect of the degree of polymerization or degradation or devitrification, etc., can be better understood by observing the internal friction spectrum changes. An important consideration is the background level of internal friction apart from the peaks. The background gives an indication of the general molecular mobility. In the case of a completely cross-linked glass—i.e., fused quartz—there is virtually zero damping.

2. A Contrast of High and Low VOlGT MAXWELL Frequency Testing Figure 1. Voigt and Maxwell models It is of interest to contrast the use of high and for, mechanical behavior. low frequency testing. Each has advantages, and the of frequency measure- use both high and low de , (1) ments will give ^^ery valuable data relative to the activation energy of phenomena because of the where large spread in rate of loading. There are also, of ff= stress course, some very specific properties of importance elastic modulus in the end use for which damping properties meas- e = strain ured at one end of the spectrum would be more 77= viscosity important. The principal advantage of low frequency testing is in the resolution of dispersions It can be seen that if a force is cyclic (sinusoidal) or of internal friction peaks. The lower temper- the term for the elastic component will be in ature peaks are usually of lower activation energy, phase, whereas the viscous flow term will be and as the frequency- is increased, they shift to 90° out of phase. Mgher temperatures at a more rapid rate than the In a real solid, the resultant complex shear will be the resultant of the real higher frequency peaks. At sufficiently high fre- modulus of G* of the solution, G' (representing the elastic quencies this will lead to a blending of peaks. part portion) and the 90° out-of-phase, imaginary Nielsen [1] points out that the torsion pendulum portion, G" as is shown in figure 2. is useful over a very wide range of moduli, and from very low frequencies to moderately high. He felt the best frequency range was from a 100th of a cycle per second to 10 cycles per second. For determining low frequency damping data there are three general types of equipment. The first, and the one which will be emphasized in this paper, is the freely oscillating pendulum operating to O". at_ the system resonance. A second type is a Figure 2. Relation of tan 8 G' and driven system involving its natural resonance, such as a vibrating reed. Here the internal friction The angle 8 represents the lag of the strain to is determined by the shape of the amplitude versus stress. Tan S is equal to G"/G'. temperature, frequency curve in the resonance area. The Zener [3] showed that for a given third type, a driven nonresonant rotating beam, tan 5 is a maximum when the mean relaxation developed by Maxwell, will be discussed briefiy time (the mean of constant stress relaxation below. time and constant strain relaxation time) is equal

146 : :

3. Elasticoviscous effects where is frequency. Thus, if the tempera- to jr-^ F 4. Viscous flow 5. Precipitated crystallites ture is held constant and the frequency of loading varied, at one frequency there will occur a max- For the typical alkali silicate ceramic there is a imum in the damping. This will yield the relaxa- distinct tan S peak around room temperature that tion time of the material at this temperature. has been associated with the diffusion under stress Zener also developed the equation for tan 5 as of the sodium ion. A higher peak has been asso- a function of the log decrement of the amplitude. ciated with the nonbridging oxygen ion. The elas- The equation was arrived at by considering the ticoviscous effects he relates to movements of large rate of energy dissipation per unit volume (in segments of the glassy network as the transforma- the range where tan 5 is approximately equal to tion region is approached. He explained the diffu- sin 5) . He arrived at eq 2 sion of the sodium ion by considering that under a stress there will be a redistribution of electrical charges, and in tliis way the equilibrium position Tan s^IeAM^ (2) of the sodium ion will shift. In some cases the ion must get past an obstacle in order to reach its new Ao is the initial amplitude; A„ is the amplitude equilibrium position. The particles with sufficient after n cycles. energy will be able to get past an obstacle. The distribution of particle energy is related to temper- 3.1. Activation Energies ature. Thus the diffusion rate will be temperature dependent and the typical activated process equa- Since the relaxation phenomena express rate tions will result. processes that are temperature dependent, the Kirby showed that soda lime glasses and activation energy can be determined by testing at [7] borosilicates have distinct peaks and different several frequencies. If a viscoelastic solid is tested background values. Of interest was that fused at a constant frequency, there will be a tempera- silica, which has nonbridging oxygen ions and, ture showing a maximum in damping. By testing of course, no alkali or alkaline earth ions, has at several frequencies and determining the peak extremely low damping coefficient until the temperatures, the activation energy can be elasticoviscous effects are reached. Blum in a determined for a specific relaxation process by [8] very carefully controlled experiment showed a eq 3. value of 0.000012 for tan 8. 1 EIRT Avith 7 = 7^6 (3) Forry [9] worked three compositions of 2ttF' soda in sodium silicate glasses. As the soda per- Where: 2^= frequency centage is increased, the low temperature peaks T=relaxation time occurred at lower temperature, were higher, and were sharper. activation energy He found the height above back- i?=gas constant ground was almost linear with soda content, ex- T= absolute temperature trapolating to zero at about 10 percent soda content. The in-phase component or the elastic compo- Hoffman and Weyl [10] reviewed the effect of nent of the complex modulus stores energy as alkali and alkaline earth oxide on silicate glasses. potential recoverable energy. The out-of-phase They found that lithium had the highest activa- imaginary component results in an energy loss, tion energy (the smallest ion and thus the most principally in internal heat. tightly bound), sodium the next, and potassium For real materials there is no single relaxation the lowest. The alkaline earth oxide additives, be- time as would be calculated for a simple Voigt cause they have network bridging properties, solid. There will exist a distribution of relaxation tighten the structure and suppress the alkali dif- times, hence a range of temperatures sho^ving fusion peak. high damping for a single frequency. Ryder [4] Shelby and Day [11] studied the effect of mixed pointed out that the area under a peak of tan 6 alkalis and silicate glasses. Figure 3 shows some versus will temperature curve be related to a of their findings. As the second alkali is added, number of diffusing species. For the same number the alkali peak shifts to a higher temperature and of diffusing elements a narrow distribution of becomes smaller. The nonbridging oxygen peak relaxation times will result in a high narrow peak became slightly higher. A new damping peak, compared to a low broad peak for a broad distribu- large and close to the nonbridging ion peak, was tion of relaxation times. The effect of changing developed and, in fact, can mask it. From their compositions can, therefore, be reviewed for this work they found that it is easy to detect a small factor as well as the locations of the peaks. amount of a second alkali. Fitzgerald [5, 6] listed some of the sources of Fitzgerald [12] showed the effect of annealing energy loss in ceramic materials. He included versus chilling of sodium silicate glasses. In the 1. Flow of heat annealed state—that is, the tighter structure 2. Diffusion of ionic species state—the activation energy for the sodiimi ion

147 : :

AFTER KAEBLE, 1965

T T g m

TEMPERATURE (°K)

Figure 4. Modulus of polymers as a function of temperature at constant frequency.

results in very low internal friction and high modulus. As the temperature is increased, side chain motion or very local chain motion can take place

I I 1 I I -200 -100 0 +100 +200 +300 +40C and the modulus will go down. At the glass tran-

TEMPERATURE "^C sition temperature, segments of the chain will start in motion and there will be a large drop in modu- Figure 3. Internal friction of mixed alkali-silicate lus. There will be a rubbery plateau above the glasses. glass transition temperature, depending on the degree of cross-linking and the molecular weight. diffusion was 20 kilocalories per mole versus 16 The last reduction in modulus above the glass kilocalories per mole for the chilled glass, thus transition temperature for amorphous materials giving some indication of the effect of amiealing comes when the chains themselves actually slip as on the network. in molding. This region is very dependent on cross-linking, and if the material is cross-linked 3.2. Viscoelastic Behavior the slippage does not occur until decomposition starts. For linear materials, if there is high molec- Fichter [13] listed a number of possible causes ular weight, entanglements will act as temporary of viscoelastic relaxation for metals. damping cross-linking sites and will raise the temperature These include of the slip point. In the case of crystalline mate- 1. Interfaces grain boundaries. and rials, the glass transition temperature is not ac- 2. Changes in lattice order due to stress. companied by a great change in modulus until 3. Magnetoelasticity, particularly in ferro- the crystallites melt. magnetic materials. Boyer [16], in discussing the plastic transitions 4. function local tempera- Phase changes (a of below the glass transition temperature, lists the ture) . following 5. Precipitation and recrystallization. (a) Side chain group motion such as the ester Nowick [14] also reviewed anelastic phenomena groups in polyalkyl methacrylates. with emphasis on metals. He cited the example of (b) In-chain motions of subgroups as in the aluminum where there is a substantial peak in tan polycarbonates. S at low frecjuency at about 300 °F (150 °C) in a (c) "Crank shaft" motion as in polyethylene polycrystalline material. This peak is absent in a when there are about four consecutive single crystal. He attributes this to viscoelastic methylene groups. behavior in the grain boundaries. He describes the glass transition temperature Kaelble [15] presented the diagram in figure 4 phenomenon as large "crank shaft" motions in- giving an overview of the phenomena in polymers. volving 30 to 40 carbons in the chain. Above the The plot is of modulus versus temperature at con- glass transition temperature the transitions occur stant frequency. The relaxations accompanied by with coordinated motions of entire chains. changes in modulus level will also be accompanied McCrum et al. [17] point out that the assign- by peaks in the internal friction curve. At the very ment of mechanisms of relaxation to polymers has low temperature, polymer strain in response to been done qualitatively by experiments rather than stress is due to primary bond angle bending and quantitatively. The convention used for polymeric stretching. Hence there is no diffusion and this materials is that the alpha peak is the highest

148 : —

temperature, and beta the next lowest, etc. With copolymers of methyl and cyclohexyl methacrylate. the amorphous materials the alpha peak is usually The change in the nature of the curves is nearly the glass transition temperature when there is a quantitatively additive. Poly (methyl methacry- modulus change of three orders of magnitude or late) has no low tempei'ature peak but a broad beta more. The lower temperature transitions are due peak in the 40 °C region, whereas polycyclohexyl- to the local motions in the polymers while they methacrylate has a similar glass transition tem- are still in the glass-like condition; that is, the perature to poly (methyl methacrylate) and a low larger segments of the chains are frozen in the temperature beta peak below 0 °C. The intermedi- glassy state, but the small groupings, are able to ate copolymers are intermediate in both these move above their own transition temperatures. In respects. The glass transition temperature is partially crystalline materials, such as polyethyl- relatively unchanged since the end members were ene, the alpha peak is usually the melting of the similar. crystallites and the beta the glass transition tem- Nielsen [1, 20] pointed out that if the rates of perature of the amorphous phase. Thus the magni- reaction are unequal, the initial polymer formed tude of this peak will be alfected by the degree of will be rich in the faster reacting material, and crystallinity, and indeed this can be used as a the final polymer formed rich in the slower mate- measure of the degree of crystallinity in a study rial. This will result in broader and lower peaks of a particular polymer. McCrum mentions that than in homogenous copolymers. there is a low temperature absorption peak in With graft or block polymers, where there are nylons and methacrylates due to absorbed im- two separate phases insoluble in each other pres- purities (e.g., water). He also suggests that the ent, the resultant tan S versus temperature curve ''crank shaft" motion can be applied to other will show the seperate features of each phase in groupings in a chain other than four methylenes, proportion to their concentration. Similarly, poly- provided they allow simultaneous rotation of two blends made by mechanical mixing will show the colinear bonds to give the "crank shaft" effect. character of both phases. Nielsen gives a qualitative picture for the [1] Deutsch et al. [19] in comparing methyl meth- energy absorption at the transition of a polymer, acrylate and methyl a-chloro-acrylate, showed that wherein as the temperature increases, frozen seg- in replacing the methyl group with the more polar ments start to move. Below that temperature the substituent, the glass transition temperature in- damping is low because deformation is bond angle creased from 110° to 140° in the case of the bending, and hence highly elastic with little heat C C dissipation. At temperatures well above the tran- methyl a-chloro-aorylate. Also, the beta peak tem- sition, damping is low because the chain seg- perature was increased and the activation energy ments are free to move, but the resistance is low of this peak significantly increased due to the and hence the modulus is low. Thus if the segments stronger interchain bonding caused by the polar involved are either frozen or very free to move, group. damping would be low. The damping peak will 1 1 1 occur at the intermediate condition. COPOLYMERS Nielsen [18] points out that in polymeric mate- METHYL METHACRYLATE CYCLOHEXYL METHACRYLATE rials, damping losses will be affected by the following 1.00 0.00 — . 0.80 0.20 Crystallinity 0.60 0.40 — — 0.40 0.60 Cross-linking 0.20 0.80 Number of phases 0.00 1.00 State of aggregation AFTER McCRUM, READ AND WILLIAMS, 1967 SOURCE - HEIJBOER, 1956 ? Various chain motions Deutsch et al. [19] noted that plasticizers, residual monomer, absorbed water, etc., complicated the systems. A 1 / / 3.3. Internal Friction and Copolymerization O ' J f Internal friction measurements are useful for the interpretation of copolymerization. If the K / • starting materials are comonomers of close reaction rates, the end product will be homogenous /A y ""S^ with a distribution of effects such as glass transition temperature intermediate between the end 1 1 1 1 -100 -50 0 +50 +100 members. The sharpness of the tan S versus tem- TEMPERATURE °C perature peaks will tend to be intermediate between the end members. McCrum, et al. [17] show Figure 5. Relation of tan 6 to temperature for varying in figure 5 the additive nature of a series of methacrylate copolymer ratios.

149 3.4. Impact Resistance and Low Temperature increase in the temperature of the tan 8 peak rela- Peaks tive to the matrix resin. Reinhardt [29] was concerned with the means authors 21-26] have considered the Many [1, of measuring the progress of the final stages of relationship between the presence of a low tem- cross-linking that would be more meaningful than perature peak in a rigid material and impact some of the indirect physical property tests. The strength. In early observations there seemed to be validity of the chemical tests for unsaturation is a correlation between the presence of such a peak doubtful because of the insolubility of the cross- and impact resistance. Later data show that the linked polymers. His study was based on dynamic low temperature peak is neither a necessary nor mechanical tests with specimens of varying cure. sufficient condition for impact resistance. However, During the termination of cure, as the network it is of significance in many cases and can lead to lightens there is an asymptotic increase in modulus an interpretation of the mechanisms involved. and a decrease in tan 8. He showed that curing Matsuoka [25] in his review of the dynamic for 2 hr at 100° C produces a higher modulus and properties of glassy and partly crystalline mate- lower tan 8 (presumably more cross linking) than rials versus resistance, points out that in impact in the self-curing system cured 500 hours at 35 °C. polymers, ahead of a propagating crack tip, there Malpass [30] showed that high strains (orienta- IS plastic deformation. The greater this deformation in the polymer) produced a definite peak that tion the more energy is required to propagate a correlated with physical properties (they helped crack and the tougher the material will be. low The when in the direction of the stress, but were temperature signifies molecular motion transition detrimental when at 90° to the stress). Annealing and should relate to impact resistance. However, eliminated these peaks and brought the properties low temperature peaks due to side chain motion in line with the normal material. do not relate to the ability of the chains to draw under stress and, hence, are not related to impact 4. Experimental Results resistance. Peaks due to local motion in the main chains are related to impact resistance. 4.1. Glasses In partially crystalline materials the beta peak usually signifies the glass transition temperature A series of tests on sodium phosphate glasses of the amorphous phase, and thereby molecular were conducted by this author [31]. The glasses motion. These materials have good impact resist- were drawn into fibers and internal friction versus ance between the glass transition temperature and temperature curves made at frequencies around the temperature corresponding to the alpha peak one cycle per second. Sodium phosphate glasses indicative of the melting of the ciystallites. will vary in structure depending on the ratio of The relationship between the secondary low sodium oxide to phosphorous pcntoxide. When the temperature peak and impact resistance is indirect. molar ratio is 1, the theoretical structure will be When it signifies side chain motion, as in a cyclo- a long chain. As the ratio becomes greater than 1, hexyl methacrylate, it is of no significance. In mix- chains become shorter and shorter. If the ratio is tures of two materials (or in some block and graft less than 1—that is, the molar proportion of phos- copoilymers) wherein there are two phases, if one phorous pentoxide is greater than the molar pro- phase is a rubber there will be a low temperature portion of sodium oxide—the structure is cross- peak. This does not necessarily mean impact re- linked. The greater the percentage of phosphorous sistance. If the rubbery particles are of the proper pentoxide, the greater the degree of cross-linking. size and distribution and well bonded to the ma- In the sodium phosphate glasses the backbone trix, they will nucleate drawing of the glassy phase structure is an alternating series of oxygen phos- under stress and impart impact resistance. phorous atoms. Each phosphorous atom is sur- Heijboer [23] and Turley [22] point out that rounded by four oxygen atoms, three of which are the relation between the temperature of the peak capable of bonding with another phosphorous or and the tough brittle transition is not clear. Also, sodium atom, and the fourth is a nonbridging oxygen atom with a double as in the case of poly (2-6 dimethyl-p-phenylene bond to the phosphorous. Thus there is present in this structure the sodium oxide) , one can have impact resistance with no low ion and the nonbridging oxygen ion as in the temperature peak. It should be borne in mind in sodium silicate glasses. In figure 6 it can be seen the correlation of tan 8 and impact resistance, that there are indeed two peaks below the glass that the torsion pendulum data is in the linear transition temperature; the lowest temperature viscoelastic range, whereas impact failures involve peak is probably the sodium ion diffusion peak, large scale molecular movement. the intermediate the nonbridging oxygen ion. As 3.5. Composites, Crosslinking, and Orientation the sodium to phosphorous ratio is dropped, the low temperature peak drops. Conversely, as the sodium Maeda [27] and Galperin [28] show that in the to phosphorous ratio drojDS, the medium tempera- case of composites there is a tendency towards a ture peak increases in height. Both of these decrease in the height of the tan 8 peak and an phenomena are in line with the fact that as the

150 —

1 1 1 1 T 1 1 r

TEMPERATURE °C

Figure 6. Internal friction of glasses with varying Na/P ratios.

TEMPERATURE OC sodium to phosphorous ratio decreases, the sodium Figure 7. Internal friction as by crossUnking in ion concentration decreases, but the nonbridging effected a hexyl methacrylate-ethylene dimethacrylate system. oxygen ion concentration increases. It is of interest to note that the glass transition temperature goes down as the phosphorous pent- beta peak of methyl methacrylate at about 40 °C. oxide content goes up and as the cross-linking goes The internal friction in the rubbery region falls up, contrary to what is seen in organic glasses. This ofi' to a low value, lower than that in the glassy suggests that the sodium ion, in coordination with region. oxygen ions of other chains, bonds the chains more Ethylene dimethaci-ylate showed a veiy broad strongly than does the actual chain cross-linking. peak at 45 °C. The rate of fall of the modulus This is supported by the data of van Wazer and through this transition Avas small, indicating that Hoist [32] who found that when three oxygens it is not the normal glass transition phenomenon. were shared with other PO4 tetrahedra the com- Again this may be similar to the methyl meth- pound w^as less stable than when only two or one aciylate beta peak. Decomposition of the materials oxygens were shared with other PO4 tetrahedra. started above 200 °C. The modulus of the ethylene dimethacrylate polymer at -1-200 °C w^as approxi- 4.2. Polymers mately half that at nearly —200 °C, thus showing a remarkably small rate of fall with temperature. Work done by this author on the effect of ci'oss- A recent series of experiments were conducted linking witli a hexyl methacrylate-ethylene di- with a number of dental materials. They were methacrylate system was presented to the Dental fabricated and machined to specimens approxi- Materials Group of the International Association mately 2 in (50 mm) in length between grips, 0.200 for Dental Research in 1962 [31-33]. Figure 7 in (5 mm) wide and 0.062 m (1.5 mm) thick. The shows the salient features of that series of tests. specimens were maintained free from water by There is a reduction in background as cross-linking storage in a desiccator until testing was done. increases. For the hexyl methacrylate formulations The tests "were conducted on the Plas-Tech. there is a low temperature dispersion scale off Direct Recording Torsion Pendulum. The fre- that is, —200 around °C—probably related to the quency was in the range of 2 to 10 cycles per sec- hexyl group. There appears to be, in the hexyl ond, and the temperature ranged from —120 °C to methacrylate, a beta peak at about —100 °C. The -I- 120 °C. The tests were of a scanning nature. The glass transition temperature was raised consider- chamber was cooled to abou.t —120 °C with liquid ably by addition of 25 percent ethylene di- nitrogen, and then the temperature raised about methacrylate. It should be noted in this case that 1 °C per minute. Readings were taken every 10 °C. there is a very broad peak, possibly similar to the Equilibrium was not reached for each temperature

15J .

and the results, therefore, are to a certain extent relative. Only one specimen for most experiments was tested; for more precise data of certain key- 0 CAST POLYMEThVL MFTHACRYLATE points the temperature should be controlled and a A CAST POLYETHYL METHACRYLATE .26 - niunber of runs made. a CAST COPOLYMER (.5 MMA - .5 EMA) Samples of methyl methacrylate polymer, ethyl methacrylate polymer, and an equal-by-weight copolymer of the two were cast in test tubes and machined to the specimen geometry. Figure 8 shows tan 8 versus temperature curves for the three materials. It can be seen that the ethyl meth- .18 acrylate and methyl methacrylate produce similar polymers, with, as expected, the ethyl methacryl- Z O ate having a lower softening temperature. The M n copolymer is intermediate although much closer H .14 « to the ethyl methacrylate than might be expected. Em Figure 9 shows the modulus versus temperature Z data. The copolymer is intermediate between the s two end members in this prt>perty. To test the effect of a mixture rather than a copolymer of ethyl methacrylate and methyl methacrylate, 50 parts by weight of ethyl methacryl- .06 ate monomer and 50 parts of methyl methacrylate .04 polymer were rnixed together in a dough and molded in the usual dental fashion, machined .02 and tested. Figure 10 shows data for the copolymer and the mixture plotted together. It can be + 100 seen that the mixture has a broader softening TEMPERATURE °C transition point, as is shown by the lower rate of Figure 8. Internal friction of methyl methacrylate, ethyl increase of tan 8 with temperature. Of great in- methacrylate and an equal weight copolymer of the terest in this figure is the prominent peak at ap- two methacrylates

-r -r T"

• CAST POLYMETHYL METHACRYLATE 4.0 ^ CAST POLYETHYL METHACRYLATE

O CAST COPOLYMER (.5 MMA - .5 EMA)

3.0

2.0

1.0

-60 -40 +120

TEMPERATURE °C

Figure &. Effect of temperature on modulus of methyl m,ethacrylate, etlvyl methacrylate and equal weight copolymer of the two methacrylates. To convert from psl (pounds per square Inch) to N/m^ (newtons per square meter), multiply by 6895.

152 proximately —10 °C. This was unexpected and, as will be shown below, is in some way associated with the polymer phase. Figure 11 shows the modulus versus temperature data for the copolymer and the mixture. Over most of the range the modulus of the mixture is greater than the copolymer. There appears to be a change in modulus level at about the same temperature as the low temperature maximum. To determine the effect of the polymer-monomer fabrication versus casting, a mixture of the above poly (methyl methacrylate) powder and methyl methacrylate monomer was made of equal parts by weight and molded. In addition a compression molded sample of the polymer was also made. In figure 12, the internal friction versus temperature curves are plotted. The data from the compression molded sample are remarkably similar to the data of the cast sample with the exception of the pronounced low temperature peak. It is clear that this phenomenon is related to this polymer, possibly to the complex surface of the material. The polymer-monomer mix material showed a similar low temperature peak and a lower

I 1 I I I glass transition temperature. It can be seen in -100 -50 0 +50 +100 figure 13 that the cast specimen produced the high- TEMPERATURE °C est modulus of rigidity presumably because of a Figure 10. Internal friction versus temperature for higher molecular weight produced the slow ethyl methacrylate-methyl methacrylate copolymer and by mixture. cure.

CAST COPOLYMER (.5 MMA - .5 EMA) 4.0

.5 EMA MONOMER a POWDER-LIQUID MOULDED .5 MMA POLYMER

3.0

Ho X H W 2.0 -

1.0

-L. -120 -100 -80 -60 -40 -20 -0 +20 +40 +60 +80 +100 +120

TEMPERATURE °C

Figure 11. Modulus versus temperature for ethyl methacrylate-methyl methacrylate copolymer and mixture.

To convert from psl (pounds per square Inch) to N/m» (newtons per square meter), multiply by 6895.

153 I T A comparison of the internal friction curve of "powder-liquid" denture base material of • CAST POLYMETHYL METHACRYLATE conventional impact resistance with cast poly (methyl 75 MMA M0N0^4ER methacrylate) is .26 POWDER-LIQUID MOULDED shown in figure 14. The low tem- ,5 M^4A POLYMER perature properties are similar with the exception ^ COMPRESSION MOULDED MMA POLYMER of the polymer associated peak described above. Since the impact resistance is not enhanced, this .22 peak is not likely to be an "in the chain" molecular phenomenon. The slightly subdued beta peak would indicate the presence of a nonmethacrylate component. The internal friction curves for two impact resistant denture materials compared to cast poly (methyl methacrylate) are shown in figure 15. Im- pact resistant material II shows a glass transition temperature similar to poly (vinyl chloride). Poly (vinyl chloride) has a broad low temperature peak at about -25 °C and a minimum at about +20 °C. There is no evidence of this in this curve ; however, a blend with poly (methyl methacrylate) would mask the peak because of the poly (methyl meth- .06 acrylate) beta peak. Impact resistant denture base I is obviously .04 closely related to poly (methyl methacrylate). There is a slight indication of a low temperature peak superimposed on a poly (methyl methacrylate) curve between —50 °C and —20 °C. Also, -100 -50 HOO tliere appears to be a slig'ht low temperature TEMPERATURE °C "polymer" peak near 0 °C. The modulus versus temperature plot of the same three materials Figure 12. Internal friction versus temperature for shown in figure 16 shows a dispersion starting methyl mcthacrylate polymer prepared casting hy from about —60 °C with impact resistant material I liquid, ty molding powder-liquid mixture and hy compression molding powder only.

CAST POLYMETHYL METHACRYLATE

4.0 . 5 MMA MONOMER D POWDER-LIQUID MOULDED .5 MMA POLYMER

COMPRESSION MOULDED MMA POLYMER

3.0

o rH

2.0 0<

1.0

-L. -L. -120 -100 -80 .60 -40 -20 -0 +20 +40 +60 +80 +100 +120

TEMPERATURE °C

PiGUEE 13. Modulus versus temperature for methyl methacrylate polymer prepared by casting from liquid, iy molding powder-liquid mixture and hy compression molding powder only. To convert from psl (pounds per square Inch) to N/m' (newtons per square meter), multiply by 6895.

154 1 "T" 1 1

• CAST POLYMETHYL METHACRYLATE • CAST POLYMETHYL METHACRYLATE POWDER-LIQUID DENTURE BASE MATERIAL D O IMPACT RESISTANT DENTURE BASE I .26 ^ IMPACT RESISTANT DENTURE BASE II

.22 .22

.18

.14

.10

.06

.04 .04

.02 (- .02

-50

TEMPERATURE °r TEMPERATURE °C

Figure 14. Internal friction versus temperature for oast Figure 15. Internal friction versus temperature for cast poly {methyl methacrylate) and poivdcr-liquid denture poly (methyl methacrylate) and impact resistant den- base material. ture base materials.

T 1 1 1 1 1 1 r—

• CAST POLYMETHYL METHACRYLATE 4.0 O IMPACT RESISTANT DENTURE BASE I A IMPACT RESISTANT DENTURE BASE II

3.0

2.0

1.0

-120 -100 -60 -40 -20 +20 +40 +60 +80 +100 +120

TEMPERATimE "C

Figure 16. Modulus versus temperature for cast poly (methyl methacrylate) and impact 7'esista7it denture base materials.

To convert from psl (pounds per square Inch) to N/m^ (newtons per square meter), multiply by 6895.

452-525 0—72-^ 11 155 that would, support theory that there is some low temperature transition taking place that might relate to the toughness of the material. From the stress strain curves shown in hgure 17 of the two impact resistant materials, it can be seen that there is very little drawing during the tensile failure, hence one w^ould not expect evidence of pronounced low temperature phenomena.

IMPACT RESISTANT DENTURE

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

INCHES PER INCH

FiGUKE 17. Stress strain curves for two impact resistant denture base materials.

To convert from psl (pounds per square Inch) to N/m^ (newtons per square meter), multiply by 6895.

5. Equipment for Torsion Pendulum Measurements

Many authors have written about equipment for measuring low frequency mechanical properties [8, 11, 17, 22, 26, 31, 34-42]. In March, 1962 a basic torsion pendulum design was described, before the Dental Materials Group of the International As- sociation for Dental Research. This device utilized a light beam and a galvanometer scale for visual data recording. Figui^e 18 is a photograph of the torsion pendulum enclosed in a Lucite Avail chamber to prevent drafts. Figure 19 shows a cutaway 19. shown in sketch with the pendulum attached to the top of Figure Cutaway sketch of equipment figure 18. the machine and the sample surrounded by a chamber which can be cooled by liquid nitrogen to about -180 °C or heated to -f-300 °C. Beneath the the galvanometer scale. Figure 20 shows a photo- temperature control chamber an inertia bar is at- graph of the equipment with a zirconimn arc lamp tached to the pendulum via a pin vice. On the light soui'ce. The frequency was determined with inertia bar there is a mirror to deflect a light beam a stop watch and the readings on the galvanometer and thereby show the amplitude of vibration on scale were recorded vocally on a tape recorder.

156 have their own friction) and by damping oils. Damping oils will eliminate precession but offer virtually no resistance to the axial rotation of a smooth cylindrical rod. The challenge of recording the oscillations of a torsion pendulum offer the ingenious instrument designer ample opportunity. For most sensitivity the optical system recorder seems best. One technique is to send a light beam by a mirror attached to the inertia bar to a sheet of photographic paper on a revolving drum. Development of the paper will then produce a record of the oscillations of the pendulum. A variation of this used by Turley [22] employed ultraviolet light reflected 'to ultraviolet sensitive paper. This can eliminate the need for development of the paper. Weir [34] described in detail an optical Figure 20. Torsion pendulum equipment with zirconium device arc lamp. for recording the amplitude of oscillation using two photocells close together near the zero deflection point. The principle of his recording is based on the fact that for constant frequency in a sinusoidal oscillating system, the velocity at zero deflection is proportional to the amplitude. Thus the time it takes the light beam to travel between cells Oscillations were started by the electro-magnetic is proportional to the amplitude. The period can coils located near the ends of the inertia rod. Pre- be determined by noting the total elapsed time in cession in the oscillation was damped out by the one cycle. A digital recorder in conjunction with damping oil at the bottom of the pendulum. The a microsecond interval time meter was used to pin vices supporting the specimen were attached to record the data. thin-walled stainless tubing which had the strength Plajek et al. [37] used a constant light source and rigidity to function properly mechanically, aimed at the mirror with a shadow device that but were poor conductors and minimized thermal would cut off part of the beam in proportion to variations. the amplitude. Thus the emf that was generated In designing equipment to operate above two from the photocell was of a sinusoidal form in cycles ]per second, it is necessary to use a record- proportion to the frequency and amplitude of the ing device. oscillation of the pendulum. For measuring very low damping materials, high Gillham [41] utilized an inertia disc of glass sensitivity is needed. This can be provided with varying opacity such that a light beam pass- _ by operating in a vacuum and thereby eliminating ing through it to a photocell will be of varying air resistance. Blum [8] developed a very sensitiye intensity with the angle of rotation. instrument and recorded with fused quartz a value Klein [40] described a torsional pendulum for tan S of 0.000013 at a vacuum of 0.5 jum of utilizing an electromagnetic pickup. This was ac- mercury. This suggests that fused quartz is an complished by attaching a permanent magnet to excellent material for calibrating the internal fric- the pendulum which induced a current in a coil tion of an instrument itself. mounted adjacent to it on the fixed portion of the For further improvement in temperature distri- instrument. The induced voltage was then bution, Douglas et al. [36] recommended the use recorded. of a chromel-alumel thermocouple rather than Nielsen [42] allowed the "fixed" end of the copper-constantan because of the high conductiv- pendulum to have a small motion by restraining it ity of the latter. To eliminate residual magnetic with relatively rigid spring rods. The motion is effects from the starting electro magnets, they used proportional to the amplitude of the inertia mem- copper wire coils wound on plastic forms and a ber and is measured by a linear differential vari- piece of silicon iron transformer sheet attached to able transformer. He reported recording values the inertia bar. of tan S as low as 0.003. Nielsen [4r2], in his design to eliminate the effect The Plas-Teoh unit shown in figure 21 was used of tensile stress, placed the fixed end at the bottom for some of the data reported in this paper. The of the pendulum, with the inertia member at the instrument has a fixed bottom member with a top, supported by a fine wire and counter weight. for It is desirable to have all nonaxial rotation counterweight support system to compensate eliminated. This requires good alignment, a good length changes. The inertia member at the top is starting system, and a means of maintaining con- supported by a frictionless bearing in a rotary dif- centricity. This has been done by bearings (which ferential variable transformer which is used to

157 tend to take place. This deflection can be eliminated by applying a force perpendicular to the initial force direction. The ratio of the force normal to the desired deflection to the deflecting force itself yields tan 8. These can both be measured by strain gauges. The advantages of this system are a wide range of frequencies with a single specimen geometry and with simple recording devices. Gillham [41] described the torsion braid pendulum. Basically the technique used is a multifilament substrate with the sample disposed through the strands by melt or solution (or by other means). This allows a material to be tested beyond coiiditions wherein it can support its own weight (i.e., at high temperatures or in its re-

action periods) . The use of a multifilament system allows more material to be tested, better stress transfer, and minimizes the modulus of the substrate. 6. Future Work

One obvious project from the brief work reported in this paper is to further understand the nature of the low temperature peak in the pearl polymer used. If this is a surface phenomenon it should be proportional to particle size. A series of tests with dift'erent sieve fractions from the same polymer should verify this. If it is a bulk prop- Figure 21. Example commercially availahle torsion of erty, then a systematic review of the suspension pendulum equipment. agents, stabilizers, catalysts, etc., will have to be (Courtesy of Plas-Tech Equipment Corporation, Natlck, Mass.) made to determine just Avhat produces this effect. Of interest to dentistry is energy absorption. It supply a voltage proportional to the amplitude to has often been speculated that the ridges could a recorder. be i^rotected if the shock of mastication could be Heijboer [26] utilized a recording system absorbed. Indeed soft liners have been proposed wherein a specially conductive paper on a rotating for this. With internal friction measurements it drum was marked by an electric spark from the would be posible to develop a material of very oscillating inertia member of the pendulum, thus high damping capacity at the mouth temperature recording the motion of the pendulum. region with a broad range of frequency absorp- Kodi-iguez et al. [35] and Plajeck et al. [37] tion. Whether this is better than rubbery mate- presented designs of equipment for recording the rials that are of low damping capacity is not internal friction of weak elastic materials such as known. Clinical tests would be required. However, gels. In one case a conical plate type of viscosim- recording the internal friction of the denture eter is used to hold the sample, and an air bear- materials at a variety of frequencies in the mouth ing is used to support the weight and minimize temperature range Avould allow a characterization friction. In the other case a disc-like specimen was of the material that might profitably be coordi- held between two plates and was the elastic mem- nated with clinical data. ber of the pendulum system. The torsional braid pendulum should be of value Maxwell described a nonresonant driven [39] in characterizing all types of setting reactions. It device to allow continuous variation of tempera- would be possible to get nearly continuous data on ture and frequency at a controlled strain. The use the modulus and the damping capacity of the ma- of a controlled strain instead of a controlled stress terial versus time at a fixed temperature or at a is important in nonlinear viscoelastic materials. programmed rate of temperature change. This The instrument tests a cantilever beam of circular cross section, held in a collet and rotated at the should be applicable to dental cements, amalgam, desired speed about its cylindrical axis. A load is impression materials, resin, etc. applied at the unsupported end causing a fixed de- There is evidence that the internal friction of a flection. Each portion of the beam then will be composite material is affected by the bond between alternately under tension and then compression. the resin and the matrix. This would offer a means If the beam is made of a viscoelastic material there of quantitatively evaluating the bond and then will be a lag in the strain relative to the force, testing again after exposing the composite to and a deflection normal to the applied force will various environments, thermal cycling, etc. The

158 . :

data will give not only the internal friction aiding in the bond. The torsion pendulum is quite changes but also tlie modulus clianges. sensitive to ionic diffusion in glassy structures as High strength materials for denture base use are was shown above. of obvious interest. A significant amount of work This suggests that a method could be developed has been done correlating toughness with internal to identify the degree of diffusion and identify friction curves. Experiments might be done de- the activation energy for diffusion. Experiments veloping low temperature peaks of various magni- in metal and ceramic design could then be con- tudes and correlating the data against the tough- ducted to optimize this diffusion process. The de- ness of the material, the tensile strength of the gree of diffusion in the porcelam gold system is material, and, most important, against the creep probably small and located near the interface. In of the material. When there is significant drawing a system of high interface area, it might be pos- taking place in absorbing the energy of impact, sible to get sufficient diffusion to be measurable. there will also be distortion. There is a level of One approach would be to disperse in a ceramic distortion that is not acceptable, and some brittle- mix a fine gold alloy powder and then fire at the ness would be more suitable. appropriate temperature. The electron probe could An intriguing possibility of utilizing the torsion be used as before to determine the species of ions pendulum is as a means of following the progress taking part in the diffusion, and the internal fric- of polymerization in self-cure systems. Following tion curves at several frequencies could be used to the shrinkage is unrewarding because the specimen give a measurement of the amount of diffusion and will shrink by polymerization but also by loss of the activation energy of diffusion process. monomer either through volatility or leaching in It should be possible to use internal friction saliva. The modulus and internal friction curves measurements to help determine the causes of will be very sensitive indicators of the degree of physical property changes noted in implant mate- polymerization. It would be interesting to make rials. For instance, embrittlement through cross- self-cure and heat-cure samples and note the linking would yield a different spectrum than em- change in internal friction curves of both with brittlement through degradation. It would be time after storage in 100 °F (38 °C) water. Pre- worthwhile testing the internal friction of mate- sumably, in the case of the heat-cure material rials as a function of time of implantation. where polymerization is virtually complete, the principle effect will be a plasticizing or the system by absorbed water. In the case of the self -cure 7. References material there will be that plasticizing effect, but [1] Nielsen, L. E., Mechanical Properties of Polymers coupled with it will be a leaching of the monomer (Reinhold Publishing Corp., N.Y., 1962). and a continued polymerization of the monomer [2] Braden, M. and Stafford, G. D. Viscoelastic proper- with time. ties of some denture base materials, J. Dental Res. 47, 519 (1968). In analyzing ceramic materials there is a prob- [3] Zener, C, Elasticity and Anelasticity of Metals fre- lem of the specimen geometry. With a low (University of Chicago Press, Chicago, 1948). quency pendulum, specimens are needed of the [4] Ryder, R. J., The Internal Friction of Simple order of one square millimeter in cross section and Alkali Silicate Glasses Containing Alkaline Earth perhaps 5 to 10 cm in length. Most of the work done Oxides, Ph. D. Thesis, Pennsylvania State Uni- versity, 1959. in ceramics has been done with drawn glass fibers. [5] Fitzgerald, J. V., Anelasticity of glass : I. Intro- To fire a ceramic to these dimensions is difficult. duction, J. Am. Ceram. Soc. 34, 314 (1951). To draw a glass fiber from a ceramic melt would [6] Fitzgerald, J. V., Anelasticity of glass II. Internal not be satisfactory because the microstructure of a friction and sodium ion diffusion in tank plate glass, a typical soda-lime silica glass, J. Am. material would be significantly altered. One ap- Ceram. Soc. 34, 389 (1951). proach is to fire a specimen oversize machine and [7] Kirby, P. L., Internal friction in glass Part II it to the proi^er dimensions. Another approach Flexural and torsional vibration, J. Soc. Glass would be to use refractory investments to make a Technol. 38, 383 (1954). mold of about the right dimensions that would [8] Blum, S. L., Some physical factors affecting the internal damping of glass. J. Am. Ceram. Soc. firing support the ceramic during the operation. A 38, 205 (1955). third approach Avould be to use a two component [9] Forry, K. E., Two peaks in the internal friction as a system—e.g., porcelain on gold. By knowing the function of temperature in sodium silicate glasses, J. Am. Ceram. Soc. 90 (1957). data of the gold system by itself, the information 40, [10] Hoffman, L. C. and Weyl, W. A., A survey of the relative to the porcelain can be approximated. In- effect of composition on the internal friction of ternal friction measurements should be an excellent glass, Glass Ind. 38, 81 (1957). method of determining changes in firing and re- [11] Shelby, J. E. and Day, D. E., Mechanical relaxa- peated firings such as solution, crystal melting, tions in mixed alkali silicate glasses, I. Results, J. Am. Ceram. Soc. 52, 169 (1969) and devitrification. [12] Fitzgerald, J. V., Anelasticity of glass No. Ill, There has been a great deal of interest to date effect of heat treatment on the internal friction in the nature of the porcelain to gold bond. Evi- of tank plate glass, J. Am. Ceram. Soc. 34, 388 (1951). dence has been presented to show that there is ionic [13] Fiehter, R., Damping in metals. Schweiz. Arch. -Ann. diffusion taking place during the firing reaction Suisses 24, 65-78 (1958).

159 . . .

[14] Nowick, A. S., Anelastic phenomena in metals and [28] Galperin, I., Dynamic mechanical properties of a nonmetallics. Internal Friction, Damping, and TiOa filled cross-linked epoxy resin from 20-90° C, Cyclic Plasticity, ASTM iSpecial Technical Puto- Am. Chem. Soc. Div. of Plymer Chem. Preprints lication No. 378 (American iSociety for Testing 7, 890 (1966). and Materials, Philadelphia, Pa., 1965). [29] Reinhardt, J., Dynamic mechanical behavior of a [15] Kaelble, D. H., Micromechanisms and phenomenol- polyester during cross-linking. Plastics & Poly- ogy of damping in polymers. Internal Friction, mers, 36, 541 (1968). Damping and Cyclic Plasticity, ASTM Special [30] Malpass, V. E., Damping behavior and in-service Technical Publication No. 378 (American Society performance of impact plastics, Paper presented for Testing and Materials, Philadelphia, Pa., at Seventh International Symposium on High 1965). Speed Testing, Boston, Mass. (March 1969). [16] Boyer, R. F., The relation of transition temperature [31] Myerson, R. L., The Internal Friction of Linear to chemical structure in high polymers. Rubber and Cross-Linked Phosphate Glasses and Ohem. and Technol. 36, 1303-1421 (1963). Methacrylate Polymers, M.A. Thesis, M.I.T. May [17] McCrum, M. G., Read, B. G., and Williams, G., 1961. Anelastic and dielectric effects in polymeric solids, [32] van Wazer, J. R., Phosphorous and its Compounds, (John Wiley and Sons, Inc., N.Y., 1967). I ( Interscience Publishers, New York, 1958). [18] Nielsen, L. B., Dynamic mechanical properties of [33] Myerson, R. L., Effect of cross-linking on mechanical high polymers, SPE J. 26, 525 (1960). damping and stiffness of methacrylates, I.A.D.R. [19] Deutseh., K., Hoff, E., and Reddish, W., Relation Abstracts of the 40th General Meeting, p. 89 between the structure of polymers and their (1962). dynamic mechanical and electrical properties, [34] Weir, F. E., Use of a new torsion pendulum for

Part I : Some alpha substituted acrylic ester poly- polypropylene property evaluation, SPE Trans- mers, J. Polymer Sci. 13, 565 (1954). actions 2, 302 (1962). [20] Nielsen, L., Effect of chemical heterogeneity in [35] Rodriguez, F., Van Brederode, R. A., and Cocks, (copolymers on some physical properties, J. Am. G. G., A recording air-ibearing torsion pendulum. Chem. Soc. 75, 1433 (1953). J. Appl. Polymer 'Sci. 22, 2415-2420 (1968). [21] iCuddahy, E. F., and Moacanin, J., Dynamic Mechan- [36] Douglas, R. W., Duke, P. J., and Magurin, O. V., On ical Properties of iSome Cured Epoxy Resin. Am. the network contribution to the anelasticity of Chem. Soc, Division of Organic Coatings and glasses, Phys. and Chem. of Glasses 9, 169 (1968) Plastic Chemistry Papers, (April 1968). [37] Plajek, D. J., Vrancken, M. N., and Berge, J. W., A [22] Turley, S. G., Effect of Polymer Structure on Impact torsion pendulum for dynamic and creep measure- Properties. Am. Chem. Soc, Division of Organic ments of soft viscoelastic materials. Trans, of the Coatings and Plastic Chemistry Papers, (Sep- Soc of Rheology II, 39-51 (1958). tember 1967). [38] Copley, G. J., A simple photocell counter for the [23] Heijboer, J., Dynamic mechanical properties and determination of the logarithmic decrement of the Impact strength, J. of Polymer Sci. Part C 16, torsion pendulum, J. Sci. Instr. 43, 845-846 3755-3763 (1968). (1966). [24] Boyer, R. F., Dependence of mechanical properties [39] Maxwell, B., An apparatus for measuring the re- on molecular motion in polymers. Polymer Eng. sponse of polymeric materials to an oscillating

and «ci. 8, 161-185 ( 1968 ) strain, ASTM Bulletin 76-80 (July 1956). [25] Matsuoka, S., Aloisio, C. J., and Daane, J. H., Some [40] Klein, J., A method for recording torsion or bend- aspects of brittle failure in polymers, Applied ing oscillations in damping experiments on high Polymer Symp. 5, 103 (1967). polymers, Plaste Kautschuk 13, 20 (1966). [26] Heijboer, I. J., Modulus and damping of polymers [41] Gillham, J. K., Thermomechanical properties of in relation with their structure, Plastica 19 ii, polymer by torsional braid analysis, Appl. 489 (1966). Polymer Symp. 2, 45 (1966) [27] Maeda, M., Tanaka, K., Hibi, S. and Kakei, K., [42] Nielsen, L. E., A recording torsion pendulum for Mechanical properties of FRP, Part I, Journal of the measurement of the dynamic mechanical the textile machinery society of Japan 13, 58-67 properties of plastics and rubbers. Rev. of Sci. (1967). Instr. 22, 690 (1951).

160 ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings, of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Ultrasonic Methods for Determination of Mechanical Properties

George Dickson

Dental Research Section, Institute for Materials Research, National Bureau of Standards, Washington, D.C. 20234

Elastic characteristics, shear modulus, Young's modulus, bulk modulus and Poisson's ratio for materials can be determined by a variety of ultrasonic methods, most of which involve the measurement of the velocity of sound in the material. The methods are precise, rapid, nondestructive and applicable to small specimens. As a number of reports in the literature indicate, ultrasonic methods offer particular advantages to the study of dental materials and mineralized tissues.

Key words : Dental amalgam ; dental materials ; elastic properties ; mechanical properties

nondestructive testing ; ultrasonic techniques.

accomplishing this in cases is intro- 1 . Introduction of some by ducing sonic or ultrasonic waves into the material. The most commonly reported mechanical prop- Essentially this causes small volume elements of erty of dental materials is strength, usually com- the material to be cyclically strained as a series of pressive or tensile. However, strength is not a stress waves is propagated through the specimen. measure of the reaction of a material within its The velocity of the stress wave depends upon the functional range. To obtain some characterization mass or density of the particles or small volume of the mechanical reaction of a material to the elements of the material which are displaced and forces encountered in use, Young's modulus is upon the elastic characteristics of the material often determined. Generally, is Young's modulus which provide the forces tending to restore the calculated from a stress-strain curve obtained by particles to their equilibrium positions. loading a specimen in a teisting machine and meas- In addition to minimizing time dependent re- uring the change in length with increase in load. sponses, ultrasonic methods have other advantages. While the modulus determined in this way is usually considered a measure of the elastic character- They are rapid and nondestructive, and they per- istics of the material, the data often represent a mit repeated measurements on the same specimen

combination of properties : dlastic, retarded elastic, and viscous with varying magnitudes of time and stress dependence. This is illustrated in the stress- strain curves for dental amalgam shown in figure

It is evident that even with low stresses, the behavior of dental amalgam is not purely elastic. A plot of the strain of dental amalgam versus time when under constant stress (fig. 2) indicates that at least three types of response to mechanical force are present, an instantaneous elastic response, a retarded elastic response, and a viscous response

To determine the magnitude of the instantaneous elastic response, it is necessary to isolate this particular property of the material. Time dependent responses such as viscous creep

and retarded elasticity can be eliminated or re- STRAIN IN/IN. duced to a negligible level if the stress can be applied and removed rapidly enough. One method Figure 1. Typical stress-strain curves of various amalgams in tension, using 0.003 in/min head speed and 7 day old speoimens [1]. 1 Figures In brackets Indicate the literature references at the end of this paper. To convert psi to MN/ma multiply by 6.895 XIO-^.

161 :

1 1 ' r- 1 1 1 1 1 1 1 r I 1 1 2. Velocity and Modulus Determinations

3807 pil For plane waves in an elastic isotropic homogeneous medium [3], the velocities of the transverse and longitudinal waves can be shown to be

- < ir 6

/ 3458 P«l and / ••* 2959 p«< ~ E{l-y) • ^ 1900 pil - p(l -1-7) (1-27)

! 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 where 200 300 TIME - MINUTES p= density 6^= shear modulus Figure 2. Creep and recovery amalgam loaded in of E= Young's modulus tension [2J. 7=Poisson's ratio To convert psl to MN/m= multiply by 6.895X10-'. When the wave length is large compared to as it is aged or subjected to temperature change or specimen dimensions [4], the equations for pro- other treatments. Ultrasonic methods can be ap- pagation in an infinite elastic medium do not apply. plied to small and brittle specimens on which For example, for wave propagation parallel to the sides of a thin isotropic plate other methods of stress application and strain measurement may be difficult. Also, ultrasonic E methods are particularly suitable for measuring V, elastic coefficients in different directions in aniso- p(l-7') tropic materials. For propagation in an isotropic rod mth radius For measuring the mechanical properties of small compared to wavelength the longitudinal materials, low amplitude or low energy ultrasonic [4], wave velocity is waves are employed, and the effects of the medium on the wave are observed. High amplitude waves which would produce a permanent change in the specimen are avoided. There are two types of stress waves which are Where frequencies employed are high enough so particularly useful for determination of mechani- that the wavelength is small compared to specimen cal properties. These (fig. 3) are the compressional dimensions [3], the shear modulus, Young's or longitudinal wave in which particle motion in modulus, the bulk modulus and Poisson's ratio the medium is parallel to the direction of propaga- can be determined from the velocities of the tion of the wave, and the shear or transverse wave longitudinal and transverse waves, and the density in which the particles vibrate at right angles to the of the material as follows: direction of propagation of the wave. Shear modulus G=pVt'

s Direction of Wave Propagation Young modulus Ji,=pVt 2 \^ y 2_y J

Particle Vibration Bulk modulus K=p\ Transverse i-^aves

Poisson's ratio 7=

Direction of Wave Propagation From these equations, it is evident that determi- < > nation of elastic constants is essentially a problem

^ ^ Particle Vibration of determination of the velocity of sound in the material. Various methods can be employed for Lonpitudinal Waves determining the velocity of stress waves. One of the most widely used is the pulse-echo teclmique, which Figure 3. Two types of ultrasonic stress waves used for books determination of mechanical properties of materials. has been described in numerous papers and

162 — .

[3, 4, 5, 6]). In this method (fig. 4), a quartz favorable conditions, velocity determinations with crystal, or other transducer driven by a pulsed a sensitivity of 0.1 percent and an accuracy of 1 oscillator, is used to introduce the wave at one of percent can be obtained [5] two parallel faces of the specimen. The same trans- In another pulse method, the sing-around tech- ducer, or a similar one at the opposite face of the nique (fig. 6), the signal received when the stress specimen, is used to receive the mechanical vibra- wave has passed through the specimen is used to tion and produce an electrical signal which, after trigger the next pulse from the oscillator, and the amplification, can be observed on an oscilloscope. pulse repetition frequency is measured. Since time Unless the attenuation of the specimen material is delays other than that required for the stress wave very high, the stress Avave will echo back and forth to pass through the specimen are ordinarily com- through the specimen a number of times, and sig- paratively small, the period of the system is es- nals representing each double transit will be sentially the transit time. For accurate determina- observed. tions, these other delays must be taken into ac- If a piezoelectric quartz transducer is used, the count. With this simple system, velocity changes crystal may be cut with axes oriented so as to proof less than one part in lO'' can be detected [5]. duce either longitudinal vibrations (X cut) or Refinements of the sing-around method have transverse vibrations (Y or cut) in the speci- AC been made to provide sensitivity to velocity men. The crystal can be driven at its natural fre- changes of a few parts in 10^ [7]. These refine- quency or at an odd harmonic by a pulsed oscil- ments consist of using the signal after a specific lator tuned to the required frequency. The time number of echoes rather than after the first tran- interval from one pulse to the next is made long sit to refire enough to allow the echoes to die out between the transmitter and also using a specific pulses. The transducers can be coupled to the speci- cycle within the pulse. This method is particularly men with a thin film of oil or a viscous resin. A waveform generator (fig. 4) can be used to Quartz control the pulse repetition rate and supply a trig- Transducers ger signal to initiate both the pulse and the oscilloscope sweep. rectified echo signals, after The Spe c i men passage through a broadband amplifier, may be displayed on an oscilloscope (fig. 5) with a calibrated time scale the time between echoes and Wa V e form determined. The frequency, pulse length and pulse Generator repetition rate may vary over a wide range. Accuracy of the velocity determination will depend on the quality and quantity of echoes which Preamplifier can be observed, which, in turn, depend upon the X Wide Band attenuation of the signal by the specimen mate- Amplifier rial, the presence of spurious signals which may result from reflection from sides of the specimen with the probability of mode conversion from longitudinal to transverse waves or vice versa, and on the precision of specimen preparation. Speci- men parallelism to 2 parts in 10* or better is recommended, although less accurate parallelism may produce usable results for some purposes. Under Figure. 4. Typical pulse echo system.

Figure t5. Pulse echo pattern, a—input pulse, 6 pulses after passage through specimen.

163 . : ;

RF Pulse and Young's modulus is Specime Generator D- Amplifier E=p my.

Frequency One experimental arrangement [8] which has been Counter used for such resonance determination involves the use of a three-component system, shown in figure 7, Feedback consisting of a quartz crystal driver with the specimen of closely matched resonance frequency cemented to one end and a second quartz bar, also Figure 6. Simplified diagram of sing-around system. closely matched in frequency, cemented to the other end of the driver to act as a piezoelectric gauge. Another method of velocity measurement, analogous to optical index of refraction measurement, useful when the interest is in measurements of is based on the determination of the angle of re- small changes in velocity rather than in an abso- fraction of a stress wave as it passes from a liquid lute value. medium into a solid specimen or on the determin- Interference methods can be used for measuring ation of the focal length of an ultrasonic lens made velocities. In optical interferometers, such as the the type used in measuring the setting expansion of from specimen material [3]. amalgam and in ultrasonic interferometers in a Velocity measurements can also be made by liquid medium, the number of wave lengths be- reflection methods [3]. If ultrasonic pulses are tween two reflecting surfaces is changed by a con- reflected from the surface of a solid immersed in tinuous variation of path length. Such a system is a liquid medium, the reflection coefficient can be not feasible for ultrasonic measurement of the defined as follows properties of solids since path length cannot be readily varied. A similar effect can be obtained, Oir- however, by a continuous variation of frequency and consequent change in wave length. For exam- where ple, phase-coherent pulses sets or two gated, from Pr and Pi represent the pressure amplitudes of a continuoiis wave oscillator may be spaced in time the reflected and incident pulses, so that the second pulse which has passed once respectively through the specimen arrives at the receiving Vx and Vi represent longitudinal wave velocities transducer at the same time as the first pulse which in the incident (liquid) and reflecting has echoed and passed three times through the (solid) media; specimen [3]. If the oscillator frequency is then Pi and p2 represent densities of the liquid and adjusted, the phase relationship of the pulses will solid media. vary and when the signals are added a series of Instead of making absolute measurements of the nulls will be produced. The velocity can be deter- reflection coefficient, the amplitude of the pulse mined from the frequency difference necessary to pass from one null to the next. If the phase change on reflection from the ends of the specimen can be neglected Specimen

where I is the length of the specimen and /i and /a represent the frequencies at successive null positions. If the specimen material causes relatively high attenuation of the ultrasonic wave, a continuous wave interferometric method can be used [3] Here, the signal from the receiving transducer is compared with the signal being fed into the trans- mitting transducer and again phase change is observed as frequency is varied. Kesonance methods can also be used for velocity determinations. Such methods generally are used with frequencies in the lower ultrasonic range. At the natural frequency of a half-wavelength specimen bar, the longitudinal velocity is

Figure 7. Three component resonance system [8].

164 reflected from the test specimen may be compared The attenuation of ultrasonic waves in materials ^vith the amphtude of the pulse reflected from a can be determined by a number of methods. With specimen for which the reflection coefficient is the -i.ili: ""cho technique, attenuation is measured known. In this case by comparison of the amplitude of successive echoes. Losses which occur at each reflection also must be taken into consideration. Effects of of coupling losses may be determined by making measurements with a transducer at only one end of where Ai and A2 represent the amplitudes of the the specimen and comparing with results obtained pulses reflected from the known and unknown when an identical dummy transducer is coupled materials, respectively. From the reflection coeffi- to the opposite end of the specimen [5]. By making cient and the density of the test material, the measurements on specimens of different lengths, velocity of the longitudinal wave and the effects other than the increased path length can acoustic impedance, pV, can be calculated. This be kept constant and the attenuation calculated method can be used with materials which have a from measurements on the different specimens. high attenuation. It has the advantage also that Reflection of the wave from the sides of the speci- only one flat sm-face is required on the specimen. men may result in mode conversion from transverse The accuracy is reported to be of the order of 5 to longitudinal waves or vice versa, and result in percent [3]. energy loss and apparent attenuation. Such effects will depend on wavelength and specimen size. 3. Attenuation of Ultrasonic Waves When the attenuation is small, it can be determined from the decay curve of free vibrations in the specimen [6]. The resonance apparatus de- The attenuation of ultrasonic waves may be scribed for determination of velocity has as a used to determine the internal friction of a primary use the determination of internal friction. material. Internal friction may be defined as the It can be shown that the logarithmic decrement energy loss per cycle in an element of volume [8] of the specimen in such a system is proportional divided by 27r times the maximum energy stored to the ratio of the driver crystal voltage to the in the element per cycle or as the tangent of the gage crystal voltage angle by which the strain lags behind the stress. When energy losses per cycle are not high [6]

5- , AE AA 1. A, Q *=tan oL=zr-^= —j=- In -r=- ^ 2TrE ttA t A2 it where the constant K is a, characteristic of the where E= en ergy system. After K has been determined from the .4= amplitude free decay of the system, the logarithmic decre- 5= logarithmic decrement ment can be determined under different strain amplitudes or other specimen conditions by meas- The attenuation of ultrasonic stress waves in urements of the two voltages. materials and loss interactions are discussed in The internal friction can also be determined detail by Truell, Elbaum, and Chick in a recent from the amplitude frequency curve in the region book, Ultrasonic Methods in Solid State Physics of resonance from the relation [5]. In brief, energy losses may be divided into two categories, those that are dependent on the physical characteristics of the specimen material, and those that are characteristic of the method of measurement. The latter group includes such effects as coupling losses, losses due to non- where A/ is the total spread from one side of parallelism of specimen faces, phase effects in the resonance to the other where the amplitude has transducers, and diffraction and mode conversion dropped to -j=- times the resonance amplitude effects in the specimen. The losses that are of [6]. interest, those that are characteristic of the material, are of two main types, scattering effects 4. and absorption effects. Scattering losses are caused Applications of Ultrasonic Methods by lattice defects or other discontinuities in the to Dental Materials medium. The relative size of wavelength and defect and the defect density will determine Ultrasonic methods have been used to measure whether or not the scattering can be measured. the mechanical properties of a number of dental Absorption losses include dislocation damping materials. Dickson and Oglesby [9] reported the losses, thermoelastic losses, conduction electron elastic constants of dental amalgams in which the damping losses, phonon-phonon interactions, ferro- mercury content was varied by varying the magnetic resonance effects, paramagnetic reso- condensation pressure. Specimens 8mm in diameter nance effects, and nuclear spin energy interactions. by 6 to 15mm in length were prepared as shown in

165 table 1 . Measurements were made by a pulse-echo technic with a frequency of 5.5 MHz, a pulse 10.0- length of 0.5 to l.O^s and a pulse repetition rate of 6 KHz. Elastic constants obtained for an amalgam containing about 48 percent mercur}^ 9.8- .• are shown in table 2. As the plot in figure 8 shows, '^9.6- the relationship between Young's modulus and . . mercury content was essentially linear over the i. • • • - range investigated. LLJ 9.4 Using a resonance method with the three component system of specimen, driver crystal, and 9.2 - - gage crystal, Larson [10] investigated the Young's modulus and internal friction of amalgam.' The 9.0- frequency was in the 50 KHz range. Young's mod-

(5.9X10* S.sl I ! ! I I ! I L I ulus values of 8.5X10*' to 10.6 X10« psi 34 36 38 40 42 44 46 48 50 to 7.3 X 10* MN/m^) were found with mercury con- Hg Content % tents varying from 30 to 60 percent. A peak in Figure 8. Effect of mercury content on Young's modulus both modulus and internal friction was reported of dental amalgam [9]. in the 45 to 50 percent mercury ranges. To convert psi to MN/m" multiply by 6.895X10-'. As would be expected, the Young's modulus values obtained for amalgam by the rapid strain rates inherent in ultrasonic methods are consider- In our laboratory, the pulse-echo method has ably higher than those reported from conventional been employed to measure the elastic character- stress-strain curve procedures. However, using a istics of experimental composite quartz or glass- diffraction grating strain gage which permitted filled resins. The atteaiuation of the ultrasonic wave stress-strain curve data to be obtained in 3 to 4 in these materials is high and often no suitable seconds, Gardner, Dickson, and Kumpula [11] echoes are obtained so that the procedure is re- obtained a modulus value of 8X10® psi (5.5X10* duced to measurement of a single transit time. MN/m") Avhich does not differ greatly from those Effects of time delays other than transit time are obtained by ultrasonic methods. eliminated by measurements on specimens of various lengths. Data on Young's modulus for an experimental restorative material are given in table Table 1. Amalgam Specimen Preparation [9] 3 [12]. The elastic properties of apatites were investi- 11 to 8 Hg/Alloy ratio gated by Gilmore and Katz [13] using an ultrasonic interference technic. Powdered specimens 20 s trituration, Wig-L-Bug of hydroxyapatite, fluorapatite, chlorapatite, and sodium chloride-hydroxyapatite composites were 3 to 6 mixes of two pellets studied when subjected to pressures of up to 50 kbar. Transducers were mounted on the back faces Condensed in steel die of tungsten carbide pistons, as sho-wn in figure 9, and interference was obtained between reflections 8 mm diameter by 6 to 15 mm length from the near and far specimen-piston interfaces.

5,000 to 25,000 psi (35 to 172 MN/m")

Table 3. Elastic moduli of experimental composite resin 35 to 49 percent Hg content restorative material [12]

(In 10« psi)

Table 2. Elastic constants of amalgam [9] Powder-liquid ratio (grams powder to Modulus Method 0.4 ml liquid) Property Value Range*

1. 10 1. 35 1. 45 Hg content . _ . 48. 6% 0. 7 Young's modulus- 9. 09X10« psi 0. 06X10« Shear modulus. _ . 3. 41X10« psi 0. 03X10« E Stress-strain curve.. . 1. 2 1. 3 1. 4 Bulk modulus _ _ _ 9. 12X108 psi 0. 22X10« E Ultrasonic. 2. 5 2. 6 2. 4 Poisson's ratio 0. 334 0. 005 G Torsion pendulum 0. 7 0. 8 0. 8 G Ultrasonic. . 1. 0 1. 0 0. 9

Range of three values. To convert psi to MN/m^ multiply by 6.895X lO"'. To convert psi to MN/m^ multiply by 6.895X10-'.

166 -

nents of bone and the piezoelectric and pyroelectric X-cut quartz characteristics of bone suggested that bone should transducer behave elastically as a hexagonal single crystal, and the elastic stillness coefficients were determined on this basis. Young's and shear moduli found for axial and transverse bone directions, in dried bovine Piston femur and phalanx and fresh phalanx, are given in table 5.

Table 5. Elastic moduli of bovine bone [16] (In 10« psi) Specimen Phalanx Femur

Fresh Dried Dried

E (axial) . . - 3. 19 4. 42 3. 77 E (transverse) 1. 64 2. 31 2. 60 G (axial) _ 0. 78 1. 09 1. 18 G (transverse) 0. 65 0. 94 1. 07

Y-cut quartz convert psi to MN/m^ multiply by transducer To 6.895X10-^.

Figure 9. Diagram of transducer, piston and specimen 5. Summary arrangement for determination of moduli under high pressure [IS]. A variety of ultrasonic methods are available for determining the elastic characteristics of ma- Values obtained for some of the apatites, as well terials. These methods are precise, rapid, non- as values for bovine dentin and enamel, are given destructive, applicable to small specimens, and in table 4. Moduli of the apatites were found to they measure a specific property of the material. increase linearly with pressure in the higher As a number of recent reports in the dental litera- pressure ranges. ture indicate, they offer particular advantages to Lees [14] reported values for the specific the study of dental materials and mineralized of dentin acoustic impedance bovine enamel and tissues. obtained by the reflection coefficient method using specimens mounted in acrylic resin. A pulse of 6. References 60 ns length was reflected from the specimens in Rodriguez, M. S., and Dickson, G., Some tensile water. The signal amplitude was compared with [1] properties of amalgam. J. Dental Res. 41, 840 that of a pulse reflected from stainless steel under (July-August 1962). similar conditions. Values obtained for acoustic [2] Oglesby, P. L., Dickson, G., Rodriguez, M. S., Daven- impedance (product of density and sonic velocity) port, R. M., and Sweeney, W. T., Viscoelastic behavior of dental amalgam. J. Res. Nat. Bur. were in general agreement with values obtained Stand. (U.S.), 72C (3) 203-215 (1968). by other ultrasonic methods. [3] Filipczynski, L., Pawlowski, Z., and Wehr, J., The elastic coefficients of animal bone have been Ultrasonic Methods of Testing Materials (Butter- worths, London, 1966). investigated by Lang [15] using an ultrasonic [4] Hueter, T. F., and Bolt, B. H., Sonics (John Wiley & pulse method for measurement of velocities. Anal- Sons, Inc., New York, 1955). ysis of the crystallographic structure of compo- [5] Truell, R., Elbaum, C, and Chick, B. B., Ultrasonic Methods in Solid State Physics (Academic Press, New York, 1969). Principles Table 4. Elastic moduli of Apatites [iSl [6] Mason, W. P., Physical Acoustics— and (In 10» psi) Methods (Academic Press, New York, 1966). [7] Forgacs, R. L., Improvements in the sing-around technique for ultrasonic velocity measurements, E G K J. Acoust. Soc. Am. 32, 1697 (December 1960). [8] Marx, J., Use of the piezoelectric gauge for internal friction measurements. Rev. Sci. Inst. 22, 503 Hydroxyapatite. 15. 7 7. 12. 7 5 (July 1951). Fluorapatite 7. 5 13. 8 [9] Dickson, G., and Oglesiby, P. L., Elastic constants of Chlorapatite 6. 3 10. 0 dental amalgam, J. Dental Res. 46, 1475 (Nov.- Dentin 3. 0 1. 2 2. 6 Dec. 1967). Enamel 10. 7 4. 4 6. 7 [10] Larson, R. V., Internal Friction Damping in Dental Amalgam, Thesis, Dept. of Mechanical Engineer- To convert psi to MN/m^ multiply by 6.895 X 10-^. ing, University of Utah (August 1967).

167 [11] Gardner, T. V., Dickson, G., and Kumpula, J. W., [13] Gilmore, R. S., and Katz, J. L., Elastic properties Application of diffraction gratings to measure- of apatites. Proceedings of International Sym- ment of strain of dental materials. J. Dental Res. posium on Structural Properties of Hydroxyapa- tite Related 47, 1104 (Nov.-Dec. 1968). and Compounds, National Bureau of Standards, Sept. 1968. To be published. [12] Barton, J. A., Jr., Burns, C. L., Chandler, H. H., [14] Lees, S., Specific Acoustic Impedance of Enamel and Bowen, R. L., experimental intermediate- and An (Dentin. Private Communication (1968). restorative composite material. (To be published [15] Lang, S. B., Elastic coefficients of animal bone, Sci- in J. Dental Res.) ence 165:287 (July 18, 1969).

168 ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Stress Analysis of Dental Structures

R. G. Craig

School of Dentistry, University of Michigan, Ann Arbor, Mich. 48104

Because of the complex geometry of dental structures, most stress analysis studies in dentistry have been experimental. Brittle coatings have provided generally semiquantita- tive information on the magnitude, direction and sign of surface stresses of fixed bridges, removable partial dentures, complete dentures and the mandible under various loading conditions. Data from electrical resistance strain gages attached to the metal surfaces of pontics on a gold bridge with various load applications are shown. Applications of both two and three dimensional photoelastie techniques are illustrated. This method, which provides information on the magnitude, direction and sign of boundary and internal stresses is dependent upon how well the model represents the real structure. Other stress analysis methods involve photoelastie coatings, thermophotoelasticity, moire fringe patterns, holography and x-ray diffraction.

Key words : Brittle coatings for stress analysis ; dental materials ; denture, artificial, stress

analysis ; photoelasticity, dental applications ; porcelain-gold restorations, stress analysis stress analysis of dental structures.

1. Introduction 2. Experimental Stress Analysis Using Brittle Coatings A complete stress analysis of a dental structure would include the determination of the magnitude Brittle coatings have been used for many years of the principal stresses, their direction and sign by industry to indicate the position and direction all (tension or compression) at points. Fortu- of the maximum tensile strain and stress [2, 3]. nately, from the standpoint of time, a complete The early work involved the study of mill scale stress analysis is rarely required for evaluation of on steel, oxidation layers on bright steel, anodized design or for fracture analysis. layers on aluminum and the application of dried Theoretical stress analyses of highly simplified coatings of lime and Portland cement prepared dental structures have been carried out, but sim- from water slurries. Brittle coatings were not used plifying assumptions such as rectangular blocks extensively until the development of resin coatings for pontics and supporting restorations and ends which are prepared by spraying a part with a solu- of fixed bridges that are free, resulted in unreal- tion of limed wood rosin K, dibutyl phthalate, and istic calculations. The most sophisticated study carbon disulfide. The strain sensitivity of the to date, considered the roots of anterior teeth to lacquers is controlled by the amount of plasticizer be parabolas of revolution, the periodontal liga- and the method is able to detect strains of about ment to have designated physical properties, and 0.0007 in/in (700 /nm/m). The dried coating, if the boundary stresses were calculated as a function properly applied, contains uniform finely dispersed of Poisson's ratio [1].^ As more information be- air bubbles and the sensitivity is essentially in- comes available about the properties of human tis- dependent of the thickness between 0.004 and 0.008 sue and restorative materials, it is expected that in (100-200 /*m). When the coated structure is with the use of computer programs, analytical loaded, cracks appear in the coating at the point stress analysis of dental structures will be possible. of maximum tensile strain and proceed perpendic- At present, however, experimental stress analysis ular to the tensile stress. When temperature and of dental structures which have complex geometry humidity control is used and corrections are made is simpler, less time consuming, and more reliable for creep of the lacquer, quantitative values for the than analytical stress analysis. magnitude of strain can be obtained. These two Within the time available, I plan to discuss quantities may be the only measurements necessary experimental stress analysis techniques which have for the stress analysis, and the method has been been applied to dental problems, to point out their used successfully to predict fatigue failures. It has advantages and disadvantages, and to discuss the advantages that the tests can be made on the briefly experimental techniques that as yet have actual structure and with for practical purposes, not been described in the dental literature. an infinitely large number of strain gages. Com- pressive strain and stress can be determined by al- 1 Figures In brackets Indicate the literature references at the end of this paper. lowing the lacquer to creep under a load less than

169 that required to produce cracks. After sufficient number of gages may be necessary in a complex creep occurs, the load is removed rapidly and the structure to obtain a reasonably reliable stress crack patterns observed. Compressive strain also analysis, and thus, an overall idea of the stress can be measured by preparing the coating while distribution may be difficult to obtain. the structure is under load; on unloading, the A gold bridge with strain gages attached to the cracks that appear indicate the compressive strain. metal surfaces of the pontics is shown in figure 2 Thus, magnitude, direction, and sign can be deter- [7]. The sign and magnitude of the strain for mined; the method is limited to surface strain various loads and positions of loading are shown which may not be too serious since failures are in figure 3 for the gage on the molar pontic. As initiated at surfaces. the occlusal loading site was moved from the an- A typical example of the use of a brittle lacquer terior to posterior position, the gage recorded first coating in determining the surface strain on the metal framework of a mandibular partial denture when loaded 0}i the left posterior saddle area is shown in figure 1 [4]. The relationship of the load to the location of the cracks and the direction of the tensile strain in the saddle, minor connectors and major connectors can be seen. Of particular interest is that the maximum tensile strain occurred at the left finishing line and that the major connector is under biaxial tensile strain.

3. Experimental Stress Analysis Using Resistance Strain Gages

Electrical resistance strain gages, of wire, foil, or semiconductor type, are described in textbooks Figure 2. Foil strain gages attached to a four-unit gold [5] and all have been used in dental research [6]. alloy posterior bridge. The wire gages are rugged but their large size limits their use to appliances having large, relatively plane surfaces. The semiconductor gages are noted for their high output but their size again, presents problems. The foil gages have been the most practical for small areas on dental bridges and partial dentures. Unless rosette gages are used, some prior information is needed about the direction of the stress in the area where the gage is to be cemented. Of course, the problem of rosette gages is their rather large size. Strain gages can measure the magnitude, sign, and depending on the type of gage, the direction of the strain. A large

I 23456789 10 LOADING POSITION

position for the gage on Figure 1. Mandibular partial denture framework coated Figure 3. Strain versus loading gold alloy posterior with a brittle lacquer and loaded on the left posterior the molar pontic of the four-unit portion of the saddle. bridge.

170 tensile and then compressive strain. Experience tics) . They can be identified using white light since has shown replacement of the gage in as nearly as the isoclinics are black and move as the polarizer possible the same position results in strain-position and analyzer are moved. Also, in circularly curves of the same general shape, but not identical polarized light (quarter- wave plates used) only magnitude; this condition results from being un- the isochromatics are observed. The magnitude of able to cement the second gage in exactly the same the difference in the principal stresses can be meas- location and orientation as the first. Of course, care ured by detennining the order of the fringe and must be taken in cementation and the gage must knowing the fringe constant and thiclmess of the be protected if it is to be used in an adverse plastic. The sign may be determined by inspection environment. in white light by noting the order of the color of The strain gage method permits the measure- the fringes or by using various compensators such ment of surface strains of low magnitudes. (Spe- as the Cooker or Babinet. The boundary stress may cial gages are available for fatigue measure- be calculated directly, since one or the other of the ments.) As will be shown later, they are readily principal stresses must be zero. The separation of used in impact as well as static loading. the principal stresses in the interior of the model may be accomplished by additional measurements 4. Experimental Stress Analysis by of the lateral strain by determining the isochroma- Photoelastic Techniques tics using both normal and oblique incident light, or by the shear difference method. Numerous researchers in dentistry have used A dental application of the two-dimensional photoelastic methods to study stresses in dental method is illustrated in figure 5 which shows a structures [6], however, only the more recent mesial-distal section of a three-unit bridge [11]. papers will be referred to in the following The plastics were selected to have as nearly as discussion. possible the same ratio of the elastic modulii as The photoelastic method for stress analysis can tooth structure and gold. Only the isochromatics be divided into two-dimensional, three-dimen- are shown and the fringes in the areas of the sional, reflection, and scattered light techniques soldered joint show high tensile stress concentra- [8-10]. A schematic sketch of a transmission tions and high compressive stresses in the contact polariscope is shown in figure 4. In general, the area. Only minimal stress is in the gingival por- method involves the use of a plastic model or coat- tion of the pontic, although maximum tensile stress ing which is birefringent under stress. When plane would be predicted from a simple beam in trans- polarized light is used to examine the plastic, two verse bending. The boundary stresses can be ob- sets of interference lines (fringes) are observed, tained simply by noting the fringe order, the sign, one is the loci of constant stress direction (iso- and multiplying the fringe order by the fringe con- clinics) and the second is the loci of constant differ- stant and dividing by the thickness of the model. ence between the principal stresses (isochroma- The separation of the principal stresses has been

plane of polarization

plane- polarized light

circularly polarized light principal stress directions

plone of polarization

Figure 4. Schematic of a transmission polariscope.

452-525 O—72 12 171 FiGUBE 5. Isochromatic fringes in a tioo dimensional model of a mesial-distal section of a three-unit bridge. carried out using the shear difference method of graphical integration and the results are shown in figure 6. The advantages of this method are that the magnitude of boimdary and internal stresses can be measured as well as their direction and sign. The disadvantages are related to the model and how closely or remotely it represents the real structure. An additional example of two-dimensional plio- toelasticity is shown in figure 7 [12]. The model represents a porcelain fused to gold restoration cemented to dentin; it was constructed of three plastics having modulii in approximately the same Figure 7. Tico-dhnensional photoelastic model repre- ratios as the restorative materials. Of particular senting a labial-lingual section of porcelain fused to gold restorations; the plastics representing the gold and porcelain were cemented, with epoxy cement which 0 in turn was luted to the plastic representing dentin with 900- dental stone.

interest is the low state of stress in the dentin section of the model. The three-dimensional technique takes into ac- covmt the contribution of the third dimension of an object to the stress distribution. In this technique a three-dimensional model is prepared by machin- ing from plastic or by casting mto a mold. The model is subjected to a load and heated slowly under load and finally cooled to room temperature thus freezing the stresses in the model. The model may be sliced without release of the frozen stresses as seen in figure 8. Using surface slices and two- dimensional slices and sub-slices, the principal stresses may be determined vising the previously mentioned methods. The three-dimensional method may be completely reliable for homogeneous solids, but there may be serious discrepancies in the model for heter- ogenous solids or composites. In the study of dentures by Klotzer [13] the problem does not exist since the properties of denture and photoelastic plastics are similar. Papirno, Colin, and Kauf- Figure 6. Separation of principal stresses along a line midivay between A and C for the model in figure 5. man [14] have made progress toward the solution

172 «

Figure 9. Isochromatic fringes in a photoelastic coating on a gold alloy bridge.

5. Additional Techniques in Stress Analysis

brittle coatings, strain Figure 8. A slice from a three-dimensional model which The three techniques, has been subjected to a stress freezing cycle; isochro- gages, and photoelasticity can be used under condi- matic fringes are shown. tions of impact, as well as static loading. The strain gage recording system must have a high frequency response in order to measure the maximum transi- of the problem by the use of plastics of different ent strains ; a storage type oscilloscope is particu- modulii for the restoration and tooth structure. larly useful for these measurements. An example The use of photoelastic coatings on the actual of the strain on the mesial area of the molar pontic structure is a more recent method for studying sur- of a four-unit gold alloy bridge as a result of an face stresses. The coating is prepared by p6uring a impact on the lingual cusp is shown in figure 10 catalyzed resin into a thin siheet and when it has [15]. The maximum strain did not occur until the jelled it is adapted to the part to be studied. After third oscillation and the initial strain was com- the polymerization is complete, the coating is re- pressive followed by tensile strain. The maximum moved and then cemented to the structure with a strain was 192 microinches/inch ( 192 /xm/m ) and reflective plastic cement. The structure is loaded the maximum stress was 2690 lbs/in^ (18.6 and illuminated with polarized light. In this method the polarized light passes through the coating, is reflected off the cement surface and observed DROPZ. 30« with an analyzer. Fringes are observed and interpreted in a manner similar to transmission photo-

I 2 3 elasticity. A gold bridge coated with photoelastic S.G. plastic and loaded occlusally is shown in figure 9. Gold The first-order fringe is easily visible as the bright- est fringe. The highest stress concentration was observed in the area of the soldered joint vsdth the highest orders being nearest the center. The pontic and even the area near the site of loading was under low stress. If the photoelastic coating is thin and of low modulus compared to the underlying structure the reinforcing effect of the coating can be ignored and the stress in the coating will be es- • •: ;. V ^ u sentially that in the surface of the structure. If these factors cannot be avoided, corrections must be made. The magnitude, sign, and direction of the stresses can be determined as usual, and separation of the principal stresses is accomplibhed the by S.G. 2 oblique incidence method. The method has the distinct advantage that the actual structure can be Figure 10. Strain as a function of time on the mesial area of the molar -pontic resulting from used and questions about the reliability of the an impact blow on the Ungual cusp of the model are avoided. first molar pontic; total time was 4-5 ms.

173 MN/m^), and the strain decayed with time. Simi- 6. Summary lar measurements showed that torsional, as well as transverse modes of vibration occurred. In summary, a number of recently developed The simplest dynamic technique for photoelastic stress analysis techniques, as well as standard stress analysis involves a spark light source and a methods offer real promise in the solution of the delay circuit connected to the loading device. The numerous related problems in dentistry. load is applied and a photoelastic stress pattern is photographed at predetermined delays after 7. References loading. Thermophotoelasticity also has been used to [1] Haack, D. C, An Analysis of Stresses In a Model study stresses resulting in plastic models from ther- of the Periodontal Ligament, Dissertation (Kansas State Univ., 1968). mal shock [16]. Various arrangements for the ap- [2] Hetenyl, M., Brittle Models and Brittle Coatings, plication of heat have been used including liquids 636-62, In Hetenyl, M. ed., Handbook of Experi- and metal resistance strips. This method could be mental Stress Analysis (John Wiley & Sons, New used to examine the stresses in dental restorations York, 1950). [3] Dally, J. W., and Riley, W. F., Experimental Stress resulting from thermal shock. Analysis, 91-139 (McGraw-Hill, New York, 1965). Moire fringe patterns have been used to study [4] Craig, R. G., and Peyton, F. A., Strain on the surface deformation. This is simply the compari- framework of a mandibular free-end partial son between a deformed grid and an undeformed denture under load, J. Blomed. Mater. Res. 1:263-74 (Jan. 1967). master grid used as a length standard [17]. The [5] Dally, J. W., and Riley, W. F., Experimental Stress method is applicable to high temperatures, large Analysis 366-514 (McGraw-Hill, New York, elastic and plastic strains, long term strain 1965). properties of materials, and of two and three [6] Peyton, F. A., Asgar, K., Charbeneau, G. T., Craig, R. G., and Myers, G. E., dimensional analysis of transparent models using Restorative Dental Mate- rials 138-9 (Mosiby, St. Louis, 1968). embedded grids. The method is best suited to flat [7] Tlllltson, E. W., Craig, R. G., and Peyton, F. A., surfaces and the preparation of a grid on curved Experimental Stress Analysis of Gold and surfaces such as dental restorations has limited its Chromium Alloy Bridges, I.A.D.R. Dental Mate- application in dentistry. rials Group microfilm, Washington, D.C., March 1967. Holography has been used in stress analysis to [8] Dolan, T. J., and Murray, W. M., Photoelasticity I. determine isochromatic, isoclinic and isopachic Fundamentals and Two-Dlmenslonal Applications (loci of constant sum of the principal stresses) 828-923, In Hetenyi, M. ed., Handbook of Experi- fringe patterns in two dimensions [18, 19]. Com- mental Stress Analysis (John Wiley & Sons, New bination isochromatic and isopachic fringe pat- York, 1950). Drucker, D. C, Photoelasticity II. Three-Dlmen- terns have been obtained which give directly a [9] sional Photoelasticity 924-76, In Hetenyi, M. ed.. solution for the individual principal-stress com- Handbook of Experimental Stress Analysis (John ponents. A double exposure method for measuring Wiley & Sons, New York, 1950). surface stresses on dental restorations offers prom- [10] Dally, J. W., and Riley, W. F., Experimental Stress Analysis, 143-333 (^McG raw-Hill, York, ise as a means of avoiding the use of photoelastic New 1965). coatings. [11] El-Ebrashi, K. M., Craig, R. G., and Peyton, F. A., Scattered light photoelasticity has been devel- Photomechanics and Stress Concentrations in oped to the point where instruments are available Fixed Dental Restorations, I.A.D.R. Dental Mate- commercially [20, 21]. Coherent light from a laser rials Group microfilm, San Francisco, March. 1968. is focused into a thin sheet or beam of light and it [12] Craig, R. G., El-Ebrashi, M. K., and Peyton, F. A., is plastic in immersion passed through a model an Stress Distribution in Porcelain Fused to Gold tank under load and the scattered-light fringes Alloy Crowns and Preparations Using Photo- are observed. The patterns can be interpreted to elasticity, I.A.D.R. Dental Materials Group give solutions to three-dimensional problems. This microfilm, San Francisco, March, 1968. technique is comparable to locating a polarizer or [13] Klotzer, W., tiber polarisationsoptesche Unter- isuchen an Prosthesenmodellkorpern, Deut. Zahn. analyzer in the interior of the model and stress Zelt. 21:894-901 (1966). information can be obtained without stress-freez- [14] Paprino, R., Colin, L., and Kaufman, E. G., Proper- ing or slicing the model. ties and Preliminary Three Dimensional Photo- The distance between atoms may be used as gage elastic Results Using Room Temperature Curing lengths in the x-ray diffraction of crystalline sol- Resins, I.A.D.R. Dental Materials Group microfilm, San Francisco (March 1968). ids, and strains may be determined by measuring [15] Tillitson, E. W., Craig, R. G., and Peyton, F. A., changes in these interatomic distances. The method Experimental Stress Analysis of Dental Bridges can be used to nifiasure residual surface stresses Using a Dynamic Method, lADR Dental Materials Group microfilm, San Francisco, March, 1968. without drilling holes to relieve the stress. The [16] Leven, M. M., and Johnson, R. L., Thermal stresses method can give the sum of the principal stresses on the surface of tube-sheet plates of 10 and 33% in the surface layer and in fine-grained alloys, the percent ligament eflBciency, Exper. Mech., 4, 356 accuracy is about 2,000 lbs/in- (14 MN/m^) and (1964). [17] Zandman, F., The transfer-grid method, a practical in coarse-grained alloys it has a much lower moire stress-analysis tool, Exper. Mech., 7, 19 A accuracy. (1967).

174 [18] Fourney, M. E., Application of holograpliy to photo- Symposium on Photoelasticity (Pergamon Press, elasticity, Exper. Mech. 8, 33-8 (1969). New York, 277-93, 1961). [19] Hovanesian, J. D., Brcic, V., and Powell, R. L., A [21] Jenkins, D. R., Analysis of behavior near a cylindri-

new experimental stress-optic method : Stress- cal glass inclusion by scattered-light photo- holo-interferometry, Exper. Mech., 8, 362-8 elasticity, Exper. Mech. 8, 467-73 (1968). (1968). [22] Barrett, C. S., X-ray Analysis, In Hetenyi, M., ed., [20] Srinath, L. S., and Frocht, M. M., The potentialities Handbook of Experimental Stress Analysis (John of the method of scattered light, International Wiley & Sons, Inc., New York, 1950).

175

: :

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Relations Between Mechnical Properties and Clinical Behavior

David B. Mahler

University of Oregon Dental School, Portland, Oreg. 97201

A dilemma in the field of dental materials is that in many instances the critical physical properties that are relevant to the failure of restorative materials have not been identified or when identified the limiting values which separate successful from unsuccessful materials have not been established. To determine the critical property relevant to clinical marginal fracture of dental amalgam restorations, dynamic creep and other physical properties including compressive, tensile and transverse strength and transverse deflection of nine amalgam alloys were determined. Clinical evaluation of restorations made using alloys widely separated in dynamic creep showed that marginal fracture was associated with the Theological properties of dynamic creep, static creep and slow compressive strength, but showed little relationship to the other physical properties measured. Although the relationship between creep and marginal fracture may not be a cause and effect mechanism, creep appears to be reasonably predictive of clinical marginal fracture.

Key words : Amalgam, dental ; correlation of laboratory and clinical evaluations creep of ;

dental. amalgam ; dental materials; dynamic creep; fracture, marginal of dental amalgam;

mechanical properties of dental materials ; rheological properties of dental amalgam.

If an engineer were to be taken into a clinical values which can separate the successful from the operatory, a tooth which had been restored with a unsuccessful material. certain material were pointed out to him, and he Once the relevant property is identified, the were asked to comment on the efficacy of the resto- procedure for improving the material is fairly ration, what questions would he be likely to ask ? straightforward, although there may be many First he might ask the general question whether difficult and unsolvable problems along the way. the material was performing satisfactorily. Specif- The identification of this relevant property can ically he might ask the following questions about function as follows the material 1. It would allow us to identify the actual 1. Is it maintaining its shape, color and mechanism of failure. appearance? 2. It would give us a means of designing this 2. Is it biologically compatible ? weakness out of the material. 3. Is it truly restoring the tooth to its original 3. It would allow us to utilize laboratory meth- function ? ods of screening material modifications for 4. Is it doing these things over a long time possible improvement. period ? 4. It would allow us to conduct clinical trials If the answer was yes to all of these questions, he on only those modifications which show might lose interest in the conversation and leave. the greatest promise. If the answer was no, he would probably become I would like to show you an example of how we, interested and ask how is the material performing at the Dental Materials Science Department of the unsatisfactorily. Specifically what are the modes University of Oregon, have attempted to follow of failure. Let us assume that we can supply our this approach of identifying a relevant physical the engineer with an answer to this question. If so, he property. The example is that concerned with which is per- would probably start to think in terms of physical marginal fracture of dental amalgam haps the most significant failure mode for this properties and try to identify in his mind which material. property or properties might relate to the mode Initially, a laboratory test was devised which of failure outlined. was thought to parallel the situation confronted The dilemma in the field of dental materials by an amalgam restoration in the mouth. Cyclic today is that, by and large, have been we unable to compressive loading, which was considered to re- identify these critical physical properties that are late to the way in which masticatory forces are relevant to the failure of restorative materials. In applied to amalgam restorations under clinical figure 1, this dilemma is outlined. Furthermore, conditions, was applied to an amalgam test speci- where we have identified a pertinent physical men. In addition, water at 37 °C was circulated property, we have not been able to assign limiting around the specimen during the test. The deforma-

177 MECHANICAL PROPERTIES OF RESTORATIVE MATERIALS

Amalgam Modulus of Elasticity Silicate Elastic Deformation Zinc Phosphate Elastic Limit Zinc Oxide Eugenol Modulus of Resilience Synthetic Resin \ / Ultimate Strength CLINICAL Porcelain —*- 1 Plastic Defonnation PERFORMANCE Pure Gold Fatigue Strength Gold Alloys Impact Strength

Co - Cr Alloys Creep

Stainless Steel Alloys Hardness

Base Metal Alloys Wear

Figure 1. Denial restorative materials; clinical performance and mechanical properties. tion of the specimen was recorded during the 10 testing procedure using a linear transducer wired to a recorder. The parameter measured was the dynamic creep of the specimen over a fixed time interval of testing. A scnematic of the test system is shown in figure 2. The results of testing 9 commercially available amalgam alloys are shown in figure 3. The first most obvious result was that amalgam alloys having relatively small differences in conventional physical properties showed a very marked difference m dynamic creep properties. The next step was to determine the clinical sig- n nificance of these differences. In view of the many 0 E F G H I difficulties attendant with clinical testing, it is ALLOY important to maintain as simple an experimental Figure 3. Dynamic creep of nine commercial dental design as possible. Therefore, it was considered amalgam alloys. appropriate to evaluate the clinical performance of the two alloys at opposite ends of dynamic creep behavior, Alloys A and I. If no difference in behavior and there would be little point in testing clinical performance could be observed, then this alloys in between. Alloy B was also selected for property would have little relevance to clinical evaluation in case this property did prove of significance since the dynamic creep value for Alloy B was close to but significantly different from Alloy A. The procedure consisted of placing these three alloys under conditions which produced amalgam restorations having physical characteristics close DIFFERENTIAL TRANSFORMER to those of the laboratory test specimens. This was accomplished by determining the Hg content of representative restorations of the three alloys DYNAMIC CREEP TEST STATIC CREEP TE ST dentists cooperating on the project. Lab- FLUCTUATING STRESS CONSTANT STRESS placed by 500-10,000 psi 5250 psi MEAN STRESS = 5250psi were prepared at these same 1600 CYCLES/MINUTE oratory test specimens Hg contents. After one year of service, occlusal 37 C. WATER AMALGAM SAMPLE photographs were taken of these restorations. FATIGUE OR CREEP MACHINE Marginal fracture was noted in many of the restorations and the entire group of photographs Figure 2. Schematic of systems for dynamic and static creep testing. was separated into five groups of marginal frp''-

178 —

ture as shown in figure 4. The data were subjected In order to test the uniqueness of this relation- to a Ridit analysis which is shown in table 1. The ship of dynamic creep to marginal fracture, other table entries are the numbers of restorations as- mechanical property tests were conducted as well. signed to each category by three evaluators. Alloy These included compressive strength, tensile N, which corresponds to Alloy B, having a dy- strength, transverse strength, transverse deflection, namic creep value slightly higher than Alloy A in flow as determined in accordance with American figure 3, was taken as the standard distribution Dental Association specification No. 1, and static with a mean value of 0.50. Alloy D, which corre- creep. The results of these tests together with the sponds to Alloy A, having the lowest creep value, marginal fracture results are shown in figure 5. (fig. 3), had a mean value of 0.39 which was Arrows have been drawn in the direction of supe- clearly differentiable from Alloy N with a signifi- rior characteristics and the "t" values for com- cant value of 3.2. Alloy M having the highest parisons are shown with the arrows. From figure 5, creep value of all alloys tested and which corre- it is seen that only the rheological properties of sponds to Alloy I in figure 3, had a mean value of dynamic creep, static creep, and slow compressive 0.81 with a significant "z^" value of 10.2. strength relate in the proper direction to marginal In simple terms, the results of the clinical evalu- fracture. ation showed that Alloy D, having the lowest Although the relationship between creep and dynamic creep value, had the least amount of marginal fracture should not at this time be con- marginal fracture; Alloy N, having a slightly sidered a cause and effect mechanism, this prop- higher dynamic creep value, had a slightly higher erty appears reasonably predictive of the clinical incidence of marginal fracture ; and Alloy M, hav- phenomenon of marginal fracture. Additional ing the highest dynamic creep value, had an ex- work is being conducted to strengthen this tremely high incidence of marginal fracture. hypothesis.

Table 1. Ridit analysis of marginal fracture

II III IV Total

Alloy N. Evaluator A 8 35 4 1 56 Evaluator B 15 15 16 7 3 56 Evaluator C 10 20 21 4 1 56

Alloy D. Evaluator A 18 33 6 1 0 58 Evaluator B 22 19 14 3 0 58 Evaluator C 19 26 11 2 0 58

Alloy M. Evaluator A 1 5 18 22 6 52 Evaluator B 0 1 11 13 27 52 Evaluator C 2 5 19 18 52

X Sx t Alloy D compared to Alloy N 0.39 0.032 3.2 Alloy M compared to Alloy N 0.81 0.026 10.2

179 Alloy D Alloy H Alloy M

Harginal Fracture .39 .50 .81

^'^ Compressive Strength (psi) 61U00 > 59500 51900

^'^ Tensile Strength (psi) 69i»0 » 7930 < » 7770

^'^ ^''^ Transverse Strength (psi) 13220 » I8030 ^ 12U50

-'-'^ ^'^ Transverse Deflection (n) 20 •* » 23 * 32

^'^ -,J2iil ADA Flow («) 0.50 - » 0.65 3.91

Dynamic Creep {%) 0.86 ^.^j j^ g^ — 8.76

Static Creep (?) 0.76 ^ 2.36 ^ 8.37

^'^ Slow Compressive Strength (psi) 38500 * 3U6O0 < 231iOO

FiGur.E 5. Comparison of alloys with respect to marginal fracture and mechanical properties. (pslX0.0068.9 = MN/mS)

180 Development of Improved Methods for Evaluating Dental Materials

;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50tli Anniversary Symposium, Held Oct. &-8, 1969, Gaithersburg, Md. (Issued June 1972).

Need for Research To Develop Performance Characteristics

H. P. L. Schoenmakers

Tandheelkundig Instituut, Rijksuniversiteit, Utrecht, The Netherlands

To develop the most important characteristics of restorative materials, an understanding of the processes by which margins of such restorations become damaged is needed. Bulk and margin properties should be correlated. The mechanical forces developed in mastication and chemical attack by the oral environment are important factors. Tensile strength appears a more meaningful data than either shear or compressive strength. Research for selection of proper tensile test procedures is needed. Cracks and voids in specimens complicate their strength behavior. The role of temperature in laboratory evaluation is important. The mechanisms of corrosion and erosion of dental restorations requires further research.

Key words : Clinically related strength properties of dental materials ; dental materials

dental materials, correlation of bulk and marginal properties ; dental materials, performance

characteristics ; dental restorations, the role of oral environment factors in stability thereof.

fluid part of the runs faster to the margin 1 . Introduction cement than the more viscous material, which may result The purpose of a dental restoration is to rein a weaker material at the margin. It seems pos- establish the patient's ability for chewing and sible that silicate cement may show such speaking while also attention must be given to the phenomena also. esthetic requirements. The quality of the restora- The discrepancy between the margin and the tion must be on such a level that a long lifetime is other part of the filling is correlated to the clinical guaranteed. When limited only to the materials manipulation of the material. It is well known which are used to restore teeth, it may be said that that in order to get a better amalgam filling, the the lifetime of the filling is restricted by the ap- cavity must be filled to excess and then carved or pearance of recurrent caries in the adjacent tooth reduced to the right level. Tlie degree of over- structure. In many cases a correlation can be de- filling depends on several factors such as cavity tected between the recurrent caries and the adapta- design, the plasticity of the material, and the skill tion of the filling. The adaptation the filling of of the dentist. Therefore, it may be concluded that can be decreased by damage to the margin of the in clinical cases there will not be an exact cor- filling. Such a defect however is seldom discovered relation between the properties of the margin and by the patient himself. In most cases the dentist the body-part of the filling. As the properties of will trace the defect with the help of a radio- the filling materials are determined a rather graphic photograph, which shows the recurrent on caries. large specimen, which most likely shows the char- When determining the quality of a filling, at- acteristics of body-material, it is questionable if tention should be directed to the junction between the quality of the filling, particularly of the mar- filling and tooth structure. The adaptation must gin, might be predicted in this way. be so good that local caries will not occur and the Research is needed to determine the correlation quality of the margin must be on such a level between body and margin properties. If the dif- that chewing and biting forces, as well as the ferent brands of a dental material do not show the chemical attack by the oral environment, can be same correlation between the body properties and resisted. margin properties, it would be useful to test the material on a powder-liquid ratio which correlates 2. DesirabiHty of Estabhshing a Correla- with the composition of the margin, tion Between Bulk and Margin Prop- erties 3. Relation of Properties and Forces Applied Orally Many authorities have shown that the margins of amalgam fillings are inferior to the body-part. Wlien it must be decided if a material can be Also zinc phosphate cement may show such dis- used successfully for a restoration, consideration crepancies. When cementing a crown or inlay, the must be given to the chemical and physical forces

183 to which the material will be subjected. In gen- the teeth, the movement of the jaw will be stopped eral, a distinction can be made between mechani- by the reflex from the proprioceptive receptors cal forces, i.e., chewing and biting forces, which which are settled on the end of the fibers which are applied to the occlusal part of a filling, _and a support the teeth. The reflex goes through the chemical attack on the whole filling. Considera- brain and can be suppressed. When a high force tion must also be given to the fact that as long is needed to mill the food the proprioceptive re- as the filling is not hardened, the adjacent soft ceptors will react at a different level and the force tissue and the pulp may be irritated by the toxic transmitted by the unexpected hard particles nature of some ingredients. mig*ht be rather high. The materials which are used for occlusal fill- On the other hand, the movement of the jaw ings require such strength that fracture or defor- itself is realized by a very small force. In fact the mation will not occur. It is very difficult to estab- force is increased as soon as the contact between lish the exact requirements because there is a large the teeth is realized. This means that when biting diversity in tooth shape and in the forces which on a large particle, the chewing will be interrupted can be developed. Only a qualitative approach before a high force is developed. If it is accepted seems possible. In relation to the chewing muscles that hard substances in the food are a main reason average forces can be developed in the magnitude for damage to the margin of fillings, it is clearly of 100 to 200 kgf (980 to 1,960N). Also forces of desirable to develop information pertaining to the 800 kgf (7,800N) are known, but these are excep- magnitude of the developed forces. Especially tional. when the hard substances are small, a large stress When a force of about 200 kgf (1,960N) is sup- may develop. By such a phenomenon, fracture will ported by the multi-cusp contact area between the not occur very often but each incidence may result teeth and their antagonists which is approxi- in a carious attack. mately 2 cm% a stress of 100 kgf/cm== (9.8MN/m==) In many cases the filling and the margins do not results. Such a stress is far below the compressive Sossess a perfect shape. A perfect adaptation sel- strength of all filling materials. Amalgam may om occurs and many times at the margin the fill- show a flow-deformation if the stress remains for ing is not coincident with adjacent tooth structure. situation is a long time. When the force is carried by only one Such a much more dangerous for the filling tooth, a higher stress is possible. Then, the magni- occurrence of a fracture, because the is no tude of the force depends on the distance' to the longer fully supported. It is questionable if this temporomandibular joint. The maximum force is a matter of any importance to the specialist in might be about 35 kgf (340N). However, it is dental materials. When the filling extrudes because zinc- questionable if suc!h a force is, in fact, accepted the dentist incorporated moisture in the by the tooth. Normally the force is restricted by a containing amalgam, the inferior result is the re- pain reflex. sponsibility of the dentist. A more practical approach to assessing the mag- Whenever there are phenomena which cannot be nitude of the forces, developed on chewing, has controlled by the dentist, they deserve much atten- been to utilize electronic measuring devices, tion. It is very desirable to be informed in a quan- mounted in the elements of a partial prosthesis. titative way about the influence of the dimensions These measurements indicate that chewing forces of the gap between filling and tooth structure, on on first molars approximate 6 to 7 kgf (60 to 70 the loss in strength of the margin, and on the N). It seems reasonable to assume that with nat- chance for recurrent caries. Such knowledge will ural teeth higher forces might be developed. assist in the determination of the requirement for When the chewing force is doubled, e.g. 14 kgf, dental materials. (140N) and the contact area between the tooth Fillings are mainly subjected to compressive and the antagonist with food in between is estab- forces. However, even in the case of a perfect lished on 0.14 cm^, then a stress is applied of 100 adaptation, in addition to the compressive stresses kgf/cm== (9.8MN/m^). developed in the filling, tensile and shear stresses As the stresses ordinarily developed are far will also be realized. The nature of the stress caus- below the compressive strength of the filling mate- ing the fracture is difficult to tell because the ratio rials, it is most likely that fracture will occur only between the magnitude of the different stresses is in exceptional circumstances. When looking for an not known, but experience shows that the resist- explanation of a fracture of a fully supported ance against compression is much higher than the margin, it is reasonable to find it in a stress con- resistance to tensile forces. When the margin of the centration which is caused by hard substances that filling is not perfect, it is likely that greater tensile may exist in the food. For example, vegetables and shear stresses are developed. Therefore, it is may contain some sandgrit. The stress realized by difficult to decide which mechanical property is biting upon this grit is hard to calculate, because the criterion that should be used to evaluate the this depends on the size and shape of the grit and filling material. Research is needed to solve this the magnitude of the developed force. During problem. In relation to this, it might be useful to chewing, when the hard substance comes between study the phenomena which initiates the fractures.

184 4. Theory of Fracture in Compression ricating a proper specimen, the cleavage test and ' the flexure test have been used. The resulting and Tensile Testing [1-4] strength values, however, are not on the same level, therefore, research is needed to select the proper The compression test has been used most test. Also, the existence of porosities should be frequently. It is assumed in this test that fracture taken into consideration. The appearance of the is realized by shear stresses. It is believed an ideal fracture surface indication fracture shows two cones, respectively containing might give of the source of rupturing. the bottom and the top surface of the specimen. This shape of fracture is explained by shear stresses which are maximal in the planes at an 5. Role of Temperature in Property inclination of 45° with the direction of loading. Evaluation This theory, especially when applied to fracture of brittle materials, is no longer accepted by the In most cases strength measurements are made specialists in fracture phenomena. New theories at room temperature, although the materials are have been introduced since repeated failures have subjected to forces at mouth temperature. This been reported of Polaris rocket motor casings at may lead to wrong results for materials with a low stresses well below design value. An early tensile fusion temperature, such as the amalgams. It is fracture is based upon the existence of cracks, known that amalgams which go from room tem- which cause stress concentrations, according to perature to mouth temperature show a reduction the theories of A. A. Griffith [3], published in in compressive strength. The literature gives num- 1920. The materials that are used in the Polaris bers of 8 percent and 15 percent for the average rocket have a high strength but do not possess reduction. Private observations showed a reduc- much ductility. It seemed that a new property tion of about 30 percent. This discrepancy may re- must be introduced: namely the stress intensity sult from variations in the time which the amal- factor with the formula K=G^fwa, wherein "c" is gam is allowed to remain at mouth temperature one half of the length of the crack. and by the difference in the loading speed. The Because most of the dental filling materials figures regarding the reduction in strength show are brittle by nature and contain cracks and that the different brands did not behave in the porosities, it might be useful to consider these same way. Thus by this effect it might be con- newer theories. Comparing the appearance of cluded that the testing must be performed at least the fracture planes respectively caused by tensile at mouth temperature. forces and compression forces, it is suggested that In addition, it must be taken into consideration brittle fracture can only be realized by tensile that higher temperatures are developed while hot forces. Further investigations showed that most food is swallowed, resulting in a temporary de- 'ractures under compressive forces run in an axial crease in amalgam strength. While the oral cavity lirection. This was also the case with a single temperature decreases rather quickly, it is ques- crystal of sodium chloride which was oriented tionable that the phases of the amalgam come to in such a direction that oblique fracture planes equilibrium as rapidly. Where such phenomena with a certain inclination were expected. Although may cause a weakening effect during a longer a scratch on the surface initiated a crack in an time, the chance is increased for damaging of the oblique direction, the fracture propagated in the margins of the filling. In this paper fatigue direction of loading. Thus it may be stated that phenomena are not discussed. However, it may be the fracture is realized by tensile stresses and stated that these are also very important, because initiated by cracks which cause stress concentra- they may introduce the initial defect in the filling. tions. However, the influence of the crack is case. In tensile not the same in every the direct 6. Oral Environmental Effects and Labora- test, the stress concentration depends only on the tory Testing of Dental Materials longest axis of the assumed elliptical crack, which increases during fracturing, while in the com- The lifetimes of fillings which are not subjected pressive test the stress concentration depends on to chewing or biting forces will be related to the the ratio of both axes of the ellipse and on the chemical attack by the oral environments and to inclination. In the direction of fracturing under erosion created by saliva, food, and the soft tissues compression, however, only the crack itself is which slide over the fillings. The chemical attack important; and during fracturing the stress is not itself can be created by the saliva, the plaque, and increased nor decreased. eventually by the food. Because the oral environ- For dental purpose a test must be selected which ment varies from person to person, it is hard to indicates the strength of the margins. It seems very predict the clinical behavior of a filling. A proper probable that a tensile test will offer the best cor- test for measuring the solubility and disintegra- relation. Besides the direct tensile stress, which is tion of all kinds of cements is desirable. This not very favored by reason ofthe difficulty in fab- means research is needed to determine the phenomena which are responsible for the attack on the ' Figures In brackets Indicate the literature references at the end of this paper. filling materials. Special attention should be given

185 to the plaque which forms on the filling. Often to chewing forces, is dislodged from the cavity. It deep holes are discovered in silicate cement fillings is not likely that this dislodgement is a result of at places where cleaning is difficult. The basis for the formation of a corrosion layer upon the gold this elfect may be closely related to the theories alloy ; thus attention must be directed to the zinc which are developed to explain caries in the teeth. phosphate cement. When this cement is exposed Although the lifetime of silicate cement is limited, to acetic acid, a reaction product is noticed which it possesses a special benefit; namely, it decreases appeal's on the surface of the specimen. It is diffi- the chance for secondary caries. This is explained cult to tell if this also happens when the cement by the freeing of fluorides during the solution of is in contact with lactic acid. However, it seems the cement. If this theory can be proved by proper useful to research the development of reaction experimentation, it could be concluded that a cer- products on cements and on the forces which are tain solubility of the cement is required. Conse- consequently being developed. quently, then a test is also needed which will show Finally, in discussing filling materials, attention the solubility of fluorides : when the cements do not must be directed to ingredients which might ir- contain fluorides it is senseless to require a certain ritate the soft tissue and the pulp of the tooth. It solubility. is very important to know if an irritation will be It may be assumed that solubility phenomena temporary or if it will lead to a more serious com- will tend to create a decrease in the volume of in- plication. The latter will depend on several fac- serted silicate cements. An increase of the volume tors such as the concentration of the toxic ingredi- of filling materials may also occur such as is evi- ent and the time it is in an active state. Accord- denced by amalgams which are coated with a coring to these aspects research that may lead to the rosion product. If a coherent layer on the metal is creation of a safety test would be very useful and not formed, the formation of the corrosion prod- beneficial to dentistry. uct will go on. The growing forces are large and could lead to a deformation of the metal, when the 7. References corrosion product has no way to escape. Such phenomena might occur with the amalgam fillings. [1] ASTM STP 381, Symposium on Fracture Toughness When the flow property indicates that small forces Testing and its Application (1965). [2] ASTM STP 410, Plane Strain Crack Toughness Test- filling is fol- lead to a deformation of the which ing of High-Strength Metallic Material (1967). lowed by a loss of adaptation, then damage of [3] Griflath, A. A., The phenomena of rupture and flow in the margins is more likely to occur. An investi- solids. Phil. Trans. Roy Soc. London, Ser. A., 221: gation of the behavior of the corrosion product 163-198, (1920). [4] J. Gramberg : The EUipse-with-Notch Theory Ito ex- might very useful. be plain Axia Cleavage Fracturing of Rocks, (to be Many times a small inlay that is not subjected published).

186 — :

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseakch, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Need for Correlation Between Laboratory Testing and Clinical Research

Bjorn Hedeg&rd

Odontologiska Kliniken, Tandlakarhogskolan, Goteborg, Sweden

Although priority musit be given to basic and clinical research in preventive dentisitry, this does not eliminate the need for research on materials and methods for restorative dentistry which is in itself a preventive measure. There is an urgent need for collaborative and correlative studies involving properties of materials to be studied in the laboratory and tested in the clinic. Clinical research with materials is slow, and with the present incomplete information on oral environment and function, it is often difficult to assess properly the results, but there is no alternative. With sound clinical research on a larger and more penetrating scale, data and information may be obtained, that will make it possible to set

up more meaningful test procedures in the laboratory. And that is the goal : to be able to characterize the dental material in the laboratory and correctly predict its clinical performance.

Key words : Aging processes, relation to dentistry ; clinical research, dental ; correlation

of laboratory and clinical results ; dental materials dentistry, dentistry, ; preventative ;

restorative ; epidemiological information, dental.

1. Introduction to cope with the situation of the fast-growing population. I regret to have to report a failure. And this on If we thus give highest priority to prevention, behalf of the dental profession. Oral health is where does that leave us as clinicians and research- obviously our responsibility—^but with the meas- ers in the field of restorative denistry ? There are ures we have in our therapeutic arsenal we have three answers to that failed to uphold the oral health of the poptdation. 1. It will take a long time to develop effective There are many of us in the dental field—about and far-reaching preventive methods. 275,000 qualified dental practitioners spread all 2. Already today we have experience of the over the world. And we will increase our members fact that preventive measures are not fool- to 300,000 within the next decade. In that same proof ; that is, that a certain degree of decade the world population will have increased to failure has to be expected with any future and superseded the 3-billion line. method. These few figures indicate, of course, the reason 3. Traumatizing injuries, malformations, and for our failure. In fact, they imply that we as a congenital deformations will always pro- profession are facing an impossible task if we, as vide demand for restorative dentistry. we mainly do today, limit our activities to those Actually, these answers reach rather far into the which we learned and were trained for at the future. The immediate answer is simply : The re- dental schools ; that is to provide for removal and storative procedure is in itself a preventive meas- healing of pathologic processes and to restore ure and should consequently be incorporated in the anatomically—and hopefully also functionally list. And today and tomorrow the restorative pro- the damaged tooth or dentition. Within the limita- cedures are those, which will occupy most of the tion mentioned we have intricate and sometimes dentist's time. Or, I should rather say conservative in- near-to-impossible problems to face. And still this procedures, as preventive measures are always scaling. is not enough. The responsibility of our profession corporated. I have only to mention goes far beyond that. Its ultimate goal is to find ways and means to provide oral health to large 2. Epidemiological Information populations without relying upon chairside restor- There should also be another foundation laid ative procedures. down and that is our knowledge today of oral Priority should and must be given basic and status, need for restorative treatment, and possibly clinical research in preventive dentistry. There also demand for treatment. There are, imfortun- can be no objection to this obviously ; a noncarious ately, only a limited number of scientifically sound tooth is better than a restored one. It should also epidemiological studies done in this field. There- be underlined that effective preventive measures fore, we can today only give rather rough figures. provide the only means with which we can expect It could be outlined this way every fiftli person ;

4&2-525 0—72 13 187 above 16 has no natural teeth. Between one and to be solved in order to perform a correct treat- two of the remaining four wear full upper and/or ment and restoration. It should be underlined that partial removable dentures. At the most, three out all problems derive from the clinic and the answers of one thousand adults have noncarious dentitions. finally end up in the clinic. The need for treatment is high—it points to- You may comment that I am saying a mouth wards a dentist-population ratio of one to five full, and you are right. We cannot today furnish hundred. The awareness of need for treatment the laboratory researcher with sufficient clinical seems to be 60-80 percent of the need registered. and biological data, so that suitable laboratory But the demand for treatment is still considerably studies under simulated mouth conditions can be lower. It is a sad picture we gain from epidemio- performed. Actually, we know only the gross pic- logical studies in so-called well-developed coun- ture of oral function. It is still debated, whether tries. It is also a picture that heavily underlines the tooth contact occurs during the early chewing fact that restorative procedures have to be used for cycles, the time period, the direction, and the force a very long period yet—even if the incidence of during later contacts. Debate is also going on con- caries and periodontal disease is quicMy and drast- cerning mandibular movements and contact dur- ically reduced. What we actualy lack are more ing sleep. All have a bearing on methods and penetrating epidemiological studies set up on a materials in restorative dentistry. larger soale and internationally directed, so that It is only recently that some understanding has geographical as well as environmental differences been reached of tooth deformation and periodontal can be traced. performance under load. Studies of the elastic deformation of the todth, initiated by Korber and 3. Nature and Implication of Current collaborators, have indicated among other find- Restorative Treatment ings minor bending by lateral forces. This work, started in vivo and repeated in the laboratory, has Now, let us consider the restorative treatment considerable importance as background for devel- given. We define restorative in its widest sense, oping materials to be used in the tooth-restoration that is the endeavor for optimally restored func- system. tion irrespective of degree of dental deficiency. Especially in this field of elastic deformation And here the interesting question to be asked is: clinical studies and laboratory investigations of Will the treatment given have to be revised and model systems may close the now existing gap be- if so how long will it last? tween the basic materials scientist and the clinical Records on dental treatment show consistently researcher. What is absolutely necessary is a clear one and the same picture. In the adult patient the picture of the behavior of the tooth and its various restorations due to primary lesions are in the mi- components in order to define requirements for nority. Today the clinician is mainly concerned the restorative material as well as the restoration. with revision of earlier therapy. Due to secondary Here the very close collaboration between clini- lesions, restoration breakdown and changes in the cian, biologist, and physicist may yield the infor- properties of remaining tooth substances, revision mation sought after. Important steps have already of the former treatment will involve more exten- been taken; I may remind you of the work by sive restorative measures, which in due time will Stanford on the physical properties of hard tooth be followed by even more extended therapy. Thus substances. dental treatment breeds extended dental treatment. Any process or any measures that can delay the 5. Proper Assessment of Clinical Perform- deterioration of therapy given will in actual fact ance, A Requirement for Improve- increase the productivity of the profession. Just ment one example: You have in the United States around 100,000 dentists. If just one silicate filling Under the situation we have at present—^that is per dentist would have an increased longetivity incomplete information on oral environment and from four to five years, you would save 50,000 man- its changes with various factors—it is not possible hours to be used for dental treatment otherwise in laboratory testing to predict the detailed per- not available. We have to think in terms of pro- formance of a material in a tooth-restoration sys- ductivity irrespective of whether we are consider- tem. Obviously certain indications can be made, ing preventive dentistry or restorative dentistry. final an- In both instances we are after the same thing: but only the clinical test will yield the is firm belief that labora- lasting results. swer. It consequently my tory testing without evaluation of clinical per- 4. The Deficiency in Our Knowledge formance is unsatisfactory. of Oral Function It should not be necessary to point out that the same rigid research rules apply in clinical testing The problem we deal with in clinical dentistry as in laboratory testing, but evidently there is some is the impaired tooth and the consequent defective confusion among the clinicians as to this. In real- is, dentition. Here all the questions arise, which have ity, the mouth is the clinician's laboratory ; he

188 —

however, limited in his approach to evaluation ten belongs to the age group above 65. Aging can be methods, but there is no limitation to the require- described as decreasing cellular metabolism and ments on the strictness in registration of the steadily reducing function of the central nervous variables. system. We as clinicians register this in decreased As to selection of patient material for clinical function of the stomatogastric system. Saliva pro- studies, the aim of the study is important. We have duction is reduced ; the quality and composition of in a few studies deliberately selected patients with saliva changes towards more mucous consistency high carbohydrate intake with bad oral hygiene and less mineral content and reduced surface ten- this in comparative studies where two different sion. The motor activity is decreased, the threshold restorative procedures were performed intraindi- sensitivity increases, and function as a whole seems to be on a more primitive or basic level. vidually. The reason for this is obvious ; an acceleration of deteriorating factors decreases the Aging presents certain fundamental psychologi- observation period. For optimal performance a pa- cal factors as the most pronounced problems for tient group with other characteristics, for instance the dental clinician to deal with. However, in re- low caries incidence and good oral hygiene, may be cent publications on gerodontics, suggestions have the choice. been made that restorative procedures and selec- There are several studies at present where these tion of materials should be viewed in the light of circumstances are somewhat combined. In a small the properties of aging tissues. This is particularly group of patients (50) the treatment is given by important regarding the hard tooth substances, two well-trained operators, and the registrations of where changes toward increased brittleness may bilateral restorations are done collaboratively. In a provide a basis for specific properties of the re- considerably larger group of patients many clini- storative material to be used. Again, I am actually asking for future collabo- cians (15) perform the treatment according to their individual methods. In this group only a rative work regarding oral environment changes few variables are registered. All registrations are, with aging, this in order to gain a more sound however, done by two observers trained for these basis for the search for suitable materials or for studies. With such an organization we hope to ob- the selection of procedures in restorative treatment tain information on how the restorations stand up of the aging patient. under different circumstances. Laboratory tests are done according to related specifications and com- 7. Conclusion parisons are to be made later. I am mentioning these studies merely to underline the fact that the It is obvious that there is an urgent need for col- clinical investigation has to be carefully planned laborative and correlative studies involving properties in laboratory and that the patient material is one very important to be studied the and given the variable. final check and test in the clinic. It is, however, quite obvious that clinical research with dental materials is time-consuming and slow, but today 6. Consideration for Aging Processes in there is no way out of that. With sound clinical Clinical Studies research on a larger and more penetrating scale, data and information may be obtained that will There is another problem that we as clinical it possible to set more meaningful test researchers have come to be aware of during the make up last 10 to 15 years. About 10 percent of our popu- procedures in the laboratory in the future. And lation in Sweden is 65 or older. I understand that that is, of course, the goal : to be able to character- the percentage of older individuals in the U.S. is ize the dental material and in laboratory testing similar to that in Scandinavia, that is one out of correctly predict its clinical performance.

189

: ; : :

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental jVIatebials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Biological Evalvation of Dental Materials

Gunnar Ryge

Dental Health Center, Division of Dental Health

Department of Health, Education, and Welfare, San Francisco, Calif. 94118

Biological evaluation of dental materials includes (1) screening for toxic or other untoward effects of the materials, and (2) clinical evaluation of materials that pass the screening tests. Screening tests must take into account the functional requirements of various types of and service conditions for dental and auxiliary materials. Methodology for evaluation of clinical performance of materials must include examiner training and calibration in the use of rating scales. Emphasis is placed on the need for correlation of laboratory properties with clinical performance.

Key words : Biological evaluation of dental materials ; clinical evaluation of dental mate-

rials ; correlation of laboratory and clinical performance ; dental materials ; restorative

materials ; toxicity tests.

cellent conunents and criticisms—and, as you 1 . Introduction would expect, George Paffenbarger topped them Biological evaluation of materials used in den- all with his closing comment: "Don't shoot me tistry includes (1) screening for toxic, allergic, or now—wait until I'm a year older." This letter, other untoward effects of the materials and (2) incidentally was dated November 21, 1968, so I clinical evaluation of restorations, appliances, or guess Dr. George is safe for another 44 days. devices, the materials for which have passed the Previous speakers, including Dr. Schoonover and appropriate screening tests. Dr. Greulich, have made reference to Webster's and Maybe we should first discuss the definition of a Dorland's dictionaries. I am sure that I should dental material. About a year ago I sent out a have done the same before I sent out my working circular letter to initiate action on the part of the definitions. Now, however, I'm not going to quote Subcommittee on Biological Testing which was Webster and Dorland—but Paffenbarger and formed under the auspices of the FDI Commission Petzold, and I'm not talking about their abridged on Dental Materials, Instruments, Equipment and version either ! In fact, I don't believe they have

Therapeutics. Many respondents accepted these an abridged version ! According to the best au- working definitions but as those of you who have thority that I can find, this is it followed Dr. Paflenbarger's excellent and exem- Dental Material : Any substance specially intended plary work for the FDI Commission will know, for use in dentistry the greatest value in balloting of such circular (1) to replace tooth structure or tissue lost or letters lies in the votes of the dissenters, and, as it missing in part or in whole by injury, has often been demonstrated in Dr. Paffenbarger's disease, malformation, or by restorative modus operandi, the negative votes and the dental procedures thoughtful and thorough objections constituted (2) in the construction of appliances to allevi- the real substance of the discussions on definitions ate malformations and to restore and and classifications. improve esthetics and function. Although I cannot, in this short presentation, Similarly, the definition of Dental Device has pay tribute to all of the people who gave of their been clarified, improved, and refined to : "Any res- time and thought to constructive dissent, I must toration, appliance, instrument, machine, or equip- mention the excellent suggestions of Dr. Baume ment especially intended for use in dentistry." and Dr. Petzold from Switzerland, Alan Grant, Dental instrument is: "Any manual or powered Dr. Thonard, and Alan Docking from Australia, hand-held tool used in dentistry." Dr. Shoemakers and Dr. Van de Woerd from the Dental Drug: Any chemical compound or non- Netherlands, Kramer from the United Kingdom, infectious biological substance, not used for its Fischer from Germany, and several from the Scan- mechanical properties, which may be administered dinavian countries. Also, several people in the to or be used on or for patients United States, from military services, from indus- (ly as an aid in preventing, diagnosing, and try, and from universities, came through with ex- treating diseases, injuries, malfunctions.

191 ; ! —

and malformation of the teeth, jaws and plied clinical research program on Dental Mate- mouth rials and Technology, I saw the opportunity to do (2) to relieve pain or suffering or to control or something about this, and I guess that this is the improve any physiological or pathologi- reason you have to listen to me today cal condition. Most of you are probably aware that there are

Finally, Dental Therapeutic is : any dental drug, some inherent advantages in studying dental ma- dental material, or dental device used for prevent- terials in the laboratory, as compared to clinical ing, diagnosing, and treating diseases, injuries, and investigations. In the laboratory, one can design malformations of the teeth, jaws, and mouth. the experiments much neater—one can determine Obviously, there are some overlappings—these the number of specimens necessary to obtain statis- definitions do not segregate all of the categories tically significant results, and then proceed to that I had hoped to divide up into neat little pack- make the specimens under well-controlled condi- ages in my so-called working definitions. Some tions using the advice that one of my physics pro- agents will fall into more than one category, and, fessors gave me : you make all of the variables con- as I have gone through comments from all over stant and make all of the constants zero! Your the world in response to that first circular letter, specimens stay in their box until they are tested, I can see that, indeed, the dissenters carried the and you end up with a fairly well organized piece day and gave us sharper and better definitions of work that you can count on for the next lADR upon which we can build an international system meeting and a subsequent neat publication. Not so for biological testing. with clinical evaluation ! One of the first objections

The classification of dental materials on the I was faced with was the comment : "Are you tell- basis of functional requirements for the various ing me that you will evaluate dental restorations types and service conditions leads to meaningful placed by a number of operators?" Indeed I screening tests at various biological leveils. The planned to do just that. My scientific friends would classification developed by Ray Bowen in his work then say : "Don't you realize that you will have as Secretary of the Subcommittee on Toxicity operator variables?" My early answer was: "Yes Tests under the Specification Committee of the sir, I guess I will but has it ever occurred to you Dental Materials Group of the lADR does, with that operator variables do indeed exist in clinical minor modifications, appear to be internationally dentistry ?" The response from not one but several acceptable, and the next step, then, is to prescribe scientific friends can be summarized in a comment the screening methods that apply to the functional like : "Don't be smart—what I meant to say was, requirements. all you will find is that there is more difference operators between materials!" I An acute systemic toxicity test, a short duration between than conclusion inflammatory reaction test, a long duration tumor thought about that and came to the but, the other production and chronic inflammatory test, an al- that this could very well be so on this valuable information, lergic response test, an eye irritation test, a mucous hand, maybe would be don't recall finding membrane irritation test, and a pulpal irritation at least if well documented. I test can be prescribed selectively in appropriate this kind of information in the literature. Perhaps order and combinations to meet the needs on a it would be important to know that differences realistic basis. between materials were less important than operator variables particularly if one knew what were In spite of the shortcomings of that first circular — the operator variables that lead to this conclusion. letter on biological testing that was sent out about I am mentioning these comments and discussions a year ago—or perhaps because of these short- merely to illustrate the philosophy and the view- comings—this first effort has produced such a that motivated to take on the attempt to wealth of good comments that I am indeed opti- points me evaluation of dental materials a mistic about future progress in this area, and quite make clinical meaningful discipline. a number of those present here today deserve the legitimate and, hopefully, credit for it. 3. Application of Research Methodology 2. Motivation for Clinical Evaluation to Clinical Evaluation of Dental of Dental Materials Materials

Evaluation of the clinical performance of den- How does one evaluate a dental restoration tal materials has, until a few years ago, been based how does one move from testimony to research upon testimonials. The need for establishment of methodology ? Dave Mahler today showed one ap- sound clinical research methodology has been em- proach and I believe that my approach will be phasized by many investigators and I i^ersonally useful both by itself and as a supplement to his must confess to ha\dng used the phrase: "The method. In clinical practice, every dentist is mak- clinical importance of these findings is not estab- ing evaluations every time he has a patient in the lished" in several reports of laboratory studies of chair. He looks at each existing restoration and dental materials. Therefore, when the Division of makes a decision that leads to one of four courses restoration Dental Health approached me to establish an ap- of action : He either decides that the

192 !

decides that ago so that he could revise the whole system and is okay—no action needed ; or else he show the efforts of the adminisitration there is some doubt about the restoration. He may up previous decide that he isn't going to do anything now, but of the Branch that he would like to see the patient six months As a result, the clinical evaluation system now

consists of rating scales for : Color Match; from now for another evaluation ; a third decision (1) (2) would be that of replacing the restoration for Cavo-Surface Marginal Discoloration; (3) Ana- preventive reasons and, the last course of action tomic Form; (4) Marginal Adaptation, and (5) would be that of replacing the restoration because Caries. illustrate the type it is esthetically or functionally unsatisfactory. It To of judgments that are looks bad or damage is occurring around it. made to measure the clinical performance of How do we, as practicing dentists, arrive at these paired anterior and paired posterior restorations judgments? Which are the factors that we con- fabricated from contrasting dental restorative materials, reference is made to figures 1 and 2. Figure sider ? What are the criteria ? A thorough analysis of the elements that enter into such judgrnents was 1 shows the categories chosen for color match while precisely the basis for development of rating scales figure 2, similarly, describes the categories, or for clinical evaluation of the performance of re- ratings, used to evaluate marginal adaption. storative materials, as these rating scales are used A similar system is used for arriving at ratings today by the Materials and Technology Branch of for Cavo-Surface Marginal Discoloration (fig, 3), the Division of Dental Health, both in intramural Anatomic Form (fig. 4), and Caries (fig. 5), so and extramural or cooperative projects. that for each restoration, a total of 16 judgments The criteria for clinical evaluation, as we used or choices are made, resulting in 5 ratings. Each to call them, were formulated, in their first version, of two examiners carriers out such an evaluation lasted into in the fall of 1964 in a work session that independently and calls off the ratings to the re- hours of the night, and I would like to the wee corder. When the two examiners arrive at a differ- mention the names of three co-workers from my ent rating for any one category, the recorder will this session, two of whom are here today : Dr. Bjorn request a resolution of the disagreement by joint Hedegard who, as most of you know, conducted examination. Such disagreements are usually the one of the first—^and finest—studies on restorative result of one examiner discovering a discrepancy direct resin materials. Dr. Richard L. Webber and or defect that was overlooked the other Dr. R. J. McCune of the Materials and Technology by exam- Branch. Jim McCune wasn't quite satisfied with iner and are easily resolved by trained examiners. the early effort, so he proceeded to take over the In addition to the ratings for each pair of restora- Materials and Technology Branch about 9 months tions, a ranking procedure is used whenever the two

COLOR MATCH

Call Code Is ttie restorative Yes material metallic? Hotel H

Can you see it Call Code Is the restoration In Yes^>Hii_ an anterior tootli? witliout using Oscar a mirror? 0

Is there a mismatch In color, shade and/or translucency Call Code between the restoration and the Alta A adjacent tooth structure? Test: Visual Inspection at 18" without mirror on anterior restorations, with mirror on posterior restorations

Is the mismatch between restoration and adjacent Call Code tooth struchire outside the "no^ normal range of tooth color, Bravo B shade and/or translucency?

Call Code "ves^ Charlie C

FiGtJBE 1. Color match.

193 MARGINAL ADAPTATION

Test: Is there visible evidence of Call Code Lightly draw a sharp explorer a crevice along the margin hack and forth across the into which the explorer will Alfa A margin. If It "catches," penetrate? Inspect for crevice with mirror if needed

Call Code Test; Is the dentin or base exposed? ""no^ [ }.

I Visually Inspect I Bravo B

Call Code Test: Is the restoration mobile, Charlie C Visually inspect or fractured or missing test mobility of In part or in toto? restoration with explorer

Call Code. Delta D

Figure ;!. Marginal adaptation. CAVO SURFACE MARGINAL DISCOLORATION

Is the restoration Call Code metallic? Hotel H

Is there discoloration anywhere Call Code on the margin between the Alfa A restoration and the tooth structure?

Test: Visual Inspection of entire margin with mirror, If needed

Has the discoloration penetrated Call Code along the margin of the restorative Bravo B material in a pulpal direction?

^es^ Call Code Charlie C

Figure 3. Cavo-surface marginal discoloration. paired restorations receive the same rating by the evaluation records and performs complex edit- both examiners for any one characteristic. ing procedures to guarantee the accuracy of stored Clinical evaluatiQn studies are planned accord- information. The system is based upon the con- ing to a set of procedures that include strict ad- cept of a research team consisting of a dentist, a herence to the research protocol and use of specific dental assistant, and an observer-recorder, and the procedure records (figs. 6 and 7), evaluation rec- role of each person is defined in the protocol for ords (fig. 8), and other forms (figs. 9 & 10) that the study. permit effective utilization of data processing Twice annually, the Materials and Technology eq[uipment. The computer program not only pro- Branch conducts a training, calibration, and test- vides tabulation of statistical data but generates ing session for its clinical staff to maintain per-

194 ANATOMIC FORM

inspect with Visually is the restoration under-contoured if necessary mirror, i.e. is the restorative material Call Code existing anatomic discontinuous with Alfa A lorm?

Visually inspect with Is sufficient restorative material Call Code missing so as to expose the dentin mirror, if necessary or base? Bravo B

Call Code

Charlie C

FIGUEE4. Anatomic form.

CARIES

Test: Visual Inspection Is there evidence of carles Call Code No contiguous with the margin of with explorer and mirror Alfa A If needed. the restoration?

Call Code Yes X Bravo B

(1) An area at the restoration margin is carious if an explorer c. Etching or a white spot as evidence of demineralization **catches" or resists removal after insertion with moderate An area at the margin is also considered carious if the to firm pressure, and is accompanied by one or more of the explorer does not "catch," but conditions b or c are following: present.

a. Softness

b. Opacity at the margin, as evidence of undermining or demineralization

Figure 5. Caries.

formance standards. These sessions have also been scheme that has been developed over the past five attended by personnel from cooperating institu- years at the Dental Health Center. I have no hesi- tions both from the United States and other tation to admit that we have made mistakes along countries. the way—but we have benefited from some of these To supplement the clinical evaluation of paired mistakes by being forced to review procedures restorations fabricated from contrasting restora- and judgments. I hope, however, that I may have tive materials or of restorations placed with con- succeeded in arousing your curiosity or even inter- trasting techniques, an attempt is being made to est in the system and that those of you that might design test methodology that will provide mean- be inclined to venture into the magic world of ingful correlation of laboratory data with the clin- clinical research have obtained some information ical findings. Time does not permit a discussion of that may help you avoid making the same mistakes this phase of the work but we are in this respect that we have made along the way. following similar guidelines as those Dave Mahler I like to conclude on an optimistic note by laid down so beautifully this morning. would quoting a recent pronouncement by my Division

Diefenbach ; it is particu- 4. Conclusion Director, Dr. Viron L. larly apropos to describe the status of clinical re-

It is not feasible to describe fully in this short search : "The best part of our future lies ahead of presentation all aspects of the clinical evaluation us."

195 00 NOT USE PROCEDURE RECORD THIS COLUMN POSTERIOR RESTORATIVE STUDY MATERIALS AND TECHNOLOGY BRANCH, DIVISION OF DENTAL HEALTH, NIH

1-3

STUDY name: I NSTI TUTION

i-9 PATIENT'S ADDRESS 10-29 PATIENT'S NAME (type or print)

LAST Fl RST DENTAL ASSISTANT 30 SEX (circle) 31-36 DATE OF BIRTH

M F OBSERVER/RECORDER MONTH DAY YEAR 37-42 DATE OPERATOR RESTORATIONS PLACED MONTH DAY YEAR 43-44

"Type TOOTH SURFACE(S) CLASS BASE LINER MATERIAL OF NUMBER (ci rcle) (circle one) CODE CODE CODE

45 44-47 48-49 50 51 53 54-56 57 60

T 12 M 0 F L 0 3*5 61-62 63 -64 65 66 68 69-71 72 75 C 12 3*5 M D F L 0

DO NOT 76 USE THIS (T) TEST SPACE CONTROL

time trituration STARTED 77 HOUR MIN. (C) HOUR MIN.

MIX 1 MIX 1 78 2 MIX MIX 2 TRITURATION TIME SEC. SEC. SEC. SEC. '9 BLANK LAPSE TIME SEC. SEC. SEC. SEC.

INSERTION TIME SEC. SEC. 80 SEC. SEC. DELAY TIME SEC. 850, END OF KEYPUNCH TIME REQUIRED FOR INITIAL

CARV I NO SEC, SEC. TIME INITIAL CARVINQ COMPLETED HOUR MIN. HOUR MIN.

DAY FINAL POLISHING PERFORMED MO. DAY YR. MO. DAY YR,

TIME REQUIRED FOR FINAL POLI SHI NQ MIN. SEC. MIN. SEC- WAS AN IMPRESSION MADE? YES NOO

M.T. BR. FORM 65 - 9

(rev. 02 29 • 69) PENDING APPROVAL

Figure 6. Procedure record posterior restorative study.

196 00 NOT USE PROCEDURE RECORD THIS ANTERIOR RESTORATIVE STUDY COLUMN MATERIALS AND TECHNOLOGY BRANCH, DIVISION OF DENTAL HEALTH, NIH

I -3 6TU0V NAME I N6TI TUTI ON

4-S PATIENT'S ADDRESS 10-29 6-9 PATIENT'S NAME (type or print)

LAST Fl RST DENTAL ASSISTANT

30 31 -36 SEX (circle) DATE OF Bl RTH

M F OBSERVER /RECORDER MONTH DAY YEAR 37-42 DATE OP ERATOR RESTORS^TIONS PLACED MONTH DAY YEAR 43-44 TYPE TOOTH SURFACE(S) CLASS BASE LINER MATERIAL OF (ci RCLE one) PAI R NUMBER (ci rcle) CODE CODE CODE 45 46 4 7 4 8-49 51-53 54-56 57-60 T 12 3^5 M 0 F 1 L

61 -62 63-64 65 66-68 69-71 72-75 C 12 3^5 M D F 1 L

76 (T) 00 NOT TEST USE THIS CONTROL 77 (C) SPACE

TIME MIX STARTED 78 HOUR MIN. HOUR MIN, MIXING TIME SEC . SEC. 79 BLANK LAPSE TIME SEC. SEC. 80 INSERTION TIME SEC. SEC. TIME MATRIX HELD END OF KEYPUNCH SEC. SEC, DELAY TIME SEC. EEC. TIME REQUIRED FOR INITIAL FINISHING SEC. 6EC. TIME INITIAL FINI8MINQ COMPLETED HOUR MIN. HOUR MIN. DATE FINAL FINISHING PERFORMED MO. DAY YEAR MO. DAY YEAR TIME REQUIRED FOR FINAL Fl SHI NO Nl MIN. SEC. MIN. SEC. WAS AN IMPRESSION MADE? YES NO

M. T. BR. FORM 65 6 29 (rev, 02 69 ) PENDING APPROVAL

Figure 7. Procedure record anterior restorative study.

197 EVALUATION RECORD ANTERIOR AND POSTERIOR RESTORATIVE STUDIES PATIENT S NAME MATERIALS AND TECHNOLOGY BRANCH

DIVISION OF DENTAL HEALTH, U. S. P. H. S. DATE EVALUATED

STUDY INSTITUTION PATIENT S DATE 19 NUMBER NUMBER NUMBER PLACED MONTH YEAR TYPE OF CAVO-SURFACE ANATOMIC MARGINAL PAIR CRITERIA COLOR MATCH MARGINAL FORM ADAPTATION DISCOLORATION TOOTH \ NUMBER

SURFACE EXAMINER

CLASS FINAL RATING R EXAMINER [o] [2] A 0 m s m [1 a m 0 0 0 0 0 N EXAMINER K 0 m [u 0 m @ 0 m a 0 m 0 0 0 0

TOOTH PATIENT REMARKS: REASON MISSING MOVED RESTORATION OTHER NOT MISSING (EXPLAINi EVALUATED RESTORATION REPLACED

TOOTH \ EXAMINER NUMBER

SURFACE EXAMINER

CLASS FINAL RATING R EXAMINER A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N EXAMINER K 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

COLUMN 80 1 1 1 REMARKS: 1 TOOTH PATIENT REASON 1 1 MISSING 1 1 MOVED

1 1 1 OTHER NOT 1 RESTORATION 1 1 1 (EXPLAINi 1 MISSING

EVALUATED 1 RESTORATION 1

1 I REPLACED CD-

Figure 8. Evaluation record.

198 DAILY REPORT OF BATCH NUMBERS AND VARIATIONS OF RESEARCH PROCEDURES

Study Name_ Study Number _____

Institution

Date Work Performed Day Month Year

PART I: Manufacturer's Batch Numbers

Manufacturer's Batch Number N 1 im Knr Ploced Material Used Base (or solid) CaValysf (or liquid)

Test Base

Liner

Restorative

Coritro! Base

Liner

Restorative

PART IK Variations of Research Procedure

D No Variations in Research Procedures Occured on (he Above Date

Signature of nparntn.- Date

Variation of Research Procedure Check Tooth Surface Patient' s Name (Describe Variations and explain) One Number & Class

Test Tooth

Control Tooth

Test Tooth

Control Tooth

Test ToothD

Control Tooth

COMMENTS: (Include recommendations for the further conduct of this research)

Figure 9. Daily report of batch numbers and variations of research procedures.

199 M.T.Br. Form 69-1 Pending Approval PATIENT PARTICIPATION AGREEMENT

As port of an effort to develop better materials and procedures for use in dentistry, new and conventional dental materials, techniques, and devices are being investigated in controlled clinical studies. Since such studies involve considerable expense and require up to five years of periodic examinations, only persons willing to cooperate fully in this program will be included. Treatment that falls within the scope of the clinical study will be performed by staff members of

in cooperation with the Materials and Technology Branch, Division of Dental Health, National Institutes of Health. Your decision regarding participation will in no way affect your eligibility for dental care customarily provided.

If you are interested in participating in this program, please supply the following:

Name Address (last) (first) (middle) R,r*i^Anto Sex Telephone

Name and address of your dentist

Date of last dental visit

FLUORIDE HISTORY: In drinking waterD Topically Tablets None YES NO Have you now or have you ever had:

Rhpiimntir Fpvpt

Heart Trouble _ . _ ,., , ,

High or 1 nw RInnrI Prp«;5iirp

Tuberculosis - . . .

Asthma or Hayfever

Allergy to any drug or medicine

Hepatitis (Jaundice) .

Reaction to locc' anesthetic

Prolongea bleeding from an injury nr Innfl, pittmrtlnn

Are you taking any drugs or medicine?^

Are you under the care of a physician?—

If "yes" to any of the above, please explain

The attending dentist has explained the nature of the study to me and the

need to return for periodic examinations. I will do my best to cooperate.

Date Signature (patient, parent or guardian)

FiGUBE 10. Patient participation agreement.

200 ;

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Reseasch, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Corrosion Testing in the Mouth

Kazuo Nagai

School of Dentistry, Nihon University, Tokyo 101, Japan

For many years various base-metal alloys were used as substitutes for gold alloys for dental restorations in Japan. The adoption of an oflScial requirement that these alloys should not have a weight loss of more than 3 mg/cm" when immersed in solutions of 0.05 percent hydrochloric acid, 1 percent lactic acid, 1 percent sodium chloride, and 0.1 percent sodium sulfide for three days, prompted a study of corrosion of a copper-zinc alloy and dental amalgam in the mouth. Cast copper-zinc specimens had average weight losses in the mouth in the range of 0.022 to 0.034 mg/cm^ per day. Conventional amalgams had losses as small as 0.0063 mg/cm^ per day. AVeight losses in the test solutions ranged up to 150 times those in the mouth while weight losses in artificial saliva in vitro were about twice those in the mouth. It is believed that the lower corrosion in the mouth (which varies from place to place) results from a cleaning and inhibiting action by the saliva.

Key words : Base-metal alloys, dental copper-zinc dental alloy ; corrosion, clinical tests ;

corrosion of dental alloys ; dental amalgam ; dental materials.

1. Introduction their actual weight losses due to corrosion, as well as other effects. Immediately before World War II, and during the intervening years, the dental profession in 2. Corrosion Testing of Copper Alloys Japan has not been able to use gold for restorative purposes, the main reason being the critical sup- In the first part of my discussion I would like to ply of gold, as well as the economic factor. Con- introduce an investigation on corrosion testing of sequently, a variety of gold substitute alloys have copper-zinc alloy in the mouth which was reported been developed. Metallurgically, they include al- by Dr. Wakumoto. This test was carried out by loys of silver, German silver, copper, stainless steel, four different subjects using four specimens of nickel-chromium, chrome-cobalt, and others. Much various shapes and sizes. The position of the speci- research has been accomplished and published on men in each oral cavity was also varied. The speci- these alloys, and controversy still exists on the ac- men was cast and polished in the usual manner in ceptability of the alloys in dentistry. In October accordance with the manufacturer's instructions, 1950, the Ministry of Welfare of the Japanese and placed in the mouth for 50 days. Weight Government issued an official test for corrosion of change was measured by electric balance every 10 these alloys. This test required that any gold sub- days. stitute alloy used in the mouth could not have a Figure 1 shows the sizes and shapes of the speci- weight loss of more than 3 mg/cm^ when immersed mens of subjects A to D. The shaded portion in for three days in solutions of: 0.05 percent hydro- the figure indicates the copper alloy specimen chloric acid, 1 percent lactic acid, 1 percent sodium having different surface area and shape. The result chloride, and 0.1 percent sodium sulfide. of Wakumoto's report is summarized in figure 2. No reason or justification was given to the Pro- The vertical axis indicates weight loss in mg/cm^, fession by the Ministry of Welfare for the com- and the horizontal axis indicates time in days. It position of the test reagents. No information was is obvious from this figure that the result was provided as to why a dental restorative material divided into two groups. should be subjected to this kind of test, nor Table 1 shows the weight changes of the copper whether the concentrations of the chemical re- alloy specimens in the mouth after 50 days and the agents were appropriate. This is the basic reason average value per day. The largest weight loss is why we initiated a series of studies on dental alloy 2.697 mg/cm^ after 50 days and 0.0539 mg/cm^ per corrosion in the mouth, particularly with the cop- day in subject B. The least change is 0.730 mg/cm^ per-zinc alloy and amalgam. We also wished to after 50 days and 0.0146 mg/cm^ per day in subject confirm the contention that cojDper alloy Avas in- D. This variation may be due to the individual dif- jurious to the human body both locally and sys- ferences among subjects, and a change of alloy temically. Consequently, the copper alloy and structure of each of the specimens when they were amalgam were placed in the mouth to determine cast.

201 Table 1. Weight loss of Cu alloy in the mouth for 50 days CORROSION TEST IN THE MOUTH (by Wakumoto) SHAPE AND SIZE OF Cu ALLOY SPECIMEN - by WAKUMOTO - Weight loss (mg/cm^) Subject

After 50 days For a day

A Jx 2. 597 0. 0519

B 2. 697 0. 0539

C 0. 802 0. 0160

D 0. 730 0. 0146

In figure 3 the size and shape of specimens is given for the in vivo corrosion test of copper alloy as determined by Dr. Sakimia. This test was performed in order to evaluate the common opinion 17.22 cm^ that the use of copper alloy for dental purposes was harmful to the human body. He wore four specimens which have different shapes and sizes FiGTJEE 1. Corrosion test in the mouth by Wakumoto. in his mouth. The shaded portion in th« drawing Indicates the shape Figure 4 is a summary of the results of Dr. and size of the copper alloy specimen. Sakuma. According to his test, the weight loss of copper alloy in the mouth is between 0.6 and 1.4 ^—1 1 T —r mg/cm^ after 50 days. Although specimen C was broken at 40 days

(table 2) , its weight loss is the largest, being 1.383 mg/cm^ for 40 days, which corresponds to 0.0346 mg/cm^ a day. The least change in Sakuma's report is 0.611 mg/cm' after 50 days, and 0.0122 mg/cm^ per day in specimen A.

O 10 JO 30 40 50 Days Figure 3. Corrosion test in the mouth by Sakuma. FiGTJBE 2. Weight loss of copper alloy in the moutfi The shaded portion in the drawing indicates the shupe and reported hy Wakumoto. size of the copper alloy specimen

202 , -

T 1 1 1 r CORROSION TEST IN THE MOUTH SHAPE AND SIZE OF Cu ALLOY SPECIMEN Loss of Cu Alloy Weight - by NAGAI-OHASHi In the Mouth 2.0 - - by SAKUMA -

4 Shapvt of Spectman*/

1 Subjact

E » 1.5-

FiQXJBE 5. Corrosion test in the mouth by Nagai and Ohaahi. The shaded portion in the drawing indicates the shape and size of the copper alloy specimen. 0 10 30 30 40 50 Do y»

T 1 1 1 FiGUKE 4. Weight loss of copper alloy in the mouth reported by Sakuma. Weight Loss of Cu Alloy Table 2. Weight loss of Cu alloy in the mouth for 50 days {by Sakuma) in the Mooth 2.0 - by NAGAI-OHASHI - Weight loss (mg/cm') 3 Shapes of Sp«clm«n» Specimen I Subject

After 50 days For a day

1.5 A 0. 611 0. 0122 n E B 0. 941 0. 0188

C* 1. 383 0. 0346

D 1. 159 0. 0232 1 .0

After 40 days

Figures 5 and 6 relate to the in vivo corrosion test of copper alloy by our laiboratory. made We O.S three palatal denture specimens using a copper alloy, so that they have nearly the same size and shape. The position of the specimen in the mouth was also designed to be the same. The test was repeated three times in the same mouth. As is shown in figure 6, our test results have 50 less variation among specimens compared with the two reports described before. This fact proves our contention that the shape speci- and type of the FiGUBE 6. Weight loss of copper alloy in the mouth reported men is of less significance in the corrosion test of by Nagai and Ohashi.

203 452-525 0—72 14 dental alloy in vivo than the individual oral cavity difference among subjects. In our test, average Aveight loss per day ranges from 0.028 to 0.037 mg/cm- (table 3).

Table 3. Weight loss of Cu alloy in the -mouth for 50 days {by Nagai, Ohashi)

Weight (loss mg/cm^) Specimen

After 50 days For a day

FiGtTEE 8. Discoloration of copper alloy specimen after A 1. 71 0. 0342 about 50 days in the mouth.

B 1. 42 0. 0284

C 1. 86 0. 0372

Figure 7 is a photograph of the copper alloy specimen when it was placed in the mouth. It has very nice color and brightness at this stage of the testing. A slight discoloration was observed after one month. In figure 8, taken after 50 days, the copper alloy showed more discoloration and less luster. Figure 9 is a typical example of a copper alloy crown which has been cemented in a patient for about 1 year. Not much corrosion or discoloration was observed on the occlusal surface, but quite a few corrosion spots were found on the proximal surface after it was remoA'rd. Generally speaking, it is hard to find the corrosion of a copper alloy crown or inlay by visual inspection during serv- FiGUKE 9. Typical example of a copper alloy crown which ice, but the corrosion is in progress on an unsani- has been cemented in a patient for about one year. tary portion such as the proximal surface or sub- gingival area. reports (table 4), the average weight loss for a Let us reconsider the main subject of my paper. day is 0.0341 mg/cm- by Wakumoto, 0.0222 mg/ When we compare the corrosion value of three cm= by Sakimia, and 0.0333 mg/cm^ by our laboratory. 'The largest weight loss is 0.0539 mg/cm- and the smallest 0.0122 mg/cm? in 11 trials. The overall average of the three reports is 0.0299 mg/ cm^ per day. According to these results, the copper alloy which is commonly supposed to have a high degree of corrosion actually loses weight in the mouth much less rapidly than was thoug'ht.

Table 4. Comparison of weight loss of Cu alloy in 3 reports

Wakumoto 's Sakuma's Nagai-Ohashi's Report Report Report

Subject (mg/cmV Speci- (mg/cmV Speci- (mg/cmV day) men day) men day)

A 0. 0519 A 0. 0122 A 0. 0342 B 0. 0539 B 0. 0188 B 0. 0284 0. 0372 C 0. 0160 C 0. 0346 C D 0. 0146 D 0. 0232 Average 0. 0341 Average 0. 0222 Average 0. 0333

FiGXJEE 7. Copper alloy specimen when it was placed in the mouth. Overall average= 0.0299 (mg/cmVday)

204 :

3. Corrosion Testing of Conventional drochloric acid solution about 150 times as much Amalgam as the decrease in the mouth. Even in 1 percent sodium chloride solution, it is about 60 times as In the next part of this paper, I would like to great as in the mouth. The results indicating that, discuss tests carried out by our laboratory on the greater weight loss was observed in artificial saliva corrosion of a conventional type amalgam in the than in the mouth are probably due to the fact mouth. A cylindrical specimen, 4 mm in diameter, that human saliva has an inhibitive action to the was made according to the manufacturer's in- metal corrosion. structions and after 24 hr was polished to be 4 mm When we compare the alloy weight loss with that long. The specimen was imbedded in the occlusal of human enamel, the weight loss of human enamel surface of an acrylic resin artificial tooth for an is about 40 times as great as that of copper alloy upper partial denture so as not to contact the op- and amalgam in 0.05 percent hydrochloric acid, posite teeth. It was taken out and weight change and more than 100 times as great in 1 percent was measured after 40 days. The test was repeated lactic acid. Thus the results indicate that the cor- three times in one subject. rosion test in the laboratory as proposed by the The weight loss of amalgam after 40 days in Minister of Welfare is far from the actual corrosive behavior in vivo. vivo is shown in table 5. A variation of weight loss between both sides, the right and left, is noted. Average weight loss is 0.250 mg/cm^ for 40 days, 4. Summary and Conclusions which is only 0.00625 mg/cm^ per day. This value includes the weight loss by erosion or abrasion The purpose of our studies was to determine the corrosive behavior of dental restorative alloys in produced by contact with the tongue and food in the mouth and, at the same time, to examine them addition to the corrosion, so that actual weight loss from a dental health point of view. The salient by corrosion must be less than the result we conclusions are as follows obtained. 1. Even the copper alloy which is commonly supposed to have a high degi'ee of corrosion, Table 5. Weight loss of amalgam in the mouth after 40 days actually has weight loss (by Nagai, etc.) m the mouth as small as 0.022 to 0.034 mg/cm^ for a day. 2. The type of amalgam which is widely used in Weight loss (mg/cm^) Number of many countries today has a weight loss in vivo as small as 0.0063 mg/cm^ a day. test Average 3. The basic reason why these alloys are not corroded as is generally supposed is that the sali-

1 0. 35 0. 21 0. 28 vary secretion has the benefit of keeping the mouth 2 0. 28 0. 22 0. 25 clean. It is also due to the fact that the saliva 3 0. 27 0. 17 0. 22 tends to inhibit the corrosion of alloys. Average 0. 30 0. 20 0. 25 4. As it became clear by our testing that the 1950 Ministry of Welfare corrosion test was not Average weight loss per day= 0.00625 mg/cm^/day. realistic as a means of expressing the corrosive behavior of dental alloys, it was removed from the Finally, in table 6 the weight changes of hu- specifications. man enamel, copi^er alloy, and amalgam in various 5. Within the same oral cavity, the degree of environments are compared. In addition to the corrosion varies from area to area, that is, ante- specified chemical reagents, an orange juice of pH I'ior teeth, posterior teeth, occlusal, buccal to lin- 3.2 and artificial saliva prescribed by Greenwood gual, or proximal surface. Observation of the at a pH of 6.8 were used. The weight loss is far cojoper alloy restorations after they were removed greater in vitro than in vivo. By way of example, indicated that the corrosion is most pronounced on amalgam decreases its weight in 0.05 percent hy- the proximal surface.

Table 6. Comparison of weight loss of permanent tooth, amalgam, and Cu-alloy in various environments (mg/cm^/day]

Env. 1% Lactic Orange Artificial Oral 0.05% HCl 1% NaCl cavity Mat. ~ --^ acid juice saliva

Teeth 40. 60 94. 20

Amalgam. _ _ 0. 94 0. 83 0. 39 0. 145 0. 0102 0. 0063

Cu-alloy _ _ . . 1. 01 0. 68 0. 073 0. 07 0. 059 0. 0299

205

VII. Specifications

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research. Proceedings of the 50th Anniversary Symposium^ Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

International Specification Program—Australian Experience

Alan R. Docking

Commonwealth Bureau of Dental Standards, Melbourne, Australia

The Australian dental specifications as a basis for a certification program were developed through the Standards Association of Australia and not within the dental association. Products are accredited b.v the Australian Dental Association on the basis of meeting these specification tests. The Commonwealth Bureau of Dental Standards assists in preparation of specifications and in accreditation of products. Approximately thirty Australian dental standards have been issued. Participation in international specification work has proved of great benefit especially in the raising and maintenance of the quality of dental goods used by the Australian dental profession.

Key words : Accreditation of dental materials, Australian experience ; Australian dental

specifications ; dental materials ; dental standards ; specifications, dental.

1. Introduction a small number is below the critical nucleus size for the growth, on a national basis, of a systematic If imitation is the sincerest form of flattery, the research, standardization, and testing program Australian tribute on this memorable occasion is for dental materials. All the more credit is due of special significance. From the begiimings of to the early foresight of members of university dental materials research in Australia over thirty staff and the Australian Dental Association that years ago an effort has been made to emulate, in such a program was not only initiated, but a modest way, the highly successful program of flourished. the National Bureau of Standards and subsequently the system of specifications for materials 2. The Bureau of Dental Standards and of accreditation of dental products employed by the American Dental Association through the The staff of the Bureau of Dental Standards Fellowship it established at the Bureau. Thanks umnbers only eight, none of whom ai'e qualified to the inspiration and guidance of the NBS Dental dentists, unfortunately. Physicists, chemists, and Research Section and the ADA Fellowship there, metallurgists are included and for clinical opinions the Australian effort has proved to be of immense or trials the Bureau relies on the dental schools, value to the dental profession in that country too. the Services, and individual dentists either inde- The Commonwealth Bureau of Dental Standards pendently or through a Panel of Cooperating was established in 1947, but the work did not com- Practictioners throughout each State of the mence then; ten years prior to this, Dr. Howard Commonwealth. K. Worner, with his Materials Research Labora- There are essential differences in organization tory at the University of Melbourne School of too in that the Bureau of Dental Standards was Dental Science, had established a close liaison both established as an activity of the Commonwealth technically and personally with members of the Department of Health; furthermore, the stand- staff of the Dental Research Section of the Na- ard specifications on which the certification pro- tional Bureau of Standards. It was the knowl- gram is based, were developed through the Stand- edge and appreciation of the NBS success story ards Association of Australia and not within the that largely shaped the destiny of dental materials dental association as it was in America. It is in- research and specifications in Australia. teresting to note that in recent months the cor- There are many differences, of course, between i-esponding body in America, The United States the two systems both quantitatively and qualita- of America Standards Institute, has now become tively. In size, the Australian program is very very actively involved in the preparation of den- much smaller. Although Australia is of compar- tal specifications and is to become the national or- able size to the main body of the U.S.A., its ganization through which such specifications are population is only some twelve million, served by to be promulgated. about four thousand dentists. On this basis one In Australia, as in the United States, the na- could expect something of the order of a twentieth tional dental association is the body which con- in magiiitude. One would almost imagine that such trols and directs the accreditation program and

209 issues an official List of Certified Products. It has a Participation in the international sphere has proved to of great benefit to the local specifica- special comiriittee for this purpose and for taking be | tion to the raising maintenance action on other matters concerning the quality and program and and j supply of dental products available on the Aus- of the high quality of dental goods used by the tralian market. The Standards Association of Aus- Australian dental profession. tralia, through the help of the dental profession, the industry and the Bureau of Dental Standards, 4. Australian Specification Program prepares the specifications; and the Australian | Association accredits based on Dental products As to the specification program itself, some specifications they endoree. It is the these which thirty Australian dental standards have been is- Bureau of Dental Standards that does the neces- sued. These cover a wide range of items including testing, checking, investigational work sary and mercury, amalgam alloy, zinc phosphate and sili- required to assist both in the preparation of spec- cate cements, impression and laboratory plasters, ; ifications and in the accreditation of products artificial stone, inlay, sticky and modelling waxes, complying with them. modelling compound, agar and alginate impression materials, impression paste, inlay and denture 3. International Participation casting gold alloys, cobalt-chromium casting alloy, casting investment, denture base resin, synthetic Because of the limited local demand, Australia resin teeth, local anesthetic solutions, x-ray films,

has never been anywhere near self-sufficient in the hypodermic needles, hand cutting instruments, ! manufacture of dental supplies. At the outset, and gold and silver solders. ' therefore, cognizance had to be taken of the fact Currently, work is in hand to develop Austra- that a large proi>ortion of dental goods are im- lian standards for elastomeric impression mate- ported and the accreditation program had to be rials, rubber dam, orthodontic cements, matrix j adapted accordingly. It is only in recent years band material, rubber elastics, endodontic instru- ' that the American Dental Association program has ments, root canal materials, and toothbrushes. included any imported products and these are still Some of the existing specifications are under very much in the minority. There is a certain revision. amount of dental manufacturing in Australia, controlling the specification pro- particularly in relation to gypsum products, amal- The body in gam alloys, gold alloys, waxes, modelling com- gram is the Dental Materials Committee of the pounds, hydrocolloidal impression materials, Standards Association of Australia. This has equal orthodontic wires, local anaesthetic solutions, syn- representation of profession and trade together ' thetic resin teeth, and a few cements. The large with representation from the Department of bulk of equipment, instruments, and other fabri- Health. It decides on policy and priority, but the cated items are imported also, for every product ; drafting work is done through a number of sub- manufactured in Australia, there are many of the committees appointed for special fields such as same type available from overseas, in spite of gypsum products, impi'ession materials, cements, ! various protective tariffs that be imposed to may waxes, casting alloys, synthetic resins, instruments, foster local production. One of the prime functions photographic materials, and, more recently, ortho- of the Bureau of Dental Standards is to assist dontic materials, endodontic materials, and tooth- local manufacturers, many of which are not large brushes. These subcommittees also comprise the enough to support adequate laboratory staff and profession, trade, testing authorities, but also facilities. At the same time, the Bureau plays an and important part in advising importers on the qual- bring in industrial experts from the respective ! ity of their lines and ways in wiiich they could be fields. improved to meet the specified requirements. When the need for a specification is recognized, The fact that so much of investigational and a preliminary draft is prepared for the considera- testing Avork is concerned wdth overseas products tion of the appropriate subcommittee. This draft and developments has meant that Australia has is usually prepared by the Bureau of Dental maintained a keen interest in the progress on den- Standards and is based on existing overseas specifi- tal materials throughout the world; this has led cations and on its own work. When the subject has to active participation in international commit- been Avell thrashed out in committee and reviewed, tees both through the Federation Dentaire Inter- a draft is prepared for public criticism. Such nationale and the International Organization for drafts are widely circulated and are available to Standarization. This interest has been greatly all interested parties. stimulated and encouraged by personnel of the Na- After consideration of all comments received on tional Bureau of Standards and of the American the proof issued for critical review, a new draft taken. Dental Association Fellowship to whom we are is prepared and a postal ballot is finally this greatly indebted for their part in initiating inter- An effort is made to achieve unanimity at national specification programs. stage. The specification then is printed as an offi-

210 cial Australian standard, but is always subject to desire of committee members to retain some of the amendment or revision in the light of new knowl- local differences. edge and developments. The change in status of FDI specifications from The system used by the Standards Association a professional association backing to a national of Australia is modelled largely on that of the basis through ISO members should have a signifi- British Standards Institution and it appears to cant influence in dental specification j^hilosophy, work very well in British Commonwealth coun- not only in Australia, but elsewhere throughout tries particularly. the world. It now remains for the Australian specifications If any counti-y has endeavoured to emulate the to be reconsidered in the light of ISO Recommen- U.S. dental materials program of specification and dations as they appear. An attempt was made on certification, it is surely Australia and the fruits several occasions to reconcile Australian standards are apparent in the standard of dental supplies with the corresponding FDI specifications, but currently in use and in the confidence of the pro- these have not succeeded because of the expressed fession and trade in the local program.

211

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Development of European Specifications and Testing

Pierre Laplaud

L'Ecole de Chirurgie Denlaire et de Stomatologie de Paris, Paris, France

The development of a European specification program is difficult because of the many differences of race, nationality, politics, and social systems. One of the first problems is the establishment of a common dental terminology. The lack of sufficient data on the relationship of properties and clinical results is another difficulty. However, Europe has a long heritage of standardization in other fields and this has made it logical to think of the ISO as the proper organization to introduce international dental standards into Europe. Many European countries have already achieved promising results on a national basis. Cooperation on a regional or continental level in the development of a specification and certification program will help to provide better dental health for the piiblic.

Key words: Certification; dental specification; European dental specifications; FDI ISO; ; terminology, dental.

1. Introduction out difficulty. And it appeared dangerous to try to reduce them to only one; in each language, the The United. States of America seems to have a meaning of a word is, by itself, a convention. In natural propensity to be, in a large number of another country, although the term is the same, matters, first in the world. The science of dental its conventional meaning may be different. So we materials was no exception. So, we are here to have to agree, by mutual concessions, upon an in- celebrate the fiftieth anniversary of the dental re- ternational meaning, at least for important dental search at the National Bureau of Standards. The terms. Deciding this may help the countries whose joint sponsor of this celebration, the Americal Den- language is neither English nor French, to revise tal Association, has cooperated in this research for their own dental terminology, in the same way forty-one of the fifty years, and I think we all they should have revised their political terminol- agree that this was a fruitful wedding. ogy. Problems involving both terminology and Unfortunately, European weddings appear to classification belong in a specification program and be much more dilRcult. We, Europeans differ by are not simple matters either. We are confronted race, nationality, politics, and social systems. And, with just such a problem on the codification of when, by accident, we happen to believe in the dental burs and the classification of dental equip- same God, we do not worship Him tlie same Avay. ment. Moreover, I was told, that in some European coun- Apparently, there is still much work to do as ties, the dentists themselves were not very strongly concerns tei-minology, and everyone may have a united. share of it. This charming variety has made European den- 3. Philosophy tal specification programs interesting, because we had to meet nearly the same difficulties as to es- A sound and useful philosophy for international tablish a world wide plan. Firstly, we had to think dental terminology, classification, and specifica- over a terminology, a philosophy, and a policy. A tions is very difficult to establish—we are ignorant terminology for our mutual understanding, a phi- of much of the data we should know. For instance, losophy to ascertain and make our general aims experience has taught us that, wlien some physical common, and a policy to enable us to meet our properties are within certain limits, restorative ma- wide range of national and international difficul- terials, such as ones based upon zinc, silver, tin, or ties. pure gold, generally give satisfactory clinical re- 2. Terminology sults. Of course, the temptation is great to reduce every specification to these physical properties. Maybe the worst danger of our time is what we But we shall have no security whatever about the might call "babelization"—using the same word clinical behavior of such specified materials. On with different meanings. In Europe, we speak a the other hand, specifying chemical composition good number of different languages, more than may, in some cases, hamper progress. twenty. Even when these languages were restricted If we agree, for instance, upon a code number to only two, there was no true communication with- signifying that a bur is an inverted cone, we are

213 ; bound to specify the ratio between the length of With the exception of the electrotechnical j is the head and the mean diameter as well as the ratio branch, which related to CEI, all these organiza- ! between the two extreme diameters of the head. tions are very strongly linked with ISO, the In- The same concerns truncated cones. After that, ternational Standards Organization. Wlienever an should we consider as true inverted or truncated ISO Recommendation exists, and is acceptable to cones only the burs complying with the adopted the members of CEN, no further national or re- ratios ? Such a classification may come to the same gional cooperation work is initiated. inconvenience as the specification of a chemical Therefore, it was quite logical to think that ISO ' composition for any dental material. Working out was the proper organization to introduce inter- and agreeing upon a common and prospective national dental specifications into Europe. And the philosophy will certainly need wide cooperation of International Standards Organization Technical experienced specialists and learned scientists. Committee 106—Dentistry—was established. Its Secretariat was assigned to the United Kingdom, 4. A Policy for Specifications which took into account the experience of the Brit-

ish Standards Institution both in dental specifica- , It must be kept in mind that an international tions and in European standardization. Geo- specification will probably not be so easily revised graphical, historical, and intellectual conditions or amended as a national one. Once upon a time, have, in the past, made Great Britain able to serve in Washington, a senator was writing at his desk. as an abutment to various kinds of bridges estab- Suddenly, the door was flung open by a young and lished between America and Europe. Dental stand- excited secretary shouting that a Mr. Charles ardization was no exception : the nine current EDI j Lindbergh, alone in an airplane, had succeeded specifications were soon proposed as ISO Recom-

| in flying across the Atlantic Ocean, from the mendations. The EDI and ISO agreed to inform j United States to France. The disturbed senator each other of their intention to standardize such

stared calmly at his secretary and uttered : "Well, and such material, instrument, or eqviipment, thus a single man can do anything. Please call again avoiding as far as possible the duplication of work. when a commission has flown across the ocean." This agreement between the FDI and ISO, now in It looks like these days have come. Nqw, dental procedure of development, is actually a very prom-

j commissions fly across the oceans, thus making ising one. The two organizations are, I believe,

clear that we are in the era of efi'ective and more and more willing to work together. And !

worldwide cooperation in the field of dental many dentists are related to both, thus insuring a \ specificatit>ns. good conveyance of our professional ideas and If we provisionally lay aside dental specifica- needs. This might well be that fruitful European tions, international cooperation in the field of wedding we were searching for. standardization is, in Europe, a very old afi^air. Many European national standards associations 5. A Policy for Testing and Certification are more than fifty years old and celebrations of fifty years of international standardization could Another question is the testing of dental ma- have taken place during the past years. Electro- terials, instruments, or equipment according to technical international standardization began with accepted specifications. Nowadays, many Euro-

j the first years of this century. This includes long pean standards organizations have, upon a na- j established habits and rules that are far too much tional basis, their own mark of approval or j legal political certification. To these are submitted all products, implanted in the incredible and j national and imported, seeking certification. This particularisms of Europe to allow easy pacific [ work when products are exported to a rather changes. And experience proved that, in a good may limited number of countries. But this is not the number of cases, national dental federations had | case for many dental materials. And it would be a neither desire, nor power, to discuss these matters heavy charge for the manufacturers to seek certi- with sufficient force to convince national standards fication in every country. For Europe only, more organizations. tlian twenty standards organizations might be in- Moreover, these standards organizations might volved. This will probably not be tolerable for think that they were very experienced and had long. dealt with much more important matters than Europe has to work out a common program for dental standardization. They have built many certification. Perhaps on a CEN and FIDE basis, sign and European organizations : CEN, European Com- perhaps in other ways. Later on, we could other continental certifi- mittee for standardization ; COCOR which has the observe agreements with same function for the European Common Market cation programs to obtain cross examination and CENTRI, organization uniting BSI, DNA, and cross reference upon a continental basis. This AFNOR to speed up the worlc of CEN CENEL, would avoid a material being sold with the ref- ; for the coordination of electrotechnical standards, erence: "this material appears on the North dental | with CENELCOM the same for the common American—or European—list of certified i market. materials" when, in fact, the exported quality is

214 not the same, the manufacturer knowing well that might facilitate our work, and the teaching of den- his risk is nil because nobody will test his material. tists and dental hygienists. So, I would suggest that within an FDI or ISO/ Other specifications will be much more difficult TC/73/SCI schedule, continents or groups of con- to establish. For instance, it was the Swedish Del- tinents work upon this possibility of cross-testing egation to ISO/TC/106 who called attention to the and cross-referencing dental products. Anyway, need of the biological testing of dental materials. this question of continental laboratories has to be Here is a very wide and interesting issue about solved ; there is little hope that a worldwide lab- which we have not much more exploitable knowl- oratory will be convenient. Nor can it be expected edge than our forefathers. In fact, we are still that national laboratories belonging to small or de- "cementing AVooden legs to the teeth of our pa- veloping nations be efficient or economically fea- tients." There is still a lot of work to do. And sible. Europe is a good example of a continent Europe is like a mosaic, a puzzle of nations. It where there is no other possibility. The opinion of would be very surprising if each one of these na- certification specialists—particularly of the Work- tions were able to explore all the fields of science ing Group CEN/WG 42-Certification of prod- necessary to improve our specifications. Here ucts—is that such objectives will probably be again, European and even worldwide cooperation reached through a multilateral recognition of na- is necessary. tional marks indicating conformity with stand- 7. Conclusion ards. I believe that continental laboratories are of a paramount importance. Continental dental asso- In conclusion, I am obliged to say that, in the ciations could probably speed up and improve such true meaning of the term, we have then no Euro- achievements by inducing their members to exert pean specifications and testing program. But, for- pressure upon government institutions, or national tunately, many European countries have already standards associations. achieved a lot of promising results upon a national basis, and, concerning the northern countries, 6. Research, Backbone of Specification upon a regional basis. In this rather limited paper, I have cliosen not to speak of these achievements, Underlying the importance of research as the although they will certainly prove to be very use- backbone of a certification program today would ful. But, for the future of Europe, as concerns not, I am afraid, be very inventive. In fact, some dental specifications and testing, like many other very simple specifications could make our work matters, there is no better way than cooperation easier. For instance, we could decide that all the and coordination. These would be easier to attain droppers used for the liquids of our cements if national dental associations would cooperate should deliver equal drops of a standard liquid. with their national standards organizations and And we might specify that every manufacturer with government institutions while giving firm should provide a plastic or metal proportioner for support to the FDI and ISO. Organizations acting the powder, with the indication of the correspond- at tlie regional or continental level could also speed ing number of drops to obtain the standard or clin- up and improve coordination. There is no doubt ical consistency under standard hygrometric and that such cooperation will help provide what we thermometric conditions. Such a simple indication are all seeking : better dental health for the public.

215

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

Development of South American Specifications and Testing—Brazilian Experience

Leo Werner SiifTert

Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil

A successful program of specifications for dental materials requires extensive knowledge of and participation in the program on the part of teachers in dental schools. In turn the dental teachers can involve dental industry and dental societies. In Brazil a specifications program in dental materials was initiated by holding annual meetings to which persons recognized as authorities in dental materials were invited to lecture and give assistance in specification aspects. Certification testing of dental materials has been initiated in only two dental schools, but it is planned to have dental materials departments at selected dental schools responsible for three or four materials in a general certification testing program.

Key words : Brazilian experience, dental materials specifications ; dental materials ; dental

specifications ; dental specifications and testing, South American.

Experience" 1 . Introduction a brief report about our "Brazilian during the last four years, I hope this could be of During the last 35 years, in many dental meet- some benefit to other South American countries. ings throughout South America a great number of speakers talked about the importance and neces- 2. Persons and Institutions Involved in a sity of a national program of specifications for Specification Program dental materials. Sometimes theses were approved [1, 2] ^ by Gen- 2.1. The Dentist and the Dental School eral Assemblies, concerning this subject. On other occasions, specific dental materials were tested, A specification program has to involve first of according to existing ADA or FDI specifications all, the dentist. What does he think about a speci- and many articles were published, only a few of fication program ? How much does he care about which are referred to here [3, 4, 5]. a specification program ? How much does he know In Brazil, many theses with titles like "Doutora- about specifications ? mento", "Livre-Docente" or "Catedratico", from I think it is fair to state that the dentist is a If want to know teachers belonging to prosthetic dentistry or oper- product of his dental school. you about the quality of teaching of any school of den- ative dentistry departments, were completed in tistry in South America, you should analyse the dental materials departments and, in most in- curriculum, you should find out about integration stances, ADA or FDI specifications were used as a of teaching programs, you should find out about guidance. teaching facilities, but you should also be able to The effort of Degni [3] in 1949 should have evaluate the teachers themselves! How much do been the beginning of a specification program in they know about specifications and the complexi- Brazil. The effort of Pinto [4] in 1962 should also ties of a specification program ? have been the beginning of a regular specification I think that any attempt to start a specification program for dental materials in Argentina. program, in any South American country, in which Why did those two extraordinary leaders in the only a minority of the dental teachers has a fair to be a field of dental materials in South America fail, knowledge about the subject, is going with regard to a regular specification program? failure. whole teaching staff has to have, at least, Why did other people fail, each time they in- The a reasonable amount of knowledge about the prob- tended to start a specification program, in differ- lem, so that the students and later on, the dentists, ent countiies in South America ? as a consequence, will have developed the necessary The answer to these questions is precisely what mental attitude to understand and accept such a I intend to analyse and later on, when I give you specification program I This also means, of course, that the dental soci- 1 Figures in brackets indicate the Uterature references at the end of this paper. eties, consequently will easily accept and work

217 !

toward the program's acceptance. Without prepa- either supported by governments or by private ration of mental attitude there will be no accept- institutions. ance. When it comes to restorative dentistry, what As we see, our first vicious circle can only be cut, criteria do these institutions use to buy dental j by preparing the teaching staff in our dental materials? You and I know that the research , schools. program at the National Bureau of Standards , was started in 1919 to answer this question. 2.2. The Dental Dealer and the Dental Industry Especially with government-supported dental services, many programs in South America are I remember, back in 1954, in Brazil, I was criti- restricted by drastic budget cuts, and very often cized by my colleagues of the teaching staff of my as a result cheaper dental products are acquired,

own dental school, because I was introducing my unfortunately. i students through lectures, visual aids, and demon- strations to the newly developed elastomeric im- 2.5. The Government pression materials. They reasoned: "Why teach it? Those materials are not available from the I agree that the government of any country has dental dealers in Brazil." to participate in a specification program for dental , materials. it is My colleagues agreed with me, however, when I But on the other hand, I think ,

answered : "The dental dealers do not carry these unwise to think that the government of any South , products because there is no demand for them on American country should start such a program. the part of the dentist. There is no demand on the In most instances there are not enough funds part of the dentist, because we teachers, at the den- available to finance adequately the national stand- !" tal school, do hot mention the product ards institutions ! : , This is another vicious circle that very often There are still so many other important and happens in American dental schools. basic problems to be solved j South by governments of i Through the dental schools, the dental dealer, South American countries that I think we do not and especially the dental industry, will get in- have the right to expect them to start a specifica- , volved in a specification program. tion program for dental materials. The government must get involved and cooperate, but is not 2.3. The Dental Societies supposed to start, the program As we can see, specifications committees and sub-

, It used to be a policy of past-Presidents of the committees of national standards institutions need Brazilian Dental Association, to nominate a Com- | the active participation of dental schools, dental , mittee of Dental Materials: a President, a Secre- materials groups, the dental profession through , tary and usually two or three members. state and national dental associations, the associa- j ( For many years the by-laws of the Brazilian tions of dental research, the national and Latin-

| | Dental Association contained reference to specifi- American associations of dental schools, and the

\ \ cations and a specification program. The simple governments. nomination of a , Committee of Dental Materials, To be successful, any specification program for however, to the ^ used be beginning and the end of dental materials needs the active participation, the j a specification program ! I think this is true for cooperation, of the whole group of persons and , many South American dental societies. The mental institutions involved ! attitude towards acceptance of such a program, (

] was not prej^a^red. all Not even the dentists realized Specifica- 3. Brazilian Experience with a ( its importance. Materials for Dental | At the same time as the Committee was nomi- tion Program nated, new dental schools were created, with teach- In Brazil, as in any other South American ( ing staffs, in most instances, not well prepared for country, we have an excellent starting point : the i teaching. Yet those teachers, would somehow pre- tremendous amount of excellent work and experi- s pare dentists who, of course would join a dental ence accumulated during the 50 years of your ] society. I think, around 3,500 dentists graduate National Bureau of Standards; the excellence of each year from the approximately 70 dental schools postgraduate courses offered by many dental ma- i in South America. in the United States; all the terials departments ] To get the dental societies involved in a specifi- information received through your research as- cation program, we have to start the work, seri- j sociations and institutes ; the American Dental As- : ( ously, with all the dental students. offered to sociation; the excellent opportunities ;: ^ Latin American dental teachers, through fellow- 2.4. The Public Health Dental Services ^ ships sponsored by foundations such as the W. K. It is unquestionable that dentistry, all over the Kellogg Foundation. _ ( world, is seeking new methods, in an effort to ex- We also based the starting of a specification pro- ^ the experience of other insti- ' tend its service to more and more people. This gram in Brazil, on j occurs through Dental Public Health Programs, tutions, such as the Commonwealth Bureau of |

218 !

Dental Standards in Australia (22 years), the of amalgam alloys manufactured in Brazil. This operative BSI, the DNA (Deutscher Normenausschuss) , the information was sent to all teachers of AFNOR (Association Francaise de Normalisa- dentistiy in Brazil, to all the deans of dental tion), ISO-TC/106; FDI; the experience of ex- schools, "to all dental associations in Brazil, to most cellent researchers of Switzerland, Sweden, Den- dental government-supported services and to the mark, and Japan, among others, who were always dental industry. Our guest lecturer was Dr. Guil- very cooperative, sending us information. lermo INIac Pherson, from the dental school of Santiago, Universidad de Chile. Thanks to Drs. 3.1. The Brazilian Dental Materials Group Mac Pherson and Maddalena, our group has a good number of members from Argentina and especially Forty-two out of the approximately seventy den- Chile. tal schools in South America are located in Brazil. We divided Brazil into five regions, each hav- We felt that it was worthwhile to try to start a ing a coordinator who is responsible for special specification program through the dental schools activities related to dental materials in his I'egion. in Brazil. The dental school, we think, is the only We have one meeting each year and the meeting place in which we can prepare the mental attitude places are chosen two years in advance. of the futiire dentist toward the acceptance of a The fourth meeting was in Fortaleza, Ceara and specifications testing program. To achieve this with we had as our special guest lectm^er Dr. George the 2,100 students who graduate each year from the C. Paffenbarger who presented to us an excellent dental schools in Brazil, we need first of all the course in dental materials. Again we had planned understanding, the acceptance, and cooperation of group discussions, related to three main topics: all the dental teachers. The starting point, of (1) Philosophy of the teaching of dental mate- course, is the teacher of dental materials, partici- rials; (2) Integration; and (3) Teacher training pating actively in the Brazilian Dental Materials in the specialized field of dental materials : teach- Group. ing and research programs. We started the Group (GBMD,Grupo Brasileiro There were over sixty teachers present, an excel- de Materiais Dentarios) in September, 1965, when lent attendance, if you consider the great distance we had our first meeting in Porto Alegre, in the of Fortaleza (in the very northern part of the State of Rio Grande do Sul. Distances are great in country) to Central and Southern Brazil, where Brazil, and we needed from the beginning the ac- most of the dental schools are. Dr. Paffenbarger tive participation of as many dental materials has also been a great help to the Brazilian Dental teachers as possible. For our first meeting, we used, Materials Group and even though, due to his many as you might call it, a very good "bait''. To deliver assignments at the time, he could not accept our a course in dental materials and to present some invitation to visit the dental schools in the central conferences, we were fortunate to have with us at and southern parts of the country, we believe he the time, the Avorld-known authority in dental ma- has a very good idea of what we are trying to terials. Dr. Floyd A. Peyton, from the University accomplish. Through Dr. Paffenbarger we also of Michigan. Besides Dr. Peyton's course, we had wanted FDI to know a little bit more about our group discussions on the following subjects: (1) activities in Brazil.

Dental materials in the dental curriculum : lectures Our fifth meeting was held this year, in July, and laboratory course contents; (2) Integration of in Bauru, State of Sao Paulo. I think it was an dental materials with clinical departments; (3) excellent meeting too, and I am sure Dr. John W. Research in dental materials departments; The (4) Stanford could tell you more about it than I ; he establishment of a Brazilian dental materials was our special guest lecturer. We appreciated group and the adoption of specifications. very much his course and his conferences and I Over one hundred, most of Avhom were teachers would be tempted to say that he probably returned of dental materials, attended the course and the with a better impression of our group than our meeting. Thirty-one dental schools were repre- previous guests: the dental materials teacher is sented, out of a total of 38 then in existence, in now, I would say, much better prepared than he Brazil was four years ago. We notice the great difference, For our second meeting, in 1966, in Belo Hori- just reading the conclusions of the group dis- zonte, State of Minas Gerais, we had as our guest cussions ! lecturer, Dr. Hector Maddalena, Head of the Den- We make good use of the many excellent articles tal Materials Department in Buenos Aix-es, Argen- published in the Journal of Dental Education, tina. Again we had a good meeting, with group especially the ones related to teachers and training discussions related to teaching and research in programs like the Workshop on Teacher dental materials and specifications. Education [6]. Our third meeting, in 1967, was held in Piraci- Our staff of the dental school in Porto Alegi-e caba. State of Sao Paulo, and we had over 80 mem- presented to the group the results of the surveys bers of our Dental Materials Group present. At related to two additional dental materials nmnu- that opportunity Prof. Degni and staff presented factured in Brazil: investments for gold alloys to the group the results of the controlled testing and alginate impression materials.

452-525 O—72 15 219 ! :

is recognized I am certain that the dental materials teacher Our Group by the Brazilian Stand- i in Brazil is now much better prepared to exert his ards Association (ABNT) as their Committee of inflvience in preparing the mental attitude of tlie Dental Materials. other teachers of his dental school and, through We want the Government to enter the picture, them, of the dental students toward acceptance of but only later, when we have some concrete accom- a specification program. plishments to show ! '. To finish my presentation, I would like all of We are also noticing the greater interest and f greater cooperation of the dental industry in Bra- you to know that we of the Brazilian Dental Ma- zil toward our specification program. terials Group are now quite aware of the fact that So far, tests with regard to what you might call the enormous prestige of your National Bureau of a "Certification Program of Materials" in Brazil Standards, of many of your excellent dental ma- were realized only in the Dental School of the terials departments, of your dental research insti- University of Sao Paulo and in the Dental School tutions, this enormous prestige, I repeat, was of Porto Alegre, University of Rio Grande Do Sul. neither easily, nor casually acquired The dental materials departments of those two 4. References dental schools are generally considered training materials. In Sao Paulo, espe- centers for dental [1] Degni, F., Especificacoes e pesquiza, Rev. Bras. cially under the leadership of Dioracy Vieira and Odont. 3, 19-23 (Ago 1945). [ Francisco Degni, we have accomplished quite a lot. [2] Acuna, R. O., Oficina Chilena de normas para materiales dentales, Odont. Chilena 15, 11-15 Several other dental materials departments (Mar- Abr 1966). from other dental schools in Brazil are being pre- [3] Degni, F., and Pomes, C, Amalgamas Latino- Americanas, Rev. 13-24 (,Jan-Feb pared to participate actively in the specification APCD 2, 1949). Pinto, F. E., Amalgamas Dentales Argentinas, Rev. chosen dental [4] program. The idea is to have those Odonto. Arg 50, 253-259 (.Jul 1962). materials departments responsible for, say two, [5] Vieira, D. F., Pastas Zincoenolicas, Ann. Fac. Farm. three, or four specific materials, in the general Odont. S. Paulo, 10. 329-347 (1952). [6] Workshop on Teacher Education. Conclusiones testing results are distributed program. The Teacher training and education programs, J. Dent. through the Brazilian Dental Materials Group. Education 30, 54^55 (Mar. 1966).

220 :

NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972).

USA Specification and Evaluation Programs

John W. Stanford

American Dental Association, Chicago, 111. 60611

Reliable and valid laboratory tests which delineate satisfactory materials and rule out unsatisfactory ones are essential to the development of a satisfactory specifications program. Factors guiding the American Dental Association in the formulation of specifica-

tions are : Relevent tests ; correlation of laboratory, structural, and service tests ; and quality of product. An important change in organization for tlie development of dental specifications involves the formation of the USA Standards Committee for Dentistry, operating under the United States of America Standards Institute. The Certification Program provides a means for recognition of those materials complying with existing specifications. A new evaluation program on dental devices is currently issuing pertinent status reports on the safety and efficiency of such devices. Safety and usefulness are ba.sed primarily upon information established under conditions of clinical use.

Key words : Dental materials and devices, specifications for and certification of ; dental

specifications ; specifications ; standards, dental.

1. Preparation in radia- 1 . Introduction of recommendations tion hygiene and practice. Programs for the standardization and evalixa- 2. Formulation of status reports on materials tion of dental materials have been available to the and devices. profession in the United States for many years. 3. Revision of the Association's publication In fact it is well known that the basis for estab- Guide to Dental MateHals and Devices. 4. on advertising and exhibit stand- lishing dental research in 1919 at the National Guidance ards I'elated to materials and devices. Bureau of Standards was the need for a laboratory 5. Development and evaluation of new mate- standard for dental amalgam. This was the first rials and devices or the improvement of attempt to define a material by means of physical existing ones. and chemical properties. Until 1966 these programs were conducted by Before the program was instituted at the Na- tional Bureau of Standards, most evaluations of the Association's Council on Dental Research filling materials Avere based on service tests in through its Research Division at the National the mouth. Such tests were long draAvn out, ex- Bureau of Standards. The specification and cer- pensive, usually poorly controlled and, hence, tification programs for materials, along with frequently inconclusive. Materials must be tested other responsibilities, were transferred in 1966 to in the laboratory j^rior to any widespread use. a new Council on Dental Materials and Devices. The laboratory tests should be desigiied so that The formation of the new Council elevated pro- they will delineate satisfactory materials and rule grams in dental materials and defaces to a level out unsatisfactory ones. This terse dictum is dif- within the Association commensurate with their ficult to carry out because the design of tests that importance. will accomplish it requires considerable imagina- tion and the necessary research to prove that the The Council has the following By laws : tests are reliable and valid. By reliable it is meant To determine the safety and effectiveness of, that the tests can be repeated by different labora- and disseminate information on, materials and tories with satisfactory agreement in results. By devices which arc ofi'ered to the public or to the valid it is meant that the laboratory testing will profession. predict the behavior of the materials in service. The American Dental Association believes that To encourage the development and improve- only the highest quality materials obtainable ment of materials and devices for use in dental should be used and does not consider the price at practice or to improve the oral health of the is criterion of its public. which a product sold as any quality. The Association believes that this phi- Some of its duties include the following in addi- losophy is sound and dependable from a public tion to standardization and evaluation programs health standpoint, and the current specifications

221 : reflect this policy. Therefore, in designing a speci- 3. U.S.A. Standards Committee for Dental fication one has the following in mind : Materials and Devices 1. Relevant tests

2. Correlation of laboratory, structural, and Of special interest here also is the formation of service tests a USA Standards Committee for Dental Materials 3. Quality of the product and Devices. The recommendation for such a com- These factors have gaiided the Specificatioiis Com- mittee originated in the dental trade. The recom- mittee of the Dental Materials Gi'oup of the mendation as presented calls for the phasing out lADR, which acted as principal consultant to the of the duties of the Specifications Committee of American Dental Association in the formulation the Dental Materials Group of the lADR and a of specifications or standards during the period transfer of its duties to a new committee which 1953 until 1969. Some of its recent programs have would operate under the United States of America dealt with formulating specifications or standards Standards Institute (USASI). During recent for the following years the United States has become involved in an 1. Dental diamond rotary instruments international standardization program through 2. Dental excavating burs USASI.^ With formation of various committees, 3. Dental radiographic film subcommittees, and working groups there has been 4. Direct filling resins some duplication of effort and, in some instances, 5. Dui^licating materials mass confusion about the relations of all the orga- 6. Endodontic files, reamers, and points nizations involved. The new committee will be 7. Gypsum materials called the USA Standards Committee for Dental 8. Toxicity tests Materials and Devices, and all other committees 9. Zinc silico-phosphate cements will, as such, be disbanded. The American Dental 10. Base plate wax Association will act as the administrative sponsor 11. Orthodontic wire not containing noble of the committee through the Council on Dental metals Materials and Devices and will also handle the The Specification for duplicating materials be- duties of secretary. It is believed that the Associa- came official on November 1, 1968, and was desig- tion as sponsor of and the dental industry as a nated American Dental Association Specification contributor to such a committee will place orga- No. 20. Final approval of s]3ecifications for items nized dentistry in a favorable position m the event 2, 3, and 9 were given in May and June of 1969. federal legislation in the dental device area is passed. 2. Specifications and Biological Toxicity 4. The Certification Program for The formulation of a specification dealing with Dental Materials toxicity tests deserves discussion. In the past, specifications have been designed to evaluate mate- The Certification Program forms the second part rials on the basis of chemical composition and of the program for materials, after specifications physical and mechanical properties determined in have been adopted. Under this program the manu- the laboratory. These tests have been designed to facturer of a dental material certifies that his prod- simulate actual service conditions as closely as uct complies with the specifications AVhich have possible. The degree of the validity of the labora- been approved as official Sjyecifications of the tory test is determined by the degree of correlation American Dental Association and that he is in between data obtained from laboratory tests and compliance with the American Dental Association the behavior of the materials in service. The safety Advertising and Exhihit Standards. If the prod- of the materials for clinical use has in the past uct is found to comply Avith the specification, its been largely based on the fact that no serious com- name is then placed on the List of Certified Dental plications arose from clinical use. In other words, Ma.terials which is maintained by the Council and no screening tests Avere available for new mate- published periodically in Tlie Journal of the rials before they were placed in clinical use. American Dental Association. With the knowledge that new materials were From time to time, materials on the List are and are being made available to the profession at tested in the laboratories of the American Dental an increasing rate, the Specifications Committee recommended and the Association agreed that a Association. When a product is found not to com- subcommittee should be established to prepare rec- ply Avith official specifications, the manufacturer ommended sci'eening test pi'ocedures to be used in is notified, and the product is removed from the the laboratory. The subcommittee is progressing List. No manufacturer may claim paii:icii)ation in in the development of such recommendations for the Certification Program except under authoriza- guidelines to biological testing. These guidelines tion granted by the Council on Dental Materials will be of special assistance to both the maniifac- and DeAdces of the Association. The program has turer and the Association in the event of federal 2 Name later changed to American National Standard Institute legislation regulating medical devices. (ANSI).

222 : : been increasingly accepted during recent years. acquaint the dentist with the extant Imowledge. It Tables 1 and 2 show the growtli in the number of is planned that other evaluation procedures, such specifications and in the number of products on the as specification and certification or acceptance pro- List. grams, will be utilized in the future. No one type of program will apply to all devices, and, to be Table 1. of greater value to the profession, the program should include an evaluation of the usefulness and List of Certified Dental Materials effectiveness of the devices from both a clinical and a laboratory standpoint. Year Number The Council has assigned priority for evaluation of the following classes of devices 1929 40 1. Instruments for surgical cutting 1932 112 2. Devices for removing calculus 1934 145 3. X-ray machines 1936 187 4. Oral hygiene devices 1938 205 1940 225 5. Desensitizers 1942 222 6. Pulp testers 1945 245 The first application of the acceptance 1948 270 program 1951 285 is in the oral hygiene device area. The Council 1954 315 classifies an evaluated device as Acceptahle., as Pro- 1956 305 visionctUy Acceftahle or as Unacceptable under 1958 325 this 1960 350 program. Evaluation of safety and usefulness 1962 345 of a device is based primarily upon information 1964 372 established under conditions of clinical use and not 427 1967_ physical standards or specifications. Electric tooth- 1969 475 brushes (formerly a device evaluated by the Coun- cil on Dental Therapeutics) and oral irrigating Number of Manufacturers or Distributors devices are the first devices being rated by the Council undei' the Classification System which Year Number follows CLASSIFICATION SYSTEM 1969 105

1. Acceptance Program: The Acceptance Pro- gram applies to devices for which evidence of

Table 2. New Certifications safety and usefulness has been primarily established under conditions of use or to devices for During the period of 1952 to 1964 there were approxi- which physical standards or specifications do mately 120 new certifications. The following is a breakdown of results: not currently exist. Powered toothbrushes are an example of this type of device. Devices ac- Materials 120 cepted under this program shall be classified as Complied 110 described in the following section. Failed 10 or about 8% A. Classification of devices evaluated hy the Council: After consideration of a device has Compare this to surveys of certified products: been completed under the provisions of the 'Acceptance Program", the Council will Surveys 955 Failed 5 or about 0.5% classify the device as Acceptahle^ as Pro- visionally Acceptable or as Unacceptable. Devices will usually be classified as Accept- 5. Evaluation of Dental Devices able for a period of three years. Acceptance is renewable and may be reconsidered at any of tlie A new program Association concerns the time. If manufacturing ownership of the evaluation of dental devices, with priority given device changes, the period of acceptance ex- to those devices which directly relate to the health pires automatically. and safety of the patient, the dentist, and his auxiliary personnel. The program, to date, has con- 1. Acceptable devices will be listed in Guide sisted of initiation of a series of status reports on to Dental Materials and Devices and the the safety and efficacy of devices. These reports manufacturer or distributor may use an summarize the existing information and may or authorized statement as specified in Sec- may not arrive at certain specific conclusions, tion V, "Announcement and Maintenance depending upon the amount and quality of the of Acceptance or Certification", of these available evidence. At least the reports serve to provisions.

223 2. Provisionally Acceptable devices consist Another factor which should not be dismissed j of those which lack sufficient evidence to lightly is the sale of many quack devices in the } justify classification as Acceptable^ but medical field to the elderly of the country. The for which there is reasonable evidence of Food and Drug Administration currently has au- safety usefulness including and clinical thority to consider the safety of devices. These are feasibility. These devices meet the other defined as: "Instruments, apparatus, and con- qualifications established by the Council ' trivances, including their components, parts, and on Dental Materials and Devices. The accessories, intended (1) for use in the diagnosis, Council may authorize the use of a suit- . care, mitigation, treatment, or prevention of dis- able statement to define specifically the ! ease in man or other animals; area of usefulness of a device classified as or (2) to affect the structure Provisionally Acceptable. Classification or any function of the body of man or other animals." in this category is reviewed each year and is not ordinarily continued for more than Even restorative materials used in dentistry are three years. included in this definition of devices. If new legis- 3. Unacceptable devices are those which are lation similar to that Avliich has been introduced dangerous to the health of the user, ob- in the past two sessions of Congress is enacted,

| solete, markedly inferior, or useless. These the Food and Drug Administration will be given j devices do not meet the standards out- additional authority to consider the safety and

| lined in the "Provisions for Evaluation efficacy of devices and to regulate certain market- of Dental Devices." ing practices. The extent to which FDA regula- Now, one may ask why the Association has tions are applied to dental devices, including mate- ! entered the device field. There are several reasons. rials, may depend in part upon the effectiveness

First, no agency outside of the federal govern- | of an existing professional program. ment was conducting any type of evaluation of } One of the main reasons the evaluation program complex devices in relation to safety of the patient, for materials of the Association has been success- the dentist, and auxiliary personnel. The dentist ful over the last at present has no information to use in advising 35 years has been the increasing his patients on devices for use at home. The suc- cooperation of the manufacturers in standardiza- cess of the Association's programs in materials in- tion and improvement of their materials. The As- dicated that research in improvement and devel- sociation especially desires this cooperation and as- opment of new equipment could be hastened by sistance from all in the field of therapeutic agents,

\ the new program. dental materials, and devices in the years ahead.

224 VIII. Appendix. NBS Dental Research Section Personnel !

I NATIONAL BUREAU OF STANDARDS SPECIAL PUBLICATION 354, Dental Materials Research, Proceedings of the 50th Anniversary Symposium, Held Oct. 6-8, 1969, Gaithersburg, Md. (Issued June 1972)

Personnel of the Dental Research Section of the National Bureau of Standards

The Dental Kesearch Section of the National Although the present activities of many of our Bureau of Standards has served not only as a re- former personnel are not known, at least ninety search laboratory but also as a school for dental of those listed in table 1 are still in the field of researchers. Ahimni of the Section are widely dentistry. Thirty-five remain in the Dental Re- scattered in research facilities and educational in- search Section; thirty-seven are in the Federal stitutions throughout the United States and in dental services or in dental schools, and seventeen other countries. In the following tables on per- are in dental practice, in the American Dental sonnel of the Dental Research Section an attempt Association research facility in Chicago or in has been made to include everyone Avho has worked other dental activity. in the Section for six months or longer. Because Dr. Souder, the group, the tables cover a relatively long period of time Wilmer founder of our and since some of the personnel records are incom- has been retired for some time but is hail and plete or unavailable, there are undoubtedly omis- hardy, living in Landisville, Pemisylvania. Dr. sions and inaccuracies. Greorge C. Paffenbarger who first came to the Den- During the past 50 years about 175 people have tal Research Section over 40 years ago is still been members of the staff of the NBS Dental Re- working in the Section on a part-time basis, al- search Section, table 1. American Dental Associa- though theoretically retired. Dr. Irl C. Schoon- tion research associates numbered 62; there were over, one of our four Section Chiefs, became 61 military or Veterans Administration person- Deputy Director of the National Bureau of Stand- nel; 31 were Civil Service employees; and 25 ards, the second in command, before retiring from others were research associates of Weinstein Re- Government service in early 1969. Mr. William T. search Laboratories, guest workere from various Sweeney, another former Section Chief and a laboratories, or summer students. A summary is Government service retiree in 1968, is now work- given in table 2. Several members were in more ing as an associate i^rofessor at the University of than one category and because of this the total is more than 175. Alabama. Dr. John W. Stanford has become Di- Seventeen members of the staff have been in rector of the Division of Biophysics of the Amer- the Section ten to 40 years, while 23 others were ican Dental Association. Dr. Harvey W. Lyon is in the Section from five to ten years. Nearly 90 the Director of the Division of Clinical Studies staff members were military personnel, Veterans for the ADA. Dr. A. F. Forziati is Acting As- Administration employees, and other guest sistant Director of Physical Sciences, Division of Avorkers, who were stationed in the Section for pe- Water Quality Research, Federal Water Pollution riods from a few months to as long as six years. Control Administration. Dr. Robert J. Nelsen is Forty-one of these were dentists and 22 were chem- the Executive Secretary of the American College ists. Eleven of the guest workers were from outside of Dentists. Mr. Denton L. Smith is Director of the United States. Research for The J. M. Ney Company. This posi- An indication of the variety of backgrounds is tion was previously held by Mr. Richard L. Cole- shown in table 3 which lists personnel by profes- man, now retired, one of the early research asso- sion. ciates in the Dental Research Section,

227 - -. - - .-

Table 1. Personnel of the Dental Research Section, NBS, 1919-1969

Name Affiliation Profession Tenure Present location or employment

Adler, Mr. Alfred G Army Metallurgist. 1954-1956... New York. Arevjev, Mr. Vsevolod Army 1958-1959... Argentar, Mr. Harold Army-AD A. Chemist 1962-present- NBS Dental Research. Austin, Mr. Clarence ADA Metallurgist. 1960-1961...- Redstone Arsenal, Ala. AvNiMELECH, Dr. Yoram ADA Chemist 1965-1967- __ Israel Institute of Tech- nology, Haifa.

Balekjian, Dr. Aran ADA. Chemist. 1964-1965- Naval Medical Research Institute. Barber, Dr. Ronald Navy Dentist 1934-1936-.- Retired, Boston, Mass. Barone, Dr. Joseph J Army Dentist 1957-1960--- Army, Fort Sill, Okla. Barton, Dr. John A Air Force. Dentist 1965-present- NBS Dental Research. Beall, Mr. John R ADA Mechanical 1934-1947--- Army Med. R&D Com- engineer mand, Washington, D.C. Beebe, Dr. Douglas M Army_ Dentist 1953-1954. Deceased. Berger, Dr. Howard S ADA_ Dentist 1928-1929- Private practice. Falls Church, Va. Booth, Miss Esther ADA_ Typist.. 1965-1969... Washington, D.C. Bowen, Mrs. Joy S ADA. Chemist. 1968-present. NBS Dental Research. BowEN, Dr. Rafael- L ADA. Dentist. 1956-present. NBS Dental Research. Brauer, Dr. Frank J Navy. Dentist. 1955-1957--- Private practice Wautoam, Wisconsin. Brauer, Dr. Gerhard M NBS._ Chemist. 1950-present- NBS Dental Research. Brown, Mr. James J NBS.. Student 1958-1959... Washington, D.C. trainee Brown, Dr. Walter E ADA. Chemist. 1962-present- NBS Dental Research. Burns, Miss Claire L NBS. Chemist. 1951-present. NBS Dental Research. Burns, Mr. Francis R NBS. Chemist.. 1954-1956--- Chicago, Illinois. Bush, Mrs. Jesse W NBS. Secretary. 1948-1953--. Washington, D.C.

Carlson, Mr. Elmer T ADA. Chemist 1968-present- NBS Dental Research. Carlson, Mr. John C Navy. Technician. 1958 Cassel, Dr. James M NBS._ Chemist 1969-present. NBS Dental Research. Caul, Mr. Harold J ADA. Chemical 1937-present- NBS Dental Research. engineer Chandler, Dr. Harry ADA Dentist 1968-present- NBS Dental Research. Civjan, Dr. Simon Army Dentist 1961-1963... Walter Reed Hospital. Coleman, Mr. Richard L Weinstein Engineer. _ 1922-1928... Retired, West Hartford Research Connecticut. Laboratory Copeland, Dr. Henry I Air Force Dentist 1953-1956- Andrews AFB. Crist, Dr. Ray F Summer stu- Stud, techn. 1962-1963. Navy, Garden Grove, dent California.

Davenport, Mrs. Ruth M NBS-. Technician. 1950-present. NBS Dental Research. Demaree, Dr. Neil C Navy. Dentist 1959-1961--- Navy, Camp Pendelton, California. Denton, Miss Glenna M NBS Secretary. 1966-present NBS Dental Research. Dickens, Dr. Brian NBS Chemist-. 1966-present NBS Dental Research. Dickson, Mr. George NBS Physicist. 1940-present NBS Dental Research. DioRio, Mr. Alfred F ADA Chemist-. 1956-19.59-- Georgetown Univ. Hosp. Driessens, Dr Ferdinand C. N_ Guest worker. Chemist- 1968-1969.- KathoHeke University, Nijmegen, Netherlands. Duffey, Mr. Depue H ADA Chemist- 1961- 1963.. Georgetown Univ., Med. Durany, Mr. George Army-AD A.. Chemist- 1962- 1966.. ADA, Chicago, Illinois.

Eden, Dr. George T Navy. Dentist 1964-1965.. Navy, Univ. of Michigan. EicK, Mr. John D ADA. Mathematician 1963- 1966.- State Univ. of New York at Buffalo. El Sadr, Mrs. Betty Lee ADA. SecretarJ^ 1963-1965. Arlington, Virginia. Epstein, Mr. Earl F ADA. Chemist-. 1963-1964- Univ. of Wisconsin.

Fanning, Mrs. Rachel J ADA Chemist 1953-1955- University Park, Md. Farris, Dr. Lovell L Summer Student 1965, 1966- Private practice, Eaton- student. technician. town, New Jersey. Fee, Mrs. Jean G NBS Chemist 1951-1954- Philadelphia, Penna. Ferguson, Dr. George W Navy Dentist 1946-1949-- State Univ. of New York at Buffalo. Fischer, Dr. Theodore E Army & Air Dentist 1947-1951- Univ. of Alabama. Force Forsyth, Mr. John W NBS Technician 1953-1955- Howard University, Washington, D.C.

228 - .- - . .

Table 1. Personnel of the Dental Research Section, NB8, J9i9-JS69—Continued

Name Affiliation Profession Tenure Present location or employment

FoRziATi, Dr. Alfonse F ADA. Chemist- 1950-1962- Dept. of the Interior, Washington, D.C.

Gardner, Mr. Alvin F Guest worker. Chemist. 1958-1959- Gardner, Dr. Thomas V Army Dentist- 1962-1965. Army, Fort Sill, Okla. Glasson, Dr. Gilbert F Air Force & Dentist- 1947-1952. Private practice, Waterloo, ADA Iowa. Gregory, Mr. Thomas M ADA Chemist- 1963-present. NBS Dental Research. Grunewald, Dr. Alvin H Navy Dentist- 1947-1949--- Northwestern University.

Hansel, Mr. Grant, Jr Army 1956- 1957- Hansen, Mr. William C NBS Physicist. 1951- 1956. Washington, D.C. Hartley, Dr. Jack L Air Force Dentist - 1951- 1955. Consultant, San Antonio. Harvey, Mr. Jack L Army Chemist-. 1949. NBS (Tire systems). Hawkins, Mr. Norman D., NBS Physicist 1932. Deceased. Hegdahl, Mr. Trond M Guest worker. Dentist - - 1965- University of Bergen Oslo, Norway. HicHO, Mr. George E Army Metallurgist - 1962-1963. NBS (Engineering metallurgy) Hidnert, Dr. Peter NBS & ADA._. Physicist. 1957- 1960- Deceased. Hobson, Dr. Robert W Army Dentist - _ 1952- 1953- Army, Fitzsimons Gen. Hosp., Denver, Colorado. Howe, Mr. Willard B Army Chemist- 1957- 1959- Hudson, Dr. Donald C Air Force. Dentist- 1950- 1956- Univ. of Texas, Houston. Huff, Mr. Richard L Army Chemist- 1957- 1959- HuGET, Dr. Eugene F Army Dentist 1965- 1968. Walter Reed Hospital, Washington, D.C.

Isaacs, Mr. Aaron NBS Chemist- 1930-1935- Deceased.

J0RGENSEN, Prof. Kuud Dreyer_ _ Guest worker. Dentist- 1957-1958- Royal Dental College, Copenhagen, Denmark.

King, Dr. Richard_ Summer Student 1958, 1959- Private practice, California. student technician Kingsbury, Mrs. Pamela ADA Technician - _ 1967-present- NBS Dental Research. Krogh-Poulson, Dr. Willy G_ Guest worker Dentist 1947 Royal Dental College, Copenhagen, Denmark. KuMPULA, Mr. John W NBS Mech. Engr. 1952-present- NBS Dental Research. technician KuMPULA, Mrs. Marion P ADA Secretary 1952-present- NBS Dental Research.

Laueila, Dr. Riita ADA Chemist 1958- 1960. Paris, France. Lawson, Mr. Melvin E ADA Technician - - 1950- 1957. Washington, D.C. Led LEY, Dr. Robert S Army Dentist 1951- 1952- Hyattsville, Maryland. Lehmann, Mr. Frank H Army Chemist 1959- 1960. New York, N.Y.' Leibfritz, Mr. Walter A Army Metallurgist- 1953- 1954- Chicago, Illinois. Leussing, Dr. Daniel L ADA Chemist 1962. Ohio State University. Loebenstein, Dr. William V_. NBS Chemist 1964- present- NBS Dental Research. Longton, Dr. Robert W Navy Dentist- 1961- 1963--- Navy, Evanston, Illinois. Lyon, Dr. Harvey W Navy Dentist 1949- 1951--- ADA, Chicago, Illinois.

Mabie, Mr. Curtis P ADA Petrographer. 1968-present. NBS Dental Research. Macasaet, Dr. Avelino A Guest worker. Dentist 1960-1962--- Univ. of the Philippines, Manila, Philippines. Malmstedt, Mrs. Margaret ADA Secretary 1968-present. NBS Dental Research. Manceamcz, Miss Sandra A. ADA Chemist 1958-1965--- Evanston, Illinois. (Mrs. John Hefferren). Manuszewski, Mr. Richard C ADA Metallurgist 1966-present- NBS Dental Research. Margetis, Dr. Peter M Army Dentist 1953-1956--. Deceased. McConnell, Mr. Robert M._ Army Technician - - 1954-1956--- ADA, Chicago, Illinois. McDowell, Dr. Hershel ADA Chemist 1963-present_ NBS Dental Research. McLaughlin, Mr. Richard D. Summer Student tech- 1966, 1967, Georgetown University, student nician 1968. Washington, D.C. Merchant, Dr. Hubert W Army Dentist 1956-1957--- Emory Univ., Atlanta. Miller, Dr. William A. C Air Force Dentist 1955- 1958--- State University of N.Y. at Buffalo. Mitchell, Dr. James A Navy Dentist 1951-1954- Private Practice, Opelousas, Louisiana. Moore, Mr. Clarence Army Chemist 1959-1961--- Moore, Mr. Robert E Armv Chemist 1953-1955-.- Cape Kennedy Florida. Moreno, Dr. Edgard C ADA Chemist •- 1963-present- NBS Dental Research.

229 . -- - . .

Table 1. Personnel of the Dental Research Section, NBS, 1919-1969—Continued

Name Affiliation Profession Tenure Present location or employment

Morris, Mr. Richard W Army Chemist- 1960- 1962- Morris, Mr. Stephen Army 1961- 1962. MosHONAs, Mr. Manual G.-_ Army Chemist. 1956-1957. Florida. MowERY, Dr. William E Veterans Dentist 1950-1952. Veterans Administration, Administra- Dayton, Ohio. tion MuLLiNEAUx, Mrs. Anna G_ _ ADA Dental Asst. 1958-present- NBS Dental Research. MuLZET, Dr. Alfred P NBS Mech. Engr. 1966-1968- __ IBM, New York.

Nelsen, Dr. Robert J ADA. Dentist 1950-1955. American College of Dentists, St. Louis, Mo. Nelson, Mrs. Barbara ADA. Technician. 1957-1958.

Ofstead, Mr. Eilert A ADA Technician 1957-1958--- Oglesby, Mr. Philip L NBS Physicist 1957-present- NBS Dental Research. Ohashi, Dr. Masayoshi Guest worker. Physical Science. 1962-1965--- Nihon University, Tokyo, Japan. Oliver, Mr. Jerry A Army Chemist 1964 Oppenheim, Mr. William L Summer Student 1963, 1964, Georgetown University, student Technician 1965, 1967. Washington, D.C, Overberger, Dr. James E Army Dentist 1954-1956--. Univ. of West Virginia.

Paffenbarger, Dr. George C ADA. Dentist. 1929-present. Retired, NBS Dental Research. Palcic, Miss Julia C ADA. Chemist. 1967-present. NBS Dental Research. Parestzkin, Mr. Boris ADA. Chemist. 1958-1961.-- NBS (Inorganic Mat.). Patel, Dr. Prafull R ADA. Chemist. 1965-present. NBS Dental Research. Peiperl, Mrs. Martha D ADA. Chemist- 1957-1958; Silver Spring, Maryland. 1962-1963. Perloff, Dr. Alvin Army Chemist. 1954-1956... NBS (Crystallography). Perkins, Dr. Robert R Navy Dentist. 1957-1959... Navy, Norfolk, Va. Peterson, Dr. Gert Forum. __ Guest worker. Dentist- 1961-1962... Copenhagen, Denmark. Pfeiffer, Dr. Kenneth R Veterans Ad- Dentist- 1950 Deceased. ministration Piermarini, Mr. Gaspar J Army Chemist 1957-1958- NBS (Crystallography) PiNcocK, Mr. Douglas G Summer Student tech- 1966-1967. Temple University, student nician Philadelphia, Pa. POPPE, Mr. W. A Weinstein Engineer 192.5-1928. Washington, D.C. Res. Lab. PosNER, Dr. Aaron S ADA Chemist. 1950-1961- Hospital for Special Surgery, New York.

Roberts, Mrs. Cora L NBS Secretary. 1961-1966. Walter Reed Hospital. Rodriguez, Dr. Mario S Guest worker. Dentist. . 1959-1963. Loyola University, New Orleans, La. Rupp, Dr. Nelson W Navy & ADA. Dentist. 1953-1955, NBS Dental Reserach. 1969-present. Ryge, Dr. Gunnar Guest worker. Dentist 1949 U.S.P.H.S., San Francisco, Calif.

Sacchi, Dr. Hector Guest worker. Dentist- 1954-1955. Montevideo, Uruguay. Sangermano, Mr. Lawrence D Army Chemist- 1961-1962. General Electric Co. Schoonover, Dr. Iii C NBS Chemist. 1935-1953- Retired, NBS. Schouboe, Mr. Paul J ADA, NBS... Chemist. 19.54-1955. Jacksonville, Florida. Serio, Dr. Andrew F Army Dentist. 1950-1952. Private practice, Huntsville, Ala. Sheehan, Mr. William D Army. Chem. engr. 1949-1951. Simon, Mr. Lester _. Army. Chemist 1959-1961- New York, N.Y. Slade, Dr. Phihp E Army. Chemist 1956- 1958. ChemStrand, Pensacola, Florida. Smith, Mr. Denton L ADA. Metallurgist. 1947-1957_ J. M. Ney & Co., Bloomfield, Conn. Smith, Mr. W. Harold ADA Chemist. 1957- 1958. Deceased. Snover, Miss Dorothy Weinstein Chemist- 1925-1926. Res. Lab. Souder, Dr. Wilmer NBS Physicist. 1919-1945. Retired, Landisville, Pa. Stanford, Dr. John W ADA Chemist- 1952-1965- ADA, Chicago, Illinois Stephenson, Mr. S. R Army Chemist-. 1950 Steinberg, Mr. Harold L Army Chemist-. 1958- 19.59. NBS (Radioactivity) Strassburger, Mr. John NBS Chemist-. 1957-1959- Carrier, Inc. Syracuse, N.Y. Sullivan, Miss Sandra ADA. Secretary 1957-1959

230 -

Table 1. Personnel of the Dental Research Section, NBS, 1919-1969—Continued

Name Affiliation Profession Tenure Present location or employment

SusA, Mr. Martin E Army Chemist 1959-1960. Sutter, Dr. John R NBS, ADA_.. Chemist 1966-1967. Howard University Washington, D.C. SwANEY, Mr. Aubrey C ADA Chemist 1946-1950- Detroit, Michigan. SwANGER, Mr. William H Weinstein Chemist 1924-1928. Deceased. Res. Lab. Sweeney, Mr. W. Timothy. _. Summer Student 1965-1966. Medical College of student Technical! Virginia. Sweeney, Mr. William T NBS, ADA, Physicist 1922-1941, Retired, University of NBS 1949-1969. Alabama.

Taylor, Dr. Duane F NBS Metallurgist 1954-1961... Univ. of North Carolina Chapel Hill, N.C. Taylor, Dr. Norris O ADA Chemist 1928-1931. _. Deceased. Termini, Mr. Dominic J NBS Chemist 1965-present_ NBS Dental Research. Turner, Mr. John F ADA Technician 1964-1965...

ViOHL, Dr. Jochen Guest worker. Dentist 1965-1966. Free University of Berlin, Germany.

Waldron, Dr. John T Summer student. Student 1959, 1960. Private practice, Technician Pittsburgh, Penna. Wallace, Mrs. Betty M ADA. Chemist 1963-present_ NBS Dental Research. Waterstrat, Mr. Richard M ADA. Metallurgist. 1961-present. NBS Dental Research. Webb, Miss Georgia I NBS- Secretary 1953-1960... State University, Ames, Iowa. Weigel, Mr. Keith V l.. ADA Chemist 1957-1959.. Washington, D.C. Weiner, Mrs. Helen ADA Typist 1959-1963.. Washington, D.C. Weiss, Mr. Jonas Army Chemist 1957-1958.. New Brunswick, N.J. White, Dr. Eli E Summer student. Student 1955, 1956, Private practice, Merritt Technician 1957. Island, Fla. Wiedeman, Mr. William Army Chemist 1962-1963.. Buffalo, New York. Williams, Mrs. Billie S ADA Secretary 1966-1968.. Washington, D.C. WoELFEL, Dr. Julian B ADA Dentist 1957, 1958. Ohio State Universitv. WoLCOTT, Dr. Robert B Navy Dentist 1949-1951.. UCLA. WoRTHiNGTON, Dr. Chas. R-. Guest worker Chemist 1957-1958..

Yost, Mr. Ernest L Army. Physicist 1952 Pittsburgh, Pa. YuDowiTCH, Dr. Kenneth L_. ADA. 1957-1958.

Zelenka, Mr. Donald Army. Metallurgist 1960-1962. Flushing, Michigan.

Table 2. Afflliation, Table 3. Profession

American Dental Association 62 Chemists 65 NBS (Civil Service). 81 Dentists 47 U.S. Army 42 Technicians (including summer students) 19 U.S. Navy 13 Clerical (administrative) 11 U.S. Air Force ^ 7 Pliysicists 8 Veterans Administration 2 Metallurgists 9 Weinstein Research Laboratory, guest workers, Engineers 7 summer students 25 Mathematician 1 Mineralogist 1 Total 182 Other 6

Total 174

231