Frontiers in Chemical Engineering. Research Needs and Opportunities. INSTITUTION National Academy of Sciences - National Research Council, Washington, DC

DOCUMENT RESUME

ED 295 813 SE 049 194

TITLE Frontiers in Chemical Engineering. Research Needs and Opportunities. INSTITUTION National Academy of Sciences - National Research Council, Washington, DC. Commission on Physical Sciences, Mathematics, and Resources. SPONS AGENCY Department of Energy, Washington, D.C.; National Bureau of Standards (DOC), Washington, D.C.; National Science Foundation, Washington, D.C. REPORT NO ISBN-0-309-03793-E PUB DATE 88 GRANT 50SBNB5C23; CBT-8419184; DE-FG01-852260847 NOTE 232p.; Drawings and colored pages may not reproduce well. AVAILABLE FROMNational Academy Press, 2101 Constitution Avenue, NW, Washington, DC 20418 ($19.95). PUB TYPE Reports - Descriptive (141) -- Reports - Research /Technical (143)

EDRS PRICE MF01 Plus Postage. PC Not Available from EDRS. DESCRIPTORS *Chemical Engineering; *Chemical Industry; College Science; Higher Education; *Manufacturing Industry; Petroleum Industry; Power Technology; Research Needs; *Research Opportunities; Science Education; *Scientific Research; *Technological Advancement

ABSTRACT Chemical engineers play a key role in industries such as petroleum, food, artificial fibers, petrochemicals, plastics and many others. They are needed to tailor manufacturing technology to the requirements of products and to integrate product and process design. This report discusses how chemical engineers are continuing to address technological problems that are important to manufacturing interests in the United States. The research frontiers discussed in this report are grouped into four themes including: (1) starting new technologies; (2) maintaining leadership in established technologies; (3) protecting and improving the environment; and (4) developing s3stematic knowledge and generic tools. High priority areas c9nsidered include biotechnology and biomedicine; electronic, platonic, and recording materials and devices; polymers, ceramics, and composites; processing of energy and natural resources; environmental ptatection, process safety, and hazardous waste management; computer-assisted process and control engineering; and surfaces, interfaces, and microstructures. Appendices include detailed recommendations for funding, contributors, and the chemical processing industries. (CW)

*********************************************************************** * Reproductions supplied by EDRS are the best that can be made * * from the original document. * *********************************************************g************* FRONTIERS IN CHEMICAL ENGINEERING

RESEARCH NEEDS AND OPPORTUNITIES

Committee on Chemical Engineering Frontiers: Research Needs and Opportunities Board on Chemical Sciences and Technology Commission on Physical Sciences, Mathematics, and Resources National Research Council

NATIONAL ACADEMY PRESS Washington, D.C. 1988

3 NATIONAL ACADEMY PRESS 2101 Constitution Avenue, NW Washington, DC 20418 NOTICE: The project that is the subject of this report was approved by the Governing Board of the National Research Council, whose members are drawn from the councils of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of the committee responsible for the report were chosen for their special competences and with regard for appropriate balance. This report has been reviewed by a group other than the authors according to procedures approved by a Report Review Committee consisting of members of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The National Academy of Sciences is a private, nonprofit, self-perpetuating society of distinguished scholars engaged in scientific and engineering research, dedicated to the furtherance of science and technology and to their use for the general welfare. Upon the authority of the charter granted to it by the Congress in 1863, the Academy has a mandate that require.. it to advise the federal government on scientific and technical matters. Dr. Frank Press is president of the National Academy of Sciences. The National Academy of Engineering was established in 1964, under the charter of the National Academy of Sciences, as a parallel organization of outstanding engineers. It is autonomous in its administration and in the selection of its members, sharing with the National Academy of Sciences the responsibility for advising the federal government. The National Academy of Engineering also sponsors engineering programs aimed at meeting national needs, encourages education and research, and recognizes the superior achievements of engineers. Dr. Robert M. White is president of the National Academy of Engineering. The Institute of Medicine was established in 1970 by the National Academy of Sciences to secure the services of eminent members of appropriate professions in the examination of policy matters pertaining to the health of the public. The Institute acts under the responsibility given to the National Academy of Sciences by its congressional charter to be an adviser to the federal government and, upon its own initiative, to identify issues of medical care, research, and education. Dr. Sanwel 0. Thier is president of the Insiitute of Medicine. The National Research Council was organized by the Nationai Acadcmy of Sciences in 1916 to associate the broad community of science and technology with the Academy's purposes of furthering knowledge and advising the federal government. Functioning in accordance with general policies determined by the Academy, the Council has Lome the principal operating agency of both the National Academy of Sciences and the National Academy of Engineering in providing services to the government, the public, and the scientific and engineering communities. The Council is administered jointly by both Academies and the Institute of Medicine. D.. Frank Press and Dr. Robert M. White are chairman and vice-chairman, respectively, of the National Research Council. Support for this project was provided by the Amencan Chemical Society, the American Institute of Chem:cal Engineers, the Council for Chemical Research, Inc., the U.S. Department of Energy under Grant No. DE-FG01-85FE60847, the National Bureau of Standards under Grant No. 50SBNB5C23, the National Science Foundation under Grant No. CBT- 8419184, and the Whitaker Foundation.

Library of Congress Cataloging-in-Publication Data National Research Council (U.S.). Committee on Chemical Engineering Frontiers: Research Needs and Opportunities. Frontiers in chemical engineering : research needs and opportunities/Committee on Chemical Engineering Frontiers Research Needs and Opportunities, Board on Chemical Sciences and Technology, Commission on Physical Sciences, Mr.thematics, and Resources, National Research Council. p. cm. Bibliography: p. Includes index. ISBN 0-309-03793-X (paper); ISBN 0-309-03836-7 (cloth) 1. Chemical engineering -ResearchUnited States. I. Title. TPI71.N37 1988 88-4120 620' .0072dc19 CIP (Rev.)

No part of this book may be reproduced by any mechanical, photographic, or electronic process, or in the form of a phonographic recording, nor may it be stored in a retrieval system, transmitted, or otherwise copied f . public or private use without written permission from the publisher, except for the purposes of official use by the U.S. government.

Printed in the United States of America

4 I

Committee on Chemical Engineering Frontiers: Research Needs and Opportunities

NEAL R. AMUNDSON (Chairman), University JOHN A. QUINN, University of Pennsylvania of Houston KENNETH J. RICHARDS, Kerr-McGee EDWARD A MASON (Vice-Chairman), Corporation Amoco Corporation JOHN P. SACHS, Horsehead Industries, Inc. JAMES WEI (Vice-Chairman), Massac..asetts ADEL F. SAROFIM, Massachusetts Institute Institute of Technology of Technology MICHAEL L. BARRY, Vitelic Corporation ROBERT S. SCHECHTER, University of ALEXIS T. BELL, University of California, Texas, Austin Berkeley WILLIAM R. SCHOWALTER, Princeton KENNETH B. BISCHOFF, University of University Delaware L. E. SCRIVEN, University of Minnesota HERBERT D. DOAN, Doan Associates JOHN H. SEINFELD, California Institute of ELISABETH M. DRAKE, Arthur D. Little, Technology Inc. JOHN H. SINFELT, Exxon Research and SERGE GRATCH, Ford Motor Company Engineering Company (retired) LARRY F. THOMPSON, AT&T Bell HUGH D. GUTHRIE, Morgantown Energy Laboratories Technology Center, DOE KLAUS D. TIMMERHAUS, University of ARTHUR E. HUMPHREY, Lehigh University Colorado SHELDON E. ISAKOFF, E.I. du Pont de ALFRED E. WECHSLER, Arthur D. Little, Nemours and Company, Inc. Inc. JAMES LAGO, Merck and Company (retired) ARTHUR W. WESTERBERG, Carnegie-Mellon KEITH W. MCHENRY, JR., Amoco Oil University Company SEYMOUR L. MEISEL, Mobil Research and ROBERT M. SIMON, Project Director Development Company (retired) ROBERT M. JOYCE, Editorial Consultant ARTHUR B. METZNER, University of NANCY WINCHESTER, Editor Delaware ROSEANNE PRICE, Editor ALAN S. MICHAELS, North Carolina State LYNN E. DUFF, Financial Assistant University MONALISA R. BRUCE, Administrative JOHN P. MULRONEY, Rohm and Haas Secretary Company LEIGH E. NELSON, Minnesota Mining and Manufacturing Co., Inc.

iii r 1111 U Panels of the Committee

Panel on Biochemical andBiomedical Engineering ARTHUR E. HUMPHREY (Chairman), Lehigh ROBERT L. DEDRICK, National Institutes of University Health KENNETH B. BISCHOFF, University of MITCHELL LITT, University of Pennsylvania Delaware ALAN S. MICHAELS, North Carolina State CHARLES BOTTOMLEY, E.I. du Pont de University Nemours and Company, Inc. FRED PALENSKY. Minnesota Mining and STUART E. BUILDER, Genentech, Inc. Manufacturing Co., Inc.

Panel on Electronic. Photonic, and Recording Materials and Devices LARRY F THOMPSON (Chairman), AT&T RICHARD POLLARD. University of Houston Bell Laboratories T. W. RUSSELL FRASER. University of LEE F. BLYLER, AT&T Bell Laboratories Delaware JAMES ECONOMY, IBM Almaden Research MICHAEL SHEPTAK, Ampex Corporation Center DENNIS W. HESS, University of California, Berkeley

Panel on Advanced Materials ARTHUR B. METZNER (Chairman), FRANK KELLEY. University of Akron University of Delaware JOHN B. WACHTMAN. JR.. Rutgers FRANK BATES, AT&T Bell Laboratories University C. F. CHANG, Union Carbide Corporation IOANNIS V. YANNAS. Massachusetts F. NEIL COGSWELL, Imperial Chemical Institute of Technology Industries WILLIAM W. GRAESSLEY. Princeton University

6 Panel on Energy and Natural Resources Processing KEI1H McHENRY (Chairman). Amoco Oil NOEL JARRETT, Alcoa Laboratories Company FREDERIC LEDER. Dowell Schlumberger LESLIE BURRIS. Argonne National JOHN SHINN. Chevron Research Company Laboratory REUEL SHINNAR, City College of New York ELTON J. CAIRNS. Lawrence Berkeley PAUL B. WEISZ. University of Pennsylvania Laboratory

Panel on Environmental Protection, Safety, and Hazardous Materials ADEL SAROFIM (Chairman), Massachusetts THOMAS W. PETERSON, University of Institute of Technology Arizona SIMON L. GOREN, University of California, WILLIAM RODGERS, Oak Ridge National Berkeley Laboratory GREGORY J. MACRAE, Carnegie Mellon GARY VEURINK, Dow Chemical Company University RAY WITTER, Monsanto Corporation

Panel on Computer Assisted Process and Control Engineering ARTHUR W. WESTERBERG (Chairman), MANFRED MORARI, California Institute of Carnegie Mellon University Technology HENRY CHIEN, Monsanto Corporation JEFFREY J. SIIROLA, Eastman Kodak JAMES M. DOUGLAS, University of Company Massachusetts WILLIAM SILLIMAN, Exxon Production BRUCE A. FINLAYSON, University of Research Company Washington ROLAND KEUNINGS, University of California, Berkeley

Panel on Surface and Interfacial Engineering ALEXIS T. BELL (Chairman), University of KLAVS F. JENSEN, University of Minnesota California, Berkeley JAMES R. KATZER, Mobil Research and RICHARD C. ALKIRE, University of Illinois Development Company JOHN C. BERG, University of Washington LEIGH E. NELSON. Minnesota Mining and L. LOUIS HEGEDUS, W. R. Grace and Manufacturing Company Company LANNY D. SCHMIDT, University of ROBERT JANSSON, Monsanto Corporation Minnesota

1.9 Board on Chemical Sciences and Techi,ology

EDWARD A. MACON (Co-Chairman), Amoco JOHN A. QUINN, University of Pennsylvania Corporation STUART A. RICE, University of Chicago GECRGE M. WHITESIDES (Co-Chairman), FREDERIC M. RICHARDS, Yale University Harvard University ROGER A. SCHMITZ, University of Notre NEAL R. AMUNDSON, University of Houston Dame JOHN I. BRAUMAN, Stanford University L. E. SCRIVEN, University of Minnesota GARY FELSENFELD, National Institutes of DAVID P. SHEETZ, Dow Chemical USA Health LEO J. THOMAS, JR., Ea.;tman Kodak WILLIAM A. GODDARD III, California Company Institute of Technology NICHOLAS J. TURRO, Columbia University JEANETTE G. GRASSELLI, BP America MARK S. WRIGHTON, Massachusetts Institute MICHAEL L. GROSS, University of Nebraska of Technology RALPH HIRSCHMANN, University of Pennsylvania ROBERT M. SIMON, Staff Director ROBERT L. LETSINGER, Northwestern WILLIAM SPINDEL, Special Staff Adviser University PEGGY J. POSEY, Staff Associate JAMES F. MATHIS, Exxon Chemical LYNN E. DUFF, Financial Assistant Company GEORGE C. PIMENTFL, University of California, Berkeley

Vit Commission on Physical Sciences, Mathematics, and Resources

NORMAN HACKERMAN(Chairman),Robert WILLIAM D. PHILLIPS, Washington A. Welch Foundation University GEORGE F. CARRIER, Harvard University DENIS J. PRAGER, MacArthur Foundation DEAN E. EASTMAN, IBM Corporation DAVID M. RAUP, University of Chicago MARYE ANN FOX, University of Texas. RICHARD J. REED, University of Washington Austin ROBERT E. SIEVERS, University of Colorado GERP.ART FRIEDLANDER, Brookhaven LARRY L. SMARR, University of Illinois National Laboratory EDWARD C. STONE, JR., California Institute LAWRENCE W. FUNKHOUSER, Chevron of Technology Corporation (retired) KARL K. TUREKIAN, Y4le University PHILLIP A. GRIFFITHS, Duke University GEORGE W. WETHERILL, Carnegie J. ROSS MacDONALD, The University of Institution of Washington North Carolina at Chapel Hill IRVING WLADAWSKY-BERGER, IBM CHARLES J. MANKIN, The University of Corporation Oklahoma PERRY L. McCARTY, Stanford University RAPHAEL G. KASPER,Executive Director JACK E. OLIVER. Cornell University LAWRENCE E. McCRAY,Associate Executive JEREMIAH P. OSTRIKER, Princeton Director University 0;)servatory

vii Contents

ONE Executive Summary / 1

TWO What Is Chemical Engineering? /9

THREE Biotechnology and Biomedicine /17 FOUR Electronic, Photonic, and Recording Materials and Devices /37

FIVE Polymers, Ceramics, and Composites /61 SIX Processing of Energy and Natural Resources /79 SEVEN Environmental Protection, Process Safety, and Rizardous Waste Management /105 EIGHT Computer-Assisted Process and Control Engineering /135

NINE Surfaces, Interfaces, and Microstructures /153 ---TEN Recommendations /175

APPENDIXES

A Detailed Recommendations for Funding /185

B Contributors /198

c The Chemical Processing Industries /201

index /205

10 FRONTIERS IN CHEMICAL ENGINEERING

RESEARCH NEEDS AND OPPORTUNITIES O N E

Executive Summary

HEMICAL ENGINEERING occupies a spe- improving the environment, and developing sys- cial place among scientific and engi- tematic knowledge and generic tools. These neering disciplines. It is an engineering frontiers are described in detail in Chapters 3 discipline with deep roots in the world of atoms, through 9. From among these, the committee molecules, and molecular transformations. The has selected eight high-priority topics that merit principles and approaches that make up chem. the attention of researchers, decision makers in icPI engineering have a long and rich history of academia and industry, and organizations that contributions to the nation's technological needs. fund or otherwise support chemical engineering. Chemical engineers play a key role in industries These high-priority areas are described below. as varied as petroleum, food, artificial fibers, Recommendations from the committee for ini- petrochemicals,plastics,ceramics, primary tiatives that would permit chemical engineers metals, glass, and specialty chemicals. All these to exploit these areas are briefly stated in depend on chemical engineers to tailor manu- Chapter 10 and detailed in Appendix A. facturing technology to the requirements of their products and to integrate product design with process design. Chemical engineering was the RESEARCH FRONTIERS IN CHEMICAL first engineering profession to recognize the ENGINEERING integral relationship between design and manufacture, and this recognition has been one of Starting New Technehnies the major reasons for its success. This report ''lmmistrates that chemical en- Chemical engineers have an important role gineering research will continue to address the to play in bringing new technologies to com- techno:nvical problems most important to the mercial fruition. These technologies have their nation. In the chapters that focus on these origin in scientific discoveries on the atomic and problems, many of the discipline's core research molecular level. Chemical engineers understand areas (e.g., reaction engineering, separations, the molecular world and are skilled in integrating process design, and control) will appear again product design w ;:h process design, process and again. The committee hopes that by dis- control, and optimization. Their skills are nzeded cussing research frontiers in the context of to develop genetic engineering (biotechnology) applications, it will illustrate both the intellec- as a manufacturing tool and to create new tual excitement and the practical importance of biomedical devices, and to design new products chemical engineering. and manufacturing processes for advanced ma- The research frontiers discussed in this report terials and devices for information storage and can he grouped under four overlapping themes: handling. In the fierce competition for world starting new technologies, maintaining leader- markets in these technologies, U.S. leadership ship in established technologies, protecting and in chemical engineering is a strong asset.

I 14 2 PRO I IPAS IA ( 1111.111( 41. PAW: VEERING

Biotechnology and Biomedicine(Chapter 3) At the frontiers of chemical research in this area are a number of impol tant challenges: The Uidted Statese- -pies the preeminent scientific position 'n the "new" biology.If Integrating individual chemical process steps America is to derive the maximum benefit of used in the manufacture of electronic, photonic, its investment in basic biological research and recording materials and devices. This is a whether in the form of better health, improved key to boosting the yield, throughput, a). agriculture, cleaner environment, or more reliability of overall manufacturing processes. efficient F. . .ion of chemicalsit IMF' also Refining and applying chemical engineering assume a preeminent position in biochemical principles to the design and control of the and biomedical engineering. This can b ! accom- chemical reactors in which devices are fabri- plished by carrying out generic research in the cated. following areas: Pursuing research in separations applicable Developing chemical engineering models to the problem of ultrapurification. The mate- for fundamental biological interactions. rials used in device manufacture must be ultrapure, with levels of some impurities reduced to Studying phenomena at biological surfaces and interfaces that are important in the design the parts-per-trillion level: of engineered systems. Improving the chemical synthesis and processing of polymers and ceramics; Advancing the field of proc',s engineering. Important generic goals for research include the Developing better processes for deposition development of separation processes for com- and coating of thin films. An integrated circuit, plex and fragile bioproducts; the design of in essence, is a series of electrically connected bioreactors fe'plant and mammalian tissue thin films. Thin films are the key structural feature of recording media and optical fibers, culture; and the development of detailed,-n- as well. tinuous control of process parameters by rapid, accurate, and noninvasive sensors and instru- Modeling the chemical reactions that are ments. importanttomanufacturing processes and studying their dynamics. Conducting engineering analyses of complex biological systems. Emphasizing process design and control for environmental protection and process safety.

Electronic, Photonic, and Recording Microstructured Materials Materials and Devices(Chapter 4) (Chapters 5 and 9) The character of American industry and society has changed dramatically over the past Advanced materials depend on carefully de- three decades as we have entered the "infor- signed structures at the molecular and micro- mation age." New information technologies scopic levels to achieve specific performance in have been made possible by materials and use. These materialspolymers, ceramics, and devices whose structure and properties can be compositesare reshaping our society and are contributing to an improved standard of living. controlled with exquisite precision. This control is largely achieved by the use of chemical The process technology used in manufacturing reactions during manufacturing. Future U.S. these materials is crucialin many instances leadership in microelectronics, optical infor- more important than the composition of the mation technologies, magnetic data storage, and materials themselves. Chemical engineers can photovoltaics will depend on staying at the make important contributions to materials de- forefront of the chemical technology used in sign and manufacturing by exploring the following research frontiers: manufacturing processes. Chemical processing will also be a vital part of the likely manufac- Understanding how microstructures are turing processes for high-temperature super- formed in materials and learning how to control conductors. the processes involved in their formation.

is EXECUTIVE SI IMMARY 3

Combining materials synthesis and mate- industries on world markets. At a time of record rials processing. These areas have traditionally trade deficits, the chemical industry has main- been considered separate research areas. Future tained both a positive balance of trade and a advances in materia:3 require a fusion of these growing share of world markets (Figure 1.1). topics in research and practice. The future international competitiveness of these Fabricating and repairing complex mate- industries should not be taken for granted. Far- rials systems. Mechanical methods currently in sighted management in industry and continued use (e.g., riveting of metals) cannot be applied support for basic research from both industry reliably to the composite materials of the future. and government are required if this sector of Chemical methods (e.g., adhesion and molec- the economy is to continue to contribute to the ular self-assembly) will come to the fore. nation's prosperity. In a report of this scope and size, it is not possible to spell out the research challenges Maintaining Leadership in Established faced by each part of the chemical processing Technologies industries. For example, the committee has The U.S. chemical processing industries are reluctantly chosen to pass over food processing, one of the largest industrial sectors of the U.S. a multibillion-dollar industry where chemical economy. The myriad of industries listed at the engineering finds a growing variety of applica- beginning of this chapter are ;.ervasive and tions. The committee has focused its discussion absolutely essential to society. The U.S. chem- of challenges to the processing industries on ical industry is one of the most successful U.S. energy and natural resources technologies. These 20 10 0 10 20 30 40 1 A 50 '6 60 70 V aN 80 Ca -90 100 110 111 120 130 140 150 160 I I 1 1 1973 1975 1981 1982 1983 1984 1985 1986 cmAll U.S. Trade IMIU.S. Chemical Trade

FIGURE 1.1 While the overall U.S. trade balance has plummeted to a deficit of more than $150 billion, the U.S. chemical industry has maintained a positive balance of trade. Courtesy, Department of Commerce.

i4 4 FRO,V1IERS ( WC (;I:LFRA

technologies are key to supplying crucial na- materials science, metallurgy, surface and col- tional needs, keeping the United States com- loid science, and chemistry. petitive, and providing for national security. They are also the focus of substantial research Liquid Fuels Jr the Flame and development in academia and government laboratories, in addition to industry. The com- (Chapters 6 and 9) mittee has identified two high-priority initiatives Our current and foreseeable transportation to sustain the vitality and creativity of engi- technologies depend completely on a plentiful neering research on energy and naturalre- supply of liquid fossil fuels. Anticipatory resources. Thesc initiatives focus on in-situ pro- search to ensure a future supply of these fuels cessing of resources and on liquid fuels for the is a wise investment. Research of this type future. subsumes a number of generic challenges in chemical engineering, including: In-Situ Processing of Energy and Mineral Resources(Chapter 6) Finding new chemical process pathways that can make large advances in the production The United States has historically benefited of liquid fuels from solid and gaseous resources. from rich domestic resources of minerals and Processing solids, since equipment design fuels located in readily accessible parts of the and scale-up are greatly limited by our lack of earth's crust. These easily reached resources fundamental understanding of solids behavior. are being rapidly depleted. Our remaining re- Developing better separation processes. serves, while considerable, require moving Conducting research on materials capable greater and greater amounts of the earth's crust of withstanding the extreme processing condi- to obtain and process resources, whether that tion, that may be encountered when processing crust is mixed with the desired material (as in liquid fuels. a dilute ore vein) or whether it simply lies over Advancing the state of the art in the design, the resource. A long-range solution to this scale-up, and control of processes. problem is to use chemical reactions to extract underground resources, with the earth itself as the reaction vessel. This is known as in-situ Protecting and Improving the Environment processing. Enhanced oil recovery is the most successful current example of in-situ process- Responsible Management of Hazardous Substances(Chapter 7) ing, and yet an estimated 300 billion barrels of U.S. oil trapped underground in known reserves The modern world faces many environmental cannot be recovered with current technology. problems. Some of these are a consequence of Long-range research aimed at oil, shale, tar producing the ever-increasing number and va- sands, coal, and mineral resources is needed. riety of chemicals and materials demanded by Formidable problems exist both for chemists society. Chemical engineers must take ur the and for chemical engineers. Some research role of cradle-to-grave guardians for chemicals, priorities for chemical engineers include sepa- ensuring their safe and environmentally sound ration processes, improved materials, combus- manufacture, use, and disposal. This means tion processes, and advanced methods of pro- becoming involved in a range of research areas cess design, scale-up, and control. Research on dealing with environmental protection, process in-situ processing will require collaboration be- safety, and hazardous waste management. In tween chemical engineers and scientists and the following four areas, the challenges are engineers skilled in areas such as geology, clear, the opportunities for chemical engineering geophysizs, hydrology, environmental science, research are abundant, and the potential benefits mecharical .gineering, phys:,:s, mineralogy, to society are great.

15 EX:X*1711F SU,',!MARI 5

Conducting long-terpt research on the gen- under way in applying modern computational eration, control, movement. fate, detection, and methods and process control to chemical engi- environmental and health effects of contami- neering and the promise of basic research in nants in the air, water, and land. Chemical surface and interfacial engineering. engineering research should include the fundamental investigation of combustion processes, the application of biotechnology to waste deg- Advanced Computational Methods and Process Control (Chapter 8) radation, the development IA sensors and measurement echniques, and osticipation is inter- The speed and capability of the modern com- disciplinary studies of the enchonment's capacity puter are revolutionizing the practice of chemical to assimilate the broad range zf chemicals that engineering. Advances in speed and memory are hazardous to humans and ecosystems. size and improvements in complex problem- Developing new chemical engineering de- solving ability are more than doubling the ef- sign tools to deal with the multiple objc:t.:'es fective speed of the computer each year. This of minimum cost; process resilience to ctn., ;es unrelenting pace of advance has reached the in inputs; minimizatiort of toxic intermc hates stage where it profoundly alters the way in and products; and safe response to upset con- which chemical engineers can conceptualize ditions, start-up, and shutdown. problems and approach solutir s. For example: Directing research at cost-effective management of hazardous waste, as well as im- It is now realistic to imagine mathematical proved technologies (e.g., combustion) or new models of fundamental phenomena beginning technologies for destroying hazardous waste. to replace laboratory and field experi;nents. Carrying out research to facilitate multi- Such computations increasingly allow chemical media, multispecies approaches to waste man- engineers to bypass the long (2 to 3 years), agement. Acid rain and the leaching of hazard- costly step of producirq, process and product ous chemicals from landfills demonstrate the prototypes, and permit the design of products mobility of chemicals from one m...dium (e.g., and processes that better utilize scarce re- air, water, or soil) to another. sources, are significantly less polluting, and are much safer. Developing Systematic Knowledge and Future computer aids will allow design and Generic Tools control engineers to examine many more alternatives much more thoroughly and thus produce The success of chemical engineers in contrib- better solutions to problems within the known uting to a diverse set of technologies is due to technology. an emphasis on discovering and developin6 Better modeling will allow the design of basic principles that transcend individual tech- processes that are easier and safer to verate. nologies. If, 20 years from now, chemical en- Improved control methodology and sensors will gineers are to have the same opportunities for overcome the current inability to model certain contributing to important societal problems that processes accurately. they have today, then the research areas de- Sensors of the future will be incredibly scribed in the preceding sectior.s must be ex- small and capable. Many will feature a chemical plored and supported in a way that maximizes laboratory and a computer on a chip. They will the development of basic knowledge and tools. enable chemical engineers to detect chemical In surveying the field of chemical engineering, compositions inside hostile process environ- the committee has identified two cross-cutting ments and revolutionize their ability to control areas that are in a state of rapid development processes. and that promise major contributions to a wide To realize the promise of the computer in range of technologies. Accordingly, this report chemical engineering, we need a much larger cingles out for special attention the ad tqinces effort to develop methodologies for process

16 6 t RONTIER8 IN ( riEMIC AL LNGINEERING design and control. In addition, state-of-the-art Institute of Chemical Engineers (AIChE) has computational facilities aad equipment must setin motion a project to incorporate into become more widely disseminated into chemical undergraduate chemical engineering courses ex- engineering departments in order to integrate amples and problems from emerging applica- methodological advances into the mainstream tions of the discipline. The committee applauds of research and education. this work, as well as recent AIChE moves to allow more flexibility for students in accredited departments to take science electives. Surface and Interfacial Engineering A second important need in the curriculum (Chapter 9) is for a far greater emphasis on design and Surfaces, interfaces, and microstructures play control for process safety, waste minimizatit,n, an important role in many of the above-men- and minimal adverse environmental impact. tioned research frontiers. Chemical engineers These themes need to be woven into the cur- explore structure-property relationships at the riculum wherever possible. The AIChE Center atomic and molecular level, investigate elemen- for Chemical Process Safety is attempting to tary chemical and physical transformations oc- provide curricular r taterial in this area, but a curring at phase boundaries, apply modern theo- larger effort than this project is needed. Several retical methods for predicting chemical dynamics large chemical companies have significant ex- at surfaces, and integrate this knowledge into pertise in this area. Closer interaction between models that can be used in process design and academic researchers and educators and indus- evaluation. Fundamental advances in these areas try is required to disseminate this expertise. will have a broad impact on many technologies. Examples include laying down thin films for The Future Size and Composition of microelectronic circuits, developing high-strength concrete for roadways and buildings, and in- Academic Departments(Chapter 10) venting new membranes for artificial organs. A bold step by universities is needed if their Advances in surface and interfacial engineering chemical engineering departments are (1) to can also move the field of heterogeneous catal- help the United States achieve the preeminent ysis forward significantly. New knowledge can position of leadership in new technologies and help cll., , '-al engineers play a much bigger role (2) tc keep the highly successful U.S. chemical in the sync. -;:sis and modification of novel cata- processing industries at the forefront of world lysts with enhanced capabilities. This activity markets for established technologies. The uni- would complement their traditional strength in versities should conduct a one-time expansion analytical reaction engineering of catalysts. of their chemical engineering departments over the next 5 years, exercising a preference for new faculty capable of research at interdisci- HIGHLIGHTS OF THE plinary frontiers. RECOMMENDATIONS This expansion will require a major commitment of resources on the part of universities, Education and Training of Chemical government, and industry. How can such a Engineers(Chapter 10) preferential commitment to one discipline be The new research frontiers in chemical en- justified, particularly at a time of budgetary gineering, some of which represent new appli- austerity? One answer is that the worldwide cations for the discipline, have important im- cortest for dominance in biotechnology, ad- plications for education. A continued emphasis vanced materials technologies, and advanced is needed on basic principles that cut across information devices is in full swing, and the many applications, but a new way of teaching United States cannot afford to stand by until it those principles is also needed. Students must gets its budgetary house in order. As the uniquely be exposed to both traditional and novel appli- "molecular" engineers, chemical engineers have cations of chemical engineering. The American powerful tools that need to be refined and

1 7 EXECUTIVE St*VtiARY 7 applied to the commercialization of these tech- research frontiers to funding initiatives for po- nologies. A second answer is that the alternative tential sponsors. to expansion, a redistribution of existing re- A third balanced portfolio, of funding mech- sources for chemical engineering research, would anisms, is needed if the above-mentioned re- cut into vital programs that support U.S. com- search frontiers are to be pursued in the most petitiveness in established chemical technolo- effective manner. Different frontiers will require gies. The recommendation for an expansion in different mixes of mechanisms, and the decision chemical engineering departments is not a call to use a particular mechanism should be deter- for "more of the same." It is the most practical mined by the nature of the research problem, way to move chemical engineering aggressively by instrumentation and facilities requirements, into the new areas represented by this report's and by the perceived need for trained personnel research priorities while maintaining the disci- in particular areas for industry. This topic is pline's current strength and excellence. discussed in more detail in Charter 10.

Balanced Portfolios (Chapter 10) The Need for Expanded Support of liesearch in Chemistry (Chapter 10) The net result of an additional investment of resources in chemical engineering should be the Chemical engineering builds on research re- creation of three balanced portfolios: one of sults from other disciplines, as well as those priority research areas, one of sources of fund- from its own practitioners. Not surprisingly, the ing for research, one of mechanisms by which most important of these other disciplines is that funding can be provided. chemistry. A vital base of chemical science is The eight priority research areas described needed to stimulate future progress in chemical above constitute the committee's recommen- engineering, just as vital base in chem;.;a1 dation of a balanced portfolio of research areas engineering is needed to capitalize on advances on the frontiers of the discipline. in chemistry. The committee endorses the rec- In terms of a balanced portfolio of funding ommendations contained in the NRC's 1985 sources, the committee proposes initiatives for report Opportunities in Chemistry, and urges industry and a number of federal agencies in their implementation in addition to the recom- Chapter 10 and Appendix A to ensure a healthy mendations contained in this volume. diversity of sponsors. Table 1.1 links specific

TABLE1.1 High-Priority Research Frontiers and Initiatives. Relevant Audiences° Priority Research Area NSF DOE NIH DOD EPA NBS BOM CPI

Biotechnology and biomedicine 1.0 01 Electronic, photonic, and recording matenals and devices Microstructured materials In-situ processing of resources Liquid fuels for the future Responsible management of hazardous substances Woo Advarwed computational methods and process control Surface and interfacial engineenng °WI/ = major initiative recommended; 1,01 = supportiig initiative recommended. b NSF = National Science Foundation; DOE = Department of Energy, NIH = National Institutes of Health; DOD = Department of Defense; EPA = Environmental Protection Agency, NBS = National Bureau of Standards; BOM = Bureau of Mines, CPI = U.S. chemical processing industries.

I C What Is Chemical Engineering?

hemical engineering has a rich past and a bright future. In barely a century, its practitioners have erected the technological infrastructure of much of modem society. Without their contributions, industries as diverse as petroleum processing, pharmaceutical manufacturing, food processing, textiles, and chemical manufacturing would not exist as we know them today. In the ;0 to 15 years ahead, chemical engineering will evolve to address challenges that span a wide range of intellectual disciplines and physical scales (from the molecular scale to the planetary scale): And chemical engineers, with their strong ties to the molecular sciences, will be the "interfacial researchers" bridging science and engineering in the multidisciplinary environments where a host of new technologies will be brought into being.

9 19 10 FROSTIERS (HEM( 1I. ENGIVEERI

MAGINE AWORLD where penicillin and to visualize modern life without the large-vol- other antibiotics are rarer and more ume production of antibiotics, fertilizers and expensive than the finest diamonds. agricultural chemicals, special polymers for Imagine countries gripped by famine as dwin- biomedical devices, high-strength polymer com- dling supplies of natural guano and saltpeter posites, and synthetic fibers a:d fabrics. How cause fertilizers to become increasingly scarce. would our industries function without environ- Imagine hospitals and clinics where kidney mental control technologies; without processes dialysis is as nsky and as uncertain over the to make semiconductors, magnetic disks and long term as today's artificial heart program. tapes, and optical informatiLstorage devices; Imagine serving on a police force or in the without modern petroleum processing? All these infantry without a lightweight bulletproof vest. technologies require the ability to produce spe- Imagine your closet with no wash-and-dry, cially designed chemicalsand the materials wrinkle-free synthetic garments, or your home based on themeconomically and with a min- without durable, easy-cleaning, mothproof syn- imal adverse impact on the environment. De- thetic rugs. Imagine cities choked with smog veloping this ability and implementing it on a and soot from millions of residential coal fur- practical scale is what chemacal engineering is naces and millions of automobiles without emis- all about. sion controls. Imagine an "information society" The products that depend on chemical engi- trying to function on vacuum tubes and ferrite neering emerge from a diverse array of indus- core storage for data processing. Imagine paying tries that play a key role in our economy (Table $25 or more for a gallon of gasoline, if you can 2.1). These industries produce most of the even buy it. This world, in which few of us materials from which consumer products are would want to live, is what a world without made, as well as the basic commodities on chemical engineering would be like. which our way of life is built. In 1986, they Chemical engineers have made so many im- shipped products valued at nearly $585 billion. portant contributions to society that it is hard They had a payroll of 3.3 million employees, or

ABLE2.1The Chemical Processing Industries in the United States° Number of Value of Value Added by Employees Shipments Manufacture Industry1' (thousands) (S millions) (S millions)

Food and beverages 378 73.633 24.370 Textiles 99 7.6'9 2.897 Paper 7:2 51.145 19.871 Chemicals 1.023 197,932 95.258 Petroleum 169 129.365 17.112 Rubber and plastics 340 31.078 15.390 Stone. clay, and glass 354 34.372 17.449 Nonferrous metals 49 21.920 524 Other none...rables 577 37.594 24.291

TOTAL 3.321 584.689 217.161 Chemical processing industries' share of total manufacturing 17.5% 25 7% 21 7% " Data for employment and value of shipments are for 1986. Data for value added by manufacture are for 1985. SOURCE: Data Resources. Inc. b The definition of the chemical processing industries (CPI, used in thi' :dime is the one used by Data Resources and Chemical Engineering in compiling their statistics on these industnes. For several of the Industries listed, only a par: is considered to be in the CPI and data are presented for this part only A list of the Standard Industrial Classification codes used to define the CPI for this table is given in Appendix C.

20 WHAT IS CHEMICAL. ENGINEERING?

17.5 percent of all U.S. manufacturing employ- verizing, drying, roasting, crystallizing, filter- ees. They generated over $217 billion in value ing, evaporating, distilling, electrolyzing, and added in 1985, or 21.7 percent of all U.S. so on. Thus, for example, the academic study manufacturing value added. The chemicals por- of the specific aspects of turpentine manufacture tion of the CPI is one of the most successful could be replaced by the generic study of U.S. industries in world competition, producing distillation, a process common to many other an export surplus of $7.8 billion in19t!6, in industries. A quantitative form of the unit op- contrast to the overall U.S. trade deficit of $152 erations concept emerged around 1920, just in billion. time for the nation's first gasoline crisis. The But chemical engineering is more than a group rapidly growing number of automobiles was of basic industries or a pile of economic statis- severely straining the production capacity for tics. As an intellectual discipline, it is deeply naturally occurring gasoline. The ability of involved in both basic and applied research. chemical engineers to quantitatively character- Chemical engineers bring a unique set of tools ize unit operations such as distillation allowed and methods to the study and solution of some for the rational design of the first modern oil of society's most pressing problems. refineries. The first boom of employment of chemical er.gineers in the oil industry was on. During this period of intensive development TRADITIONAL PARADIGMS OF of unit operations, other classical tools of chem- CHEMICAL ENGINEERING ical engineering analysis were introduced or Every scientific discipline has its character- were extensively developed. These included istic set of problems and systematic methods studies of the material and energy balance of for obtaining their solution---that is, its para- processes and fundamental thermodynamic digm. Chemical engineering is no exception. studies of multicomponent systems. Since its birth in the last century, its fundamental Chemical engineers played a key role in intellectual model has undergone a series of helping the United States and its allies win dramatic changes. World War II. They developed routes to syn- When the Massachusetts Institute of Tech- thetic rubber to replace the sources of natural nology (MIT) started a chemical engineering rubber that were lost to the Japanese early in program in 1888 as an option in its chemistry the war. They provided the uranium-235 needed department, the curriculum largely described to build the atomic bomb, scaling up the man - industrial operations and was organized by spe- ufactu.ing process in one step from the labo- cific products. The lack of a paradigm soon ratory to the largest industrial plant that had became apparent. A better intellectual founda- ever been built. And they were instrumental in tion was required because knowledge from one perfecting the manufacture of penicillin, which chemical industry was often different in detail saved the lives of potentially hundreds of thou- from knowledge from other industries, just as sands of wounded soldiers. An in-depth look at the chemistry of sulfuric acid is very different this latter contribution shows the sophistication from that of lubricating oil. that chemical engineering had achieved by the 1940s. ' Penicillin was discovered before the war, but Unit Operations could , My be prepared in highly dilute, impure, The first paradigm for the discipline was based and unstable solutions. Up to 1943, whe. chem- on the unifying concept of "unit operations" ical engineers first became involved with the proposed by Arthur D. Little in 1915. It evolved project, industrial manufacturers used a batch in response to the need for economic large-scale purification process that destroyed or inact manufacture of commodity products. The con- vated about two-thirds of the penicillin pro- cept of unit operations held that any chemical duced. Within 7 months of their involvement, manufacturing process could be resolved into a chemical engineers at an oil company (S!aell coordinated series of operations such as pul- Development Company) had appliedtheir

21 12 t RO Vin.R S IA ( II!. %II('I. h: Wercl,i;RI V6

knowledge of generic engineering principles to chemical engineering curriculum in its present build a ful:y integrated pilot plant that processed form. Perhaps more than any other develop- 200 gallons of fermentation broth per day and ment, the core curriculum is responsible for the achieved nearly 85 percent recovery of penicil- confidence with which chemical engineers in- lin. When this process was installed by four tegrate knowledge from many disciplines in the penicillin producers, production soared from a solution of complex problems. rate in 1943 capable of sustaining the treatment The core curriculum provides a background of 4,100 patients per month to a rate in the in some of the basic sciences, including math- second half of 1944 equivalent to treatments for ematics, physics, and chemistry. This back- nearly 250,000 patients per month. ground is needed to undertake a rigorous study A second challenge in getting penicillin to the of the topics central to chemical engineering, front was its instability in solution. A stable including: form was needed for storage and shipment to hospitals and clinics. Freeze dryingin which multicomponent thermochnamics and ki- the penicillin solution was frozen to ice and netics, then subjected to a vacuum to remove the ice transport phenomena, as water vaporseemed to be the best solution, unit operations, but it had never been implemented on a pro- reaction engineering, duction scale before. A crash project by chem- process design and control, and ical engineers at MIT during 1942-1943 estab- plant design and systems engineering. lished enough understanding of the underlying phenomena to allow workable production plants This training has enabled chemical engineers to be built. to become leading contributors to a number of interdisciplinary areas, including catalysis, col- The Engineering Science Movement loid science and technology, combustion, electrochemical engineering, and polymer science The high noon of American dominance in and technology. chemical manufacturing after World War II saw the gradual exhaustion of research problems in conventional unit operations. This led to the A NEW PARADIGM FOR CHEMICAL rise of a second paradigm for chemical engi- ENGINEERING neering, pioneered by the engineering science Over the next few years, a confluence of movement. Dissatisfied with empirical descrip- intellectual advances, technological challenges, tions of process equipment performance, chem- and economic driving forces will shape a new ical engineers began to reexamine unit opera- model of what chemical engineering is and what tions from a more fundamental point of view. chemical engineers do (Table 2.2). The phenomena that take place in unit opera- A major force behind this evolution will be tions were resolved into sets of molecular events. the explosion of new products and materials Quantitative mechanistic models for these events that will enter the market during the next two were developed and used to analyze existing decades. Whether from the biotechnology equipment, as well as to design new process dustry, the electronics industry, or the high- equipment. Mathematical models of processes performance materials industry, these products and reactors were developed and applied to will be critically dependent on structure and capital-intensive U.S. industries such as com- design at the molecular level for their usefulness. modity petrochemicals. They will require manufacturing processes that can precisely control their chemical composition THE CONTEMI ORARY TRAINING OF and structure. These demands will create new CHEMICAL ENGINEERS opportunities for chemical engineers, both in product design and in process innovation. Parallel to the growth of the engineering A second force that will contribute to a new science movement was the evolution of the core chemical engineering paradigm is the increased

2'4 WHIT IS CHEMICAL IAGI.V1.1.11ING? 13

TABLE2.2Enduring and Emerging Characteristics of Chemical Engineering Endunng Charactenstics Emerging Charactenstics Serves industries whose products remain unchanged on Serves industnes whose products are quickly superseded tLe market for long penods in the marketplace by improved ones Serves industnes that compete mainly on the basis of Serves industries that compete on the basis of quality and price and availability product performance

Expertise in the manufacture of homogneons materials Expertise in the manufacture of composite and structured from small molecules matenals from large molecuk s Exp,:rtise in the manufacture of commodity materials Expertise in the manufacture of high-performance and specialty materials Expertise in process design Expertise in designing products with special performance characteristics Expertise in designing large-volume processes Expertise in designing small-scale processes Expertise in designing continuous processes Expertise in designing batch processes Expertise in designing industrial plants dedicated toa Expertise in designing flexible manufactunng plants single product or process

Practitioners use simple models and approximations to Practitioners use more complete models, better solve problems approximations, and large computers to solve problems rigorously Practitioners have access to only a few simple analytical Practitioners have access to many sophisticated analytical instruments instruments Practitioners build their careers around a single product Practitior -rs have multiple career changes line or process

Academic research is mostly performed by sing!° pnncip.i Academic research is also performed by multidisciplinary investigators within chemical engineering departments groups o' principal investigators, sometimes in centers or other organizational environments Research and education focus on the mesoscale Research and education also include the microscale (equipment level) (molecular level) and microscale (systems level)

competition for worldwide markets. Product The third force shaping the future of chemical quality and performance are becomingmore engineering is society's increasing awaren.ss of important to global competition than ever be- health risks and environmental impacts from fore. If the United States is to remain compet- the manufacture, transportation, use, and ulti- itive in world chemical markets, it must find mate disposal of chemicals. This will be an new ways to lower costs and improve product important source of new challenges to chemical quality and consistency. Similarly,a strong engineers. Modern society will not toleratea domestic energy-producing industry is needed continuing o' currence of such incidentsas the to preclude foreign domination of thisvital release of methyl isocyanate at Bhopal (in 1984 sector of the economy. The key to meeting and the contamination of the Rhine (in 1986). these challenges is innovation in process design, It is up to the chemical engineering profession control, and manufacturing operations.It is to act as the cradle-to-grave guardian for chem- particularly important that the United States icals, ensuring their safe and environmentally maintain a vigorous presence in commodity sound use. chemical markets. Commodities are at the base The fourth and most important force in the of industries that employ millions of Americans, development of tomorrow's chemical engineer- provide basic neces rtios for our society, and ing is the intellectual curiosity of chemical generate valuable export earnings. Thriving engineers themselves. As they extend the limits commodity businesses are also vital to specialty of past concepts and ideas, chemical engineering chemical businesses. The technical expertise researchers are creating new knowledge and and financial r sources that commodiiit.spro- tools that will profoundly influence the trainirg vide is crucial to the long-term research and and practice of the next generation of chemic...! development efforts that specialties require. engineers.

23 14 FRONHEIV, I\ CHEMICAL ENGLNLERING

When a discipline adopts a new paradigm, studies of the atieroscale dynamics of complex exciting things happen, and the current era is liquids (see Chapter 5). probably one of the most challenging and po- Thus, future chemical engineers will conceive tentially rewarding times to be entering chemical and rigorously solve problems on a continuum engineering. How can the unfolding pattern of of scales ranging from microscale to macroscale. change in the discipline be described? They will bring new tools and insights to re- The focus of chemical engineering has always search and practice from other disciplines: mo- been industrial processes that change the phys- le.:ular biology, chemistry, solid-state phys;cs, ical state or chemical composition of materials. materials science, and electrical engineering. Chemical engineers engage in the synthesis, And they will make increasing use of computers, design, testing, scale-up, operation, control, and artificial intelligence, and expert systems in optimization of these processes. The traditional problem solving, in product and process design, level of size and complexity at which they have and in manufacturing. worked on these problems might be termed the Two important developments will be part of mesoscale. Examples of this scale include re- this unfolding picture of the discipline: actors and equipment for single processes (unit operations) and combinations of unit operations Chemical engineerswill become more in manufacturing plants. Future research at the heavily involved in product design as a com- mesoscale will be increasingly supplemented by plement to process design. As the properties of studies of phenomena taking place at molecular a product in performance become increasingly dimensionsthe microscaleand the dimen- linked to the way in which it is processed, the sions of extremely complex systemsthe ma- traditionaldistinction between product and croscale (see Table 2.3). process design will become blurred. There will Chemical engineers of the future will be be a special design challenge in established and integrating a wider range of scales than any emerging industries that produce proprietary, other branch of engineering. For example, some differentiated products tailored to exacting per- may work to relate the macroscale of the en- formance specifications. These products are vironment to the mesoscale of combustion sys- characterized by the need for rapid innovation, tems and the microscale of molecular reactions as they are quickly superseded in the market- and transport (see Chapter 7). Others may work place by newer products. to reline the macroscale performance of a com- Chemical engineers will be frequent partic- posite aircraft to the mesoscale chemical reactor ipantsinmultidisciplinary research efforts. in which the wing was formed, the design of Chemical engineering has a long history of the reactor perhaps having been influenced by fruitful interdisciplinary research with the chem-

TABLE 2.3Microscale-Mesoscale-Macroscale: Illustrations Mit,roscale (s10-' m) Atomic and molecular studies of catalysts Chemical processing in the manufacture of integrated circuits Studies of the dynamics of suspensions and microstructured fluids Mesoscale (10-3-10, m) Improving the rate and capacity of separations equipment Design of injection molding equipment to produce car bumpers made from polymers Designing feedback control systems for bloreactors Macroscale (>10 m) Operability analysis and control system synthesis for an entire chemical plant Mathematical modeling of transport and chemical reactions of combustion-generated air pollutants Manipulating a petroleun reservoir during enhanced oil recovery through remote sensing of process data, development and use of dynamic models of underground interactions, and selective injection of chemicals to improve efficiency of recovery

24 4 IIAT LS CHEMICAL ENGINEERING? 15

ical sciences, particularly in industry. The po- are relatively immune to changes in field of sition of chemical engineering as the engineering applicationmust remain constant if chemical discipline with the strongest tie to the molecular engineers of the future are to master the broad sciences is an asset, since such sciences as spectrum of probleLis that they will encounter. chemistry, molecular biology, biomedicine, and At the same time, the way in which this philos- solid-state physics are providing the seeds for ophy finds concrete expression in course offer- tomorrow's technologies. Chemical engineering ings and requirements must be responsive to has a bright future as the "interfacial discipline" changing needs and situations. that will bridge science and engineering in the multidisciplinary environments where these new NOTE technologies will be brought into being. 1. Additional background and references on chemic^' engineers and the effort to win World War II may Some things, though, will not change. The be found inSeparation and Purification: Critical underlying philosophy of how to train chemical Needs and Opportunities(Washington, D.C.: Na- engineersemphasizing basic principles that tional Academy Press, 1987), pp. 92-100.

25 THREE

Biotechnology and Biomedicine

Advances in molecular biology and medicineare spawning new technologies andnew opportunities for chemical engineer s. Potentialareas for contributions to human health include the design andmanufacture of artificial organs, diagnostic tests, and therapeuticdrugs. In agricalturc, the man4acture of veterinarypharma- ceuticals and the scalingup of plant cell-culture techniques represent new applications for chemicalengi- neering principles. Other opportunitiesinclude using genetically engineered systems for the synthesis ofchem- icals and the biological treatment ofwaste. This rich potpourri of technological possibilities has attractedin- tense interest on the part of the United States' technological competitors. Theyare putting in place substantial research programs and facilities to exploit the potential of biotechnology. This chapter describesintellectual frontiers that chemical engineers shouldpursue. They include modeling of fundamental biological interactions, investigating surface and interfacial phenomena important to engineering design in living systems, expanding the scope of process engineering into biologicalsystems, and imnducting engineering analysis of whole-organor whole-body systems. Implications of thesenew challenges for chemical engineering research andeducation are discussed.

26 1$ FRONHERS IN CHEW(' il, ENGINEERING

MERICA LEADS the world in the bio- ical engineer, one with a solid foundation in the sciences, thanks largely to 25 years of life sciences as well as in process engineering major support for fundamental research principles. This engineer will be able to bring by the federal government. This research in the innovative and economic solutions to problems "new" biologyaspects of which are popularly in health care delivery and in the large-scale known as biotechnologyis providing the basis implementation ofadvances in molecular biology. for revolutions in health care, agriculture, food The biologically oriented chemical engineer processing, environmental improvement, and will focus on areas ranging from molecular and natural resource utilization. The new technol- cellular biological systems (biochemical engi- ogies that will be made possible by advances in neering) to organ and whole-body systems and the biosciences, and particularly in molecular processes (biomedical engineering). Biochemi- biology, will be applied to the search for solu- cal engineers will focus on the ...agineering tions to come of the world's most pressing problems of adapting the "'new" biology to the problems They will, in addition, create new commercial production of therapeutic, diagnos- industries and spur economic growth. Estimates tic, and food products. Biomedical engineers of the potential annual market for new products will apply the tools of chemical engineering from these technologies range from $56 billion modeling and analysis to study the function and to $69 billion for the year 2000 (Table 3.1). response of organs and body systems; to elucidate the transport of substances in the body; and to design artificial organs, artificial tissues, CHALLENGES TO CHEMICAL ENGINEERS and nrostheses. These exciting opportunities for The commercialization of developments in chemical engineers are described in more detail biotechnology will require a new breed of chem- below, first in terms of the potential impact on

TABLE 3.1Estimated World Markets for the Products of Biotechnology (millions of dollars). Year Market 1985 1990 2000 Medical products Pharmaceuticals 3,500 20,000-30,000 Diagnostics 100 1,500 5,000 Vetennary products 100 1,500 5,000 Other (materials, sensors, etc) -/5 500 Chemicals Fine and specialty chemicals 500 2,000-4.000 Commodity chemicals 1,000 Agricultural products Chemicals and biologicals 20 500 2,000 Plants and seeds 2! 1,500 5,000-6,000 Improved animal breeds 20 500 1,000 Food and rnimal feed products Additives and supplements 200 1,500 4,000 Flavors and fragrances 10 100 500 Associated equipment and 1,500 4,000 10,000 engineering systems TOTAL 1,975 15,175 56,000-69,000 4 Dollar values are at manufacturer's level. Inflation is estimated at 6 to 8 percent per year. SOURCE: SRI International.

2' 4 BIOTECILNOL0(,1 1:11) BIOMEDI( INT. 19

the first successful artificial or- ganthe artificial kidneywas Improving Platelet Storage the result of an innovative NIH program in the early 1960s that Trauma, leukemia, and hemophilia patients commonly require brought together an interdisci- Irdusions of platelets to control bleeding. These platelets are plinary team of chemical engi- obtained from separation from donated whole blood and we stored neers, materials scientists, and in special plastic storage bags. Using current storage methods, physicians. Chemical engineers these platelets survive only 3 days. This results in a chron: applied the fundamental con- shonege of platelets, particularly alter weekends or holidays when cepts of fluid mechanics, mem- donations decline. A program that has applied chemical engi- brane transporttheory,mass neering principles to this problem has demonstrated a new material transfer, and interfacial physical for storage bags that can keep platelets viable for more than a week. chemistryto systems design. The chemical engineering approach began with an analysis of They developed predictive cor- the biochemistry of platelet metabolism. Like many cells, platelets relations relating the blood-clear- consume ghooee by two pathways, en oxklative pathway and an ance performance of a dialyzer anaerobic pathway. The oxidative pathway produces carbon diox- to operating parameters such as ide, which makes the solution containing the platelets more acidic membrane area, channel dimen- (low pH) and promotes anaerobic metabolism. This second sions, blood and dialysate flow metabolic pathway produces large amounts of tactic acid, further rates, pressure drop in the sys- lowering pH. The drop in pH from both pathways kills the platelets. tem, and temperature. Within 5 The chemical engineering solution was to design a new material years,severalsoundlyengi- for the storage bag that was canable of "breathing"of allowing neered prototype systems, using carbon dioxide to diffuse out and oxygen to diffuse in. This prevents disposable membrane cartridges the drop in pH. Platelets stored In this new bag survive 10 days and sophisticated monitoring and or more. control equipment, were in advanced stages of development. By the mid-1970s, hemodialysis society and then in terms of intellectual frontiers had graduated from an experimental procedure for research. to a well-established, re';able, and safe means of sustaining patients suffering from acute and chronic renal failure. Today, hemodialysis and Human Health its companion process, hemofiltration, are stan- Chemical engineers are needed to help trans- dard hospital and clinical procedures and are form the results of basic health research into responsibL for major reductions in mortality practical products. They have been instrumental and morbidity due to kidney failure (Plate 1). in designing processes for the safe and econom- The success of the artificial kidney can be ical production of extremely complex therapeu- attributed to the relative simplicity of its task. tic and diagnostic agents (e.g., insulin and hepati- Unwanted substances are removed through a tis-B surface antigen). The insert boxes in this membrane separation carried out in a device chapter on platelet storage (p. 19), tissrplas- external to the body. Some of the targets for minogen activator (p. 21), interferons (p. 29), future artificial organs, such as the pancreas and kidrry f'jnction (p. 32) illustrate the signif- and the liver, are much more complex systems icance of chemical engineering research in this in which significant numbers of chemical reac- area. tions are carried out. In these cases, replace- ment might take the form of hybrid artificial Artificial Organs, Artificial Tissues, and organs containing living and functional cells in Prostheses an artificial matrix. Development of such sys- Chemical engineers can also make an impor- tems will be 0W :catty dependent on the contri- tant contribution to the development of art!ncia butions of chemical engineers to interdisciplin- organs, artificial tissues, and prostheses. In fact, ary teams.

r P. Rs r r rrr oil ir I wax! I R/\(, 20

The concept of :he artificial pancreas illus- diagnostic systems and devices Molecular biologists have discovered or created a variety of trates how chemical engineers candeelop new antibodies that arc artificial or semiartificial organs, parilarlyif enzymes and monoclonal -liseases, they are grounded in whole-organ physiology car...ble of detecting a wide range and biochemistry and capable ofcommunicating disorders, and genetic defects. Chemical engi- these materials fluently with endocrinologists and physiologists neers are needed to incorporate A chemical engineer working alone might con- into devices and systems that are fast,inexpen- ceive of an implantable power-driven insulin sive, accurate, and not susceptible to error on pump. for instance, controlledby feedback from the part of the person carrying out the test. For an electronic glucose sensor.In talking with an example. although an enzyme-linked immuno- endocrinologist, the engineer might devise an sorbant assay (ELISA) exists for detecting the implantable device containing viable pancreatic antibodies to cytomegalovirus (CMV) in blood islet cells that functions as a normal pancreas. samples, it cannot be reliably used in practice Working with a subcellular physiologist and to follow the course of a new CMVinfection. having enzymologist. the chemical engineer might come The error introduced into the test by up with what is, in effect, anartificial islet cell different operators perform it on each new blood highly a "smart membrane" devicethat senses blood sample in the series is sufficient to render questionable the interpretation of trends in the glucose levels and response releases insulin the magnitude from a reservoir encapsulated by the membrane series, particularly if changes in able Each of these design concepts is potentially use- he result are small. It is important to be antibodies because ful: the one that ultimately succeeds in practice to follow trends in CMV will be the one that is easiest to make. most CMV infections can be life-threatening to in- reliable, and most durable under the actual condi- dividuals with compromised immune systems, single tions °fuse. The wide choice of options andalter- and congenital CMV infections are the natives makes this field of research particularly largest cause of birth defects exciting and rewarding for chemical engineers. Chemical engineering research leading to the Artific;11 organs that perform the physical and design of devr es and systems that are fast and biochemical functions of the heart, liver, pan- "accurate" includes the following. creas, or lung are one classof organ replace- adsorbent, ments. A rather different target of opportunity developmentofselectively functionalized porous media to which inimu- is the development ofbiological materials that amenable play a more passive role in the body; for example. noreagents can be affixed and that are to speedy optical assay after contactwith body biocompatible polymer solutions whose fluids; rheological proper*- lake them suitable as design of fluid-containing substrates that replacements for synovial fluids in Joints or the allow small volumes of test fluids to contact aqueous and vitreous humors inthe eye: reagents efficiently and with highlyreproducible temporary systems that stimulate the re- assay respoo,e; and generation of lost or diseased body massand design of flexible manufacturing systems to then are absorbed or degraded by thebody make the wide variety cf expensive monoclonal (e.g.. an artificial "second skin" forburn pa- antibodies needed for ciagnostic test kits. tients): and electrochemical signal transduction sys- Ch,:mi-al engineers at several pharmaceutical tems that would allow the body's nervous ,ys- firms are using hollow fiber reactors to grow tem to-inmunicate with and control muscu- monoclonal antibody-producing hybridomas in loskeletal prostheses an in vitro hatch process.Research -)n reactor design to optimize the production of monoclonal Diagn0A !I( S antibodies will have a significant impact on the di- A second area rich in opportunities forchem- future development, economy, and use of ical engineers is the design and manufactureof agnostic tests

2 1 RO\ /// RS I\ (III 11/l 11, /. \ (,/ \ / / RIV,

PIATE IThe kidney dialysis machine (artificial kidney) is responsible for major reductions in deaths and adverse health consequences from kidney failureIts development required a team effort that brought together chemical engineers, physicians, and materials scientists The design of the disposable filter cartridge, shown attached to the front left side of the dia,ysis machine. was a major contribution by chemical engineers to the project Courtesy. National Institute of Di thetes and Digestive and Kidney Diseases

-4_

PLATT, 2 these crystals of human insulin are made by bacteria whose genetic Instructions have been altered using recombinant DNA techniques Human insulin is needed by diabetics who develop allergies to the animal- derived insulin that has been used to treat the disease since 1921 Without chemical engineering contributions such as process design and punfication technology. though. the large-scale production of human insulin would not he possible Courtesy. Eli Lilly and Company

3(1 i Al,\ Hiit',

PLATE 3Cell culture of plants now takes plav: in individual containers cared for by hand. Automation of plant cell culture using chemical engineering techniques could improve the yields and economics of plant cell culture and expand the range of its applications in the Production of aeNN species and hybrids Courtesy. Monsanto Company

P1,1-if 4Iiioreactors that use mammalian cells. likethis tower fermentor are on the cutting edge of nokbiotech - nology manufacturing processes Courtcsi., ( etasCorpo- ration BIOTECHNOLOGY AND BIOVEDICINE

will depend on precisely de- Tissue Plasminogen Activator: Superior signed three-dimensional config- Therapy For Heart Attacks urations. These configurations can be most easily created by syn- Many serious health problems result from abnormally located thesizing the compounds 1:colog- blood dots: heart attacks (dots In coronary arteries), pulmonary wally or from biologically de- embolism (clots in the lungs), and peripheral arterial occlusion rived precursors, using cells that and deep vein thrombosis (clots in the limbs). Each year heart have been altered through re- attacks alone afflict over a thill!In peode in the United States, and combinant DNA techniques (Plate almost half of them die as a result. 2). The manufacture of these In the past, only two treatments have been available for breaking compounds, examples of which down blood dots: streptoldnase and urokinase. Both treatments are listed in Table 3.2, will entail lack specificity for clots, so they can cause a general breakdown new challenges for chemical en- tithe hemostatic system, sometimes leading to generalized bleedirvi. gineers. For processes involving Recently, a superior therapy has been approved for use by the bacteria oryeastasproduct federal government: tissue-type plasminogen activator (tPA). This sources, the manufacture of mol- naturally occurring enzyme dissolves blood clots as part of the ecules with the correct three- normal healing process. By administedng relatively large quantities dimensional configuration may of it, act breakdown time can be shortened from about a week to require additional steps to mod- under an hour. ify or refold the proteins. Pro- Normal circulating levels of tPA are low, so that to accomplish cesses involving plant and mam- this dramatic clot breakdown one would need the amount of tPA rn 'Ilan tissue cultures as product contained in 50,000 liters of blood. This is clearly not practical. sources will require new types Instead, the molecule has been cloned and expressed in mam- of reactors capable of growing malian cells so that it can be produced in quantity. Using cells the specialized cells, control pro- from mammals, rather than bacteria, results in a product molecule cedures and sensors tailored for that has the same folding, Internal bonding, and coat of sugar biological processing, and ex- residues as the natural protein. tremely special and gentle puri- Producing the kilograms of tPA necessary to satisfy the world's fication procedures to ensure that therapeutic needs requires the special skills possessed by modem products of adequate purity can biochemical engineers. Sophisticated engineering of the fermen- be produced without chemical tation vessels, culturing conditions, and media compositions is change or loss of configuration. required to culture thousands of liters of mammalian cells. In These are formidable engineer- addition, new extremes of purity must be achieved in order to ing problems. Chemical engi- assure the safety of proteins derived from mammalian cells. The neers, long involved in the man- cost of the starting materials and the capacity constraints of the ufacture of antibiotics, peptides, present-day equipment require that yields from each fermentation and simple proteins, have signif- batch be as high as possible. icant experience to apply to these The current cost per dose of tPA (about $1,000) has already problems. emerged as an important barrier `3 its widespread use in hospitals Providing new modes of deliv- and clinics. Continued research in chemical engineering will be ering drugs presents almost as crucial to finding more economical processes for the production important an opportunity as proof this breakthrough therapeutic. viding new ways of making them. The standard practice or period- ically administering drug doses can k adto initial concentrations in the body Preventing and Curing Disease that may be sufficiently high to induce undesir- The biological activity of many of he next able side effects. Later, as the drug is metab- generation of compounds needed to prevent olized or eliminated, its concentratit n can drop disease (e.g., vaccines) or to cute it (e.g., drugs) below the effective level (Figure 3.1). This 22 RON/I/RS 1\ ( /// i11( I/ I \ (,//./?/(/

TABLE3.2Important Therapeutic Targets of Opportum'y Therapeutic Action

Antigens Stimulate antibody response Interferons Regulate cellular response to viral infections and cancer proliferation Tissue plasminogen anvators Stop thrombosis by dissolving blood clots Human growth hormone Reverse hypopituitansm in children Neuroactive peptides Mimic the body's pain-controlling peptides Regulatory peptides Stimulate regrowth of hone and cartilage Lymphokines Modulate immune reactions Human serum albumin Treat physical trauma Gamma globulin Prevent infections Antihemophik factors Treat hereditary bleeding disorders Monoclonal antibodies Provide site-specific diagnostics and drug delivery

problem is particularly important with drugs or taped to the skin (Figure 3.2). Chemical that are metabolized or eliminated rapidly from engineers have been instrumental in designing the body and with drugs that have a narrow and manufacturing volymers that are capable therapeutic range (the span between the thera- of such controlled release over long periods of peutically effective and the toxic concentra- time. tions). The optimal pharmacological effect can Another approach to delivering drugs is to sometimes be attained by establishing and main- target the administration of a drug to a specific taining a steady-state concentration of the drug site in the body. This might be accomplished or by time- senuencing its administration.The by coupling a drug to an antibody that has been controlled release of short-half-life drugs over cloned to attack a specific receptor at the disease a long period of time can be effected by admin- site. This approach would make possible, for istering the drug through low-flow pumps, as a example, he selective exposure of tumor-bear- mixture of capsules that disintegrate at different ing tissues to high concentrations of toxic drugs. rates, or in pouches inserted under the eyelid Chemical engineers are needed to produce such targeted drugs and to elucidate the kinetics of monoclonal antibody transport through 'he body to target sites. z TOXIC 0 Other areas in therapeutics that are ripe for LEVEL interdisciplinary collaboration include the design of special-purpose pumps and catheters, sterile implants that allow access from outside the body to veins and body organs, and imaging 0 techniques for monitoring drug levels. Efforts

MINIMUM by chemical engineers to provide improved data EFFECTIVE LEVEL acquisition and quantitative modeling of phar- macokinetics can lead to the design of better drug administration procedures and better tim- TIME ing to maximize the delivery of drugs to the organs that need them while minimizing the FIGURE 3.1When a tablet of medicine is taken. or an injection given, sharp fluctuations of drug levels in the body exposure of other organs. can resultAt the peak level. indesirable side effects of the drug can manifest themselves Unless the tablet or injection is given very frequently, the level of the dnig in Agriculture the body can fall ton low to be effective Chemical engineers Major opportunities exist for chemical engi- are working on ways to deliver drugs that maintain asteady, effective level of the drug in the body neers to help develop agricultural biochemicals. BIOTECHNOLOGY AND B1011EDICII:

pharmaceuticals and vaccines, and the requirements for chemical engineering expertise are similar. A third focus is the development of large- scale plant-cell culture techniques. These technique, convert undifferentiated cell clumps into differentiated cells of genetically selected roots and stems ready for planting. Such plant cell clones are already being used to produce new crop varieties that are more resistant tc adverse environmental conditions or disease. Examples include disease-resistant trees and virus-free potatoes. Cell culture techniques will continue to be used to increase crop productivity by allowing horticulturists to propagate quickly new plant strains showing increased resistance to pests, drought, or soil salinity; higher productivity or enhanced growth FIGURE 3.2This transdermal (through the skin) product delivers a steady level of nitroglycenn to the body, pre- rates; venting the pain of angina The thin, adhesive unit admin- ability to produce increased amounts or isters the drug directly to the bloodstream when applied to higher quality of seed proteins and other plant the skin This once-a-day patch provides medicatinr: without pioducts such as alkaloids, carotenes, latex, interfenng in a patient's daily activities or .,thouthaving to take pillsseveral times a dayIt does not require and steroids; and punctunng the skin with a needle Chemical engineers are improved efficiency of nitrogen fixation and Involved in the design and manufacture of new polymer photosynthesis. systems for medical applications such as thisCourtesy A LZA Corporation At present, cell culture work is done mostly by hand by horticulturists in large greenhouses (Plate 3). Chemical engineers could greatly in- These opportunities roughly parallel the fron- crease the usefulness of this method of plant tiers that have opened up in the human health propagation by developing efficient automated area. In agriculture, a deeper understanding of processes for producing plants from cloned biological processes in plants has paved the cells. way for biologically derived fungicides and herbicides that are highly potent, species spe- Biochemical Synthesis cific, and environmentally safe. The rapid introduction of these compounds into widespread By manipulating the genetic machinery of the use will require expertise in process design. cell, it is possible to cause mo%t cellular systems process control, and separation technology to to produce virtually any biochemical material. ensure that they are manufactured free from Unfortunately, the growth of cellular systems contaminants that would threaten the environ- (particularly in tissue cultures) is constrained ment or worker safety. by end-product inhibition and repression; hence, A second foclis for chemical engineers in it is difficult to produce end products in high agriculture is the improvement of ve,,rinary concentration. Furthermore, cells are always pharmaceuticals (e.g., peptide horrnon, .t grown in aqueous solution, so biochemicals promise to stimulate growth, fecundity, and produced by cellular routes must have intrin feed efficiency in farm animals) and vaccines. high value in order for the cost of recovery The prospects for improvement of these com- from dilute aqueous solution to be minimized. pounds parallel the bright rospects for human Thus, most biochemicals of commercial interest 24 FRONTIERS IN CHEMICAL ENGINEERING to be produced by biotechnology will be high-value products such High-Fructose Corn Syrup: Biotechnology on as enzymes, biopolymers, or a Billion- Dollar Scale metabolic cofactors. In general, their potency is so high that only When you crack open a can of Coca Cola or Pepsi, you are small quantities will be needed. tasting some of the fruits of biochemical engineering, Most non- Accordingly, the challenge to diet soft drink. sold in the United States we vises led with high- chemical engineers in producing fructose Olin syrup (HFCS), a substitute for the neural sugar that these products is not so much ir. comes from cane and beets. HFCS, produced by an enzymatic process scale-up but rather in reaction, is an example of the surmised application of chemical obtaining high process yield and engineeringprindpies to bioehernical synthesis. So successful, in minimal process losses. fad, that more than $1.5 billion of HFCS was sold in the United Enzymes are an important class States last yeti'. of biochemicals; they are the ca- To make HFCS a commerde reality, Wro *spirals bioprooseses talysts needed in the chemical had to be developed, 'Oiled up, and brought on One in a reaction cycles of living systems, manufacturing plant. The lirei biomass was a fannentation to and they execute their catalytic manufacture the necessary enzyme. The second process used role with exquisite chemical pre- the enzyme b convert de:dross to HFCS. The eery knrohrement cision. Enzymes have great po- of chemical engineers In the design of these proc:esses, and their tential in synthetic chemistry be- fruitful interaction with biologists, was a key lo the success of causetheycaneffectster- these two endeavors. eospecific reactions, avoiding the The fermentation for making isomerase enzyme is releively fast production of an unwanted iso- and can be carded out in a number of process configurations. mer of a complex molecule. Cur- Basic to all of the process configurations are the problems of rently, many of the enzymes used n'Aintakling sternly and containment; engineering heand mass inindustrial processing (e.g., transfer; controlling levels of 0 and CO2 in the fermentation those used to convert starch into solution; and regulating temperature, pressure, and levels of sugar or milk into cheese) are dextroee. The sotution must be carefully mixed during the fermen- derived from microbial sources teflon; damage b the cells by agitation can either mechanically because they are beyond the kill the microorganisms that produce the enzyme or hopeiftealy practical reach of current syn- complicate the recovery of tne enzyme from the fermentation thetic chemical technology. Bio- broth. Chemical engineers we sidled in solving all of these problems. technology offers the potential, The conversion of dextrose syrups to high-fructose corn syrups through cellular genetic control, for making enzymesnot only those that are now used industrially but also by traditional synthetic chemistry. Addressing others for new uses in synthetic chemistry. The this challenge will bring the chemical engineer synthesis and processing of these complex mol- into close contact with hiochemists and syn- ecules require conditions that will maintain their thetic chemists. specific three-dimensional structures. One challenge for chemical engineers will be to develop Environment and Natural Resources processes that can meet the rigorous requirements for optimally producing and recovering Biotechnology offers promise for improving enzymes. the quality of our environment through the intro- Another challenge will be to understand the duction of new microbial and enzymatic tech- chemical transformations that enzymes cata- niques for removing and destroying toxic pollu- lyze. The goal would be to determine how these tants in municipal and industrial wastes. This transformations can be used or tailored through opportunity is discussed in detail in Chapter 7. changes in enzyme structure to produce com- The depletion of domestic high-grade ore pounds that are difficult or costly to produce deposits has made the United States vulnerable

3,) BIOTECHNOLOGY AND BIOMEDICLVE 25

when the earth itself is the bioreactor. requires two additional chemical engineering steps. The first is the Another opportunity for bio- rigorous purification of the dextrose to remove any contaminants technology may be to provide a that could inactivate the isomers* enzyme. The dextrose syrup new source for certain petro- is dgerthilly demkieralized, filtered, and Mined over carbon, chemicals. Biological routes to a treetedvsith- a magnesium op-catalyst, and brought to the appro- number of organic chemicals cur- priate tviperature arid levelof acidity. The dextrose its then ready rently derived from petroleum for the second step. It is passed down a column i:ontainkig the have been demonstrated (Table Worm* enzyme isolated on a older. Enzyme loadings of over 3.3). For structurally complex 10 tidlikinunits of active enzyme per cubic foot are not uncommon. chemicals,theseroutes may The immerization process can betionducted in either of two ways, prove more economically effi- depenclingon market conditions. in the summer, when the demand cient than alternative routes (e.g., is gait, It is common to run dextrose through the columns as those using synthesis gas from quickly as pestle to supply the customer. This method of coal gasification as a starting ma- opera** maul % in higher column ieraperattgeS that shorten the terial). Whether this will be the Ile of the -enryme.,, The cokanns need to be changed more case depends largely on engi- frequently, but the market demand can be met without building neering research efforts'bio- mess minOlecturing capacity. In the winter, when the demand proccssing and in other resource is law,' the manufacturer mull we the lbw to get maximum enzy- areas. matic Illetthie and, corregiondingly; lowed operating costa En- zymic liana can be detrimentally _affected by poor control of INTERNATIONAL tempivalter and acidity, mat air or sow theoluble material that can COMPETITION plug the alum will shadow the lifetime of the column regardless of enly/110 activity. The economics of HFCS production, thus, are Who will lead the commer- based on the ability to Maintain stoat process control. cialization of the "new" biol- Thc record time in which HFCS was developed and brought to ogy? The answer is not yet clear. high levels of production and sale. is a testament to the versatility "Nr principal technological com- and pow of shendcal engineering principles. No new chemical petitors in the world, the Euro- engineering principles had to be discovered to make HFCS a peans and the Japanese, are ag- commercial reality. They were waiting to be applied to a biological gressively expanding their efforts system. to commercialize the results of basic biological research. West Germany, Japan, and the United Kingdom each have three large to shortages of metals (e.g., chromium, man- government-supported institutes dedicated to ganese, and niobium) that are important to the biotechnology. The United States has only one production of high-strength steel and other al- center of comparable magnitude (the MIT Bio- loys. Biological systems with a high affinity for technology Process Center). Not surprisingly, metals are known, and genetically engineered our competitors are establishing commercial microorganisms might be used to sequester positions by practicing effective and forward- metals from highly dilute waste streams (see looking biochemical and biomedical engineer- Chapter 6), from dilute sources underground ing. Some examples of their recent accomplish- (see Chapter 6), or from the sea. To make such ments attest to their aggressiveness. recovery concepts practical, chemical engineering will be needed to design systems that Basic technology for membrane separation allow these microorganisms to function opti- of biomolecules was invented in the United mally and to efficiently contact large volumes States, but the West Germans and the Japanese of dilute solutions, or, in the case of in-situ lead In its application to separations of enzymes metalsextraction,tooperateefficiently and amino acids from complex mixtures. Jap-

3 G 26 R011 II RS It ( 111111( 11 1 1 1 R1Vi,

TABLE 3 3Potential Routes to Commodity petition in biotechnology and medicine. The Chemicals by Microbial Fermentation of potential economic rewards for success are very Glucose great, as shown in Table 3.1First entry into these markets will be critically important in Chemical Microorganismts) international competition, and major shares in Ethanol Sas haromssesere vissac it-worldwide bioproducts market will be cap- lvnionionos mohdis tured by those countries who possess the needed Clostrultunt u, elubuts lu sun Butanol research infrastructure Adipic acid Pseuslomenas species Methyl ethyl ketone Klebssella pneumonia(' Glycerol .Sasharoniss es sere% Issue INTELLECTUAL FRONTIERS Dunahella species Citric acid Aspergsllus 'tiger The intellectual frontiers for chemical engi- be S(RCE T K Ng, R NIBusche, C C McDonald. neers in biotechnology and biomedicine can and R W F Hardy Sc lens e, 219, 1983, 7B Copyright described on a continuum from microscale 1983 by the AAAS Excerpted with permission through mesocale to macroscale. At either end of this spectrum are highly interdisciplinary research topics that will require modeling and anese government support of membrane sepa- analytical tools currently used by chemical en- ration research and development alone amounted gineers in other contexts. The important me- to $21 million in 1983. This is many times the soscale challenges of bioprocessing will require level of comparable effort expended by the U.S. chemical engineering expertise in reaction en- government, One impact of the well-funded gineering, process design and ccntrol, and sep- Japanese effort can be seer. in the increasing arations. The following sections discuss these number of Japanese kidney dialyzers appearing challenges in greater detail. in U.S. hospitals. Technology for very large (400,000-gallon) Models for Fundamental Biological continuous fermenters was developed and is Interactions being practiced in the United Kingdom. This development pushes biochemical engineering to The living microbial, animal, or plant WI can limits not yet explored in the United States. be viewed as a chemical plant of microscopic Although the use of fermentation to pro- size.It can extract raw materials from its duce ethanol is an ancient technology, more environment and use them to replicate itself as efficient immobilized-cell, continuous processes well as to synthesize myriad valuable products have been conceived, and Japan has established that can be stored in the cell or excreted. This the first demonstration-scale plant. microscopic chemical plant contains its own power station, which operates with admirably According to the Office of Technology As- high efficiency. It also contains its own sophis- sessment (OTA), Western Europe and Japan ticated control system, which maintains appro- have historically maintained a large and stable pnate balances of mass and energy fluxes through funding pattern for biochemical engineering. the links of its internal reaction network. This is not so for the United States. The existing Cell membranes are not simply passive con- base of biochemical engineers in other coun- tainers for the cell's contents Rather, they are tries, and their strong interest in exploiting the highly organized, dynamic, and structurally discoveries of the "new" biology, are reflected complex biological systems that regulate the by ex' ensive government funding and facilities transfer of specific chemicals through the cell support. It is clear that countries such as West wall. UN-many and Japan are laying a foundation of One important constituent of cell membranes engineering research and training as part of their is a class of moleculesthephospholipids overall strategy for intense international com- that spontaneously form two-layer films in a

3 1NI) 27 number of geometries. Many of the important of ions across membranes, antibody-antigen physical properties of cell membranes, such as interactions, cellular protein synthesis, and nerve two-dimensional diffusion and differentiation impulse transmission. Progress in achieving between the inside and the outside of a tube or similar efficiencies in engineered enzyme pro- sphere, can be studied with these spontaneously cesses, bioseparations, and information transformed structures. mission can be aided by acquiring more sophis- If we can develop accurate quantitative models ticated knowledge of biochemical processes at that simulate how cells respond to various interfaces. With this knowledge, such products environmental changes, we coon better utilize as synthetic antibodies for human and animal the chemical synthesis capabilities of cells. antigens, or synthetic membranes that can serve Steps toward this goal are being taken. Models as anificial red blood cells or transport barriers, of the common gut bacterium Escherichia coli could be developed. have been developed from mechanisms of sub- Surface interactions play an impo,tant role in cellular processes discovered or postulated by the ability of certain animal cells to grow and molecula: biologists. These models have pro- produce the desired bioproducts. An under- gressed to the point where they can be used standing of the dynamics of cell surface inter- with experiments to discriminate among pos- actions in these "anchorage-dependent" cells tulated mechanisms for control of subcellular (cells that function well only when attached to processes. a surface) will be needed, for example, to Some of the most promising potential appli- improve tl,e design of bioreactors for growing cations of biotechnology involve animal or plant animal cells. cells. Models for these organisms, which have Interactions at surfaces and interfaces also greater internal complexity as well as more play an essential role in the design and function demanding environmental requirements than of clinical implants and biomedical devices. simple cells, are not yet available. It will prob- With a few recent exceptions, implants do not ably be necessary to incorporate the structure attach well to tissue, and the resulting mobility of functional subunits of the cell (organelles) of the tissue-implant interface encourages chronic into models for complex cells in addition to the inflammation. The result can be a gathering of chemical structure that is used in bacterial cell platelets at the site, leading to a blood clot or models. Cellular reactions are subject to the to the formation of a fibrous capsule, or scar, limitations imposed by he laws of thermody- around the implant (Figure 3.3). namics, by diffusion, and by reaction kinetics. A number of fundamental questions about Chemical engineers are familiar with the tech- biological changes at the tissue-implant interface niques for designing mathematical models that challenge chemical engineers in the design of involve these parameters and should be able to medical implants and devices. How do cells make major contributions to the development interact with the surfaces of well-characterized of cellular models. The development of reliable materials? Which receptor sites on cell mem- models hinges on acquiring accurate data bases branes interact with which functional groups on on enzymes, biologically important proteins, the surfaces of biomedical materials? What is and cellular systems. The data should include the effect of othmorphological features of the physical properties, transport properties, chem- surface, or of the mechanical properties of the ical properties, and reaction rate information. material? How does the metabolic activity of the cell change after a reaction with a material interface? What new enzymes or chemicals arc Biological Surfaces and Interfaces produced by the cell after such a reaction? How Many biological .,.actions and processes oc does information gained in this area lead to cur at phase boundaries and are thus controlled better materials, or to the development of new by surface interactions. Examples include such methods for attaching biomedical materials to highly efficient processes as selective transport tissues? How can chemical engineers contribute

3 28 FAO17 IERS IA ( IMAM' 11, I.NGINTI,RING to better ways of monitoring implanted materials noninvasively'' .,) i

Bioprocessing / Three major intellectual fron- 4 tiers for chemical engineers in bioprocessing are the design of biore ctors for the culture of plant and animal cells, the development of control systems along 1. with the needed blosensors and 4 analytical instruments, and the Ilk development of processes for separating and purifying products. A critical component in I each of these three research areas is the need to relate the microscale to the mesoscale. FIGURE 3.3Implanted matenals and devices in the body that are perceived as foreign objects will encourage the formation of scar tissuesurrounding Bioreactors fOr them In this photomicrograph, a nonporous membrane (G), implanted between Manalatittring Prot e.s A CA the skin and some subcutaneous tissue ID and PC). generates the synthesis of granulation tissue (GT) and a fibrotic sac (scar tissue, FS) within 4 weeks Much of the early work in Courtesy. loannis Yannas, Massachusetts Institute of Technology applying recombinant DNA technology to the production of bioactive substances has used microbial cell systems to transfer the cells from one vessel to species such as bacteria, yeasts, and molds another. These microbes are fairly easy to manipulate Plant cells tend to aggregate, and large ag- genetically and are hardy under adverse con- gregates nose problems in maintaining a supply ditions. Unfortunately, animal or plant proteins of nutrients to all cells and in removing wastes. produced by clones of microbial cells often lack The development of bioreactors for plant cells the critical three-dimensional structure that is will require an understanding of limitations on formed when the same proteins are produced mass transfer in such aggregates. by animal and plant cells. For this reason, these Some bioreactoi systems must be completely proteins may not be biologically active even protected from microbial contamination, mean- thougL they have the correct sequence of amino ing that not a single alien bacterium or virus acids. One important future area of biotechnol- particle can be allowed to penetrate the system ogy lies in using plant and animal cells inplace Reliable and economical systems need to be of microbial cells. The large-scale use of plant developed to achieve this level of contamination and animal cells in tissue culture raises impor- prevention. Along with the need for prevention tant problems in the design and operation of is the need to be able to detect contamination bioreactors (Plate 4). at a level of a few microorganisms in a hundred One problem mentioned earlier is that cert _tin kiloliters of medium. This degree of detection animal cells are anchorage-dependent. Also, is not yet achievable. Research could vastly plant and animal cells are easily ruptured by improve the crude detection methods that are mechanical shear. Bioreactors for handling such used today. cells must be designed so that the contents of Most industrial bioprocesses are now oper- the reactor can he mixed without disrupting the ated in a batch mode Batch processing is the cells. A similar problem exists in the design of method of choice for small-scale production, BIOTECHNOLOGY AND BIOMEDICINE 29

tory, beer is still brewed and aged in batches. Attempts to use a Interforens continuous process to manufacture a product as well understood Cinder new arnicht one out of fax adults. One of the more as beer have not produced a pitecishiglharaples for coif*, .ifirkle of cancers involves the use beverage with acceptable taste ecilleeteleOt inninute quantities in the body wFr4filieerNPc '.'tkoitualevistern. Interferon Process Monitoring and In' Ohne of bansfonned Control sulk If ifs**, possible to culture . up,to Arisichandred 'milliliters. Continuous and detailed knowl- 401144, 4$111***Vik -#00ditiPOlote-br the aseptic edge of process conditions is nec- -104,00..,0010elarger, making it essary for the control and opti- PC0014.19.**** Op*Ocaittiehtl quantities. mization of bioprocessing oper- 'ealltiiiilluittrlie the dim- ations. Because of containment 10%40.71*-1111i itlepassible to achieve and contamination problems, this a #i holdup-Cell culture. Yet, the knowledge must often be ob- alOPYields of less than 5 tainedwithoutsamplingthe 'rthe cell culture was a process stream. At present. con- pratesf ii Ihist000p 4!ith:protolnetiochenietisuchinasdesioninter- ditionssuch astemperature, %MP: pressure, and acidity (pH) can ing ,and, thofe a ixualed coral to the be measured rapidly and accu- Prablint)71 tieettatinititialktutati, enroll -nat process that mini- rately. It is more difficult to mon- mized elleellate the- interferon to the protease. Chemical itor the concentrations of the engine:VS how* also developed doontatographic separation sys- chemical species in the reaction tems thdt IiiIpmite the yield by Sepai*tkig Interferon from the medium, to say nothing of mon- reactitincibdure quiddy Arid efficiently. itoring the cell density and intra- As medical researMere.continue to explore the therapeutic cellular concentrations of hun- properties of interferon., 'chordal engineers will cootinu1 to dreds of compounds. provide the expertise needed to make available the quantities of The development of rapid, these molecules necessary for &lice! En aluation. accurate, and noninvasive on- line measurement sensors and instrumentsisahigh-priority and it has the advantage that the equipment can goalinthecommercializationof biotech- be used for intermittent production of more than nology (Figure 3.4) Some of thes.: .nstruments one product. An intriguing future possibility is will build on analytical methods now used t;iat chemicals and biochemicals will be pro- in catalysis and other surface sciences, such duced by biotechnology on a large-scale, con- as tinuous basis. Continuous processing frequently offers advantages in economy and uniformity Fourier transform infrared spectroscopy, of product quality. However, the engineering fluorospectrometry, problems involved in converting from batch to mass spectrometry, continuous biological processing are not trivial. nuclear magnetic resonance (NMR) spec- Continuous processing of biological systems trometry, and places stringent demands on equipment design, combinations of some of the above-men- instrumentation, and operation for maintaining tioned methods with chromatography. aseptic conditions and biological containment. One indication of these difficulties is the fact These methods will be applied by chemical that although processes for fermenting natural engineers to monitor and control reaction and materials to produce beer predate written his- recovery systems

4 I RON /// RS 11 ( ///.11/( 1// N(iINI PRIV,

problems requires generic research on hig:ly selective separations, as well as on the problems of concentrating materials from very dilt te solutions (Figure 3.5). These and other generic research opportunities in separations have been described in detail in a recent report from the National Research Council.' Pursuing these opportunities will result in a better understanding of separation processes now used for the large- scale purification of proteins (e.g., precipitation and process chromatography). It may also result FIGURE 3.4Several approaches to developing analytical in novel separations involving aspects of tech- instrumentation for bioreactors are shown in this figure (I) Gases being fed to the bioreactor must be analyzed to niques such as determine then flow rate and composition Flow rates cat, be measured with mass flowmeteis or rotameters, the chromatography, concentrations of oxygen and carbon dioxide in the gas membrane separation, mixture can be determined by electrochemical methods or fractionation in electric and gravitational IR analysis, respectively. These data, when combined with similar measurements on the gases exiting from the bio- fields, reactor (2), provide information on oxygen uptake and immunoadsorption, carbon dioxide evolution in the bioreactor(3) Vanous extraction with supercritical fluids, sensors may be placed in the bioreactor. Properties that two-phase aqueou_ solution extraction, and might be measured Include temperature, pressure, pH, separation by use of microemulsions. dissolved oxygen, and liquid feed rates Sensors are under development '^ measure glu. Jse, ethanol, various ions (e g , NFL', , K', Na', a1id PO.' -), and other The development of such new separations is important bio.00lecules (e.g., ATP, ADP, AMP, DNA, crucial to the development of industrial bio- RNA, and NADH) (4) The rotating shaft of the impeller technology. can be used to measure viscosity. (5) Varous spectropho- Another approach to separation problems lies tometnc cells can be used to measure properties such as turbidity, if the culture medium in the bioreactor a not too in the development of modified organisms that denseFrom L E Enckson and G Stephanopoulos, produce the target proteins in high yield and "Biological Reactors," ch12 inChemical Reactwn and concentration, thus reducing the time and cost Reactor Engineering,J. J Carberry and A Varma (ed.; ), of separating the proteins from large amounts Marcel Dekker, Inc., New York, 1986 of water. This is an area where early involvement of chemical engineers in designing genetically engineered organism-, would be taluable. Separation of Bioproducts With their insights into the requirements of Cell culture bioreactors produce a dilute mix- downstream processing of biologically synthe- ture of cells in an aqueous medium. Recovery sized substances, chemical engineers could be of the product proteins from these cells may valuable members of an interdisciplinary team require disruption of the cells. This creates I of molecular biologists and biochemists seeking host of problems. Cell walls and organelles must to tailor the genetic code of cells. be removed. Proteins must be concentrated is mostly from a highly dilute solution that Engineering Analysis of Complex Biological composed of water and other small molecules. Systems The desired proteins must be separated from other macromolecules with similar physical The development ofnew therapeutic proce- properties. For biological'y active proteins, sep- dures will be aided by a better understanding arations must not only be specific for the target of physiological and pathological processes in proteins, but also gentle enough to pt.( ent the body. One area to which chemical engineers denaturation and loss of biological activity and can contribute is the application of engineering suitable for large-scale operation. Solving these analysis to systems found in the body. The

q BIOTECHNOLOGY AND BIOMEDICINE 31

tagenic alterations. pathological processes such as birth defects or cancer can result. We need a better basic understanding of this control process. Theoretical and systematic advances by chemical IC engineers in process control may be applicable to the study of this I problem. While the mechanical performance of artificial materials in the human body can be predicted with some reliability, forecasting their biological performance is difficult. The problem of interactions at surfaces has already been mentioned. Research frontiers also include FIGURE 3.5 The desired product, usually a protein, produced by a genetically engineered microorganism must be separated and purified before use. The developing ways to simulate in centrifuges shown nere separate the components of the microorganisms. and vivo processes in vitro and ex- then furtht; separation is tamed out to isolate the one protein that is desired tending the power and applic- from the thousands of othe- proteins produced by the microorganism. The ability of such simulations to Isolated protein must be rigorously purified to eliminate contaminants from allow for better pdiction of the final productIn many cases. separation and purification is the most exrensive part of the production process Ccurtesy. Genentech the performance of biomedical materials and devices in the patient. Fundamental information study of the transport of substances across on the correlation between the in vivo aid in membranes is an example. There is considerable vitro responses is limited. Chemical engineers knowledge of the transport of small molecules might also make contributions to the problem across living membranes; this should be ex- of noninvasive monitoring of implanted mate- tended to studies of larger molecules. A more rials. complete understanding of the transport of biologically active agents would be particularly IMPLICATIONS OF RESEARCH important in diagnosis and therapy. FRONTIERS Biochemical proces:,,ts in humans can now be measured by such techniques as positron The most successful efforts on problems such emission tomography, magnetic resonance im- as those listed above will come from a new aging, and x-ray computer-assisted tomography. breed of chemical engineer. fluent in the lan- and the measurements can be enhanced by guage and concepts of modern biology and digital subtraction methods. Chemical engineers medicine. Currently. few chemical engineers can help elucidate the data obtained by such are sufficiently knowledgeable in the principles techniques by developirquantitative models of modern molecular biology, microbiology, thatincorporate thermodynamics,transport genetics. and biochemistry to permit their ef- phenomena, fluid mechanics, and principles fective collaboration with life scientists Con- of chemical reaction engineering. These adversely, few life scientists are sufficiently aware vances will lead to improved therapeutic proof the engineering principles and practical prob- cedures. lems associated with the scale-up of biological The normal growth of tissues and organs is processes, the large-scale processing of bio- under a remarkable degree of natural control. products, or the development of artificia: bio- When this is compromised by genetic or mu- logical devices All the participating disciplines 32 FRONTIERS IN CHEMICAL ENGINEERING must recognize the importance of the innovative synthesis 3f Kidney Function Through new concepts that unite life sci- Research ence theory and fact with engineering principles, or that com- rOle in regulating the composition bine an engineering idea with a OkOtincti: hoer body fluids. In doing so, they help biological speculation. Suck in- "onfilliiiten*onment k the bcdy. The first step novative synthesis is likely to IS thailitratIon of blood across the wails of come about only in an environ- ceetelal*1$4301-reiels in the Iddney. This filtration ment where research needs and Waller and low-moleurlar-weigh sub- unsolved problems can be iecu- ill 0d to enter the renal tubules, while tifiedthatbridge disciplinary 441kanikamertkily plasma prokine within the boundaries and compel represen- bloothiffilikitthia100014kieelairkethilii, tailed *manlier caplet- tatives of all relevant disciplines aiklayisiel***111011tif tulle `a itiCti011 of a milikneker in to work together to find the best diamaisfitilik4111011114111)1001WYOlhealllionalioglorreituN solutions. Prompt and effective Present ineath*OleYilkitlniquelit reeks* of renel deem. exploition of the "new" biol- OUring thleafy tvroi-, techrisif innovetioi4 by renel-physiology is c pendent on the improve- °gists permitted the *at drool mernsUrements of the pressures ment cif this disciplinary inter- responsible for tiomeruler Malawi in mammals. This created the face; and this is one of the most opportunity to- use engineering principles in antty...ing kidney criticalproblemsconfronting resuNnirodahiiiii&firitweenPhislakiiiistailid bioengineering today. chemical wok:ears left to several major new insights. The need to develop a new Hydrudic permeabilitytile rate of filtration per unit of applied fusion with modern biology has pressure --was one of the phyrelal properties that were studied. important implications for chem- It was discovered that the hydraulic penneabdity of the glomerular ical engineering education and capillary -wag was twice that of capillaries in other organs. This research: surprising finding impliec that *lc !Milting factor in iddney function was not hydraulic permeability, as had been assumed, but rather The development of fruitful the Ms of plasma flow to the kidney. It was further discovered education and research programs that the design of the glomerulus ingeniously avoided one of the in biochemical and biomedical classic problems of man-made filtration processes: concentration engineering cannot take place in isolation from the life sciences; strong, complementary academic programs in plysiology should be part of the course require- the h;ological or medical sciences are essential. ments for chemical engineers specializing in Institutions that do not have strong research bioengineering. Such courses should be struc- activities in the life sciences should probably tured specifically for engineers, include mean- not be encouraged to develop programs in ingful laboratory experience, and provide the biochemical or biomedical engineering. prerequisite background for the engineering stu- Curricula at the undergraduate and gradu- dent to take advanced biology and nicdical ate levels need to be modified so that students science courses, if desired. will gain sufficient 1owledge of the biological Ph.D. students must be prepared for the sciences to apply engineering methods of anal- interdisciplinary environment in which they will ysis and design to solve problems that originate likely spend their careers as biochemical or in the biological science. Chapter 10 discusses biomedical engineer,. The best way to do this general principles for modifying the undergrad- is to expose therd to interdisciplinary research uate curriculum to respond to emerging appli- as graduate students. To facilitate this, a broad cations for chemical engineering. At the grad- and stable base of research support targeted at uate level, in-depth courses in molecular biology, interdisciplinary research must be created. Par biochemistry, and cellular and mammalian ticularl;' valuable would be support targeted to BIOTECHNOLOGY 4N!) BIOMEDICINE 33

training is initially in the medical and life sciences is one step that polarization, or the buildup at the surface of a filter of substances might be encouraged. The pres- fo0 larillf*.ilef!I through- ence of a strong research biology 'ffinglinniOnonwilly Produce large volumes of filtrate without department or a nearby' aching lignilicanftlaskitatitplaeMa protebe into the urine. The amount hospital/medical school is prob- of Buifialtialipaapif the *mien* each day is sone 50 times ably needed to furnish an envi- the yoltana df, bleOdiffeeteiv , Because Vie body cannot rapidly ronment that will attract and re- 111000041,1001.40.0101, eVen awl defects In the glomerular tain the best such faculty There iseetretw. r tier to the Involvement of are many practical obstacles to ffidnersludies, it.vras thought-that protein be overcome in making such ap- because the Mies in the giomerular pointments successful. A pro- rhr -than the major proleill 11101010118, so that gram to encourage "pioneers" Cajillanii4atistal of sievaAl Idle sieving la ono important who wish to cross the disciplin- peorojOnieMploriodlonHfurther collebtiralft, IWO* t,j PbY8101- ary divide into chemical engi- 0011014#144111ff#0041101neers _revealed anothei; unexpected phe- neering departments is outlined naluiROWiriedue was found *diCHtinirwlecaong circu- in Chapter 10. Some bioengi- isangpaVaint :lints* cf. Iwir, OfectrInitchirje, in addition to neering departments haveal- theirl******910141Yzinaged **cube were repelled by ready made joint appointments nags** chit/gide. Tanen**, the onpliarif wall, and thus were with biological and medical fa- ---retairedlnilartgoodsbeentmoreaffsdkeirlair uncharged mole- culties. Where the organizational cules 011ie eame size. The neadiiiii clhedleed mokcules attached problems inherent in such ar- to ihe Canary walls dew gionistull hedbeenidentified previously, rangements can be avoided or but their tendons( eignificence Was not appreciated. resolved,suchappointments Think. studies led to the realization that proteinuriathe abnor- should be encouraged. mal appearance of protein in the urinecould result not only from A number of other factors will the artatgement of submicrosocpic Woe in the giumerular capillary be important in sustaining a vital welt but also from the ices or neutralization of its negatively research effort in biochemical and charged components. This finding as provided a new direction biomedical engineering. These for research co, the molecular basis for the nephrotic syndrome, include: a group of kidney diseales all characterized by massive oroteinuria. Instrumentation: and facilities. Suitable instrumentation and medium-sized research coi'aborations bringing facilities for education and research in bioen- together two or three to- principal investigators gineering can be very expensive. For example, whose backgrounds and expertise cross the equipping a state-of-the-art tissue-culture facil- bount'ai y heten chemical engineering and the ity for engineering studies costs in the range of life xiences, including medicine. (Se "Cross- $500,000. Other costly equipment required in- discipli,...ary partnership awards" in Chapter 10.) cludes ultracentrifuges, electron microscopes, While large centers certainly can provide an mass spectrometers, NMR spectrometers, scin- erdisciplinary research environment, a greater tillation counters, and specialized instrdments number of medium-sized collaborations might to study surfaces (see Clsapter 9). Some of this foster a faster growth of U.S. capabilities in equipment must be specially modified and ded- critical 1..ioedgineering areas. icated to a particular group's use. Other instru- is :ty expert in both the engineering ments can be shared among a coterie of chemical and the by_,.ogical aspects of the research fron- engineering, biological, and medical research- tiers described in this chapter is needed to mount ers. Chemical engineers should make use of a siinificant educational program in biochemical existing f.sties in life sciences and medical and biomedical engineering. The hiring of fac- departments wherever possible, particularly in ulty into chemical engi neering depart ments whose the case of animal facilities. 34 FROMMRS IN CHEMICAL ENGINEERING

National research centers.Special, highly understaffed in biologically coaversant engi- sophisticated ensembles of analytical and com- neers. These agencies (e.g., EPA, USDA, and putational equipment and expertise might be FDA) should also support chemical engineering brought together in national research facilities research to obtain the data, models, and insight available to academic, government, and indus- necessary for effective risk assessment and trial groups for limited t'me periods. Some management. potential areas of specialization for such centers It is characteristic of U.S. labor markets for include modeling and control of bioreactors, scientific and engineering personnel to experi- measuring and modeling pharmacokinetic data, ence severe shortages and overcompensating measuring actual and simulated bioflows in cesses. Now is the time for the federal gov- living systems and reactors, and studying ki- ernment and universities to build a research and netics of biological reactions and related pro- education base in academia that can respond cesses. flexibly and efficiently to the personnel demands Effective coupling to industry.Effective that will inevitably come. Now is the time to links between universities and industry are es- prepare a cadre of chemical engineers who will sential to successful research and education in interact as easily and successfully with life biochemical and biomedical engineering. In this scientists as chemical engineers currently 2o rapidly growing technological area, a particular with chemists and physicists. need is effective contact and interchange between chemical engineering departments and NOTES smaller venture-capital firms specializing in biotechnology or biomedical products. Liaison pro- I. National Research Council, committee on Sepa- grams and other mechanisms that promote in- ration Science and Technology.Separation and teractions between active researchers, and Purification: Critical Needs and Oppc7;uniiies. opportunities for students to spend time in Washington, D.C.: National Academy Press, 1987. industrial laboratories, should be encouraged. 2 U.S. Congress, Office of Technology Assessment. CommercializingBiotechnologyAnInterna- Better communication among professional tional Analysis.Washington, D.C.: U.S. Govern- societies. The field of biochemical and biomed- ment Printing Office, 1983. ical engineering is in danger of fragmentation among a plethora of professional societies, some of which are quite narrow in focus. Literally SUGGESTED READING dozens of such organizations are currently on J.Feder and W. R. Tolbert. "The Large-Scale the scene. The AIChE could play a valuable Cultivation of Mammalian Cells." Sci. Am., 248 role in ameliorating this situation by promoting (I), January 1983, 36. better co imunication and cooperation among E. L. Gaden, Jr. "Production Methods in Industrial societies and researchers in other disciplines. Microbiology." Sci.Am.,245 (3), September 1981, 180. A. E. Humphrey. "Commercializing Biotechnology: The biochemical and biomedical engineers of Challenge to the Chemical Engineer."Chem. Eng. the future will be in great demand by industry, Prog.,80 (12), December 1984, 7. academia, and federal and state government A. S. Michaels. "Adapting Modern Biology to In- agencies. Already, there is a strong demand by dustrial Practice." Chem. Eng. Prog.. 80 (6), June 1984, 19. universities for faculty in biochemical and A. S. Michaels. "The Impact of Genetic Engineer- biomedical engineering. While recent demand ing."Chem. Eng. Prog.,80 (4), April 1984, 9. from industry has not been as intense, itis National Academy of Sciences-National Academy of projected to increase strongly as products are Engineering-Institute of Medicine, Committee on better defined and move closer to commercial Science, Engineering, and Public Policy. "Report production.2 Federal and state agencies that will of the Research Briefing Panel on Cnemical and Process Engineering for Biotechnology," in Re- be responsible for regulating the introduction searchBriefings 1984.Washington, D.C.: National of new bioproducts into society are woefully Academy Press, 1984.

4'5 BIOTECE'...9LOGY AND BIOMEDICINE 35

National Research Council, Engineering Research visory Board. Bioprocessing for the Energy-Effi- Board. "Bioengineering Systems Research in the cient Production of Chemicals. Washington, D.C.: United States: An Overview," in Directions in National Academy Press, 1986. Engineering Research. Washington, D.C.: Na- R. A. Weinberg. "The Molecules of Life." Sci. Am., tional Academy Press, 1987. 253 (4), October 1985, 48. National Research Council, National Materials Ad-

4E FOUR Electronic, Photonic, and Recording Materials and Devices

he information technologies on which modern society depends would not be possible without integrated circuits, optical fibers, magnetic media, devices for electrical interconnection, and photovoltaics. In the future, these technologies may undergo another revolution as superconductors find use in devices for information storage and handling. Chemical processes are the means by which the physical properties and structural features of these materials and devices are established and tailored. Chemical engineers are beginning to play an important role in process design, optimization, and control in the electronics industry. Their contribution to improving manufacturing technology is particularly wel- come as the United States faces fierce international competition from Japan and Korea. These two countries have already gained the edge in some manufacturing areas and are making inroads into U.S. markets for electronics and recording devices. This chapter describes intellectual frontiers for chemical engineers that, if successfully pursued, can improve the U.S. competitive position. These frontiers include process integration, reactor design and engineering, vltrapurification, materials synthesis and processing, thin film deposition, modeling, chemical dynamics, and process design and control for safety and environmental protection. Implications cf these challenges for the profession are discussed.

37 38 FRONTIERS L\ CHLV/C4/, ENGINEERM

LMOST EVERY ASPECT of our livesat community leads the world in size and sophis- work, at home, and in recreationhas tication. The United States is in a position to been affected by the information rev- exploit its strong competence in chemical pro- olution. Today, information is collected, proc- cessing to regain leadership in areas where the essed, displayed, stored, retrieved, and trans- initiative in manufacturing technology has passed mitted by an array of powerful technologies that to Japan and to maintain or increase leadership rely on electronic microcircuits, light wave in areas of U.S. technological strength. communication systems, magnetic and optical Table 4.2 illustrates some of the ways in data storage and recording, and electrical inter- which chemical engineering can contribute to connections. Materials and devices for these research on information and photovoltaic mate-hnologies, along with photovoltaic materials terials and devices. To fully achieve its potential and devices, are manufactured by sophisticated contribution, though, the field of chemical en- chemical processes. The United States is now gineering must strongly interact with other dis- engaged in fierce international competition to ciplines in these industries. Chemical engineers achieve and maintain leadership in the design must be able to communicate across disciplinary and manufacture of these materials and devices. lines, as the technologies discussed in this The economic stakes are large (see Table 4.1); chapter involve solid-state physics and chem- national productivity and security interests dic- istry, electrical engineering, and materials sci- tate that we make the strongest possible effort ence. to stay ahead in processing science and tech- Electronic, photonic, and recording materials nology for this area. and devices may seem to be an exceedingly In the manufacturing of components for in- diverse class of materials, but they have many formation and photovoltaic systems, there has characteristics in common: their products are been a long-term trend away from mechanical high in value: they require relativly small production and toward production by chemical amounts of energy or materials to manufacture; processes. Chemists and chemical engineers they have short commercial life cycles; and have become increasingly involved in several their markets are fiercely competitiveconse- areas of research and process development. quently, these products experience rapid price Worldwide, however, many high-technology in- erosion. The manufacturing methods used to dustries, such as microelectronics, still have produce integrated circuits, interconnections, surprisingly little strength in chemical process- optical fibers, recording media, and photovol- ing and engineering. The United States has a taics also share characteristics. All involve a particular advantage over its international com- sequence of individual, complex steps, most of petitors in that its chemical engineering research which entail the chemical modification or synthesis of materials. The individual steps are designed as discrete unit or batch operations TABLE 4.1 Worldwide Mal ket for Materials and, to date, there has been little effort to and Devices for Information Storage and integrate the overall manufacturing process. Handling (billions of 1986 dollars) Chemical engineers can play a significant role Year in improving manufacturing processes and tech- Technology 1985 1990° 1995° niques, and investments in chemical processing

Electronic semiconductors 25 60 160 science and engineering research represent a

Light wave fiber and devices 1 3 5.5 potentially h:0,h-leverage approach to enhancing Recording materials 7 20 55 our competitive position. Interconnections 10 21 58 Photovoltaics 0.3 0.8 3 Total electronics 397 550 n a. CURRENT CHEI,...CAL MANUFACTURING ° Market projection. PROCESSES SOURCE: AT&T Bell Laboratories. Compiled from var- Before the invention of the transistor in 1948, ious published sources. the electronics industry was based on vacuum

4E; TABLE4.2Chemical Engineering Aspects of Electronic, Photonic, and Recording Materials and Devices

Technologies Optical and Magnetic Chemical Engineenng Storage and Light Wave Media Contributions Microcircuits Photovoltaic Devices Recording and Devices Interconnection

Large-scale synthesis of Ultrapure single- Ultrapure amorphous Optical recording matenals Ultrapure glass Bending matenals materials crystal silicon silicon E-beam recording matenals Fiber coatings Polyimides Ill -V compounds Ill -V compounds Magnetic panicles Reinforced composites 11-VI compounds 11-VI compounds Film and disk substrates Ceramic substrates Electrically active polymers Other chemicals used in processing of electronic matenals

Engineenng of reaction and Chemical vapor Chemical vapor Sputtenng Modified chemical Plating deposition processes deposition deposition Plasma-enhanced chemical vapor deposition Physical vapor Physical vapor vapor deposition Sol-gel processing deposition deposition Plating Plasma vapor Plasma vapor Solution coating deposition deposition Wet chemical etching Wet chemical etching Plasma etching Plasma etching

Other challenges in Effluent treatment Effluent treatment Fiber drawing and Ceramic processing engineenng chemical coating processes processes

Packaging and/or assembly Process integration Process integration Process integration :lab ling Wafer-scale and automation and automation Laser packaging automation New materials for packaging Modeling of heat transfer in design of packaging 40 FRON mks ICHEMICAL ENGINEERING

tube technology, and most electronic gear was tors in which they are laid down, the choice of assembled on a metal chassis with mechanical chemical reagents, separation and purification attachment, soldering, and hand wiring. All the steps, and the design and operation of sophis- components of pretransistor electronic prod- tica'ed control systems. Microelectronics based ucts vacuum tubes, capacitors, inductors, and on microcircuits are commonly I.sed in such resistorswere manufactured by mechanical consumer items as calculators, digital watches, processes. personal computers, and microwave ovens and A rapid evolution occurred in the electronics in information processing units that are used in industry after the invention of the transistor and communication, defense, space exploration, the monolithic integrated circuit: medicine, and education. Microcircuitry has been made possible by our Today's electronic equipment is filled with ability to use chemical reactions and processes integrated circuits, interconnection boards, and to fabricate millions of electronic components other devices that are all manufactured by or elements simultaneously on a single sub- chemical processes. strate, usually silicon. For example, a 1-million- The medium used for the transmission of bit dynamic random access memory device information and data over distances has evolved (Figure 4.1) contains 1.4 million transistors and from copper wire to optical fiber.It is quite 1 million capacitors. with some chemically etched likely that no wire-based information transmis- features on the chip being as small as 1.1 p.m. sion systems will be installed in the future. The This stunning achievement is just one step in a manufacture of optical fibers, like that of mi- long-term trend toward the design and produc- crocircuits, is almost entirely a chemical pro- tion of integrated circuits of increasing com- cess. plexity and capability. There is still considerable Early data storage memory was based on room for further increases in component density ferrite core coils containing a reed switch that in silicon-based microelectronics (Figure 4.2), mechanically held bits of information in either not to mention possible advances in component an on or off state. Today, most data is perma- density that world result from alternative meth- nently stored through the use of magnetic materials and devices, and the next generation of data storage devices, based on optoelectronic materials and devices, is beginning to enter the marketplace. Ferrite cores were manufactured by winding coils and mechanically mounting the individual memory cells in large arrays. Mag- netic and optical storage media are manufactured almost entirely by chemical processes. The importance and sophistication of current chemical manufacturing processes for electronic, photonic, and recording materials and devices are not widely appreciated. A more detailed description serves to highlight their central role in these technologies.

Microcircuits

A semiconductor microcircuit is a series of FIGURE 4.1Chemical reactions are used to achieve the electrically interconnected films that are laid fine structures seen in modern integrated circuits. This down by chemical reactions. The successful electron micrograph shows a transistor in a "cell" of a (- megabit dynamic random access memory chip The distance growth and manipulation of these films depend between features is aboutIp.m. Courtesy, AT&T Bell heavily on proper design of the chemical reac- Laboratones

5 u F1.ECTRON1C, PHOTON1C, AND RECORD/A.4; IER1ALS AN!) DI:lien 41

0 1,400 -I ,500 °C under an argon at- lee mosphere. Tiny quantities ofdop- antscompounds of phospho- 107 rus, arsenic, or boronare then added to the melt to achieve the w io desired electrical properties of

105 the finished single-crystal wafers. A tiny seed crystal of silicon 2 2 0 with the proper crystalline ori- 8 entation is inserted into the melt 103 and slowly rotated and with-

102 drawn at a precisely controlled rate, forming a large cylindrical 10 single crystal 6 inches (14 cm) in 1900 diameter and about as tall as an YEAR adult human being (1.8 m) with FIGURE4.2The chemical processes used for the manufacture of microcircuits the desired crystalline orienta- have become progressively more sophisticated. This development is respon- tion and composition. Crystal sible for the large increases in the number of components that can be placed growth kinetics, heat and mass on a single chip. Trends in Increasing component density are shown from transfer relationships, and chem- 1960 for silicon chips (top line) and from 1975 for developmental chips based ical reactions all play important on gallium arsenide (bottom line) The rates of growth shown have been remarkable. From 1962 to 1972. silicon compo ent density increased a roles in this process of controlled thousandfold and from 1972 to 1982. a hundreo,,ild From 1975 to 1985. growth. The resultingsingle- component density in developmental gallium arsenide devices grew by a factor crystal ingots are sawed into waf- of 40,000. Courtesy. AT&T Bell Labortories. ers that are polisheJ to a flatness in the range of from Ito 10 Rm. The next steps in device fab- ods of storing and transferring information (e.g., rication are the sequential deposition and pat- three-dimensional circuits and Josephson (quan- terning of thin dielectric and conducting films tum) or optical devices). (Figure 4.5) The polished silicon wafer is first Chemical reactions and processes in the man- oxidized in a furnace at 1,000-1,200°C. The ufacture of microcircuits (Figure 4.3) begin with resultant silicon dioxide film is a few hundred the basic material for integrated circuits, high- nanometers thick and extremely uniform. The purity (less than 150 parts per trillion of impur- wafer is then coated with a photosensitive ities) polycrystalline silicon. This ultrapure sil- polymeric material, termed a resist, and is icon is produced from metallurgical grade (98 exposed to light through the appropriate pho- percent pure) silicon by the following steps tomask. The purpose of the photolithographic (Figure 4.4): process is to transfer the mask pattern to the reaction at high temperature with hydrogen thin film on the wafer surface. The exposed chloride to form a complex mixture containing organic film is developed with a solvent that trichlorosilane; removes unwanted pertions, and the resulting separation and purification of of trichloro- pattern serves as a mask for chemically etching silane by absorption and distillation; and the pattern into the silicon dioxide film. The reduction of ultrapure trichlorosilane to resist is then removed with an oxidizing agent polycrystalline silicon by reaction with hydro- such as a sulfuric acid-hydrogen peroxide mix- gen at 1,100-1,200°C. ture, and the wafer is chemically cleaned and ready for other steps in the fabrication process. To prepare single-crystal silicon ingots suit- The patterned wafer might next be placed in able for use as materials in semiconductors, a diffusion furnace, where a first doping step is polycrystalline silicon is melted in a crucible at performed to deposit phosphorus or boron into 42 FRONTIERS IN CHEMICAL ENGINEERING

IC 041111111 ipso Mallon Sisatiorad MAIN and Isyout SOW and EN Photosvislis

Front end processing

Assembly and test

FIGURE 4.3The manufacture of integrated circuits requires both expertise in electronic design and chemical processing. Chemical process steps are Important to the preparation of silicon materials, to the steps from oxidation of silicon wafers through establishment of bonding pads, and to the final assembly of chips in individual packages Excerpted by special permission from Chemical Engineering. June 10. 1985. Copyright 1985 by McGraw-Hill, Inc.. New York, NY 10020. the holes in the oxide. A new oxide film can glass fiberfor purposes that include telecom- then be grown and the photoresist process munications, data and image transmission, en- repeated. As many as 12 layers of conductor, ergy transmission, sensing, display, and signal semiconductor, and dielectric materials are de- processing. Optical fiber technology is less than posited, etched, and/or doped to build the three- 14 years old and only became a commercial dimensional structure of the microcircuit. reality in the early 1980s. It is now a $1 billion per year industry. The data-transmitting capacity of optical fiber systems has doubled every Light Wave Media and Devices year since 1976 (Figure 4.6). In fact, optical Photonics involves the transmission of optical fiber systems planned on the basis of the pre- signals through a guiding mediumgenerally a vailing technology at that time are often obsolete

52 ELECTRONIC, PHOTONIC, ,AND RECORDING $1,111:IllALS ,1,\P DEUCES 43

1,100'C Si + Ha sma, + [2] SiH03 + H2 Nes Si + %HA HOI + SiiHyaz Distillation

HCI AliHr

Meta IkegIcal 'ikon [11 [3] Polycrystalline silicon

FIGURE 4.4 The production of polycrystalline silicon for the electronics industry Involves several chemical steps aimed at the reduction of impunties. These include (I) reaction of metallungcal grade silicon to produce a mixture of chlorosilanes, (2) distillation of trichorosilane, and (3) reduction of trichloro- Wane to polycrystallinz silicon. Excerpted by special permission from Chem- ical Engineering, June 10. 1985. Copyright 1985 by MGraw-Hill, Inc, New York, NY 10020. by the time they are implemented. Typically, a tronics emphasize reactive ion etching, epitaxy given light wave technology is supplanted by (e.g., metalorganic chemical vapor deposition an improved technology after one year. (MOCVD), vapor-phase epitaxy, and molecu- Other applications for light guides, such as lar-beam epitaxy (MBE)), and photochemical optical fiber sensors and transducers, are re- and beam processing techniques for writing ceiving a great deal of attention. Image trans- circuit configurations. All these processes are mission (e.g., endoscopes), energy transmission based on chemical reactions that require precise (e.g., light pipes), and display (e.g., decorative process control to produce useful devices. signs) are growing commercial areas. Optical fibers are made by chemical process- Light wave media and devices include the es. The critical feature of an optical fiber that guiding medium (optical fibers), sending and allows it to propagate light down its length is a receiving devices, and associated electronics core of high refractive index surrounded by a and circuitry. The transmission of light signals cladding of lower index. The higher index core through optical fiber? must occur at wavelengths is produced by doping silica with oxides of where the absorption of light by the fiber is at phosphorus, germanium, and/or aluminum. The a minimum. Typically, for SiO2P?..02 glass, the cladding is either pure silica or silica doped with best transmission windows are at 1.3 or 1.5.m fluorides or boron oxide. (Figure 4.7). There are four principal processes that may Optical signal processing for integrated optics be used to manufacture the glass body that is and optical computing is in a rudimentary state drawn into today's optical fiber. "Outside" but is certain to be an important area for future processesoutside vapor-phase oxidation and technological development. The processing in- vertical axial depositionproduce layered de- volved in making optoelectronic devices is very posits of doped silica by varying the concentra- similar to that used in microcircuit manufacture, tion of SiC14 and dopants passing through a but with considerable utilization of Group III- torch. The resulting "soot" of doped silica is V compound semiconductors, lithium niobate, deposited and partially sintered to form a porous and a%, ariety of polymeric materials. Devel- silica boule. Next, the boule is sintered to a opmental manufacturing processes for optoelec- ore-free glass rod of exquisite purity and trans- 44 FRONTIERS IV CHEMICAL FNGINEERLVG

SILICON WAFER SILICON INGOT

Resist 1110 1915 IHO Coat Base YEAR FIGURE 4.6Since 1975, both the capacity of optical fiber and the distance a signal can be carried on optical fiber have steadily increased Courtesy, AT&T Bell Laborato- ries. 4Mask "iraworsoe Expose parency. "Inside" processessuch as modif chemical vapor deposition (MCVD) and plasma chemical vapor deposition (PCVD)deposit doped silica on the interior surface of a fused silica tube. In MCVD, the oxidation of the NNW Develop halide reactant- is initiated by a flame that heats the outside of the tube (Figure 4.8). In PCVD, the reaction is initiated by a microwave plasma. More than a hundred different layers with different refractive indexes (a function of glass Etch composition) may be deposited by either process before the tube is collapsed to form a glass rod. In current manufacturing plants for glass fiber, rr-1,477..."4 Strip the glass rods formed by all the above processes are carried to another facility where they are

FIGURE 4.5 Chemical steps in photolithography. A sim- 10 plified series of steps in photolithography is shown A 1250nm 1390nm 1600nm silicon wafer, taken from a single-crystal silicon ingot, is 5 Si-OH COMBSi-OH P-OH coated with a polymer resist that is sensitive to light A SHARP SHARP BROAD E mask is placed over the wafer and the resistis thus JC selectively exposed to light Depending on the type of CO cc, OH ADDED polymer coating used, two things can happen. If the polymer LOSS is a positive resist, exposure to light makes the polymer O easier to dissolve in a solution during the development step 05 After the development step, a protective film is left on the wafer that is the Image of the mask used. If the polymer is RAYLEIGH SCATTERINGX-4 a negative resist, exposure to light makes the polymer more 01 08 10 12 14 16 difficult to dissolve dunng the development step. After- WAVELENGTH (p,rn) wards. a protective film is left on the wafer that is the opposite of the image on the mask used. A corrosive gas FIGURE 4.7Transmission losses in glass fibers carrying or liquid is then used to etch away those parts of the wafer optical signals are due to the interaction of light with unprotected by the resist film The resist film is removed chemical bonds. From 1.2 p.m to 1.6 p.m, losses due to after etching in preparation for other process steps. Ex- Rayleigh scattering in the fiber are minimized, but trans cerpted by special permission from Chemical Engineering. mission losses from Si-OH and P-OH bond3 become large. June 10, 1985. Copynght 1985 by McGraw -Hill. Inc New The lowest transmission losses occur at wavelengths of 1.3 York, NY 10020. and I 5 p.m. Courtesy, AT&T Bell Laboratones.

54 ELECTRONIC, PIIOTONIC, AND RECORDINGA1ATER;tLSANDDEVICES 45

FLOW METERS, MASS-FLOW CONTROLLERS 02 AND MANIFOLD FUSED SILICA TUBE

MULTIBURNER DEPO:ilrin LAYER TORCH OF .;ORE MASS

POC SeC24 02 bCi. SIF4

H2 SICi4 SF, C-l2FREON 4 TRANSLATION

BUBBLERS FIGURE 4.8Modified chemical vapo- deposition (MCVD) isone of the principal processes use') to manuf - wtical fiber. In MCVD. a mixture of gases (0., POCI,, F GeCI4,JCl,,SiF4. SF6, Cl.. and freon) pass down the interic of a hot' tube that is being externally heated by a moving flame. The gases reat. to form a fine 'ayer of silica glass doped with constituents of the gaseous mixture. Many layers can be deposited before the silica tube is collapsed and drawn into optical fiber. Courtesy. AT&T Bell Laboratories drawn into a thin fiber and immediately coated could reduce the cost of glass fiber by as much with a polymer. The polymer coating is impor- as a factor of 10, a step that would greatly tant; it protects the fiber surface from micro- :ncrease the scope, availability, and competi- scopic scratches, which can seriously degrade tiveness of light wave technologies. the glass fiber's strength. At present the chemical steps involved in sol- Current manufacturing technologies for op- gel processes are poorly understood. Methods tical fiber are expensive con-oared with the low are being sought to manipulate these processes cost of commodity glass. U.S. economic com- to produce precisely layered structures ina petitiveness in optical technologies would be reliable and reproducible way. greatly enhanced if low-cost means were found for producing wave guide-quality silica glass. The manufacture of glass lends itself to a fully Recording Media integrated and automated (i.e., continuous) Recoruing media come in a number of formats process. One can envision a fiber manufacturing (e.g., magnetic tape, magnetic disks, or optical plant that moves from purification of chemical disks) and are made by a variety of materials reagents to a series of chemical reactions, glass- and processes (e.g., evaporated thin filmso. forming operations, and, finally, fiber-drawing deposited magnetic particles in polymer ma- steps. Intermediate products would never be trixes). The next generation of recording media removed from the production line. Sol-gel and will be based on optical storage of data (Figure related processes (see Chapter 5)are attractive 4.9). Already, read-only optical disks (or CD- candidates for such a manufacturing technology, ROMs) are on the market for applications such which would start with inexpensive ingredients as search and retrieval of information from large and proceed from a sol to a gel, to a porous uata bases. And optically based compact disks silica body, to a dried and sintered glass rod, (CDs) are available in every record store. The to drawn and coated fibers. Such a process possibility of creating optically based recording 46 FRONTIERS IN CHEMMAL EN6INEERING media for read-write storage of information has disk or tape are very important, for they deter- generated a tremendous amount of industrial mine the density at which information can be research but, so far, no commercially viable recorded. Paramount among these properties products. are the shape, size, and distribution of the Since a practical read-write optical disk has magnetic particles. An extremely na.-row size not yet been invented, it is hard to describe the range of magnetic particlesthemselves only a processing challenges involved in making it. few tenths of a micrometer in diametermust Thus, the remainder of this section examines be achieved in a reliable and economic manner. the most challenging (from a processing stand- Furthermore, the particles must be deposited point) of the remaining foms of recording me- in a high! ' oriented fashion and lie as closely dia: magnetic disks and tape. Magnetic media together as possible, so that high recording are still an economically important part of the densities can be achieved. To accomplish this, recording market and have a rich array of a variety of challenging problems must be solv-i processing challenges with which chemical en- in me chemistry and chemical engineering of gineers have been involved. These challenges barium ferrite and the oxides of chromium, are relevant to the emerging technologies and cobalt, and iron (e.g., the synthesis and pro- materials in recording. cessing of micrometer-sized materials with spe- In the manufacture of magnetic recording cific geometric shapes). media, the chemical and physical properties of The manufacture of magnetic tape illustrates the magnetic particles or thin films coated on a an interesting sequence of chemical processing challenges (Figure 4.10). A carefully prepared dispersion of needle-like magnetic particles is 1010 coated onto a fast-moving (150-300 m/min) poly- ester film base 0.0066-0.08 mm thick. The ability Optical disks to coat tun, smooth layers of uniform thickness is crucial. The particles, after being coated onto 51Ivor hairdo Limit the film, are oriented in a desired direction either magnetically or mechanically during the coating process. After drying, the tape is cal- endered (squeezed between microsmooth steel 108 and polymer rolls that rotate at different rates), providing a "microslip" action that polishes the tape surface. These manufacturing steps (i.e., materials synthesis, preparation and handling 107 of uniform dispersions, coating, drying, and calendering) are chemical processes and/or unit operatio.is that are familiar territory to chemical engineering analy ,,s and design. 106

Materials and Devices for Interconnection and Packaging 105 1965 1970 1980 1990 2000 Interconnection and packaging allow electronic devices to be usefully incorporated into FIGURE 4.9 The density at which information can be products. The manufacture of complex elec- written on optical disks (measured in bits/in2) was demon- tronic systems requires that hundreds of thou- strated in the early 1980s to be 10 times greater than the sands of electronic components be efficiently current highest performance magnetic disk. The gap between optical and magnetic storage carabilities is projected connected with one another in an extremely to increase over the next 20 years '!".c..printed with permis- small space. In the past, this was accomplished sion from Electronic Design, August :8. 1983, 141. by hand wiring discrete components o.i a chassis

56 ELECTRONIC, PHOTONI(', ANI) RECORDING MATERIALS AM) DEVICES 47

(4) (2) Drying oven (1) Coating head air flotation (5) Coating reverse roll (shown) (shown) Calender mix gravure knife 0 i Coating mix Orientation (3) Ferrite magnets Sand," III Dispersing eiectro magnets Unwind Base film Breakaway of base film to show air slots

Screen

Mix supply to coats,

Base film ____----- Rotating Blades Enlargement of air-flotation oven creating sinusoidal wave form of base film

Mix Inlet

Enlargement of sandmIll dispersing chamber

FIGURE 4.10 Themanufacture of magnetic tape involves a series of steps including (I) forming a uniform dispersion of coating mix, (2) applying this coating to the film base. (3) orienting the magnetic particles, (4) drying the magnetic coat in an air-flotation oven, and (5) calendering and final wind-up on spools. Chemical processes are central to several of these steps. Excerpted by special permission from Encyclopedia of Chemical Technology, 3rded., Vol. 14, p. 745. Copyright 1978 by McGraw-Hill, Inc., New York, NY 10020. assembly. Today, interconnection technology The dielectric and the conductors are selected is based on high-density printed wiring boards, to maximize data transmission speed while min- often with as many as 30 parallel layers of imizing signal loss. In aldition, dissipating heat interconnection. The boatd insulation substrate generated by the microcircuits is raoidly becom- may be either polymer or ceramic, with appro- ing an important consideration. It too much priate metal conductors. heat builds up in the microelectronic device, 48 FRONTIERS I'% CHEW(' ENGINEERES(, the device will start to fail. The next generation of interconnection substrates will likely require A Top surface metallurgy new materials and assembly techniques to cope with this challenge. Plastic packaging of microelectronic devices is the most Discrete wing cost-effective means of provid- Top surface ing electrical, mechanical, and metallurgy and environmental protection for an redistribution integrated circuit device. Ap- layers Serial proximately 80 percent of the distribution and billions of these devices used in - =.7r_717- - signal reference the United States each year are laYeT packaged in plastic, and most are Power distribution packaged in thermoset molding and module compounds by a conventional I/O layers transfer molding process. This process must be carried out with exceptionally high yield, productivity, and reliability if the United States is to achieve a competitive F A power plane advantage in the packaging of layer these devices. E A reference plane layer Allmodern interconnection devices are manufactured by FIGURE 4.11Cross-section of the IBM Multilayer Ceramic interconnection chemical processes such as etch- pac!.::;... Various layers in this interconnection device are shown. Copyright 1982 by International Business Machines Corporation Reprinted with per- ing, film deposition, and ceramic mission. forming. Substrate formation is a vital part of the manufacturing process for fiberglass-impregnated printed wir- Chapter 5). A good example of ceramic intercon- ing boards. It utilizes chemical processes such iw'tion boards are the multilayer ceramic (MLC) as metal deposition, lithographic patterning, structures used in large IBM computers (Figure etching, and chemical cleaning. Process design, 4.11). These boards measure up to 100 cm2 in to improve quality and decrease cost, continues area and contain up to 33 layers. They can in- to be a challenge, particularly in terms of greater terconnect as many as 133 chips. Their fabrica- uniformity in plating and etching and the envi- tion involves hundreds of complex chemical ronmental problems posed by the disposal of processes that must be precisely controlled. spent etching and plating baths. Better insulating substrates will be required Ceramic boards are currently widely used in for the ultrahigh-speed interconnection modules high-performance electronic modules as inter of the future. Although organic polymers appear connection substrates. They are processed from to offer cost advantages for substrate applica- conventional ceramic precursors and refractory tion, the large-scale production of organic sub- metal precursors and are subsequently fired to strates with sufficiently low dielectric constants the final shape. This is largely an art; a much bet- has not yet been realized. Achieving this will ter fundamental understanding or the materials require the scale-up of new chemical reactions and chemical processes will be required if low- and the development of process methods for cost, high-yield production is to be realized (see new polymer substrates. ELECTRONIC, PHOTONK:, A,\9 RECORDING .14,tTERIALS AA D DE% ICES 49

Photovoltaics exhibit superconductivity at temperatures up to 120 K, the so-called high-temperaturesuper- A photovoltaic cell is a solie -state device that conductors, has sparked a tremendous amount transforms solar energy into electricity. Signif- of scientific activity and commercial interest icant research on photovoltaic cells began in around the world. the early 1970s and until now has generally The key to the superconducting properties of focused on the invention and improvement of these ceramics seems to be the presence of specific devices, including single-crystal silicon planes of copper and oxygen atoms bondedto cells, amorphous silicon cells, heterojunction one another. The significance of the other atoms thin-film cells, and gallium arsenide cells. in the lattice seems to be to provide a structural Gallium arsenide cells have achieved high- framework for the copper and oxygen atoms. energy efficiencies (in excess of 20 percent) and Thus,inthesuperconductingcompound have been successfully used in systems where YBa,Cu30, the substituti^r of other rare earths there is magnification of the sun's rays, that is, for yttrium results in little change in theprop- concentrator systems that increase the light flux erties of the material. per unit area. Substantial research is being Currently, superconducting materials are pro- devoted to improving the conversion efficiency duced by standard techniques from the ceramics and lowering the cost per kilowatt of such industry: mixing, g.inding, and sintering. The devices. basic structure of the 95 K superconductcrs is The materials and processes used in the formed at temperatures above 800-900°C, and manufacture of photoelectric energy conversion then annealed with oxygen at a temperature devices are almost identical to those used in below 500°C. More needs to be known about manufacturing microelectronic devices and in- the effect of various synthesis conditions on the tegrated circuits. microst-octure found in these materials. Alter- The photovoltaics industry could expand rap- native routes to ceramic-, such as sol-gelpro- idly if the efficiency of polycrystalline modules cesses, may lead to significant improvements in could be increased to 15 percent, if these mod- the production of these materials. For micro- ules could be built with assurance of ieliability electronic applications, various chemicalvapor over -a 10- to 20-year period, and if they could deposition routes to these materials need to be be manufactured for $100 or less persquare investigated. MOCVD might be a particularly mete:. Solar energy research has been largely promising route if volatile precursor compounds directed toward only one of these issues: effi- could be discovered and synthesized. Chemical ciency. All research aimed at reducing manu- engineers have the needed background for de- facturing costs has been done ;n industry and veloping this technology (see Chapter 5)as well has been largely empirical. Almost no funda- as finding the optimal procedure for drying, mental engineering research has been done on sintering, and calcining the final product. either the laboratory scale or the pilot plant scale for cost-effective processes for theproduction of photoconverters. INTERNATIONAL COMPETITION In each of the technologies described in the Superconductors preceding section, U.S. leadership in both fundamental research and manufacturing isse- Superconductivity has been known since 1911, verely challenge;, and in some cases the United and superconducting systems based on various States is lagging behind its foreign competitors. metal alloys (e.g., NbTi and Nb3Sn) are currently used as magnets and in electronics. These materials exhibit superconductivity only at tem- Microcircuits peratures below 23 K and require cooling by A recent report of the National Research liquid helium. The discovery of ceramics that Council' has assessed the comparative position 50 t RON MKS IA (IMAM AL LACIALEIIIA6

if the Uni'zd States and Japan in advanced origin gain a competitive advantage stemming processing of electronic materials. The report, from this early access. A look at the installation whit .1 focuses heavily on evaluating Japanese record for JEOL focused ion beam instruments research on specific process steps in the man- (which are the best in the world) illustrates this ufacture of electronic materials, provides sig- phenomenon (Table 4.3). nificant background for the following observa- Much remains to be done, in both the United tions: States and Japan, to solve the problems of process integration in microcircuit manufacture. The U.S. electronics industry appears to Effort is being expended on equipment design be ahead of, or on a par with, Japanese industry for specific processing steps, but a parallel effort in most areas of current techniques for the to integrate the processing of semiconductor deposition and processing of thin filmschem- materials and devices across the many individ- ical vapor deposition (CVD), MOCVD, and ual steps has received less attention in blth MBE. There are differences in some areas, countries. Yet the latter effort may have signif- though, that may be crucial to future technol- icant payoffs in improved process reliability and ogies. For example, the Japanese effort in low- efficiencythat is, in "manufacturability." The pressure microwave plasma research is impres- United States, with the strongest chemiza! ensive and surpasses the U.S. effort in iome gineering research community in the world, has respects. The Japanese are ahead of their U.S. the capability to take a significant lead in this counterparts in the design and manufacture of area. deposition equipment as well. Japanese industry has a very substantial Light Wave Media and Devices commitment to advancine, high-resolutionli- thography at the fastest possible pace. Two The Japanese are our prime competitors in Japanese companies, Nikon and Canon, have the development of light wave technology. They made significant inroads at the cutting edge of are not dominant in the manufacture of optical optical lithography equipment. In the fields of fiber thanks in part to a strong overlay of patents x-ray and electron-beam lithography, it appears on basic manufacturing processes by U.S. com- that U.S. equipment manufacturers have lost panies. In fact, a major Japanese company the initiative to Japan for the development of manufactures optical fiber in North Carolina for commercial equipment. shipment to Japan. This is the only example to Japanese researchers are ahead of their date of Japan importing a high-technology prod- U.S. counterparts in the application of laser and uct from a U.S. subsidiary. Nonetheless, the electron beams and solid-phase epitax; for the Japanese are making strong efforts to surpass fabrication of silicon-on-insulator structures. the United States and are reaching a par with The United States leads in basic research us in many areas. related to impiantation processes and in the The United States stillsignificantly leads development of equipment for conventional ap- Japan in producing special purpose and high- plications of ion implantation. Japan appears to strength fibers, in preparing cables from groups have the initiative in the development of equip- of fibers, and in research on hermetic coatings ment for ion microbeara technologies. for fibers.

There is a penalty to be paid for falling behind Recording Media foreign competitors in process equipment design and engineering. Early access to new prototypes Japan is the America's principal technological of equipment allows a manufacturing firm to competitor in the manufacture of magnetic me- concurrently troubleshoot the equipment and dia, and Korean firms are beginning to make integrate it into its existing process linc. When significant in oads at the low end of the magnetic the state-of-the-an processing equipment comes tape market. U.S. companies producing mag- from overseas, companies in the country of netic tape use manufacturing processes that

60 ELECTRONIC. PHOTONIC. 4N7) RECORDINGMATERbiL.S .111) MILK L.S 51

TABLE 4.3Installation Record for Focused Icn Beam instrumens Made by JEOL Semiconductor Equipment Division

Instrument Instrument Year Number Customer Country Type Installed

I The Institute of Physical and Japan JIBL-34 1982 Chemical Research 2 The Institute of Ph: :mai and Japan JIBL-100 1983 Chemical Research 3 Optoelectromcs Joint Japan JIBL-100 1983 Laboratory 4 LSI R&D Lab. Mitsubishi Jap. JIBL-100 198' Electric Corporation 5 Fujitsu Laboratories Ltd. Japan JIBL-100A 1984 Atsugi 6 Optoelectronics Joint Japan JIBL-100A 1984 Laboratory 7 institute of Laser Engineenng, Japan ".BL-30 1984 Osaka University 8 The Institute of Physical and Japan JIBL-200 1984 Chemical Research 9 NIT Musashmo Electrical Japan JIBL-100 1984 Communication Laboratories 10 Institute of Industrial Science. Japan JIBL-106 1984 Tokyo University I I Fujitsu Laboratories Ltd. Japan JIBL -GPI 1984 Atsugi 12 Dainippon Screen Japan JIBL-100 1984 13 Fujitsu Laboratories Ltd. Japan JIBL-100 1984 Atsugi 14 NIT Atsugi Electncal Japan JIBL-100 1984 Communication Laboratones 15 Tsukuba Research Center. Japan JIBL-100A 1984 Sanyo Electnc Co.. Ltd. 16 NEC Corporation Japan JIBL-140 1984 17 LS1 R&D Lab. Mitsubishi Japan JIBL-140 1984 Electnc Corporation 18 Optoelectronics Joint Japan IPMA-10 1984 Laboratory 19 Institute of Industnal Science. Japan IPMA -I0 1986 Tokyo University 20 Matsushita Laboratory Japan JIBL-106 1986 21 Nihon Denso Japan J1BL-106 'I 22 Sony Japan JIBL -GPI q 23 Denka Japan JIBL-100 q 24 Max Planck Institute Germany JIBL-100A 1 25 IBM U.S.A. JIBL-106 1988 SOURCE: AT&T Bell Laboratories and JEOL. achieve higher integration through combined firms have closed this gap in recent years and unit operations, but Japanese companies have are now capturing worldwide market shares a higher degree of automation in these separate from the Japanese, even in Japan. operations. U.S. companies lead the Japanese The most significant development in Japan is in the use of newer thermoplastics in calender- the entry of photographic film companies (Fuji compliant roll materials. Japan used to surpass and Konishuroku) into the manufacture of mag- the United States in the product uniformity of netic media. They are having a large impact magnetic tape for professional applications; U.S. because the heart of the manufacturing process

6... 52 FRONTIERS I\ (lIEMICII, E,VGINEERING is the deposition of thin layers, and chemical petition as intense, but the U.S. competitive processing technology from the photographic position in science as good. Japan, China, a film business can be used to improve the quality number of European countries, and the USSR and yield of magnetic tape and disks. are putting in place significant scientific and The United States still lags behind Japan in technological efforts. In Japan, industrial con- the treatment and manufacture of magnetic sortia are being organi,:ed by the government particles (except possibly for 3M, which man - to begin initial development activities. The re- ufactc s its particles internally). There are port concludes, "Japan offers perhaps the disturbing signs that the Japanese may be ahead strongest long-range competitive threat to the of the United States in the next generation of U.S. position." film base, especially for vapor-deposition magnetic media. The situation is not entirely clear, because 3M and Kodak make their own pro- General Observations prietary film. Other 1J.S. magnetic media com- The industries that manufacture materials and panies, though, may be buying their film tech- components for information applications are nology from Japan in the future. characterized by products that are rapidly Optical recording media for read-write appli- superseded in the market by improved ones. cations are still in the research stage. U.S. This rapid turnover stems from the intense companies are roughly on par with European competition among these industries and results and Japanese companies in such research. Read- in rapid price erosion for products, once intro- only applications (e.g., CD-ROM disks and duced. These industries also require rapid tech- compact a'idio disks) are largely dominated by nology tran4er from the research laboratory manufacturing technology from overseas. onto the proouction line. Many of their products cannot be protected by patents, except foF minor features. Therefore, the key to their competitive Interconnection and Packk 'hg success is thoroughly characterized and inte- The United States leads its competitors in the grated manufacturing processes supported by design of central processing unit packaging for process innovations. In the past, much of the large computers. Companies such as IBM, Cray, process technology on which these industries and Amdahl are on the cutting edge of inter- depend has been developed empirically. If the connection design and manufacturing. Japanese United States is to maintain a competitive po- companies are ahead in some interconnection sition in these industries, it is essential that it technologies found in mid-sized and smaller develap the fundamental knowledge necessary computers (e.g., phenolic paper boards and to stimulate further improvement of, and in- epoxy-resin boards). novation in, proc,:sses involving chemical reactions that must be precisely controlled in a manufacturing environment. In the next section, Photovoltaics the principal technical challenges are set forth. The U.S. photovoltaics industry serves more than 100 different countries. Major competition INTELLECTUAL FRONTIERS comes from Japan and, to a limited extent, from Europe. U.S. firms have a dominant position A variety of important research ',sues require in the power module market (devices with much more work if U.S. companies are to photovoltaic areas greater than 0.5 m2) while establish and maintain dominance in information Japanese firms have dominated the consumer storage and handling technologies. These issues market for small-photovoltaic goods (e.g.. cal- are quite broad and cut across the spectrum of culators, watches, and radios). materials and devices.

St perconductors Process Integration A recent report on high-temperature super- Process integration is the key challenge in the conductivity= characterizes international com- design of efficient and cost-effective manufac- 6'4 ELECTRONIC, PHOTONIC. AND RECORDING MATERIALS AM) I)EV 53

turing processes for electronic, photonic, and tions. For example, in the manufacture of mi- recording materials and devices. Except for crodrcuits, chemical engineers can provide magnetic tape. these products are currently mathematical models and control algorithms for manufactured through a series of individual, the transient and steady-state operation of in- isolated steps. If the United States is to retain dividual chemical process steps (e.g., lithogra- a position of leadership, it is crucial that its phy, etching, film deposition, diffusion, and overall manufacturing methodology be exam- oxidation), as well as interactions betweenpro- ined and that integrated manufacturingap- cess steps and ultimately betwt.en processing proaches be implemented. Historically, all in- and the characteristics of the final device. As dustries have benefited both economically and another example, in microcircuit manufacture, in the quality and yield of products by theuse chemical engineers can provide needed simt of integrated manufacturing methods. As indi- lations of the dynamics of material movement vidual process steps become more complex and through the plant and thus optimize the flow of precise, the final results of manufacturing (e.g., devices (or wafers) through a fabrication line yield, throughput, and reliability) often depend (Figure 4.12). The continuous production of critically on the interactions among the various photovoltaic devices will require similar studies steps. Thus, it becomes increasingly important with even more emphasis on automation. to automate and integrate individual process steps into an overall manufacturing process. The concepts of chemical engineering are Reactor Engineering and Design easily applied in meeting the challenge ofpro- Closely related to challenges in process in- cess integration, particularly because many of tegration are those in reactor engineering and the key process steps involve chemicalreac- design. Research in this area is important ifwe

MICRa a0CESSOR COPITROL UNIT

RESIST PATTERN WAFER LASER DEVELOP FILM COAT & TRANSFER ION IMPLANT ENTRY READER ETCH TEST +DEPOSITION BAKE (EB, X-RAY 4* STRIP (SPUTTER, PHOTO, THERMAL CLEAN MBE, CVD ION) DR*VE EVAPORATE)

'PRODUCT' FIEJECT L lI

WAFER TRANSPORT ELECTRONIC CONTROL RE-ENTRY PA1H EXIT PATH

FIGURE 4.12 The integrated semiconductor processing line of the future will be a fully automated suns of chemical processing steps. Chemicalengineers will be needed to integrate individual process steps intoa manufacturing line that can be operated free from human handling, and possiblecontamination, of the devices. Courtes,, AT&T Bell Laboratories. 54 I. ROA I MRS IN (lIE.111( AI. IIAGIVEI.R1 are to automate manufacturing processes for require reactors with a different design from higher yields and improved product quality those currently used, especially for epitaxial Processes such as CVD, epitaxy, plasma-en- growth processes. The challenge is to be able hanced CVD, plasma-enhanced etching, reac- to control the flow of reactants to build layered tive sputtering, and oxidation all take place in structures tens of atoms thick (e.g., superlat- chemical reactors. At present, processes and tices). To achieve economic automated pro- reactors are generally developed and refined by cesses, the reactor design must allew for the trial and error. A basic understanding of fun- acquisition of detailed real-time information on damental phenomena and reactor design would the surface processes taking place, fed back facilitate process design, control, and reliability. into an exquisite control system and reagent Because all these processes involve reaction delivery sy stem. This problem gives rise to an kinetics, mass transfer, and fluid flow, chemical exciting series of basic research topics. engineers bring a rich ba-kground to their study and improvement. For example, high-yield, Ultrapurifir *ion continuous processes for film deposition and packaging are required if photovoltaic devices A third research challenge that is generic to are to be manufactured at costs that are com- electronic, photonic, and recording materials petitive with other energy technologies. New and devices stems from the need for starting reactors and a better understanding of chemical materials that meet purity levels once thought dynamics in reactors are central to achieving to be unattainable. this. This need is particularly acute for semicon- An important consideration in reac,.or design ductor materials and optical fibers. For semi- and engineering is the ultraclean storage and conductor materials, the challenge is to find transfer of chemicals. This is not a trivial prob- new, lower cost routes to ultrapure silicon and lem; generally, the containers and tonsfer me- g :Ilium arsenide and to purify other reagents dia are the primary sources of contamination in used in the manufacturing process so that they manufacturing. Methods are needed for storing do not introduce particulate contamination or gases and liquids, for purifying them (see the other defects into the device being manufac- next section), and for delivering thorn to the tured. For optical fibers, precursor materials of equipment where they will be usedall the high purity are also needed. For example, the while maintaining impurity levels below 1 part SiCI4 currently used in optical fiber manufacture per billion. This requirement puts severe con- must have a total of less than 4 parts per million straints on the types of materials that can be of hydrogen-containing compounds and less used in handling chemicals. For example, ma- than 2 parts per billion of metal compounds terials in reactor construction that might be (Figure 4.13). Either impurity will result in chosen primarily on the basis of safety often strong light absorption in the glass fiber. For cannot be used. Designs are needed that will magnetic media, the challenge is to separate meet the multiple objectives of high purity, and purify submicrometer-sized magnetic par- safety, and ' 'w cost. ticles to very exacting size and shape toler- The ulti. limit to the size of microelec- ances. tronic devices is of molecular dimensions. The A variety of separation research topics bear ability to "tailor" films at the molecular level on these needs, such as generation of improved to deposit a film and control its properties by selectivity in separations by tailoring the chem- altering or forming the structure, atomic layer ical and steric interactions of separating agents, by atomic layeropens exciting possibilities for understanding and exploiting interfacial phe- new types of devices and structures. The fab- nomena in separations, improving the rate and rication of these multilayer, multimaterial struc- capacity of separations, and finding improved tures will require deposition methods such as process configurations for separations. These MBE and MOCVD. Depositing uniform films are all research issues central to chemical en- by these methods over large dimensions will gineering.

6,3, ELECTRONIC, PHOTONIC, AND RECORDING MA-II:HMOAND DEVICES 55 communications inlocal area network (LAN) applications, because their large core size makes it relatively cheap to attach connectors to them. There is a need for polymer fibers that have low losses and that can transmit the bandwidths needed for LAN applications; the acrylate and methacrylate polymers now under study have poor loss and bandwidth performance. Research on monomer purification, polymerization to precise molecular-size FIGURE 4.13Schematic diagram fora purification plant for producing "optical distributions, and well-con- fiber grade" SiCI4 The feed matenal is passed through a reactor ( I) where trolled drawing processes is rel- chlorination takes place. Excess HCI generated in the reactor is removed (2) and the product stream is passed through two distillation columns (3,4) where evant here. There is also a need ontaminants are removed. On-line IR spectroscopy (5) is used to monitor forprecisionplasticmolding final product purity. Plants built wing this design currently produce about 27 processes for mass production of kg/h of ultrapure SiC14. Contaminants (e.g., compounds containing R-H, optical fiber connectors and splice compounds containing C-H. and Fe) are reduced to below the limits of hardware. A tenfold reduction in detection. Courtesy. AT&T Bell Laboratones. the cost of fiber and related devices is necessary to make the Chemical Synthesis and Process* of utilization of optical fiber and related devices Polymeric Materials economical for local area networks and the telecommunications loop. Although chemical engineering challenges re- Another challenge for polymer research in lated to polymeric materials are discussed in light wave applications is in the use of active Chapter 5, the special challenges for polymers coatings on optical fibers as transducers for in materials and devices for information storage sensors. Such coatings may have magnetostric and handling deserve some mention here. tive or piezoelectric properties. These coa..inw For the processing of microcircuits and in- or the fiber itself, may also incorporate dyes terconnecting devices, improved radiation-sen- that would respond to chemicals, light, radia- sitive polymers are needed for the formulation tion, or other stimuli to produce transmission of better photoresists. Resists must be highly loss changes in the fiber. Such systems have sensitive to the radiation used for exposure, but enormous potential as sensors that would be not to the microwave radiation used after de- ultrasensitive, capaLl. of distributed sensing, velopment for other process steps such as plasma able to operate in harsh environments, and etching. Chemical engineering studies of poly- unaffected by electromagneticinterference. mer behavior during development steps are also Specialty fibers such as polarization-maintain- needed. Details c,f the dissolution of the exposed ing fibers, which have an asymmetric core and (or unexposed) regions of the resist are at can double the bandwidth by transmitting two present poorly understood. There is a need for modes at once, may also play an important role fundamental studies and modeling of the for- in sensor technology. mation of a swollen gel layer at the solvent/ Techniques foi fabricating low-cost optical polymer interface and the subsequent diffusion components such as graded index lenses, mi- of polymer chains into solution. crolenses, couplers, splitters, and polarizers are Light wave technologies provide a number of needed to support optical fibe: technology. Special challenges for polymeric materials. Fo!!!- Traditionally, amorphous inorganic materials mer fibers offer the best potential for optical have peen used, but there are tremendous

65 56 FRONTIERS IN CHEMICAL ENGEVEEKING opportunities for innovation with poly ii.ers, create a uniform microstructure. Chemical routes which offer manufacturing versatility that is not to better ceramics have the advantage of being available with glass. For example, photoselec- more amenable to continuma and automated tive polymerization techniques can be used to processing. Among the typical examples of such make branching wave guide circuits such as approaches are sol-gt-1 and related processes. splitters and couplers. Photopolymerization and (See Chapter 5 for a more detailed treatmeot of copolymerization of multiple monomer systems sol-gel processing.) have been used to make radial, axial, and Deeper involvement of chemical engineers in spherical graded-index lenses with a high degree manufacturing processes for ceramics may be of perfection (e.g., freedom from aberration). particularly important to the eventual commer- Large-scale,well-controlled chemicalproc- cialization of metal oxide superconductors. The esses will be needed to fabricate these struc- current generation of such superconductors tures. consists of planar structurz_ formed dui .g a For recording applications, new approaches conventional ceramic synthesis. The ability to to high-quality polymeric film substrates are precisely control complex phase structure and needed. Improved automation and control of phase boundaries seems critical.It is by no thin-film coating are also important. means clear that the formulations and structures For interconnection and packaging technol- that may produce optimal performance in su- ogies, an important goal is to achieve high- perconducting ceramics (e.g., room-tempera- purity molding and dielectric materials. Epoxy- ture superconductivity, capacity for high-cur- Novo lac prepolymers with ionic impurity levels rent density) are accessible by these techniques. below 20 ppm offer one approach. There is a Rational synthesis of structured ceramics by further need for low-viscosity molding com- chemical processing may be crucial to further pounds to minimize the development of flow improvements in superconducting properties stresses during processing. Continued devel- and to efficient large-scale production. opment of thermally stable polymers with low dielectric constants (such as the polyimides) is Deposition of Thin Films also necessary. Advances in our fundamental understanding of polymer chemistry and rheol- Precise and reproducible deposition of thin ogy are crucial for all these areas (see Chapter films is another area of great importance in the 5). chemical processing of materials and devices for the information age. In microelectronic devices, there is a stcady Chemical Synthesis and Processing of trend toward decreasing pattern sizes, and by Ceramic Materials the end of this decade, the smallest pattern size Challenges for chemical engineering related on production circuits will be much less than to ceramic materials are also discussed in Chap- Iiim (Figure 4.14). Although the lithographic ter 5, but the potential contribution of chemical tools to print such patterns exist, e - exposure engineers to this area cannot he emphasized too step is only one of a number of processes that strongly. A tremendous oppurtunity exists for must be performed sequentially in a mass pro- chemical engineers to apply their detailed duction environment without creating defects. knowledge of fundamental chemical processes Precise and uniform position of materials as in the development of new chemical routes to very thin films onto substrates 14 cm or more high-performance ceramics for electronic and in diameter must be performed in a reactor, photonic applications. The traditional approach usually at reduced pressure. Particulate defects to creating and processing ceramics has been larger than 0.1µm must be virtually nonexistent. through the grinding, mixing, and sintering of Low-temperature methods of film deposition powders. Although still useful in many appli- will be needed so that defects are not generated cations, this technology is being replaced by in previous or neighboring films by unwanted approaches that rely on chemical reactions to diffusion of dopants.

66 ELECTRONIC. PHOTONIC, IND RECORDING MATER! US .01) 1)E1 IlLS 57

10 a0 and chalcogenide glasses that may e find use in optical fibers of the 7 0 future because of their compati- 6 bility with transmission at longer 5 0 wavelengths. Considerable progress in the I4 0 0 science and technology of de- 3 positing thin films is necessary if the U.S. recording media indus- i try is to remain competitive with 0 2 foreign manufacturers. New, fully 0 automated coating processes that will gene:ate high-quality, low- defec media are needed. Not only must considerable effort be 1 li t t IIII t 74 74 74 'SO '62 III'64 mounted in designing hardware '66 YEAR and production equipment, but FIGURE 4.14Feature sue on rivcroclectronic devices has steadily declined complex mathematical models over the years as improved chemical etching processes have been developed must be developed to study the This graph shows feature size as a function of the year in which the device kinetic and thermodynamic with the smallest feature size was first producedCourtesy, AT&T Bell properties of film coating and the Laboratones. effect of non-Newtonian flow and polymer and fluid rheology. A betterunderstandirgofdispersion For optical fibers, improved control over the stability during drying, as well as of diffusion structure of the thin films in the preform will mechanisms that result in intermixing ofse- lead to fibers with improved radial gradients of quential layers of macromolecules, is important. refractive index. A particular challenge is to Thin films are also critical to the performance achieve this sort of control in preforms created of electrical interconnection devices (Figure by sol-gel or related processes. 4.15). Better methods fer depositing thin films Another challenge in depositing thin films on .Jnforn ably (for good sidewall coverage) and optical fibers occurs in the final coatilig step. for achieving high-aspect-ratio trenchesare Improved coating materials that can be cured needed for the interconnection of electronic very rapidly, for example, by ultraviolet radia- devices for the high-frequency transmission of tion, i to needed for high-speed (>10 m/s) fiber- data. New processing strategies and device drawing processes. Both glassy and elastomeric stiuctures are required that use compa3lle polymers are needed for use over temperatures layers of materials to minimize undesirable ranging from 60 to 84°C or higher. Hermetic phenomena such as contact resistance; elec- coatings are required to avoid water-induced tromigration; leakage currents; delamination; stress corrosion of silica ',lasses, which pro- and stress-related defects such as cracks, voids, ceeds by slow crack growth. Materials under and pinholes. study include silicon carbide and tit-tr;umcar- bide applied by chemical vapor deposition,as well as metals such as aluminum. A tenfold Modeling and the Study of Chemical increase in the rate at which such coatingscan Dynamics be applied to silica fiber during drawing is A challenge related to the problems of reactor needed for commercial success. Coatings must design and engineering is the modeling and study be free of pinholes, have low residual stress, of the fundamental chemistry occurring inman- and adhere well. Hermetic coatings will also be ufacturing processes for semiconductors, opti- needed to protect the moisture-sensitive halide cal fibers, magnetic media, and interconnection. 58 flION1 MKS /A Cllli 1//('/1 /. ENGINEERING

For example, mathematical models originally developed for continuously stirred tank reac- Pb-Sn Cu-Sn intermetallic tors and plug-flow reactors are solder pad Phased Cr-Cu applicable to the reactors used Cr for thin-film processing and can be modified to elucidate ways to 3 8 pm S102 improve these reactors. For these models to reach their full descrip- 2.3 pm AI-4% Cu tive potential, detailed studies of 0.85 pm AI-4% Cu 2.8 gm Si02 the fundamental chemical reac- 2.4 pm Si02 tions occurring on surfaces and ,1.4 urn AI-4% in the gas phase are required. For example, etching rates, etch- if /4511MAN/IF iiii ing selectivity, line profiles, de- Thermal Si02 SiN4 PtSi posited film structure, film bend- 3 ing,andfilmpropertiesare Silicon 0.15 1.4M Cr-C.1)(0y determined by a host of variables, including the promotion of FIGURE 4.15Cross-section of multilevel interconnections for advanced bi- surface reactions by ion, elec- polar devices. Fourteen separate Ia)ers are laid down in the fabrication of tron, or photon bombardment. interconnections such as the one shown. The precise orientation and composition of these layers are controlled by chemical process steps. Copyright The fundamental chemistry of 1982 by the International Business Machines Corporation. Reprinted with these surface reactions is poor)/ perni,ssion. understood, and accurate rate expressions are particularly needed for electron-impact reactions (i.e., dis- etching process and their reaction pathways. In sociation, ionization, and excitation), ion-ion addition, this work led to the discovery that the reactions, neutral-neutral reactions, and ion- organic polymer photoresist contributed to neutral reactions. The scale and scope of effort plasma chemistry and selectivity in important devoted in recent years to understanding cata- ways. This in turn led to new, improved plasma lytic processes need to be given to research on processes that are currently being used in pro- film deposition and plasma etching. Until we duction. have a basic understanding of chemical reac- For magnetic media, mathematical models tions occurring at the surface and in the gas could enhance our fundamental understanding phase, it will be difficult to develop new etching of the manufacturing processes used to make systems. uniform high-purity magnetic particles. Models Research in this area has had a demonstrable for the kinetics and mechanisms of reactions impact on recent innovations in plasma process- and ar improved understanding of the thermo- ing. Five years ago, it was well known that a dynamics of producing inorganic salts are re- fluorine-containing plasma etches silicon at a quired. rate significantly greater than the rate for Si02, Modeling to describe the flows of viscous thus offering significant advantages for fabri- fluids could lead to better packaging of inte- cating integrated circuits. However, well-con- grated circuits by assisting in the development trolled processes could not be developed that of molding compounds and processes that will would perform in a production environment. provide for lower thermal shrinkage stresses, The work of chemists and chemical engineers lower permeability, and lower thermal conduc- in elucidating the relevant chemical reactions tivity. Such modeling could also contribute to and their kinetics was crucial to the identifica- the development of packaging materials and tion of the important chemical species in the processes amenable to automation.

6 6 ELECTRONIC, 11110TONI(', .1,NI) RITORDING II tri:R: IS I ND Di. l /( ES 59

Engineering for Environmental Protection TABLE 4.4Empl,yment of Chemical and Process Safety Engineers in the Electronics Industry, 1977-1986" Safety aliu environmental protection are ex- Number of tremely important concerns that present de- Year Chemical Engineers mandi..g intellectual challenges. The manufac- 1977 700 ture of materials and devices for information 1980 960 handling and storage involves substantial quan- 1983 1.648 tities of toxic, corrosive, or pyror horic chemi- 1986 2,100 cals (e.g., hydrides and halides of arsenic, Employment figures for Standard Industrial boron, phosphorus, and silicon; hydrocarbons Classification code 367. "Electronic corn onents and organic chlorides, some of which are sus- and accessories." pected carcinogens; and inorganic acids). The SOURCE: 14ational Science Found hon.' expertise of chemical engineers in the safe handling and disposal of highly reactive materials is much needed in the electronics industry. 1977, the number of chemical engineersem- Recent studies in California indicate that the ployed by the industry has tripled (Table 4.4). semiconductor industry has an occupat:onal Up to 25 percent of the recent graduating classes illness rate three times that of general manufac- of several leading chemical engineering depart- turing industries. Nearly half of these illnesses ments have been employed by the electronics involve systemic poisoling from exposure to indust-;es. The increasing demand of these in- toxic materials. Problems with groundwater dustries for chemical engineers is one factor to contamination in Santa Clara County, Califor- consider in planning for the support of the field. nia, have also raised concerns about how well Any new mechanisms proposed must address the semiconductor industry is equipped to han- this need. dle waste management and disposal. If the In the electronics industry, a large number of semiconductor and other advanced materials relatively small firms play a key role ingener- industries are to continue to prosper in the ating new process concepts and equipment. United States, it is important that the expertise These firms face important research problems of chemical engineers be applied to every aspect in fundamental scic.tce and engineering that of chemical handling in manufacturing, from would benefit markedly from the insights of procurement through use to disposal. academic chemical engineering researchers. Ac- ademe: researchers should seek out aid forge IMPLICATIONS OF RESEARCH ;inks to these small firms that stand at the FRONTIERS crucial step between laboratory research and production processes. Potential mechanisms for Industry has been the prime mover in ad- accomplishing this are described in Chapter 10. vancing technology in electronic, photonic, and The current undergraduate curriculum in recording materials and d.evices. It will remain chemical engineering, although it providesan so for the foreseeable future. University re- excellent conceptual b..se for graduates who starch groups need to develop and inaintain move into the electronics Indus' .z.s. could be good communication with "ounterpart research improved by the introduction of instructional groups in industry. Collanorative mechanisms material and example problems relevant to the are needed to promote academic-industrial cou- challenges outlined in this chapter. This world pling. not require the creation of new courses, but This coupling will become even more im- rather the provision of material to enrich existing portant as the electronics industry hires ever ones. This theme is echoed, more broadly, in greater numbers of chemical engineers. Sir Chapter 10. 60 FRONT/FRS IN CHEMICAL EvGINEERING

NOTES (d) Pre'iminary data from the 1986 survey of manufacturing industries provided by the NSF I. National Research Council, National Materials Division of Science Resources Studies. Advisory Board. State of the Art Reviews: Ad- vanced Processing of Electronic Materials in the SUGGESTED READING Ur'ted States and Javan. Washington, D.C.: M. Bohrer, J Amelse, P. Narasimham, B. Tariyal, National Academy Press, 1966. J. Turnipseed, R. Gill. W. Moebuis, and J. Bo- 2. National Academy of sciences- National Academy 'leker. "A Process for Recovering Germanium of Engineering- Institute of Medicine, Committee from Effluents of Optical Fiber Manufacturing." on Science, Engineering, and Public Policy. "Re- J. Lightwave Tech., LT-3 (3), 1984, 699. search Briefing on High-Temperature Supercon- T. Li. ed. Optical Fiber Communications, Vol. I. duct;vity," in Research Briefings 1987. Washing- Orlando, Fla.: Academic Press, 1984. ton, .C.: National Academy Press, 1987. P. D. Maycock and E. N. Striewalt. A Guide to the 3. (a) National Science Foundation. Division of Sci- Photovoltaic Revolution. Sunlight to Electricity in ence Resources Studies. Employment (.., Sci- One Step. Emmaus, Pa.: Rodale Press, 1984. R. H. Pk..-ry and A. A. Nishimura. "Magnetic Tape entists, Engineers, and Technicians in Man- Production," in Kirk-Othmer Encyclopedia of ufacturing Industries: 1977 (NSF 80-306). Chemical Technology, 3rd ed., Vol. 14, p. 744. Washington, D.C.: U.S. Government Printing New York: Wiley-Interscience, 1979. Office, 1980. Solar Engineenng Research Institute. Basic Photo- (b) National Science Foundation, Division of Sci- voltaicPrinciples and Methods (FT-290-1448). ence Resources Studies.Scientists,Engi- Washington, D.C.: U.S. Government Printing Of- neers, and Tet link:tans in Manufacturing and fice, 1982. Non-Manufacturing Industries: 1980-81 (NSF S. M. Sze. Semiconductor Desires: Physics and 83-324). Washington, D.C.: U.S. Government Technology. New York: John Wiley & Sons, 1984. Pnnting Office, 1983. W. Thomas, ed. SPSE Handbook of Photographic Science and Engineering. New York: Wiley-Inter- (c) National Science Foundation, Division of Sci- science, 1973. ence Resources Studies. Scientists, Engineers, L. F. 7 -,npn, C. G. Willson, and M. J. Bowden. and Technic,' - . in Manufacturing Industries: Intrt, lion to Mu rolithographv (ACS Sympo- 1983 (NSF 85-328). Washington, D.C.: U.S. sium Series No. 219). Washington, D.C.: American Government Printing Office, 1985. Cheifiical Society, 1Q83. F ! V E

Polymers, Ceramics, and Composites

heroical enginc4rs have long been involved with materials science and engineering. This involvement will increase as new materials are developed whose properties depend strongly on their microstructure and processing history. Chemical engineers will probe thenature of microstructurehow it is formed in materials ant; what factors are involved in controlling it. They will provide a new fusion between the traditionally separate areas of materials synthesis and materials processing. And they will bring new approaches to the problems of fabi icating and repairing complex materials systems.

61 7i 62 FRONTIERS IN CHEMICAL ENGINEERING

FEW YEARS AGO,Who would have dreamed that an aircraft couldcir- cumnavigate the earth without landing or refueling? Yet in 1986 the novel aircraftVoyagerdid just that (Figure 5.1). The secret ofVoyager'slong flight lies in advanced materials that did not exist a few years ago. Much of the airframe was constructed from strong, lightweight polymer-fiber composite sections assembled with durable, high-strength adhesives; the engine was lubri- cated with a synthetic multicomponent liquid designed to maintain 'ubricity for z long time under FIGURE 5.1The first airplane to circk the globe without refueling was continuo'soperation. 'These Voyager, which accomplished this feat in 1986. This novel aircraftwas made possible by high-performance lightweight materials and adhesives that were special materials typify fie ad- used in its construet;on. Chemical engineering reF arch is crucial to the design vances being made by scientists of such new matt.uls and thi.r large-scale, efficient production. Copyright and engineers to meet the de- 1986 by Doug Shane. Visions. mands of nuy.lern -ociety. The future of industries such as .. ansportation, communications, electronics, molecular conformation, mi,:roscopi: and mac- and energy conversion hinges on new and im- roscopic structure, and methods of combining proved materials and the processing technolo- the components in a way that will maximize gies required to produce them. Recent years product performance. have seen rapid advances in our understanding Chapter 4 discussed chemical engineering of how to combine substances into materials challenges presented by materials and chemi- with special, high-performance properties and cally processed devices for information storage how to best use these matenals in sophisticated and handling. In this chapter, five additional designs. classes of materials are covered: polymers, Chemical engineers have long been involved polymer composites, advanced ceramics, ce- in materials science and engineering and will ramic composites, and composite liquids. become increasingly important in the future. 'linen. contributions will fall in two categories. e.HALLENGES TO CHEMICAL ENGINEERS For commodity materials, which are nonpro- prietaryforme-' itionswithwell-establis.ied The revolution in materials science 1.nd en- chemical compositions and property standards, gineering presents both opjortunities and chal- chemical engineers will help maintain U.S. com- lenges to chemical engineers. With their Insic petitiveness by creating and improving pro- background in chemistry, physics, and m dile- cesses to Lake these chemicals as pure as matics and their understanding of t: ansport possible and in high yields at the lowest possible phenomena, thermodynamics, reaction engi- investment and operating costs. For advanced nering, and process design, chemical engineers materials, which are generally multicomponent, can bring innovative solutions to the problems often proprietary, compositions designed to have of modern materials technologies. But itis very specific performance properties in specific imperative that they depart from the traditional uses, the competitive edge will come from "think big" philosophy of the profession; to chemical engineers who excel in controlling participate effectively in modern materials sci-

74('' POLYMERS, CERAMICS, AND COMPOSITES 63

ence and engineering the must learn to "think polybenzothiazole, as well as traditional poly- small." The crucial phenomena in making mod mers such as polyethylene. Ultrahigh-strength ern advanced materials occur at the molecular fibers of polyethylene have been prepared by and microscale levels, and chemical engineers gel spinning. The same concept, controlling the must understand and learn to control such molecular orientation of polymers to produce phenomena if they are to engineer the new high strength, is also being achieved through products and processes for making them. This other processes, such as fiber-stretching carried crucial challenge is illustrate in the selected out under precise conditions. materials areas described in the following sec- In addition to processes that result in mate- tions. rials with specific high-performance properties, chemical engineers continue to design new pro- Polymers cesses for the low-cost manufacture of polymers. The UNIPOL process for the manufacture The modern era of polymer science belongs of polyethylene is a good example of the con- to the chemical engineer. Over the years, poly- tributions of engineering research to polymer mer chemists have invented a wealth of novel processing. Polyethylene is probably the quin- macromolecules and polymers. Yet understand- tessential commodity polymer. It has been man- ing how these molecule: can be synthesized and ufactired worldwide for decades, and current processed to exhibit their maximum theoretical U.S. production exceeds 15 billion pounds per properties is still a frontier for research. Only year. Considering the global capital investment recently has modern instrumentation been de- in existing plants for making polyethylene, it veloped to help us understand the fundamental could be argued that inventing a new process interactions of macromolecules with them- for its manufacture is a waste of time and money. selves, with particulate solids, with organic and Not so. Chemical engineers at Union Carbide inorganic fibers, and with other surfaces. Chem- designed a proprietary catalyst that allowed ical engineers are using these tools to probe the polyethylene to be made in a fluidized-bed, gas- microscale dynamics of macromolecules. Using phase reactor operating at low temperature and the insight gained from these techniques, they pressure (below 100°C and 21 Bar). The resulting are manipulating macromolecular interactions process produces a polymer with exceptional both to develop improved processes and to uniformity and can precisely control the molec- create new materials. ular weight and density of the product. The The power of chemical processing for con- advantages of the process (including a low sa ety trolling materials structure on the microscale is hazard from the mild operating condition_ and illustrated by the current generation of high- minimal environmental impact since there are strength po!ymer fibers, some of which have no liquid effluents and unreacted gases are strength-to-weight ratios an order of magnitude recirculated) are such that, in 1986, UNIPOL greater than steel. The best known example of process iicensees had a combined capacity suf- these fibers, Kevlar, is prepared by spinning ficient to supply 25 percent of the world's an aramid polymer from an anisotropic phase demard for polyethylene. This is remarkable (a liquid phase ii. which molecules are sponta- mark'.t penetration for a new process technol- neously oriented over microscopic dimensions). ogy for a mature commodity, particularly in This spontaneous orientation is the result of light of the tremendous existing (and fullyam- ...th the processing conditions chosen and the ortized) worldwide capacity for polyethylene. highly rigid linear molecular structure of the In 1985, Union Carbide and Shell Chemical aramid polymer. During spinning, the oriented successfully extended the UNIPOL process to regions in the liquid phase align with the fiber the manufacture of polypropylene, another ma- axis to give the resulting fibei high strength and jor polymer commodity. Interestingly, the first rigidity. The concept of spinning fibers from two licensees for the new polypropylene process anisotropic piases has been extended tc both were a Japanese chemical company and a Ko- solut;9ns and meits :A newer polymers, suet, as rean petroleum company.

7 " 64 FRovriERs iv ciiilitiicti. 1: \'G /.VEER /.VG

Polymer Composites chemical engineers and textile engineers might lead to ways of selecting the warp, woof, and Polymer composites consist of high-strength weave in fabrics of high-strength fibers to end or high-modulus fibers embedded in and bonded up with trussworks for composites with highly to a continuous polymer matrix (Figure 5.2). tailored dimensional distributions of properties. These fibers may be short, long, or continuous. They may be randomly oriented so that they First-generation polymer composites (e.g., fiberglass) used thermosetting epoxy polymers impart greater strength or stiffness in all direc- reinforced with randomly oriented short glass tions to the composite (isotropic composites), or they may be oriented in a specific direction fibers. The filled epoxy resin could be cured so that the high-performance characteristics of into a permanent shape in a mold to give lightweight, moderately strong shapes. the composite are exhibited preferentially along one axis of the material (anisotropic compos- The current generation of composites is being made by hand laying woven glass fabric onto a ites). These latter fiber composites are based mold or preform, impregnating it with resin, on lie principle of one-dimensional microstruc- and curing to shape. Use of these composites tural reinforcement by disconnected, tension- bearing "cables" or "rods." was pioneered for certain types of military aircraft because the lighter airframes provided To achieve a material with improved prop- greater cruising range. Today, major compo- erties (e.g., strength, stiffness, or toughness) in nents for aircraft lnd spacecraft are manufac- more than one dimension, composite laminates tured in this manner, as are an increasing can be formed by bonding individual sheets of number of automobile components The current anisotropic composite in alternating orienta- tions. Alternatively, two-dimensional reinforce- generation of composites are being use iin automotive and truck parts such as body panels, ment can be achieved in a single sheet by using fabrics of high-performance fibers that have hoods, trunk lids, ducts, drive shafts, and fuel tanks. In such applications, they exhibit a better been woven with enough bonding in the cross- overs that the reinforcing structure acts as a strength-to-weight ratio than metals, as 4,01 as i nproved corrosion resistance. For exampk, a connected net or trusswork. One can imagine polymer composite automobile hood is slightly that an interdisciplinary collaboration between lighter than one of alliminurn and more than twice as light as one of steel. The lzvel of energy required to manufacture this hood is slightly lower than that requilsx1 for steel and about 20 percent of that for aluminum; meitrig and tooling costs are lower and permit more rapid model changeover to accommodate new de- 6pns. Polymer co.nposite hoods and trunk lids are commercial on the 1987 models of one major U.S. automobile line, and the early problems of higher manufacturing cost and of achieving adequate production have been largely overcome. The mechanical strength exhibited by these FIGURE 5.2Fibers that are either very strong or very stiff composites is essentially that of the reinforcing can be used to reinforce polymers and ceramics. The glass fibers, although this is often compromised revolting materrils. known as composites. usually have one by structural defects. Engine -ring studies are of foie structures depicted in this figure. Clockwise from the yielding important infurmation about how the upper left, reinforcement may be accomplished by embed- properres of these structures are influenced by ding randomly oriented fibers. by orienting fibers along a particular ax. by assembling reinforced layers into lami- the nature of the glass-resin interface and by nates, or by embedding fabrics of reinforcing fibers in the structural voids and similar defects and now material nicrodefects can propagate into structural fail-

74 POLYMERS, CERAMICS, AN!) COMPOSITES 65

ure. These composites and the information gained in turbines permit the construction of engines from studying them have set the stage for the that operate at much higher temperatures than next generation of polymer composites, based metallic engines, thus greatly increasing their on high-strength fibers such as the aramids. thermodynamic efficiency and compactness. Ceramic liners and other ceramic components Advanced Ceramics in diesel engines provide added benefits, such as the elimination of the need for water cooling For most people, the word "ceramics" con- and the prompter ignition of the fuel. An in- jures up the notion of things like china, pottery, vestment in wear-resistant ceramic cutting tools tiles, and bricks. Advanced ceramics differ from can be more than repaid by the decrease in these conventional cerara:::s by their ( Imposi- downtime for sharpening or replacing a dulled tion, processing. and microstructure. For ex- et- worn metallic tool. ample: Given these advantages, it is not surprising Conventional ceramics are made from nat- tnat market forecasts f r advanced ceramics ural raw materials such as clay or silica; ad- (including ceramic composites) are optimistic: vanced ceramics require extremely pure man- in fact, sales in the year 2000 are predicted to made starting materials such as silicon carbide, be $20 billion. The market for advanced ce- silicon nitride, zirconium oxide, or aluminum ramics in heat engines is slated to grow by 40 oxide and may also incorporate sophisticated percent per year to a total of $1 billion in 2000. additives to produce specific microstructures. The use of advanced ceramics is predicted to Conventional ceramics initially take shape grow 16 percent per year over the next 5 years, on a potter's wheel or by slip casting and are and sales for automotive applications are fore- fired (sintered) in kilns; advanced ceramics are cast to increase frog- '''''4 million per year in formed and sintered in more complex processes 1986 to $6 billio :ar by the end of t.,e such as hot isostatic pressing. century. The microstructure of conventional ce- Uniform microstructure is crucial to the su- ramics contains flaws readily visible under op- perior performance of advanced ceramics. Ina tical microscopes. the microstructure of ad- ceramic material, atoms are held in place by vanced ceramics i ar .nore uniform and typically strong chemical bonds that are impervious to at-1 by corrosive materials or heat. At the is examined for defects under electron micro- s -open capable of magnifications of 50,000 times same time, these bonds are not capable of much or more. "give." When a ceramic material is subjected to mechanical stresses, these stresses concen- Advanced ceramics have a wide range of trate at minute imperfections in the microstruc- application (Figure 5.31. In many cases, they ture, initiating a crack. The stresses at the top do not constitute a final product in themselves, of the crack exceed the threshold for breaking but are assembled into components critical to the adjacent atomic bonds, and the crack prop- the successful performance of some other com- agates throughout the material causing a cata- plex system. Commercial applications of ad- strophic brittle failure of the ceramic body. The vanced ceramics can be seen in cutting tools, reliability of a ceramic component is directly engine nozzles, components of turbines and rclated to the number ar 1 type of imperfections turbochargers, tiles for space vehicles, cylinders in its microstructure to store atomic and chemical waste, gas and oil A. the requirements for greater homogeneity drilling valves, motor plates and shields, and in ceramics become more stringent, and the electrodes for corro ive liquids. scale at which imperfections °cm- becomes Because advanced ceramics provide key com- smallet, the need for chemical processing of ponents to other technologies for major im- ceramics becomes more compelling. Traditional provements in performance, their impact on the approaches to controlling ceramic microstruc- U.S. ecc.nomy is much greater than is indicated ture, such as the grinding of powders, are by their sales figures. Ceramic compe-,..nts used reaching the limits of their utility for mic rostruc-

7;J 66 FRO1I1ERS (111:311(11. FVGINEERING

Nuclear fuel Mph - temperature Industrie' Nuclear fuel fumes lining cladding

Electrode material Control material Moderating material Neat sink for electronic parts Radiation Remo, mining Reractoriness resistance Insulation Fleet collection Toolsand II. Thermal conductivity Laser bode

Nigh strength Optics' condensing

Light- emitting Fluorescence diode

Tufting blades Optical

Low therms' Translucence expansion Hcet- Soled resakent translucent lubricants porcelain Biological Electric chemical megnetic Optical Lubrication conductivity Precision instrument Electrical Optical Biological insulation cunmunianion compatibility Electrical conductivity cable Catalysis adsorption Senbconductivity Artificial Piezoelectric integrated bons and Corrosion Dielectric circuit resistance tooth Magnetic substrate Resistance Meting Fixed catillYSIcattiatGeothermal end element Met machinistofhhore devilloPrrhint Varistor equipment sensor Piezoelectric filter Chemical atinlmimil Memory element

FIGURE 5.3The myriad functions, properties, and applications of advanced ceramicsReprinted from High-Tr( Itnelagy Cerannt srn Japan. National MateriaN Advisory Board. National Research Council. 1984. tura) control. Chemical engineers have an un- Sol-Gel Processing paralleled opportunity to contribute their expertise in reaction engineering to problems that Tic use of sol-gel techniques to prepare are in need of new analytical, synthetic, arid ceramic powders has recently attracted much processing tools. These include sol-gel process- interest in academia and industry. Sol-gel tech- ing and the use of chemical additives in ceramic niques involve dissolving a ceramic precursor processing. (e.g., tetramethyl orthosilicate) in a solvent and

u POLYMERS, CERAMICS. AND COMPOSITES 67

FIGURE 5.4Stages in sol-gel processing are captured by a new electron microscopy technique. (I) Spherical particles tens of nanometers across can be seen in a colloidal silica sol. (2) Addition of a concentrated salt solution initiates gelation. (3) The gelled sample. after drying under the electron beam of the microscope. shows a highly porous structure. Courtesy. J. R. Bellare. J. K. Bai's. and M. L. Mecanney. Unversity of Minnesota. subjecting it to a carefully controlled chemical dense amorphous materials whose synthesis reaction, hydrolysis (Figure 5.4). When the might r ,t otherwise be possiere, hydrolysis products first al-pear as a separate materials with controlled degrees of poros- phase, they are fine colloids consisting of small ity and possibly tailored surfaces within pores, particles, some with radii as small as a few and nanometers. This colloidal suspension (the sol) ceramics with surfaces modified to alter further reacts and polymerizes to form a porous their response to mechanical forces or 'o pro- high-molecular-weight solid (the gel) that con- mote their adhesior, to other materials. tains the solvent as a highly dispersed fluid component in its internal network structure. Sol-gel processes also allow the manufacture of Removal of the solvent leaves behind so!ids preforms that, upon sintering, collapse to a final product with the proper shape. with a wide variety of macrostructures depend- There are many unresolved problems in sol- ing on the solvent and the way in vhich it was removed. These macrostructures can be sin- gel processing, many of which revolve around tered to convert them to dense ceramics. the poorly characterized chemistry of the pro- Sol-gel techniques are of interest because cess. Understanding and controlling the poly- they can be used to pr :pare powders with a merization reactions that produce the gel are key challenges, as are characterizing and opti- narrow diAribution of particle size. These small particles undergo sintering to high density at mizing both the removal of fluid from he gel and the subsequent sintering of the porous solid temperatures lower by several hundred degrees to a fully dense ceramic 'oody. Solving these centigrade than those used in conventional ceramic processing. Sol-gel processes may also problems will make sol-gel processing the process of choice for the synthesis of a v it;. variety be used to prepare novel glasses and ceramics of ceramics, glasses, and coatings. such as ceramics with novel microstructures and Chemical Additives in Ceramic Processing distributions of phases, amorphous powders and dried gels that can Another area to which chemical engineers be processed without crystallization to fully can contribute is the use of chemical additives 68 FRONTIERS IN CHEMICAL ENGINEERING

to improve the properties of ce- i-amic materials. For example, zirconium oxide can form a metastable state in ceramic bodies that is denser than its normal state. The incorporation of suitable chemical additives stabilizes the metastable state sufficiently to allow the fabrication of parts containing it. When a crack forms in such a ceramic part, the zirconium oxide region at the crack tip changes to the less dense form.he rebuffing expansion blunts the crack tip and stops its propagation (Figure 5.5). This strategy for using a chemical ad- ditive to improve ceramic resistance to cracking is called transformation toughening.

Ceramic Composites FIGURE 5.5Zirconia ceramics can be made strongzr and less brittle by using Like polymer composites. ce- chemical additives to stabilize a more compact tetragonal structure that does ramic composites consist of high- not naturally occur at room temerature. When such a phase is subjected to stress, it can change phases. expanding to the monoclinic structure. This strength cr high-modulus fibers expansion fills any stress-initiated crack and prevents it from moving. This embedded in a continuous ma- micrograph shows a zirconia ceramic composed of lozenge-shaped grains. A trix. Fibers may be in the form pore with grain boundaries radiating from its top and base dominates the of "whiskers" of substances such picture. Courtesy, National Physical Laboratory (United Kingdom). as silicon carbide or aluminum oxide that are grown E. s single crystals and that in its fluid form, be capable of wetting the fibers. therefore have fewer defects than the same Chemical bonding between the two componen.s substances in a bulk ceramic (Figure 5.6). Fwers can then take place. in a ceramic composite serve to block crack C...camic matrix composites are produced by pror gation; a growing crack may be deflected one cf several methods. Short fibers and whis- to a fiber or might pull the fiber from the matrix. kers can be mixed with a ceramic powder before Hot'. processes absorb energy, slowing the the body is sintered. Long fibers and yarns can propagation of the crack. The st-eng.h. stiffness, be impregnated with a slurry of ceramic particles and toughness ofceramic composite is prin- and. after drying, be sintered. Metals (e.g., cipally a function of the reinforcing fibers. but aluminum. magnesium. and titanium) are fre- the matrix makes its own contribution to these quently used as matrixes for ceramic composites properties. The ability of the composite material as well. Ceramic metal-matrix composites are to conduct heat and current is strongly influ- fabricated by infiltrating arrays of fibers with enced by the conductivity of the matrix. The molten metal so that a chemical reaction be- interaction between the fiber and the matrix is tween the fiber and the .-aetal can take place in also important to the mechanical properties of a thin layer surrounding the fiber. the composite material and is mediated by the As with advanced ceramics, chemical reac- chemical compatibility between fiber and matrix tions play a crucial role in the fabrication of at the fiber surface. A prerequisite for adhesion ceramic composites. Both defect-free ceramic between these two materials is that the matrix. fibers and optimal chemical bonds between fiber POLYMERS. CERAMICS. AND COMPOSITES 69

chemical processing industries, and chemical engineers face continuing challenges in tailoring their end-use properties. Some of these challenges are illustrated in the following examples:

Motor lubricants are complex liquid com- Posites in which components provide different performance characteristics. The basic component is a hydrocarbon oil with a fixed boiling range. It must have sufficient viscosity at engine operating temperatures to prevent the friction and wear of moving surfaces, but must be fluid enough below freezing temperatures for winter start-up. Viscosity modifiers are high-molecular-weight polymers that reduce the temperature coefficient of viscosity (viscosi y index). Sus- pended colloidal particles of celcium ormag- FIGURE 5.6This is a fractured sample of a ceramic nesium carbonate are added to neutralize engine composite (alumina with 30 volume-percent silicon carbide acids and are stabilized by adsorb--1 polymers whiskers). The lighter regions of circular or cylindncal and surfactants to prevent coalescence. Solids shape are randomly oriented whiskers protruding from the dispersants are low-molecular-weight polymers fractured surface. The rod-like depressions in the surface mark places where whiskers nearly parallel with the fracture with functional groups that pick up carbon were pulle., out. Courtesy. Roy W Rice. W. R. Grace and particles generated in combustion and maintain Company. them in suspension. At low temperatures, the waxes (straight-chain paraffin hydrocarbons) in the lubricant form long crystals to set up a solid and matrix are required for these composites to gel. To prevent this, low-molecular-weight poly- exhibit the desired mechanical properties in mers. called pour point depressants, are added use. Engineering these chemical reactions in to co-crystallize with the wax; the resulting reliable manufacturing processes requires the smaller crystals do not gel. Finally, there are expe-tise of chemical engineers. antiwear additives and antioxidants to reduce engine wear and deposits. Lubricants with out- Composite Liquids standing viscosity indexes enable an engine to start when the lubricant temperature is as low Afinal important class of composite materials as 40 °C and yet operate well when the lubri- is the composite liquids. Composite liquids arc cant temperature is as high as 200°C. Other highly structured fluids based either on particles additives allow broadening tht. :nperature range or droplets in suspension, surfactants, liquid further by providing increased thermal and ox- crystalline phases, or other macromolecules. A idative stability. The use of synthetic base oils number of composite liquids are essential to the allows still broader ranges of operating temper- needs of modern industry and society because atures, up to 500°C. they exhibit properties important to special end Advanced adhesives are composite liquids uses. Examples include lubricants, hydraulic that can be used, for example, to join aircraft traction fluids, cutting fluids, and oil-drilling parts, thus avoiding the use of some 30,000 muds. Paints, coatings, and adhesives may also rivets that are heavy, are labor-intensive to be composite liquids. Indeed, composite liquids install, and pose quality-control problems. Ad- are valuable in any c se where a well-designed hesives research has not involved many chem- liquid gate is absolutely essential for proper ical engineers, but the ger eric problems include delivery and action. surface science, polymer rheology and ther- All composite liquids are produced by the modynamics, and molecular modeling of ma-

'1 r. RON I IERS IV (ill if IC %I, k VG! \FERIVG terials near or at interfaces. The scientific and under stand these materials are mounting major engineering skills needed are very similar to efforts to exploit these developing markets. those needed for polymer composites and multicomponent polymer blends. The time-tested Polymers and Polymer Composites mechanical methods developed for joining metals are not satisfactory for composite and other The Panel on Advanced Materials of the NSF advanced materials, and chemical engineers JapaneseTechnologyEvaluationProgram skilled in interfacial science are well qualified (JTECH) issued a report in 1986 assessing the to contribute to this area. status and direction of Japanese research and Another class of liquid composites is that development efforts in several high-technology of coating compositions used to deposit thin polymer areas.' The panel noted that the major films on a substrate or other films; these com- Japanese chemical :ompanies already manufac- posites have evolved from typical paints and ture most of the commercially available poly- varnishes into multilayer films in which each mers and that "since 1970, there has been an layer contributes specific p rties to the en- increasing flow of upgraded technology from semble. Such films may paints used for span to the United States." For engineering sealing and decorative purses, films used for plastics and resins, the panel judged the United printing or packaging purposes, of multilayer States to be ahead in basic research (although products used in recording tapes and photo- the lead is diminishing), on a par with Japan in graphic produc'3. All are based on generic advanced development, and behind Jar n in scientific principles that include many common product implementation. elements from thermodynamics, polymer sci- The JTECH panel also compared the U.S./ ence, rheology, and fluid mechanics. Japanese position in high-strength/high-modulus polymer research and development. Its conclu- Liquid composites seldom behave as New- sions substantially agree with the following tonian fluids. These complex mixtures usually statements, drawn from a -ecent report of the contain macromolecules, suspended particles, NAS/NAE/10M Committee on Science, Engi- and surfactants. They are frequently multi- neering, and Public Policy.2 The United States phase, and changes in phase composition o- has a strong position in the development of formation of new liquid phases may occur ove- high-strength polymers, but compardhle activity the range of operating conditions. Phase com- in this area exists in Japan. The Japanese position may be shifted by chemical reaction, Ministry of Interuatioi.al Trade and Industry by shear forces, or simply by changes in tem- (MITI) has designated the development of a perature or pressure. Liquid-liquid and liquid- "third-generation" fiber as a government-sub- solid equilibria are crucial. Detailed molecular sidized project, beginning in 1983 and targeted understanding of the interactions among such for practical application by 1988. Because of components as surfactants, polymers, and par- the overwhelming importance of the load-bear- ticles is essential for the rational design of liquid ing fibers in composites, the field is sensitive to composites. Much of this design is now accom- breakthroughs in stronger fibers. SW 's a break- plished by informed empiricism, which is useful through could come from Japan; for example, for the incremental improvement of current a Japanese group was first to patent a process products but inadequate for major changes and for making high-strength fibers from poly(ethylene innovation. terephthalate). Much of the technology used for manufac- INTERNATIONAL COMPETITION turing carbon fibers in the United States is licensed from Japanese companies. The high The potential markets for the advanced ma- level of Japanese carbon-fiber technology sug- terials discussed in this chapter are lucrative, gests that Japanese companies may produce and most nations that possess the technoiotocal many of the expected future advances in these infrastructure needed to invent, develop, and materials.

80 POLYMERS, CERAMICS, 4,\'D COMPOSITES 71

The United States currently has a strong of our society and to maintain our international position in composite manufacturing and pro- competitive position in technology. cessing technology and leads the world in developing major applications in aircraft, sporting goods, and automotive components. There is Composite Liquids growing overseas activity in composites tech- The field of composite liquids has not received nology, however, particularly among European much attention outside the industries associated aircraft Lompanies. with specific liquid products (e,g., the petroleum industry). In areas such as lubrication, the United States has clear technological leader- Ceramics and Ceramic Composites ship. The situation is less clear for liquid crystals Japan is the United States' chief competitor and adhesives, where there is greater compe- in ceramics. There is a widespread but false tition from Europe and Japan. perceriion that Japan leads the United Slates in ceramics in general. Nevertheless, it is clear INTELLECTUAL FRONTIERS that the Japanese effort in ceramics is comrre- hensive and long term and that several Japanese A wide variety of chemical engineering re- companies lead their U.S. industrial counter- search frontiers involve advanced materials and parts in specific technologies, There is tremen- belong in the mainstream of academic chemical dous enthusiasm in Japan for the potential of engineering departments. The ''flowing list of ceramics, and a recent report of a U.S. visiting frontiers ranges from the molecular level to the team to Japan' reached the following conclu- systems level. sions: Microscale Structures and Processes The Japanese are committed to vigorously developing and dominating the field of advanced The study of materials has traditionally cen- ceramics. They have put in place a well-inte- tered on the influence of molecular composition grated national effort primarily based on goy- and microstructure on mechanical, electrical, ernvit-industry interactions. optical, and chemical properties. At the raolec- industrial management comm. ,,ent to long- ular level are a variety of research frontiers that term research and development at,:ars to be can profitably draw chemical engineers into more solid in Japan than in the United States. close collaboration with physical and theoretical Japanese research is focusing increasingly chemists. They include the following research on fundamental research issues. Government- areas. inspired basic research programs are being put in place, and Japan's position as a net consumer New Concepts in Molecular Design of of basic ceramics research may chanbe in the Composite Materials coming years. The toughest challenge and the greatest op- Given the potential future importance of ce- pui kinity in chemical engineering for high-per- ramics in areas as diverse as electronics (see formance materials lie in the development of Chapter 4), machine tools, heat engines, and wholly new designs for composite solids. Such superconductors (see Chapter 4), the United materials are typified by composites reinforced States can ill afford t- surrender technical lead- bythree-dim( isionalnetworks andtruss- ership to its compte :ors. The dominant trent, worksmicrostructures that zre multiply con- in the field is toward materials with finer micro- nected and that interpenetrate the multiply con- structures, fewer defects, and better ;nterac- nected maul), in which they are embedded. In tions at interfaces (particularly in composites). such materials, both reinforcement ilnd matrix Chemical processes provide important tools to are continuous in three dimensions; the com- capture the promise of ceramic.; for the benefit posite is bicontinlous, Geometric prototypes of 72 I. RON HERS IN ENGINEFRING

such structures are found in certain liquid crys- can be deactivated when they are intercepted tals and colloidal gels at.d in bones and shells. at solid boundaries, the high interfacial area of The challenge is to break away from today's a prepolymerized composite represents a radi- technology, to go beyond today's research in cally different environment from a conventional two-dimensional fabric-like reinforcement, and bulk polymerization reactor, where solid bound- to determine how to create truly three-dimen- aries are few and very distant from the regions sional microstructures and design, construct, in which most of the polymerization takes place. anti process them so as to control the properties The polymer molecular weight distribution and of the composite. cross-link density produced under such diffu- A potentially promising area is "molecular sion-controlled conditions will differ apprecia- composites,- in which the fiber and its sur- bly from those in bulk polymerizations. rounding matrix have the same composition and Second, the molecular orientation of the differ only in molecular structure or morphol- fiber and the prepolymer matrix is important. ogy. This might involve forming the composite The rate of crystal nucleation at the fiber-matrix from very stiff, linear polymer molecules, some interface depends on the orientation of matrix -` which are aligned during the forming step ac molecules just prior to their change of phase riforcing crystallites in the amorphous re- from liquid to solid. Thus, surface-nucleated gionsthe matrix. An analogous ceramic com- morphologies are likely to dominate the matrix posite may be envisioned. There are difficult structure. engineering problems to be solved in learning Third, the ultimate mechanical properties how to control the orientation of the crystalline of a composite will be strongly influenced by regions and the ratio between crystalline and the degree to which the matrix wets the fiber amorphous regions in the material. surface and by the degree of adhesion between the two after curing. Both phenomena depend on intimate details of the surface science of the The Role of Interfaces in Materials two phases, about which little is known. Chemistry Molecular modeling techniques, augmented Two general problems relate to the role of by careful measurements o structure of the interfaces in advanced materials. The first is interfacial regions, hold promise for ^lucidating simply that we do not have the theory or the details of these three aspects of terfacial computational or experimental ability to under- control of matrix polymerization. stand the interatomic and microscopic interactions at 'he interfaces between components of an advanced material, on which its properties Understanding the Molecular Behanthr of are critically dependent. There is a general need Complex Liquid% for research on processes at interfaces and on thestructure-property-performancerelation- Basic understanding of the liquid state of ships of interfaces. matter still lags behind that of the solid and The second problem relates to the role that gaseous states. Our knowledge of interactions interfaces play in mediating chemical reactions in multicomponent liquids containing macro- in the synthesis of composite materials. This molecules and suspended solids is extremely problem has three parts, which are illustrated limited. Thus, the study of complex liquids, here for polymeric composites. including polymer solutions, cols, gels, and composite liquids, is a significant challenge for First, in cc nposites with high fiber con- chemical engineers. centrations, there is little matrix in the system The ability to predict liquid-liquid and liquid- that is not near a fiber surface. Inasmuch as solid equilibria in complex systems is still rather polymerization processes are influenced by the undeveloped, in part because of the lack of diffusion of free radicals from initiators and systematic and molecularly interpreted experi- from reactive sites, and because free radicals mental information. Considerable reset h has

04 POLYMERS, CERAMICS, 4ND COMPOSITES 73

been conducted an the by ':a for of liquids near for ration is the chemistry of sol -gel sys- their critical po nts, on lower critical solution te 1,..tnight lead to nonoxide ceramics.

1 -omena, on spinodal decomposition, and on related dynamics such as the growth and morphology of new phases, but generalized corre- The Intimate Connection Between Materials lations and connections of theory to practice Synthesis and Froc'ssing are few. Materials synthes 'and materials processing Molecularly motivated empiricisms, such as have classically been thought of as set, .e the solubility parameter concept, have been activities, and in the days of simple, horn_ ,e- valuable in dealing with mixtures of weakly neous materials, they were. But today's com- interacting small molecules where surface forces plex materials are bringing these two areas are small. However, they are completely ir2d- closer together in research and in practice. Four equate for mixtures that involve macromole- outstanding intellectual challenges demonstrat- cules, associating entities like surfactants, and ing this connection are described in this section. rod-like or plate-like species that can form ordered phases. New theories and models,... needed to describe and understand these sys- Processing of Complex Liqu"s tems. This is an active research area where Complex liquids are ubiquitous in materials advances could lead to better ,..,.ier3tanding of manufacture. In some cases, they are formed the dynamics of polymers and colloids in so- and must be handled at intermediate steps in lution, the theological and mechanical proper- the manufacture of materials (e.g., sols and gels ties of these solutions, and, tore generally, the in the making of (...ramics, mixtures of monomer fluid mechanics of non-Newtonian liquids. and polymer in reactive processing of polymers). In other cases (e.g., composite liquids), Chemical Dynamics and Modeling, of they are the actral products. Un&rstanding the Molecular Processes properties of complex fluids and the implications of fluid properties for the design of materials Chemical dynamics and modeling were iden- processes or end uses presents a formidable tified as important research frIntiers in Chapter intellectual challenge. 4. They are critically important to the materials Complex liquids seldom behave as classical &cussed in this chapter as well At the inolec- Newtonian fluids; thus, analysis of their behav- ular scale, important areas of i, F ti gati on in- ior requires a thorough understanding of non- clude studies of statistical mechanics, molecular Newtonian rheology. The importance of this and particle dynamics, dependence of molecular knowledge is illustrated by the following two motion on intermoler lar and interfacial faeces, examples: and kinetics of chemical processes and phase changes. The problem of processing complex liquids Mechanistic studies are particularly needed while they are undergoir, rapid polymerization for the hydrolysis and polymerization reactions is an important challenge in reactive polymer that occur in sol-gel processing. Currently, little processing (e.g., reactive injection molding and is known about these reactions, evenitsimple reactive extrusion). In these processes,the systems. A short list of needs includes such viscosity of a reaction mixture, as it proceeds rudimentary data as the kinetics of hydrolysis from a feed of monomers to a polymer melt and polymerization of single alkoxide sol-gel product, may cLange by7decades or more in systems and identification of the species present magnitude. Fluid mixtures flowing into a mold at various stages of gel polymerization. A study of complicated geometry may exhibit large tem- of the kinetics of hydrolysis and polymerization perature gradicat- from the highly exothermic of double all:oxide sol-gel systems m,, .lead chemical reactions taking place and significant to the production of more homogeneous ce- spatial variations in viscosity and molecular 1 artics by sol-gel routes. Another major area weight distribution.

83 FRONTIERS IN CHEMICAL ENGINEERING

Rheology is especially important to the routes include refinements of older processes, understanding of corr 'osite liquids in many such a-- .ipitation and thermal decomposi- applications a-, products (e.g., lubricants, sur- tion, as well as newer processes, such as plasma face coating agents, and additives for enhanced processing and chemical vapor deposition. The oil recovery and drag reduction). Such liquids nucleation and particle-growth processes in such usually contain polymers, and their behavior is systems need to be described quantitatively to frequently viscoelastic under use conditions. enable better process development and scale- While data on linear response and relatively up. Chemical engineering frontiers include the mild shear flows are available for nonassociating development of new chemical proccsses for polymer solutions in the relevant ranws of producine ceramic raw materials, such as sub- molecular size and concentration, far fewer data micron, spherical, uniform powders, and high- are available on liquid systems that contain strength fibers and whiskers. particles or micelles, palicularly those in which Chemical engineers could also work to devise therearestronginterparticleinteractions. processes to improve the flow characteristics Knowledge of the fluid mechanics of ordered of powdeafter they are formed. 'uch research liquids is similarly sparse. Information on the would help control agglomeratitof particles response to rapid shear flows and extensional in subsequent processing steps z, well as facil- flows, even in simple polymer solutions, is very itate ,e production of compacted ceramic pie- limited. Thus, we are far from having depend- forms. For example, gas-solid chemical reac- able equations from which models of such fluids tions might be used to tailor the chemical could be developed and farther still from a composition of powders. As another example, generalized molecular understanding of the better methods of compounding powders with structure-property relations of these fluids and binders might be achieved by processes that from extrapolations of the flow patterns and mix powders with suitable binders in a liquid stress distributions in such fluids in geometries and then spray dry the resulting suspension. close to those in which they are used. For Powder processing is also one element in the example, there is significant divergence between engineering of grain boundaries in large, com- theoretical prediction and empirical observation plex parts. Such engineering would allow sin- of the flow of lubricants in journal bearings. tering ceramics to fill: density without degrading oxidation resistance and long-term strength. Even if satisfactory equations of state and const;tutive equations can be developed for complex fluids, large-scale computation will still Processing of Polymers be required to predict flow fields and stress Other important research challenges confront distributions in complex fluids in vessels with chemi-ml engineers in the area of polymer pro- complicated geometries. A major obstacle is cessing. One concerns the inseractiens of poly- that even simple equations of state that have mers with their environment. For example, been proposed for fluids do not always converge contacting a glassy polymer with a solvent or to a solution. It is not known whether this swelling agent may 'mad to unusual diffusion difficulty stems from the oversimplified nature characteristics in the polymer, stress I'm:nation, of the equations, from problems with numerical crazing, or cracking. Such phenomena are poorly mathematics, or from the absence of a laminar understood because glassy polymers may ex- steady-state solution to the equations. hibit complex viscoelastic behavior in the presence of a liquid or during their second-order Processing of Powders (g'ass) transitions. The study of diffusion in glassy polymers is a virgin research area for One route to better ceramic powders, sol-ge chemical engineers. A better understanding of processing, has already been described in tt polymer-solvent interactions could have impor- chapter. There are, however, many other pos- tant payoffs in the development of posi.ive sible routes to improved ceramic powders. These resists for microcircuit manufacture (see ' p-

8'i POLYMERS, CERAMICS, AND ('OMPostin 75 ter 4), because the, dissolution characteristics It is particularly important L. study process of polymeric res:sts are crucial to their appli- phenomena under dynamic (rather than static) cation and ren _Iva! during microlithography conditions Most current analytical techniques The focus research on engineering ther- are designed to determine the initial and final moplastics with enhanced mechan;cal, thermal, states of a material or process. Instruments electrical, and chemical properties has shifted must be designed for the analysis of materials away from synthesizing rovel polymers tcward processing in real time, so that the crucial combining existing polymers. Multicomponent chemical reactions in materials synthesis and polymer blends pose interesting and challenging processing can be monitored as they occur. new problems for chemical engineers. Many Recent advances in nuclear magnetic resonance multicomponent polymeric melts are homoge- and laser probe.; indicate valuable lines of de- neous at processing temperatures but separate velopment for new techniques and comparable during cooling. Judicious choice of stress levels instrumentation for the study c: interfaces, during cooling and of the cooling rate can effect complex liquids, microstructures, and hierar- changes in the structure and morphology of the chical assemblies of materials. Instrumentation end product and hence in its properties. needs for the study of mic' structured materials The fluid prepolymer in which the load-bear- are discussed in Chapter 9. ing fibers of a polymer composite are placed undergoes fur,her polymerization and cross- linking during the thermal curing of the com- Fabrication and Repair of Materials r is posite. The chemical reactions that occur during Advanced materials systems ba' .poly- curing are exothermic and are difficult to con- mers, ceramics, and composites are . _instructed trol. Some regions in the composite material by assembling components to create structures react adiabatically while others lose he by whose properties and performance are deter- conduction to their surroundings. The resu,, ag mined by the form, orientation, and complexity point-to-point variations in polymer matrix mo- the composite structure. The properties of lecular weight and cross-link density result in .nese assemblages are determined not by the cnanges in the composite's properties and qual- sum of weighted averages of the components ity. We need to better understand and control but rather by synergistic effects in intercon these variations in well-characterized processes .lected phases. For this reason, the study c and to deduce how to change the geometry of fi brication of hierarchical assemblages of ma- the finished object or the distribution of fibers rials,as well as the study of mechanisms for within it to compensate for the variations in polymer stcture that might inherently arise repairing defects in assembled structures, must be supported by fundamental research. dnring processing.

Process Design and Cont Designing Systems from the Molecules on Up Because processing conditions and history have such an important influence on the con- Successful systems design and fabrication formation and properties of materials, there is depend on understanaing the connections be- a need to develop models and systems for the tween microscale phenomena and macroscale measurement and control of materials manufac- behavior af materials. For example, with suf- turing processes so that processes can be better ficient insight into intermolecular interactions, designed, more precisely controlled, and auto- appropriate models, and the computational power mated. Opportunities for chemical engineers in of supercomputers, it may be possible to predict process design and control, including advanced changes in macromolecular configurations when mathematical modeling of polymer processing, loads are imposed on polymers or changes in are explored in depth in Chapter 8. the properties of a material as a resul, of

. 85 76 FROVTIERS IN CHEMICAL E VGIATERIAG

branching or cross-linking the material's ma- of large molecules or particles, also holds prom- cromolecular structure. ise for solving se ae fabrication problems. A related problem in comnosites is the nee.: 'o design optimal fiber ork. ations for a composite part given the set of stress vectors and Detection and Repair of Flaws in levels to which the part will be subjected. These Materials Systems design coliderations would be useful in de- A central problem in complex materials sys- signing airframe components such as parts for tems of any kind involves testing to detect the tail, wing, or fuselage. A similar problem is flaws, analysis to predict their effect on re- assessment of the n-'rmance penalties that maining service life of the system, and repair might resull fre/n imperfections in manufacture. strategies to overcome them. For the structural Solution:, to tl problems lie in the realm materials discussed in this chapter, these prob- of computer-aided design and manufact' lems are uncharted territory in need of explo- (CAD/CAM). This area of technology is being ration by chemical engineers. developed rapidly by mechanical engineers, but There is a general need for nondestructive the problems encountered include many that test methods capable of determining whether are logical extensions of polymer process en- the manufacturing process for a polymer, ce- gineering. Interdisciplinary collaborations be- ramic, or composite has achieved the desired tween mechanical and chemical engineers should microstructure. Chemical engineers can profit- be fostered for problems where chemical ex- ably contribute to interdisciplinary efforts to pertise would be valuable. lust as chemical cievelop such test methods. For example, one engineers of r. previous era contributed exten- type of defect in a composite arises because the sively to the knowledge of heat and mass trans- placement of the fibers is different from what fer by collaborating with mechanical engineers, was intended. This may reflect perturbations in so are they now well positioned to contribute the filament-winding operations on a mandrel to composites CAD/CAM and tc the education or fiber movement during curing in response to of students who may one day use and oversee Mere: itial stresses. The processing expertise these processes in industry. of chemical engineers could be useful in developing instrumentation to detect such flaws dur- _hemical Processing in the Fabrication of ing the manufacturing process so that automatic Materials Systems control and coaection of O.:. process can be in, ked to avoid or compensate for flaws. One might imagine that the fabrication of There is also a need for :nef ods to predict materialssystemsinvolving polymers,ce- the effects of flaws and the rediaining service ramics, and composites would Lz principally a life of a flawed or degraded part in ese. For concern of mechanicrily Aented materials en- example, because of limited basic knowledge gineers. This is not true. For example, the about composites, structures based on them are mechanical attachment of composites to other now overdesigned for considerably greater mar- materials (e:g., metal parts) by drilling holes in gins for error than those required for mf,t21 the composite and attaching mechanical fas- structures, thereby losing some of the inherent teners can alter and degrade the performance superiority of composites. of the composite. In a number of situations, Finally, attempts to repair composite stric- joining and fabrication processes involving tures will become increasingly common in future chemical reactions with the material will be years as the use of composites spreads. At this 'seeded in systems fabrication. point, a fundamental repair science for com- Fundamental research to support materials posites is completely lacking. Since se/eh strat assembly and fabrication probably centers on egies are likely to depend heavily on chemical the science and technology of adhesion, al- reactions to heal breaks and flaws chemical though research on mechanical assembly driven engineers should be at the forefront of this by chemical action, such as the self-assembly emerging field.

86 POLYMERS, CERAMICS, AN!) C'OMPOSITES 77

IMPLICATIONS OF RESEARCH tors with interests in materials. There is a FRONTIERS perception in the chemical engineeringcom- Chemical engineers are already equipped to munity that MRLs are more physics directed, pursue the frontiers outlined in this chapter. and probably not open to .ignificant participa- The core indergraduate curriculum provides tion by chemical engineers. One way ofat both a science base and an engineering knowl- dressing this problem would be for NSF to edge base for approaching problems in materials target more funds to the Division of Materials phenomena and processing. At the same time, Research with an emphasis on interdisciplinary undergraduate chemical engineering students and process studies. Another way might be to would benefit from a broader exposure to prob- develop mechanisms intermediate bcween MRLs lems in materials science and engineering. What and E; Cs that would promote engineeringre- is needed is a better integration of such problems search on materials. into the curriculum at all levels. This is best A final goal for improving the chemical en- achieved by deve'oping better instructional ma- gineering contribution to materials research would terial and example problems for existing courses be to develop focused continaing education in thermodynamics, transport, and reaction en- programs to help qualified chemical engineers gineering. What is not needed is a proliferation move aggressively into materials-related areas. of general, ency-lopedic materials courses for Such courses might take a number of forms. undergraduates. The AIChE might take the lead in sponsoring All the scientific and engineering disciplint-s short courses within the context of its existing involved in materials research are in need or continuing education program. Universities might better instrumentation and facilities. Suitable provide co. iplementaiy more intense exposure equipment for chemical engineers interested in to the pi oF,,ems and opportunities in materials .daterials quest ans might include the following: research t v initiating speci: I workshops.masters degree programs, or sabbaticals for indus- solid-state NMR spectrometry; trial researchers. spin-echo NMR spectrometry; Raman spectroscopy; NOTES secondary ion mass spectrometry; X-ray photoelectron spectroscopy: I. ITECH Panel Report on Advanced Materials in laser light scattering; Japan. La Jolla, Calif.: Science Applications In- advanced dynamic rheometers; ternational Corp., 1986. computer-con. oiled, fully equipped poly- 2. National Academy of Sciences-National Academy merization reactor of Enpiaeering-Institute of Medicine, Committee directional irradiatic i devices; and on Science, Engineering, and Public Policy. "Re- port of the Research Briefing °anel on High- dynamic mechanical property measure- Performance Polymer Composite ,," in Research ment equipment. Briefings 1984. Washington, D.C.: National Acad- emy Press, 1984. Special efforts to help academic institutions 3. National Research Council, National Materials acquire these instruments are needed. Future Advisory Board. High-Technology Ceramics in chemical engineers will be required to under- Japan (NMAB-4I8). Washington, D.C.: National stand the design and operation of sophisticated Academy Press, 1984. equipment in the analysis of materials proper- tie- An early exposure to these tec' tiques is highly desirable, and is probably indispensable SUGGESTED READING to quality research at the graduate level. C. G. Gogos, 1. Tadmor, D. M. Kalyon, P. Hold, The Materials Research Laboratories (MRLs), and J. A. Biesenberger. "Polymer Processing: An Overview." Chem. Eng. Prog., 83 (61, June 1987, sponsored by the NSF, have been one mecha- 49 nism for providing instrumentation awl facilities National Research Council, Engineering Research support to smil groups of principal investiga- Board. "Materials Systems Research in the United

8"4 78 P R011ERS IV ( HEIM' U. LW:LW:WV,

States." in Directions in Engineering Research. U S. Congress Office of Technology Assessment. Washington. D.C.: National Academy Press. 1987. New Structural Materials Technologies: Oppor- D. R. Uhlmann. B. J. J. Zelinski. and G. E. Wnek. tunities for the Use of Ad. anted Ceramics and The Ceramist as ChemistOpportunities for New CompositesA Technical Memorandum (OTA- Materials." Mat. Res. Soc. Symp. Prccc.. 32. 1984. TM-E-32). Washington. D.C.: U.S. Government 59. Printing Office. 198E S I X

Processing of Energy and Natural Resources

nergy, minerals, and metals are three basic building blocks of our technological society. Chemical engineering has long been involved in the technologies used to convert natural resources into energy and useful products. The expertise of chemical engineers will be needed more than ever if we are to make progress on problems such as enhanced oil recovery, shale oil production, coal conversion, ele h ochemical energy storage, solar power, and turning waste into a useful source of energy and metals. Important intellectual challenges await them in in-situ processing, processing solids, developing better separations, finding better materials for use in energy and mineral applications, and advancing the knowledge base for process design and scale-up. Implications of these challenges for chemical engineers are discussed a, the end of the chapter.

71" 89 80 I. RON% IER., IN (ill: Viet!. ENGINEER! VG

HE MORAL EQUIVALENT OF WAR." This resoui zes industries have experienced cutbacks summons to respond to the energy in support for industrial and academic research crisis, uttered by President Jimmy Carter that are similar to those in the energy industries. less than 10 years ago, today seems so remote Energy processing and natural resources that it might be the subject of a good trivia proce "sing share numerous fundamental tech- question. The mid-1980s are witness to a vast nical problems, many of which fall squarely in deniobilization of resources and personnel once the domain of the chemical engineer. The cur- committed to energy research. Armies of re- rent retrenchm' in research in both fields does searchers have been redirected, retired, or laid not imply that .he problems have largely been off. Pilot and demonstration facilities have been solved. Indeed, tin; problems are as challengirK, mothballed. Spending on alternative fuels re- as ever, and they are not going to be solved search has dwindled. The cumulative effect of overnight. Now is the time to conduct the this rapid negative change is to discourage fundamental research needed for their solution. academic investigators and young graduates How do we take a fut!ire-oriered approach from seriously considering the formidable en- to research on energy and metals? What criteria ergy research problems that remain unsolved. do we use to set research priorities? Short-term Yet there is nothing trivial about energy. The projections of prices and availability of re- cost of energy over the last decade has had a sources are poor guides to a national policy for significant influence on the rate of inflation in research. It:.i virtually impossible to predict the United States. The energy processing in- the course that energy prices will take over the dustries constitute one of the country's largest next few years or to prognosticate political industrial segments. In 1985, shipments of pe events that might affect the supply of key troleum and coal products amounted to $194 minerals to the United State.. The most anyone billion,' and the value of natural gas produced can say is that oil prices will rise and that a real in the United States exceeded $42 billion.2 The threat exists to the stability of our supply of availability of secure fuel sit7plies is vital to several key minerals. national defense. Cheap and abundant energy What is predictable is that the cost of pro- supplies are just as important to national eco- ducing oil and minerals in the United States will nomic competitiveness in peacetime. No indus- be infenced by the rate of depletion of re- trial segment has a greater impact on national coverable domestic sources. This rate will in well-being, jobs, defense, and economic com- turn be influenced by the economics of recover- petiti reness. No industrial segment is more ing usable materials from these domestic sources dependent on chemical engineers for its health It is certain that there will be a need for engineers and progress. Research, development, and com- to develop and manage technologies that can mercial operations in the energy processing slow the depletion rate by permitting recovery industries all draw heavily on the knowledge of a greater fraction of the resources that are ano echniques of chemical engineers. there. There is nothing trivial about our nation's The United States is the recognized world primary metals industries eitherin 1985 they leader in energy and minerals technology; chem- accounted for $126 billion in shipments.' This ical engineering has been and must continue to industrial sector and the energy industries have be the key discipline in maintaining that posi- a number of characteristics in common. both tion. The primary .hrust of chemical engineering face an uncertr, n future shaped by the declining research in these areas should be to provide the quality of domestic raw materials reserves and basic knowledge that Mould give the nation the substantially higher quality reserves in countries capability to react to future shifts in the prices that need to obtain foreign exchange through and availability of energy and minerals. This commodity sales, regardless of market price or capability can be o, . doped by pursuing fun- profit considerations. The United States de- damental research in the priority at as laid out pends heavily on foreign imports for both energy in this chapter. Our ultimate objectives should and metals resources (Figure 6.1). The natural be to extract and nrocess current resources

90 PROCESSING OF ESERGIA.i D it.RAI. RENO( lat..1 81

and cleaner than coal, they made great itavads on the use of coal as aprimary energy source. However, in recent years, it has become apparent that ..:silt' accessible, high quality reservoirs of gaseous and liquid fossil fuels are not inexhaustible. Moreover, the demand for energy continues to increase as the population increases and labor-sparing machines are developed. It is a basic premise that this 1970 1975 1980 1985 1990 1995 2000 challenge can best be met by developing technologies for more efficient use of existing resources and for utilizing previously un- Percentage tapped sources of energy. A Niobium 100 Brazil, Canada, Thailand Manganese 100 South Africa, France, Brazil, Gabon number of approaches to these Kea (sheet) 100 India, ilekalunt, France goals appear to hold promise for Stronlium 100 Mexico.Spain. Mead% i . amine 97 Australia, Jamaica, Guinea. Surinam success in the next few decades. Cobol 05 Zaire, Zambia, Canada, Nanny Plainest group 02 South Africa, Great Britain. Soviet Union Tantalum 02 Thailand, Brasil, Wawa; Australia Potash 77 Canada, Israel Enhanced Oil Recovery Chromium 73 &math Mica, 2lmbabw41, Yugoslavia, Turkey Tin r2 Thailand, Malaysia, Bolivia, Indonesia Asbestos 71 Canada, South Wes Technologi:s for oil produc- Mahe a9 China, Morocco. Chic, Pens tion can be divided into three talc ee Canada, Pent, Mexico. Auetralla Nickel at Cana* Australia. flotworana, Norway classes: primary recovery, sec- Tungsten IS Canada, China, Bolivia, Perim. Silver Se Canada, Mexico, Peru, Great Britain ondary recovery, and enhanced Mercury sr Spain. Algeria. Japan, Turkey oil recover' Csdmkim 55 Cana* Australia, Peru, Mexico In primary recov- Selenium 54 Canada, Great Britain, Japan ery, the oil ind gas ;low naturally through the reservoir rock to the FIGURE C I The United States is becoming ever more dependent on foreign productiawell,impelled by sources of oil and minerals. The top graph displays trends in U.S. production and consumption of petroleum feedstocks from 1970 to 2000. It ...lows the subterranean pressure. For typ- growing contribution of imported oil to U.S. consumption, a contribution that ical light oils, only 15-20 percent is projected to increase rapidly in the 1990s. The bottom table shows that the of the oil iq the formation is United States depended in 1985 on foreign suppliers for 10 minerals and extracted in primary recovery. metals, some of which arc critical to national security. Courtesy. Chevron Oil Secondary recovery processes Company (top) and the U.S. Bureau of Mines (bottom). extend primary recovery by injecting water or gas to maintain reservoir pressure as the oil is more efficiently and to deve:op ways ')f exploit- removed. These processes are well established ing alternative resources. and generally recover an additional 15-20per- cent of me original oil. Thus, for conventional TECHNOLOGIES FOR EXPLOITING medium and light crude oils, which have rela- ENERGY SOURCES tively low viscosity at reservoir conditions, about one-third of the original oilcan be re- In the early part of this century, technologies covered by primary and secondary methods. were developed for eAploiting apparently lim- Enhanced oil recovery (EOR) processes (also itless reser"oirs of gaseous and liquid fossil celled tertiary recovery processes)are used to fuels. Because, these fuels are easier to handle recover a portion of the remaining two-thirds

94. 82 t ROATII:RA IA (IMAM 41. I: N6INhIRI.NG of the original oil. Adverse reservoir properties and conditions limit both the applicability of EOR processes znd the extent of recovery from those reset voirs to which the processes can be applied effectively. In addition, there are entire oil deposits so viscous"ultraheavy" crudes, bitumens, and tarsthat primary recovery is not possible and secondary recovery processes are generally ineffective. Such deposits in the United States approach the potentia, of conventional oil reserves, and on a worldwide basis they are several-fold greater than total reser/es of the lower viscosity conventional crude oils. Much of the total resource of these c:itremely viscous oils is in Canada and Venezuela. The ability to recover as much as possible of the remaining two-thirds of conventional oils in known formations and to utilize ultraheavy crude deposits will become increasingly important as U.S., and ultimately worldwide, reserves of conventional crude oils are depleted The three classes of E0`.: technologies that have been studied extensiv -iy are thermal recovery, miscible flooding, and chemical flooding. For each of these methods, the following two basic problems must be overcome if we are to recover a significant part of the remaining oil.

Oil deposits are found in porous sedimen- FIGURE 6.2Interfacial tension imprisons residual oil in tary rocks with limited pathways (permeability) rock. preventing its displacement by water. without inter- for flow of oil through the reservoir formation facial tension. oil flows freely. leaving no residual portion to a producing well. The first problem is to in the rock. Courtesy. Ammo Production Company. achieve microscopic recovery efficiencyto displace the oil from the rock matrix and cause phenomc :la such as fingering and allows pockets it to flow through the formation along a specific of oil to be bypassed, dramatically lowering the pathway. Figure 6.2 depicts an oil droplet in sweep efficiency of all curreot EOR methods porous sand. Because of interfacial tension, the (Figure 6.3). A basic challenge for chemical oil cannot be moved by water pumped into the engineers is how to detect and/or model the formation. If the interfacial tension of the oil is movement of oil, water, gas, and injected chem- loweredwhether by iNcreased temperature, icals in the heterogeneous environment of the by an oil-miscible sweeping fluid, or by chemical reservoir. This problem is discussed in detail in additives--the oil can move fort its original Chapter 8,whicl.is devoted to computer- location through the porous sand. assisted process and control engineering. The second problem is to achieve macroscopic sweep efficiencyrecovery of oil frolo Thermal recovery methods involve the use a significant fraction of the reservoir formation. of steam and in-situ combustion. Thermal EOR This problem is generally more intractable than processes add heat to the reservoir to reduce the first The reservoir formation is not homo- the viscosity of the oil or to v ioorize it. In geneous in porosity or perr ieability, which causes addition, these processes use steam w oil com-

oG 4n PROCESSi VG OF ENERG) i.11) N Ail '1( 4 I. RESOL R( 83

which is driven fOTArd by a combination of steam, hot water, and gas. The earat itself is the reaction vessel and chemical plant. The complicated reaction chemistry and thermodynamics involve mixers, reactors, heat exchangers, separeors, and fluid flow pathways that are a scrambled design by nature. Only the sketchiest of flowsheets can be drawn. The chem.;:at reactor has complex and ill-defined geometry and must be operated in intrinsically transient modes by remotecon- FIGURE 6.3Oil can be recovered from reservoirs by trol. Overcoming these difficulties is a true pumping in steam, gas, or specialty chemicals. All methods frontier for cinical engineering research. face common problems posed by inhomogeneities in the Another EOR approach to reducing the vis- rock containing the oil. Some major problems include poor vertical coverage. inefficient sweeping that bypasses pock- cosity of oil in the restrvoir is miscible flood- ets of oil, and severe channelling of fluids along fissures or ingthe injection of fluids that mix with the oii highly permeable layers of rock. Chemical methods of under reservoir conditions. Such fluids include compensating for these inhomogeneities would boost the carbon dioxide, light .gdrocarbons, and nitro- yields and cut the operating costs of enhanced oil recovery. gen. Supply and cost of carbon dioxide are often Courtesy, Amoco Production Company. more favorable than for injectants. Ex- tensive research and field testing have established the technical viability of miscible flood- bastion products as a drive fluid to move oil to ing, and a number of commercial carbon dioxide producing wells. Thermal processes are most miscible flooding projects are in operation. often used in reservoirs of viscous oils and tars Chemical EOR methods are based on the on which tests have established that primary injection of chemicals to develop fluid or inter- production will be small and wa:erttooding will facial properties that favor oil production. The be largely ineffectual Thus, thermal methods three most common of these methods are poly- are usually used in place o", rather than after, mer flooding, alkaline flooding, and surfactant secondary or primary methods. flooding. The first EOR method to achieve widespread Commercialimplementationof polymer commercial acceptance, steam injection (Figure flooding and alkaline flooding is in progress, and 6.4), has been used commercially in CaIfernia there is confidence that research can make these for more than 20 years and now accounts for processes more cost-effective and extend their more than 80 percent of U.S. EOR production. applicability to a greater fraction of the known A less well dev dope," thermal method, in- reservoirs. The research focus is on improving situ comustion, holds much promise but also the thermal, chemical, and biological stability poses a tremendous challenge to the theory and of polymers and making them more cost-effec- practice of chemical engineering. The in-situ tive. Research targets in alkaline flooding in- combustion process involves many processes clude better definition of th reactions of alkaline occurring simultaneously. Heat is generated materials with the rock formation and theuse within the reservoir by injecting air or oxygen of ancillary chemicals to improve performance. to burn part of the reservoir oil (Figure 6.5). It As with polymer flooding, a primary objective is common practice to co-inject water with or is to extend the applicability to more severe above the oxidant to scavenge energy from hot conditions. rock lying behind the burn front. In-situ com- Of the chemical EOR technologies, surfactant bustion often achieves considerably higher tem- flooding is the most comple'., the farthest from peratt les than steam flooding, and the heat not commercial feasibility, and the most challenging only physically and chemically reduces the oil in terms of research needs, yet it has the greatest iscosity but also partially vaporizes the oil, ultimate potential. It involves injecting surfac-

fi FRO:11 II.R.S EV CHEMICAL ENGINEEREVG

PRODUCTION FLUIDS COIL. GAS, WATER) SEPARATION AND PRODUCTION WELL STACK GAS STORAGE FACILITIES SCRUBBER

OOIL AND WATER ZONE NEAR ORIGINAL RESERVOIR TEIAPERATURE 0 HOT WATER ZONE 0 HEATED OIL ZONE 0 STEAM AND CONDENSEDWATER ZONE

FIGURE 6.4 Steam flooding is one of two principal thermal methods for oil recovery and has been commercially applied since the early 1960s. A mixture of steam and hot water is continuously injected into the oil-bearing formation to displace mobilized oil to adjacent production wells. Repnnted with pei - mission fromEnhanced Oil Recovery.Copynglit 1984 by the National Petroleum Council. tants, such as sulfonated crude oil, to mobilize total ultimate EOR potential for the United the oil for subsequent recovery by a waterflood. States is 14.5 billion barrels (Figure 6.6), which The most recent National Petroleum Council is more than 50 percent of the U.S. total (NPC) study of EOR3 estimates that chemical estimated future recovery by primary and sec- EOR technologies constitute more than 60 per- ondary processes. The NPC study also esti- cent of the additional EOR potential for ad- mates that successful development of projected vanced technology and that surfactant flooding advanced EOR technologies could make pos- represents more than 90 percent of this poten- sible the recoveryf 27.5 billion barrels of tial. Research objectives are to extend surfac- domestic oil (Figure 6.6). This is more than 10 tant flooding to more severe conditions and to years of U.S. production at current rates and make it more cost-effective. could provide an important augmentation of The stakes in continued research and devel- domestic supplies well into the next century. opment of EOR technologies are enormous. They invoive decreasing U.S. dependence on Shale Oil Production imported oil and extending the useful lifetime of the wor!d's exhaustible supply of petroleum. Oil shales are a large, virtually untapped The NPC study of EOR estimates that with source of hydro( arbons. U.S. reserves repre- currently implemented EOR technologies the sent several hun ed billion barrels of oil and PRO( I, SSINI, Of 11l R6/ N 111 R 1i RI SIE 14 I S S5

COMPRESSOR PRODUCTION WELL

INJECTION WELL WATER PUMP COMSUSTION ,"Ilmer 444f: t'Ut

1 0 coon meow 0 114/RNMO FRONT AND COMOUSTION MOW 1200F1 0 AM AND VAPORIZED WATER ZONE ®INJECTED AIR AND WATER ZONE (BURNED OUT)

FIGURE6.5 1n-situ combustion is a major thermal means of oilrecovery. Heat is generated in the reservoir by injecting air and burning part of the oil. This partially vaporizes the remaining oil. which is then driven forward bya combination of steam. hot water. and gas. Any oil left behind becomes fuel for the in-situprocess. Water is also injected into the well: it improves the efficiency of theprocess by transferring heat from the rock behind the combustion zone (71 to the rock immediately ahead of the combustion zone (4). Repnnted with permission from Enhanced OilRecovery. Copyright 1984 by the National Petroleum Council. are located in both the western and eastern shale, this retorting method is the onlyrecovery states. Eastern oil shales are intimate mixtures process that has been developed. of inorganic silts and insoluble organic material Retorting can be carried out above groundor that have been consolidated Into rock. In west- in situ. The former process involves mining the ern oil shales, the matrix is a carbonate-based shale and heating it in"essel. Process devel- marlstone. The organic content of both shales opment has been concentrated on three imnor- is typically 5-30 percent, and most of it isa tant engineering problems. The first is how to polymeric, insoluble petroleum precursor called handle the large quantities of solids thatmust kerogen. Shale oil can be recovered by heating be processed. The second involves howto the rock through the range from 250 to 500°C, transfer heat to those solids. And the third where the kerogen is thermally decomposed to concerns the effect of shale particle size on the liquid and gaseous products, leaving 20-35per- efficienzy of oil recovery. In-situ retorting isan cent of the organic matter as coke (Figure 6.7). alte native to mining the shale, but only if the Because of the insolubility of kerogen and the shale bed can be made sufficientlyporous to difficulty of physically separating it from the allow injection of air to burn part of the kerogen

S 86 ER0. I IER.S IA CHEMICAL E,VGINEERING and the resulting coke and to permit outflow of the retorting products. In shallow beds, blasting can lift the overburden and fracture the shale to permit these necessary flows. In deeper deposits, partial mining followed by blasting shale into the resulting space is used to create a porous rubble bed underground. Engineering researchhasfo- cused on ways of producing porosity, on gas flow and combus- ADVANCED TECHNOLOGY IMPLEMENTED TECHNOLOGY tionin porous beds, and on 27 5 BILLION BARRELS 145 BILLION BARRELS recovery of products from tht. FIGURE 6.6Prospects for enhanced oil recovery using Implemented and large quantities of off-gas. Much advanced technologies are shown above. The ultimate amount of U.S. oil that of this research and development can be recovered by "implemented technology." technology that presently must be done in the field on a exists in at least the proven field test stage, is estimated to be 14.5 billion large and costly scale. barrels. Using "advanced technology." technology that might be conceivably It is possible that greater po- developed before 2013. adds another 13 billion barrels of oil to the estimate. for a total of 27.5 billion barrels. A comparison of theAribution of ultimate rosity in shale beds could be recoveries by method is also shown. Most of the increase in the estimate achieved by chemical commi- from applying advanced technology comes from improvements in chemical nution of the shale. For example, flooding methods. The protections assume that crude oil has a nominal price the treatment of western oil shales of $30 per barrel and that the minimum rate of return on capital is 10 percent. with acid solutions might result Repnntcd with permission fromEnhanced 011 Recovery.Copyright 1984 by the National Petroleum Council. in comminution by inducing corrosive stress fracture of the carbonate rock. Chemical engineering research in Coal is gasified by heating it in the presence this area, as well in the elucidation of oil-rock of steam to make synthesis gas (syngas), a interactions, might provide insights for new mixture of carbon monoxide and hydrogen. A strategies for oil shale production. variety of processes for coal gasification have evolved, and several U.S. pilot facilities have Conversion of Coal to Gaseous and been built on scales of 100 to 1,000 tons per I 1 -nid Fuels day during the last 10 years. The Great Plains Coal is the giant of fossil fuel resources. coal gasification plant in North Dakota is op- World reserves are many times those of petro- erating at 10,000 tons per day. Coal syngas has leum, and the United States is one of the major lower energy eontent than natural gas for fuel resource holders. Coal can be used directly in use, but is widely used for the synthesis of combustion or converted to gas or liquid. Only liquid fuels and other chemicals. For example, combustion consumes significant amounts of Tennessee Eastman is operating a commercial coal today. plant that converts 900 tons of coal per day into Coal is currently economically useful only in syngas that is in turn converted into acetic an- plants that are equipped for large-scale handling hydride and other chemicals by a series of cata- of solids, and it is used only indirectly as a raw lytic reactions. The Tennessee Eastman process material for chemical synthesis. Accordingly, is an excellent example of innovative chemical there has been considerable research on pro- engineering in the design and construction of cesses for converting coal into gaseous or liquid an efficient plant to synthesize organic chemi- fuels and chemicals. Only gasification has ad- cals from nonpetroleum raw m Iterials. (See vanced to commercial status. "Acetic Anhydride from Coal" on pp. 88-89.) PROCE.SSING OF E.NER61 .1:11) .1',11 t'R11, REcOtRCLS 87

al. GAS TO CLEANUP

CRUSHED OIL & FINES SOLID RETORT OIL SHALE--. LIQUID SYNCRUDE SEPARATION

FINES COMBUSTION

SPENT SHALE

UNDERGRADE SOUDS MATERIAL DISPOSAL

FIGURE 6.7Steps in oil shale retorting are shown. Oil shale is crashed and then heated in a retort to drive off the oil that is trapped in the rock. Any oil left behind, as well as particulates returned to the processas the recovered a is processed, is burned to provide heat for the retorting. The oil thatis recovered from the shale is chemically treated to produce synthetic crude for further processing in conventional refineries. Courtesy, Amoco Oil Company.

Coal syngas can be converted into liquid the problems associated with materialscorro- hydrocarbon fuels by catalytic reactions. One sion and mechanical solids handling; the main process for this conversion, the Fischer-Tropsch environmental problem is groundwater contam- process, was developed in Germany during ination. A further benefit of in-situ coal gasifi- Woild War II and is being operated on a large cation is the potential for exploiting coalre- scale in South Africa. Today's pilot facilities serves that cannot be mined economically. and pioneering uses of syngas are establishing However, itis no less complex than in-situ a technical and economic basis for the genera- recovery of heavy tars or shale oil, and the tion of commercial coal gasification projects engineering challenges are comparable. that is expected to emerge in the 1990s. The conversion of coal into liquid materials Coal gasification research and development can be accomplished by pyrolysis or by direct have concentrated on handling of solids, prob- liquefactionheating coal in the presence ofa lems with ash, and dealing with the sulfur and hydrogen source Neither of these routes is yet nitrogen compounds present in coal. The newest economically feasible. pilot plants are investigating catalytic gasifica- In pyrolysis, coal is split into a hydrogen-rich tion. The integration of coal gasifiers with elec- liquid and a hydrogen-depleted solid char. The tric power generators in a combined-cycle mode liquid contains significant amounts of nitrogen (Figure 6.8) is an emerging field for design and sulfur compounds, as well as high-molec- studies and economic evaluation. Much of the ular-weight aromatic compounds such as as- combined-cycle equipment is being perfected in phaltenes. It is difficult to upgrade this liquid natural-gas-burning plants that will come on to a fuel suitable for transportation uses. Fur- stream in the next few years. thermore, while the liquid might be used in Iii-situ coal gasification has Men demon- boilers, it would pose severe problems of nitro- strated in small -scale field tests. Compared with gen and sulfur oxide emissions from power aboveground gasification, in-situ gasification has plants. Research to date has uncovered few the potential advantage of mitigating many of uses for the char, and it must be disposed of. 88 FRONTIERS IN CIIEM::'11, ENGINELRIAG

There are challenges and opportunities in developing a process for in-situ pyrolysis of coal in Acetic Anhydride from Coal which the char is the principal fuel. The most successful example of generating chemicals directly In direct liquefaction, coal is from coal is the Tennessee Eastman integrated process for heated in the presence of hydlo producing acotic anhydnde. The commercial plant gasifies ap- gen and a catalyst such as cobalt - proximately 900 tons of coal per day and performs four chemical molybdenum or nickel-molyb- steps to yield annually 500 million pounds of acetic anhydride, denum on alumina to give a 390 million pounds of methyl acetate, and 365 million pounds of greater yield of high-quality hy- methanol. In addition, 150 metier pounds per year of acetic acid drocarbons than that produced may be produced from acetic anhydride. by pyrolysis. This hydrogenation The process begins with a gasification process that converts process has been demonstrated coal into carbon monoxide and hydrogen. Part of this gas is sent in several 50- to 250-ton-per-day to a water-gas shift reactor to increase its hydrogen content. The plants. purified syngas is then cryogenically separated into a carbon Chemical engineering research monoxide feed for the acetic anhydride plant and a hydrogen-rich on direct liquefaction has fo- stream for the synthesis of methanol. cused on improving hydrogena- Eastman uses acetic anhydride primarily for esterification of tion efficiency, for example, by cellulose, producing acetic acid as a by-product. This acetic acid treating a coal slurry in a hydro- is ussd to convect the methanol into methyl acetate in a reactor- gen-donor solvent in a high-pres- distillatirn column in which acetic acid and methanol flow coun- sure reactor. Basic knowledge of tercurrently. coal structure and reactivity as The final step in the process involves reacting purified carbon well as scientific understanding monoxide from the gas separation plant with methyl acetate to of hydrogen-transfer reactions has form acetic anhydride, using a proprietary catalyst system and been crucial in improving the process. Part of the acetic anhydride is reacted with methanol to process. Equally important has produce acetic acid and methyl acetate, and the latter is recirculated been he reali: ation that the de- to the csutonylation step. siredreactionproductscan Eastman chemical engineers provled innovative solutions to undergo secondary reactions that Key steps in the methyl acetate process. Computer simulations diminish yields and quality of were used extensively to test ways to minimize the size of the final pr-ducts. Accordingly, two- reactors and recycle streams, to maximize yields and conver- stage liquefaction with a catalyst sions, and to refine the Whyl acetate in a minimum number of in one or both stages is being steps. tested on a small scale. Although The novel approach finally taken was to conduct the reaction catalyst performance is improv- and purlication steps in a reactor-distillation column in which ing, there is still a need for cata- methyl acetate could be made with no additional purification steps lysts that will perform even bet- and with no unconverted reactant streams. Since the reaction is terinsuchseverelyfouling reversible and equilibrium-limited, high conversion of one reactant conditions. can be achieved only with a large excess of the other. However, if the reacting mixture is allowed to flash, the conversion is increased by removal of the methyl acetate from the liquid phase. New Raw Materials for With the reactants flowing countercurrently in a sequence of Petroleum Refineries As the domestic mix of fossil fuel resources changes over the coming years, new challenges will emerge for and gas industries, there is a need for funda- the design and ienovation of our nation's in- mental research to pro. .ae new design concepts stalled base of refineries. While the practical and for trained engineering personnel to main- aspects of this task must be left to the petroleum tain intcinational competitiveness in these in- PROCESSING OF EN'ERGY .1,`U ,V4 7-IRV. RE S01 R( Es 89

refineries have been desig ed, both heavy crudes and shale oil flashing reactor stages, the high concentrations of reactants at contain hydrocarbons of higher opposite ends of the column ensure high overal conversion. molecular weight and higher car- The methyl acetate plant is highly heat integrated. Reboiler bon-to-hydrogen ratio; more un- condensates are flashed to generate abnosphert steam for wanted sulfur, nitrogen, anu cat- preheating the acetic acid feed, and the hot by-product wastewater alyst-poisoning metals like Is used to preheat the methanol feed. A vapor sidedraw provides vanadium and nickel; and a bi- most of the heat to run the impurity stripping column. Most tumen-like residue that is diffi- importantly, the reactor column is inherently heat integrated. It can cult to refine. Such resources be thought of as four distillation columns stacked vertically, with must be upgraded by chemically the top and the bottom of each column acting as the reboiler and adding hydrogen or chemically condenser for the columns above and below, respectively. removing carbon, by redistrib- Because of the high level of heat !ntegration and the combination uting hydrogen among hydrocar- of several unit operations in the methyl acetate reactor column, bon fractions, and by removing the design of a control system was a major challenge. The system compounds of heieroatoms and had to take into account the broad range of time constants of the metals. Research has already led phenomena in the reactor column and the Interaction of the cohmn to improved thermal upgrading with impurity removal columns and heat integration devices. A methods such as fluid-bed coking control strategy was developed to solve these problems and to with coke gasification to remove allow the operation of the reactor column at an exact stoichiometric carbon (Figure 6.9) and to cata- balance of feed materials. Perhaps the most important testimony lytic hydrotreating schemes to to the engineering development work is the fact that the commercial add hydrogen and remove com- plant was a 500:1 scale-up of the largest pilot unit built. pounds of nitrogen, sulfur, vana- The plant for the catonylation of methyl acetate to acetic dium, and nickel (Figure 6.10). anhydride offered similar engineering challenges. Two pilot plants The co-processing of coal with were operated to evaluate two competing schemes for reactor heavy crude oil or its heavier configuration. Early in the pilot plant stage, computer models for fractions is being developed to calculating detailed heat and material balances, operating costs, lower capital requirements for and arpital requirements were programmed and continually up- coal liquefaction and to integrate dated from emerging pilot plant data Equipment design was processing of the products of especially challenging because the reactor system scale-up factor coal conversion into existing pe- was 10,000 times the pilot plant rate. The overall plant has several troleum refineries. This devel- environmental advantages. The gasifier operates above 1,370°C, opment appears to represent the so no tar or over hydrocarbon waste is generated. The molten main route by which coal-based slag is quenched, =shed, and removed from the gasifier as inert liquid fuels will supplement and granular pellets that qualify for nonhazardous landfill. Aqueous perhaps someday displace petro- streams containing hydrocarbons from the chemical plants are leum- based fuels. recycled to the gasifier for slurry and quench water makeup. Tail- At the low-molecular-weight gas streams from the chemical plants are burned for heat recovery end of the spectrum, a process in existing boilers and furnaces. This plant is an excellent example newly commercialized by Mobil of innovative chemical engineering in designing and buildinga for converting methanol into gas- chemically and thermally efficient plant to synthesize organic oline has significantly expanded ca remicals from nonpetroleum raw materials. opportunities in C-1 chemistry the upgrading of one-carbon molecules to multicarbon products. The process involves the use of dustries. The challenges arise from the prop- ZSM-5, a shape-selective zeolite catalyst. (See erties of the new raw materials that are already "Zeolite and Shape-Selective Catalysts"in finding their way into the process mix. In Chapter 9.) comparisonwiththecrudesforwhich S;nce methanol can be mi- le from coal, nat- 90 FRO,V111.16EV CHEAMALE,VGIAEERIM:

FIGURE 6.8 A process known as integrated gasification combined cycle (IGCC) is shown. It begins with the heating of a slurry of coal and water in an oxygen atmosphere. This produces a fuel gas composed mainly of carbon monoxide and hydrogen. After the gas has been cooled, cleansed of solid particles. and nd of sulfur it can be burned to drive gas turbines and then produce steam for a steam turbine. An IGCC plant emits fewer pollutants into the air than conventional coal-fired plants dn. Reprinted with permission from Scientific American, 257 (3). September 1987 p. 106. Copyright 1987 by Scientific American, Inc.

ural gas, or even biomass, the methanol-to- plant converts natural gas into high-octane gas- gasoline (MTG) process establishes a new link oline at the rate of 14,500 barrels a day.'' Further between these resources and liquid fuels. The developments in the process promise to extend MTG process is now operating commercially in the product slate to include other fuels as well New Zealand (Figure 6.11), where a synfuels as lubricants and chemicals. Through the coal-

1 C 0 PROCESSING OF ENERGY AND ,VA TURAL RESOURCES 91

Tertiary Cyclones Coke Gas To Cleanup

Stea.n To Fractionator Generation

.48Snubber

Slurry Dry Fines Venturi 1 Scrubber

Reactor Gasifier

\......

Steam Air Blower

FIGURE 6.9Flexicoking is a commercial process for refining petroleum that has been applied to heavy oil and tar sand fractions. Theprocess employs circulating fluidized beds and operates at moderate .emperatures andpressures. The reactor produces liquid fuels and excess coke. The latter is allowed to react with a gas-air mixture in tl.° gasifier fluidized bed to provide a low-value heating gas filet can be desulfurized and used as a plant fuel. Courtesy. Exxon Research and Engineenng Company.

to-syngas route, this C-I technology makes it these light hydrocarbons are likely to become possible for the United States to convert its increasingly abundant as their presence ingas- largest natural resource, coal, into many of the oline is reduced by more stringent regulations liquid products that are now derived frompe- to prevent hydrocarbon emissions to the at- troleum. mosphere. There are a number of related challenges in research on new refining feedstocks. One is to find selective, economically viable catalytic Municipal Solid Waste as an Energy Source methods to convert methane directly to liquids, thereby avoiding intermediate products such as The use of municipal solid waste as a source methanol. A related challenge is to upgrade of energy fills a special need by partially elim- hydrocarbons in the range C-2 to C-4. These inating urban waste in an environmentallyac- hydrocarbons are becoming more availableas ceptable manner while at the same time pro- by-products from intensive refiningprocesses ducing usable energy. used to make lead-free gasoline. Supplies of Combustion of municipal solid waste, or of

. 92 ROALIERS ( HEM V, L.Vd VITRA

"." ;-"Ta

MM.

FIGURE 6.10Hydrotreating plant. Courtesy. Amoco Oil Company. fuels derived from the waste, is a fledgling designing" the plant, with the result that com- technology, especially in this country. There bustion of municipal solid waste is not econom- are currently 63 large plants for this type of ically competitive in areas where low-cost combustion, and another 100 are starting up, electricity or landfills for waste disposal are under construction, or planned. Two different available. The future cost of electricity is diffi- technologies are used. In one, which is em- cult to predict. However, the steady decrease ployed by about 40 percent of these plants, the in the availability of landfills portends increasing raw refuse is fed direct:), to the boiler; in the use of this process to dispose of municipal other, the refuse is first processed to reduce wastes, particularly in large cities. piece size and to remove noncombustibles. About one-third,f the plants produce low- Nuclear Energy pressure steam tor heating, another one-third produce high-pressure steam for electric power The phrase "nuclear power" covers a number generation, and the balance also exhaust low- of technologies for producing electric power pres3ure steam for industrial or municipal heat - other than by burning a fossil fuel. Nuclear ing.5 fission in pressurized water-moderated reac- Process problems include slag formation, ash torslight water reactors represents the cur- removal, and process cortrol because of the rent technology for nuclear power. Down the heterogeneous solid waste feed. These problems line are fast breeder reactors. On the distant aave been managed to some degree by "over- horizon is nuclear fusion.

1(;; PROCESSING OF ENERGY 4NO NA TIR4/.. RESOt WES 93

FIGURE C. II The world's first methane-to-gasoline pleat, located in Motunui. New Zealand. In operation for about a year. this plant has already met its strategic objective of reducing New Zealand's dependence on foreign oil. Courtesy. Mobil Research a. 4 Development Company.

Nuclear Fission Fast breeder reactors continue to be developed, although the level of support has fallen Nuclear fission accounted for 13 percent of since the 1983 cancellation of the Clinch River the electricity generated in the United States in Breeder Reactor project. Since that time, in- 1985. Plants under construction in 1985 will novative fast reactor concepts like the Integral probably raise the proportion to 2J percent by Fast Reactor (IFR) have made considerable 1993. However, overexpar sion of electrical gen- headway. The IFR is a self-contained, sodium- erating capacity in this country, actual and cooled, metal-fueled reactor system. Its key imagined hazards of nuclear power plants, and features include a closely coupled fuel cycle for negative perceptions of nuclear power by the recovery, purification, and recycling of the ura- public have combined to halt commitments to nium-plutonium core fuel alloy and extraction build new plants. New construction is not ex- of a plutonium concentrate from the uranium pected to resume before the 1990s. blanket, where new plutonium is generated, for Development efforts in the nuclear industry reenrichment of the core fuel. Any fast breeder are focusing on the fuel cycle (Figure 6.12). The fuel cycle must include fuel reprocessing be- front end of the cycle includes mining, milling, cause of the inescapably high concentration of and conversion of ore to uranium iiexafluoride; fissionable materials in the used fuel. A novel enrichment of the uranium-235 isotope; con- aspect of the IFR reprocessing concept is the version of the enriched product to uranium use of electrorefining rather than solvent ex- oxides; and fabrication into reactor fuel ele- traction for recovery of fuel materials. Electro- ments. Because there is at present a moratorium refining technology will be carried out ina on reprocessing spent fuel, the back end of the molten metal-halide salt electrolyte at about cycle consists only of management and disposal 500°C. To cart y out electrorefining on a practical of spent fuel. scale will require research and development in 94 FRONTIERS IN COE UK' it ENGISEERING

1--11-1 1131 I FUEL FABRICATION SPENT FUEL STORAGE 4

1-1-1 PROCESSING SPENT FUEL SHIPMENT Abb. MILLING URANIUM DEEP GEOLOGIC ISOLATION MINING

FIGURE 6.12Chemical process steps lie at the heart of the nuclear fuel cycle used in the United States. Uranium from mining and milling operations is converted io UF, a volatile gas, which can then be separated so as o isolate the radioactive isotope of uranium. Further Laemical steps are used to convert UF into UO2 for fuel assemblies. to isolate the long-lived isotopes geneated in nuclear pow: plants. and to encapsulate these isotopes for eventual storage in nuclear waste respositories Courtesy. Argonne National Laboratory.

such fields as pyrometallurgy, thermodynamics nation of light elements into heav: r ones. The of metal-salt systems, ceramics, distillation of reaction currently under study is fusion of metals, waste processing, and remote process deuterium and tritium to produce helium and control technology. neutrons. Deuterium, an isotope of hydrogen present in low concentration in water, can be obtained by separating it from seawater. Tri- Nucleus Fusion tium, another hydrogen isotope, is generated Although generation of power from nuclear by reacting lithium with neutrons. In a fusion fusion has not yet been demonstrated, itis reactor, it would be generated in the lithium potentially a huge source of energy. Fusion blanket of the reactor by neutrons derived from power, the source of the sun's energy, results the reactor. from the release of en- rgy through the combi- Major research programs in the L nited Ste ss PROCESSING OF ENEMA' AA b A.411. RAI. RES01 95

and the Soviet Union have concentrated on forming radiant solar energy into usable elec- proof of principle and on containment of the trical energy or chemical fuels. Solar energy extremely high-temperature nuclear reaction. has inherent advantages over other energy Although the research is expensive, U.S. federal sources. It is plentiful, ubiquitous, free, and funding had been relatively ensured until recent continuously replenished, and it can be con- budgetary problems. Chemical engineering re- verted into electricity or fuels by processes that search is needed (xi the preparation of solid are environmentally benign. Yet, bc:ause it is breed -r materials, blanket tritium recovery, a dilute and highly variable source of energy, blanket coolant technology, high-temperature conversion costs are high, and the construction heat transfer, low-activation materials, and tri- of large (1,000-MW) commercial plants for elec- tium containment. tric power or fuels production is not likely in this century. Solar energy falls on the earth at Electrochemical Energy Conversion and about I kW/m2 at noon on a sunny day. A solar Storage cell area of about 40 km2 (15.5 mil) would be The last 25 years have witnessed a rapid rise required to construct a 1,000 -MW power station in the numbers and capabilities of batteries, fuel with solar cells that are 12 percent efficient, cells, and electrolysis cells. Developments in assuming an average of 5 hours of sunlight per batteries and fuel cells open the way to new day throughout the year. The cost of siting and and improved schemes for energy storage and constructing a plant of this size, based on current power generation (see "Fuel Cells for Trans- technology, would be great. On the other hand, portation" in Chapter 9). Devices invented for smaller photovoltaic arrays are being used as a military and aerospace applications have moved peaking resource by some electric utility grids. quickly into civilian usefor example, in con- For example, Southern California Edison main- sumer electronic products, stationary energy tains an array of about 0.1 km2 to produce about storage, and electric vehicles. Significant ad- 7.5 MW of power. It operates unattended and vances in electric vehicle technology could has low maintenance expenses. significantly reduce demand for transportation Solar power research should be continued to fuels derived from petroleum. make smaller scale applications of solar energy The areas of important electrochemical en- more cost-effective. Specific research areas for gineering research can be grouped by the desired chemical engineers in photovoltaics include de- results. Initial costs can be lowered by devising velopment of low-cost methods for producing better or less costly corrosion-resistant mate- the cell materials and for fabrication of the other rials for inactive components of cells, superior components that are proportional to cell area. catalysts for electrodes in fuel cells and elec- Solar-induced temperature gradients afford op- trolysis cells, and innovations that reduce the portunity for ocean thermal energy conversion. complexity of electrochemical transformation Research is needed on convective and phase- systems. The useful lifetimes of systems can be change heat transfer as well as on biofouling lengthened by reducing corrosion, improving control to minimize resistance to heat transfer electrodes, and understanding how porous elec- on t`e seawater side. Development of low-cost trodes perform and change in structure during materials and methods of construction for heat cyclic Jperatinn. Finally, battery and cell per- exchangers will be key to the success of this formance can be improved by developing new technology. materials that permit higher electrode reaction Research on the molecular basis of photoex- rates and lower internal electrochemical losses citation and electron transfer, including inter- and by new designs based on modern electro- actions of electron donor and acceptor mole- chemical engineering. cules,couldleadto new photochemicals. Development of model photosensitive com- Solar Power pounds and methods of incorporating them into The ch-Menge of solar energy research is to membranes containing donor, acceptor, or in- discover ordevelop efficient processes for trans- termediate excitation transfer molecules, and

1115 96 FRONTIERS IA CHEMICAL ENGINEERING eventual development of photochemical reactor systems, could invol re the use of sunlight to re- Geothermal Energy place less-specific energy sources Geothermal energy is derived from the heat of the earth in reservoirs of rack-trapped fluids, often at depths of hundreds of Geothermal Energy meters. Its most common use is to power electrical generating In some situations, it makes plants. Over the past 20 years, geothermsi enemy technology has good economic sense to tap the programed b the point where it occupies a position of growing earth's heat as an energy source. importance in the energy spectrum of many nations. Today, on a One approach to utilizing this worldwide beefs,deothennelenergy pavers dose to 4,000 MW resource is illustrated in the side of installed &schlo& generating capacity. bar on this page. Unocal Corporation Is the World leader In the development of geothemel energy. Most of the technology Involved in using geothermal sow has been adopted from the petroleum industry, Plant Biomass as a Fuel and the use of several sdentific disdplines has made it possible Source to meet the challenges of Mang rem °Naomi sources and of During the 1970s, considerable moving quiddy into COMMOMili operation. attention was given to fuels from M example of the rmaidiscipNnary approach used by Unocal renewable resources. Use of to exploit geolhennal energy is Medd/eminent of the Salton Sea wood-burning stoves increased geothermal field in sontheastern California. This resource is a markedly in some parts of the brine at a tea.perature of over 260°C that contains more then 20 country. A subsidized industry percent dissolved saltsnearly 10 times staler than seawater. to prepare ethanol as a gasoline Woes dated into the brine-bearing rock 110$ spontaneously, and supplement was started. But de- the brine pressure is reduced, allowing part of the brine to flash spite this, it is unlikely that plant to steam. The Maim is routed through a turbine (venerator, and biomass will ever satisfy more the spent brine and condensed steam are injected beck into the than a small percentage of U.S. geothermal reservoir. energy demands. The reasons are Unique problem; vats encountered In drilling wens in the Salton primanly technol igical and lo- Sea field. The hot salt solution in the reservoir adversely affected gistical. driiNng fluids and eqtament, requiring the We of specially for- Much of tLe plant biomass pro- mulated drilling muds and cements developed by chemists and of duced on U.S. land and water special steels recommended by comesion and materials engineers. areas is used for food or forest Facilities for separating steam from brine and for power gen- products or exists in designated- eration were designed by mechanical, thorniest and electricel use areas such as parks and na- engineer& Dissolved salts in the brine cause severe scalkkg and tiona'l forests. If 10 percent of corrosion in wells and pipelines. Chemists and chemical engineers the total biomass were available developed new production techniques to overcome these prob- for conversion to energy, it would lems, as well as pollution control technology for the operation. represent about 5.7 x 10'8 J/yr (5.4 x 1015 BTU/yr). With the generous assumption of 50 percent conversion require greater than average use of fertilizer, efficiency, this biomass would produce about irrigation, and mechanical work, all of which 2.8 x 1018 J, gr (2.7 x 10's BTU/yr) of energy as consume fuel energy themselves. The cost of fuel, or only about 3 percent of current U.S. net energy contributed by biomass will always energy demand. Furthermore, the logistics of be substantially higher than that calculable for collecting today's available biomass would be the gross (i.e., apparent) fuel product. Thus, forbidding. If a large area fo producing biomass economic considerations would appear to rule fel energy were to be set up, it would almost out production of large amounts of energy from certainly have to be on marginal land and would this source for the foreseeable future, and en- PROCESSIAG OF ENERGY A \ D AA Tt RAI. RESOt "WEN 97 gineering research for such schemes does not High concentrations of valuable elements ex- merit high priority. ist in the earth's crust but cannot ne economi- The development of bioreactor systems for cally recovered because they are buried too the production of large-solume chemicals (see deep. The challenge to the chemical engineer is Chapter 3) could be the basis for reconsidering to dIvelop methods for extracting these valuable the production of biomass in limited quantities materials in place without having to move and for fuel uses. This would require efficient mi- process enormous amounts of rock. The general crobial organisms to catalyze fermentation, concepts for in-situ recovery by solution mining digestion, and other bioconversion processes, or lea -ping are practiced for the recovery of as well as efficient separation methods to re- uranium, soda ash, and potash. Despite these cover fuel products from process streams. successes, most of the opportunities -emain untapped because of teettnological barriers. Each mineral deposit has its own characteristics, and TECHNOLOGIES FOR EXPLOITING the processing environment deep beneath the MINERAL AND METAL RESOURCES earth's surface has been const 'icted by nature; our ability to modify it is limited. The technologies involved in the minerals Many research needs in this area parallel processing inaustry can be broken down into those of in-situ processing of oil shale and those where the desired metal component is in recovery of heavy crude oils. Additional re- high concentration, such as scrap iron, iron ore, search needs cover a broad spectrum of mining, phosphate ore, and bauxite, and those where metallurgical, environmental, and chemical en- the concentration of the valuable constituent is gineering: solids handling and comminution, low, such as gold and sliver ore, lean copper separations and concentration processes for ore ore, and certain types of scrap and wastes. beneficiLtion, electrolytic processing, solvent extraction, and treatment and disposal of waste High-Concentration Raw Materials products. It will take the best-trained chemical engineers using the most sot histicated tools of The economics of extraction processes re- chemistry, physics, and computer technology quire the primary raw materials industries to to unlock economically the vast reserves of locate near the richest ore deposits. Many such metals and minerals that cannot be recovered deposits are now outside the United States. The at present. combination of high U.S. labor costs, foreign government subsidies, and very dilute domestic ore deposits (0.5 percent or less) has driven a Low-Concentration Raw Materials substantial portion of the copper mining and The ascendant method for economically pro- refining industry to foreign countries. The de- cessing deposits low in the desired component velopment or acquisition of technology abroad is solvent extraction. This is most commonly for the smelting of nonferrous metal sulfides, done in processing plants above ground, and all electrolytic extractiun of zinc, and production the spent ore must be restored to its original of steel has drawn portions of those industries locatior or otherwise disposed of in a way that overseas. Although U.S. industry has been slow meets environmental constraints. The restora- to adopt the new technologies, there are indi- tion cost can be borne by high-value products cations that the pace of U.S. researcu and like gold, silver, and uranium. In the production development in these areas may quicken. An of moderate-value products such as copper from example is the steel industry/federal government lean ores, magnesium from dolomite, and alu- initiative in steel making. Projects now under minum from raw tnaterials other than bauxite, way or planned include electromagnetic contin- the only economically viable way to process uous casting, direct reduction of ore, and de- deposits may be to extract more than one velopment of processes to remove copper and product. This requires designing and building tin from scrap steel. more complex chemical plants, with all the 98 I. ROA I IER.S I\ ( Ill...111( 1/. E.N6LNEER/V;

attendant challenges to chemical engineering of living and itspositie-in the worldwide research. economy. The demands for energy and materials con- Waste Streams as Sources of Minerals and tinue to increase, and the accessibility of natural Metals resources to meet them continues to tall as the most easily recovered fossil fuel and mineral Substantial quantities of aluminum, copper, deposits are depleted. The gap between rising and steel are reused as scrap. The challenge is demand and falling availability must be bridged to purify the scrap metal sufficiently to process by technology that improves the efficiency of it for reuse. There is opportunity for new pro- extraction, conversion, and use of energy and cesses that can remove unwanted elements materials. The development of such technology eititer alloyed or piece contaminantsmore ef- takes long lead times, and there is a paramount fectively and at lower cost than current prolong-term need to maintain momentum in re- cesses. search on the frontiers of chemical and process Many of the waste streams from U.S. process engineering. The problems enumerated here industries are water containing small quantities offer challenges equal to any that chemical of metal ions that the law requires be removed engineers have faced in the past. before the wastewater is disposed of. There is an economic incentive to recoup at least some of the cost of wastewater treatment by recover- In-Situ Processing ing and selling the metal content instead of Available resources of fuels and materials in m: rely disposing of the metals as sludge. Be- the accessible parts of the earth's crust are cause the waste streams are dilute in desired becoming increasingly scarce. The alternative materials, research is needed to devise efficient to moving greater and greater amounts of crust, extraction and separation processes. whether it is mixed with the valued substance Likewise, fly ash from power plant combus- or simply overlies it, is in-situ processing. Al- tors often contains small amounts of metals or though this technology is well establiz.hed in their oxides, which require costly disposal in petroleum recovery, the long-term incentive to the ever-shrinking number of approved hazard increase its efficiency is great. The incentives ous waste landfills. Thus, there ate economic for other in-situ technologies vary but are bound incentives to recover the metal values as well to intensify in the future. The development of as to reduce the costs of ultimate disposal. in-situ processes involves long lead times in Here, too, the metal content is low, and research research and development. Field tests are large- is needed to develop economical separation scale, prolonged projects that may last many processes. In principle, advances in this area months. The potential environmental problems could bc aanslated into recovery of metal values are considerable. By the time the need for an fro n mine tailings. in-situ process becomes acute, it is too late to commence research. The prize goes to those INTELLECTUAL FRONTIERS who are prepared. Problems with in-situ processing share certain The basic technologies used in the energy and elements. fluid phases move through a vast, natural resource processing industries have many complex network of passages in a porous me- elements in common, and the chemical engi- dium. The process is inherently nonsteady state. neering profession has a long history of finding The physical transformations or chemicalre- and adapting basic technologies to the needs of actions proceed in zones or fronts that migrate diverse industries. No profession is better suited through the porous structure. The fluids interact by tradition and training to attack the many physically with the solid walls that define the difficult technical problems of these industries. passages. The passages are irregular, and their And these problems must be attacked and solved dimensions and structure change with distance. if our country is to maintain its high standard This structural inhomogeneity imposes uncer- PROCESSING OF ENERG} AND NATURAL. RESOURCES 99 taintieg *!ta.. make processing in situ riskier than rocks) mast be broken into grains; these may processing in designed and constructed plants. have to be comminuted to ye: finer particles. Further, the potential adverse environmental Current crushing and grinding processes are impacts of in-situ prot.esing have proved to be highly energy inefficient; typi..:ally 5 percent or important barriers to the widespread commer- less of the total energy expended is used to cialization of in-situ processes for oil shale and aCCOMPIlsh solids fracture. These processes als coal. Sustained research in the following areas produce a broad distribution of particle sizes, is needed 'ace both environmental and including fines that are difficult to process fur- process risk ther. Solids comminution could be greatly improved by a process that fractured crystalline porous structures, both at the micr iscopic scale and larger; solids selectively along grain boundaries. Fundamentalunderstanding methods for creating or enhat.cing perme- ofcrushing, ability in nonporous formations of oil shale, grinding, and milling is deplorably limited. For example, there is no rational basis for the design coal, and ore bodies; of a ball mill, a commonly used industrial device. combustion processes under reservoir con - ditions: Mineral processing and chemical engineering mechanisms of oil displacement; researchers need a deeper understanding of the distribution and flow of viscous fluids solid-state science and fracture mechanics just in porous media and the motion of complex as they have mastered and are contributing to fronts; colloid and interface science. There are great challenges in devising chemical comminution surface and colloidal phenomena involved in fluid-rock interactions, such as wetting and aids as well as processes for handling solids spreading, and adhesion and release; that become plastic, sticky, or reactive at tem- phase equilibria, phase thermodynam.,, peratures reached in comminution. and chemical reactions between injected fluids Just as important as finding better ways to and solids in the reservoir; prepare granular solids and powders is finding phase behavior, colloidal aspects, adsorp- ways to move them, to contact them with fluids, tion, and rheology of surfactant fqrmulations; to allow them to react in chemical processes, rheology and degradation of hydrophilic and to separate the residues. A major study in 19816 showed that cost overruns on large proj- polymers and their interactions A ith rock; ects involving solids processing depended di- the chemistry involved in winning a desired component from a given type of deposit; rectly on the throughput rate of solids. Much separation of fines from produced mate- of the current equipment design for mineral rials; processing dates from earlier times when ores treatment and disposal of tailings; were richer and costs of processing not as high. mathematical models of such phenomena; The handling of coal, oil shale, and orzs would and be improved by research on the mechanics of process synthesis, design, management, and pneumatic and slurry transport of particulate optimization with severely limited information. solids, particularly on the mechanisms of failure through plugging, attrition, and erosion. Im- Geochemistry, geophysics, geology, environ- proved processes for coal liquefaction and gas- mental science, and chemical engineering must ification could curve from research on particu- be more closely linked if advances are to be late transport in fluidized beds, including high- made in these areas. pressure gas-fluidized beds of large particles, ebulated beds, and liquid sway reactors. We Processing Solids must also understand chemical reaction processes in systems of moving particles, especially Solids handling is ubiquitous in the processing :-.1 high temperatures and pressures. There are of energy and natural resources. To liberate the the related critical issues of particles being desired components, crystalline solids (e.g., consumed or created by chemical reaction,

i 1 o ;') 100 1 RON 111.K't ..N ( /// WC IL / l(,:\/./ RIM, particle agglomeration and sintering, and trans- ing or settling. Those that involve physically port and ,,eparation of hot sticky particles. For homogeneous mixtures must use more subtle example, Plate 5 shows the stages of retorting means to create products of different compo- of an oil zhale particle as the temperature is sition. These latter processes are pervasive in increased by an external hot gas. The kerogen industry; they consume large amounts of energy is reacted, cracked, volatilized, and coked in and require sophisticated research and design. an idealized series of concentric volumes. Lib- Separation processes are based on some ('' f- erated products must flow through the coked fcrence in the properties of the substances to zone to exit the particle. If the temperature falls be separated and may operate kinetically, as in too rapidly, particles can become wet and sticky. settling and centrifugation, or by establishing These problems are related to the more general an equilibrium, as in absorption and extraction. problem of chemical reactions involving liquids Typical separation processes are shown in Table or gases inside porous solid particles. 6.1. Better separations follow from higher se- Equipment design and scale-up present par- lectivity or higher rates of transport or trans- ticularly great challenges whenever so'rds are formation. The economics cc separation hinges to be processed on a large scale. Consequently, on the required purity of the separated substance advances in the basic understanding of solids or on the extent to which an unwanted impurity processing will be for naught if they are not must be removed (Figure 6.13). translated into practicd, reliable designs. This Most methods of separating molecules in will require close cooperation among the fields solution use direct contact of immiscible fluids of mechanical, mineral, and chemical engineer- or a solid and a fluid. These methods are helped ing and between disciplines in the earth and by dispersion of one phase in the other, fluid physical sciences. phase, but they are hindered by the necessity for separating the dispersed phase. Fixed-bed Separation Processes adsorption processes overcome the hindrance by immobilizing the solid adsorbent, but at the Separations play a vital role in the processing cost of cyclic batch operation. Membrane proof energy and natural resources.' Improved cesses trade direct contact for permanent sep- separations can lead to improved efficiency of aration of the two phases and offer possibilities existing processes or to economical means for for high selectivity. exploiting alternative resources. For example, There is already intensive research on mem- the petroleum refining industry is based on brane separations for energy and natural re- separations of natural and synthetic hydrocar- source processing. Applications have so far bons. Improved separations could lead to better centered on organic polymeric membranes for concentrations of aromatic hydrocarbons in gas- mild service conditions, but research could lead oline to enhance the octane rating and paraffinic to both organic and inorganic membranes that hydrocarbons in jet fuel to improve burning can operate under harsher cor.ditions. Zeolites characteristics. The winning of critical metals and other shape-selective porous solids like such as copper, uranium, and vanadium from pillared clays appear to offer a fertile field of low-grade domestic ores requires chemical ex- research for separation applications. Chemically traction followed by recovery from the dilute selective separation agents that distinguish be- extractant solution. More selective extractants tween absorptive, chelating, or other molecular are neede6, as are better separations to remove properties are also attracting study. fly ash, sulfur oxides, and nitrogen oxides from Research should continue on traditional sep- power plant and other gaseous emissions to aration methods. For example, there is a con- protect air and water quality. tinuing need for more selective extraction agents Every separation process divides one or more for liquid-liquid and ion-exchange extractions. feeds into at least two products of different High-temperature processes that use liquid. let- composition. Separation processes that operate als or molten salts as extraction agents should on heterogeneous feeds usually involve screen- have potential in nuclear fuel reprocessing and

1 1 0 PROCESSI,N6 IN LAI.R61 1\l) MI R 1 RI tiff( la I 101

TABLE 6.1Methods for Separating Mixtures Property Difference Examples of Processes

Particle site Screening. mechanical Jigging Magnetism Magnetic separation Density Centrifugation. settling, Jigging Solubility Extraction Surface affinity Adsorption Solid/hquid phase Filtration Molecular character Dialysis. membrane gas separation Molecular size Molecular sieve separation Dielectric constant Electrophoresis Solidification temperature Zone refining Rate of phase change Crystallitation. distillation low:: character Ion exchange

metals recovery; basic thermodynamic dat:t on ferred methods of concentrating themdepend such high-temperature systems are lacking. on both their electrochemical and colloidal prop- Many of the ores of base metals are sulfide erties. The separation processes leave a large deposits. They must be milled to exceedingly quantity of unwanted fines that must be rejected fine size in order to free the wanted grains from as slimes. Better understanding of these pro- the rest of the mineral. The desired grains are cesses should permit separation of complex semiconducting colloidalparticles, and the sulfides and discovery of paths to recovering mechanisms of leaching and flotationthe pre- individual metals from dilute, impure solutions.

Radium 106 Vitamin B-12

104 2 .-. to) Mined Gold 102 Penicillin U_ Fc I._Factor of 2 o. Uranium from Ore Tdifferential in price

1 Copper Magnesium from Seawater Bromine from Seawater Mined Sulfur Sulfur from Stack Gas

10-2IM Oxygen

1 I 100 percent 1 percent 1 thousandth of 1 millionth of 1 billionth of 1 percent 1 percent 1 percent

DILUTION (expressed as percent concentration)

FIGUer 6..3 The importance of separation processes in aetermining the eventual cost of materials and products is illustrated in this figure. Product pnces cor, elate with the degree of dilution of the raw mat' nal in the matrix from which it r be isolated. A factor of two in product price is shown in the figure . Co, y. Norman N. Li. Allied-Signal Corporation 107 FRONTIERS IN CHEMICAL ENGINEERIAG

A hypothc.ical ieparation of a homogeneous membranes and more effective in controlling mixture, carried out in a thermodynamically flow in porous media. In addition, the de% el- reversible manner, would require the theoretical opment of materials that are less energy inten- minimum expenditure of energy. In practice, sive in terms of production and use is a goal however, separations of such mixtures need 50 equivalent to other means of energy conserva- to 100 times this minimum. Thus, there is tion. significant opportunity for imi rovement of sep- The relatively mature technology of upgrading arations by creating ways to reduce energy heavy oils by reaction with hydrogen is illus- consumption %'i thout a comeoensurate increase trative. Reactors are required to withstand hy- in capital and operating costs. drogen embrittlement at high pressures and Researchers ir, separation science and tech- temperatures. Present practice is to use foot- nology draw on and contribute to a variety of thick reactors lined with alloy steels. The largest related fields, including of these can no longer be con',tructed in the United States because the cessation of nuclear phase-equilibrium therm, : namics; power construction has lel to the closing of mass transfer and transn(, phenomena; facilities capable of such fabrication. Cheaper interfacial phenomena, including surface reactor materials would improve the economics and colloid chemistry; of the process; I,' :ter materials could lead to mechanisms of chemical reactions, espe- operability under more severe conditions that cially complexation reactions; would provide higher conversions. analytical chemistry; and Materials problems abound in the energy computer-assisted process and control en- storage field. For example, cheap materials with gineering. large effective thermal capacity are needed to Future progress in separation science and tech- store thermal energy in solar heating systems. nology will require continued cooperative re- Some sy items use chemically reactive materials search between scientists and engineers in these and store energy as enthalpy of reac:ion. There fields. is an opportunity to develop photosensitive catalysts to improve the coupling between the solar energy input and the energy converter. Materials High-temperature thermal energy storage sys- Research on materials can lead to more eco- tems confront corrosion problems aggravated nomical processing under extreme conditions by thermal cycling and temperature-sensitive and to reduced capital and operating costs. solubilities that conspire to shorten system life, There are strong incentives to find construction signifying the need for better materials. Like- materials for process units that are derived from wise, battery storage of electrical energy is domestic resources, that are less contaminating limited by the cost of mate ials and by corrosion of process and environment, and that have the and microstructural changes aggravated by the following properties: inherently cyclic operation. greater strength and more resistance to Materials science is an intrinsically interdisciplinary field. Materials scientists include phys- abrasion and corrosion; icists, chemists, metallurgists, mechanical en- longer life and less subject to degradation gineers, and chemical engineers. It is the latter by cycling conditions; who have the best opportunity to establish serviceability under more severe conditions specifications for needed materials and to join of temperature, pressure, or neutron flux; and in research on ways to meet those specifications. greater resistance to hydrogen embrittlement. Advanced Methods for Design and Scale-ui There are comparable incentives to develop new process-related materials that are more Many of the shortcomings of energy and selective as cata!ysts, extractants, or separation natural resource processing arise from lack of

11 2 PROCESSING OF ENERGY AND ,VA TURA!. RESOURI ES 103

sufficiently powerful design and scale-up pro- cost implications when they select construction cedures for the practicing chemical engineer materials and prepare flow sheet s. (see Chapter 8). A goal of research is to design large units from first principles and small-scale Other Important Research experiments. This has been done in the past; scale-up factors of 50,000 are common in petro- Many additional intellectual challenges for leum refining technology. However, in much of chemical engineers are relevant to energy and energy and natural resource production, there natural resource processing. These include resis such complexity and lack of basic data, ervoir modeling (Chapter 8), combustion (Chap- especially for large-scale solids processing, that ter 7), catalysis (Chapter 9), and electrochemical empiricism will continue to prevail until pilot engineering (Chapter 9). A recent report entitled plants and demonstration projects are success- Future Directions in Advanced Exploratory Re- fully modeled. Scale-up factors in solids pro- search Related to Oil, Gas, Shale, and Tar cessing typically range from two to five. Sand Resources8 discusses some of the chemical For example, research on moving hot solids engineering research challenges described in this chapter in a broad, multidisciplinary context will require individual pieces of equipment, then whole systems, from which reliable data for that includes the earth sciences. scale-up can be obtained. Costs of such research Finally, it should not be forgotten that chem- would be out of reach for all but the largest istry plays an important role as a fundamental industrial and government laboratories. The science for these industries. Major contributions application of research results to improved com- to be expected from chemistry in energy and mercial oil shale retorting would require large natural resources are discussed in the 1985 pilot plants costing tens of millions of dollars, ort Opportunities in Chemistry.9 followed by demonstration plants or single commercial modules costing hundreds of millions. IMPLICATIONS OF RESEARCH Experimentation on an equivalent scale can be FRONTIERS imagined for a new steel-making technology, Each of the generic research areas discussed for in-situ leaching of uranium, or for solution in this chapter has a strong multidisciplinary mining of hydrothermal mineral deposits such character. While the underlying fundamentals as soda ash. Such research will require inter- of some are amenable to investigation by indi- disciplinary teams and sustained activity over vidual chemical engineers, in many cases col- periods of years. laboration will be required between chemical The problem of ever-increasing construction engineers and other scientists and engineers costs dates from the mega-project concept of skilled in geology, geophysics, hydrology, me- World War II and the race toward an overnight chanical engineering, physics, mineralogy, ma- synthetic fuels industry. Nevertheless, large terials science, metallurgy, surface and colloid construction projects will be needed to bring science, and all branches of chemistry. It will coal gasification, coal liquefaction, and oil shale be necessary to generate creative interactions processing to fruition. Construction costs are that overcome traditional academic compart- not small in the minerals processing industries, mentalization of outlook, experience, and ed- and the more dilute the ore, the larger must be ucation. Academic departments of chemical the economically viable plant. The incentives engineering should take the lead in establishing are great to develop lower cost designs and interdisciplinary teams to carry out fundamental construction methods. Noteworthy ideas in- research in these high-priority research areas. clude modular construction from preassembled They should seek ways to involve government units and the organization of construction into and industrial scientists in interdisciplinary ac- multiple small projects. Substitute materials that tivities. There must be freer flow of information can be produced with lower energy or raw between industry, university, and government; material cost need to be developed. Chemical professional disciplines; and academic depart- engineers must be cognizant of the construction ments.

1 i 3 104 ER0,\TIERS IN CHEMICAL E,VGI,NTERI.VG

The educational background of chemical en- NOTES gineers makes them particularly well suited to solve problems in the areas discussed herein. 1. U.S. Department of Commerce, Bureau of the Census.Statistical Abstract of the United States: Chemical engineers are used to working with 1987,107th ed. Washington, D.C.: U.S. Govern- concepts from all the related fields, and their ment Printing Office, 1986, Table 1310. training has evolved to cover most of the skills 2. U.S. Department of Commerce, Bureau of the needed to solve technical problems. Interdis- Census.Statistical Abstract of the United States: ciplinary research in the relevant areas can only 1987,107th ed. Washington, D.C.: U.S. Govern- strengthen the chemical engineering cadre in ment Printing Office, 1986, Tables 1233, 1215. the energy and natural resource processing 3. National Petroleum Council.Enhanced Oil Re- industry. covery.Washington, D.C.: National Petoleum The funding needs for the research described Council, 1984. in this chapter will be large and long term; they 4. J. Haggin. "Methane -to-Gasoline Plant Adds to can be met only by some combination of gov- New Zealand Liquid Fuel Resources."Chem. ernment and industry. Government support is Eng. News, 65 (25),22 June 1987, 22. appropriate because efficient processing of en- 5. E. Berenyi. "Overview of the Waste-to-Energy ergy and natural resources is key to continued Industry."Chem. Eng. Prog.,82 (11), November national growth and prosperity. Appropriate 1986, 13. initiatives for the Department of Energy, the 6. E. W. Merrow et al.Understanding Cost Growth U.S. Bureau of Mines, and the National Science and Performance Shortfalls in Pioneer Processing Foundationall in cooperation with industry Plants (Report R-2569DOE). Santa Monica, Calif.: are laid out in Chapter 10. Industry support is Rand Corporation, September 1981. mandatory because commercialization is a goal 7. This section draws, in part, on a recent NRC and because companies will be the eventual report,5, wration and Purification: Critical Needs and Opportunities.Washington, D.C.: National profit-driven proprietors of the technology de- Academy Press, 1987. veloped. There is advantage and precedent for 8. National Research Council, Board on Chemical companies to band together in consortia or Sciences and Technology.Future Directions in through institutes to provide continued funding, Advanced Explorator, Research Related to Oil. particularly of basic research. In addition, such Gas, Shale, and Tar Sand Resources.Washington, long-term commitment will allow academic re- D.C.: National Acac' ,my Press, 1987. searchers the freedom to set up ongoing pro- 9. National Research Council, Board on Chemical grams to feed basic data and concepts into the sciences and Technology.Opportunities it: Chem- centers and consortia without fear of sudden istry.Washington, D.C.: National Academy Press, shifts in funding priorities. 1985. SEVEN Environmental Protection, Process Safety, and Hazardous Waste Management

hemical engineers face important research challenges associated with the imperative to protect and improve the environment. These challenges include designing inherently safer and less polluting plants and processes, improving air quality through research on combustion, managing hazardous wastes responsibly, developing multimedia approaches to the study and control of pollutants in the environment, and assessing and managing chemical risks to human health or to the environment. The future of the profession and the industries that it serves will depend on the vigor with which chemical engineers apr:lach these challenges. These challenges have important implications for the way that chemical engineers are educated and for the means by which government and industry provide research support to the field.

ir. 106 RON MRS IN (*HE P1/4 6LN LERIA6

IR. SOIL. AND WATER arevital to life on transportation, and useand finally end with this planet. We must protect these re- the ultimate return of processed materials or sources aid use them wiselyour sur- their residues to the environment (Figure 7.1). vival as a species depends on them. Despite The resulting redistribution of chemicals within recent impressive strides in improving the en- the environment may 11:.4w. adverse impacts. vironment, evidence is overwhelming that more Harmful compounds of certain elements (e.g., effective action must be taken to address such nitrogen, sulfur, halogens) may be widely mo- critical issues as acid rain, hazardous waste bilized. Other elements may be converted from disposal, hazardous waste landfills, and ground- innocuous forms (e.g., mercuric sulfide in cin- water contamination. Itis also vital that we nabar) to highly toxic forms (e.g., methyl mer- assess realistically the potential health and en- cur) ). Chemical engineers are involved in all vironmental impacts of emerging chemical prod- aspects of chemical manufacture and process- ucts and technologies. The problems are clea_ly ing; therefore. it should be their re ,ponsibility complex and demand a broad array of new to safely manage chemicals in the environment. research initiatives. Environmental and safety concerns will be cru- Technological activitiesof which chemical cial challenges to the chemical engineers of the manufacture and processing are key parts future. No other group is better trained or more begin with the extraction of raw materials from centrally positioned in the industrial world to the environment. They then proceed through be the cradle-to-grave guardian of chemicals. numerous stepsprocessing, storage, handling, Our approach to environmental and safety

Production

Storage (--ENVIRONMENTAL TRANSPORT AND TRANSFORMATION

EXPO:ARE AND EFFECTS

(. rinl1

Transportation

FIGURE 7.1Life cycle of chemicals in the environment. ENVIRONMENTAL PROTECTION, PROCESS SAFETY. HAZARDOUS WASTES 107

MANAGEMENT DECISIONS RESPONSE OF ECOSYSTEMS in turn affect human beings and other rescarces. A risk assess- Physical Changes introduction ci new chemicals or ment of the potential health and products Iwo the env turn id Maims and energy lows Physical struck,' of system welfare effects of such changes Introduction or moat:salon of Transport and Trardonnetion can indicate whether and to what species In the e,Mronment &dopiest Changes extent mitigation measures should Technology Utillandon (Existing be taken, or original decisions or new monocle) Change in population size Modifications to interactions rethought. Each of the boxes in °lenges In resource utilization indirect seeds on speckle (whin develonment. fuels. etc) diversity the diagram, although greatly simplified, represents a cluster of questions that require answers from environmental research.

RISK ASSESSMENT AND MANAGEMENT IMPACT ON SOCIETY OF Determlnedon of risk to susceptible CHEMICALS IN THE populations ENVIRONMENT Identlkstion of mitigation measures A number of environmental issues have received widespread Publicity (Table 7.1), from major accidents at plants (e.g., Seveso and Bhopal) to the global and EFFECTS ON HEALTH AND WELFARE EXPOSURE ASSESSMENT regional impacts associated with Changes In WOK organs and the Exposure of human population energy utilization (e.g., carbon roodeted health effects dioxide, acid rain, and photo- Chinon In the "Wed environment Exposure ci ankswis and plants chemical oxidants), the improper (nwerbis, Weiblity, etc) ci cone= to humens disposal of chemical waste (e.g., Love Canal and Times Beach), FIGURE 7.2 The effects of human activities on the environment. and chemicals that have dispersed and bioaccumulated affecting wildlife (e.g., PCBs and problems must change from reactive to proac- DDT) and human health (e.g., cadmium, mer- tive; in other words, instead of responding to cury, and asbestos). crisis and public pressures, we must anticipate As a consequence, much of the public has and prevent problems. This will require an come to believe that most chemicals are haz- understanding of the detailed chemistry and ardous. A recent poll by the Roper Organization physics of processes at the molecular level. revealed that two out of three American citizens Such basic understanding is crucial if we are to expect a major chemical disaster, resulting in design plants that are safer and that cause less thousands of deaths, within the next 50 years.' pollution, devellp better ways to manage and The poll also found that a high propoi:ion of detoxify hazardous waste, and predict the fate the public lacked confidence that industry would of chemicals in the environment as well as the df .I openly with them. A public attitude toward effects of chemicals on humans and ecosystems. exposure to chemicals is developlog that can Figure 7.2 diagrams how human activities can be summed up by the words, "no risk." But, affect environmental quality and human health. as a judge recently stated, "In the crowded Introduction of new chemical products, adop- conditions of modern life, even the most careful tion of different technologies, or changes in person cannot avoid creating some risks and resourcutilization can lead to emissions that accepting others. What one must not do, and affect the physical, chemical, or biological re- what I think a careful person tries not to do, is sponses of the receiving ecosystems. This can to create a risk which is substantial."2

ii7 108 I.ROS III RS I\ ( III,%II(I. I.\(;I \I l_1(l\l,

TABLE 7.1 Some Well- PubliciztJ Environmental Issues Global Chlorofluorocarbons and their effect on ()Lone in the upper atmosphere Carbon dioxide and other "greenhouse" gases (c g.. methane) and their effect on global temperature DDT Polychlorinated biphenyls (PCBs) Regional Acid rain (NO,. SO,) Agncultural wastew;ers (Kesterson Wildlife Refuge. Chesapeake Bay) Urban Airborne lead from automobile exhlusts Carbon monoxide from automobile exhausts Photochemical oxidants. particularly ozone (Los Angeles) Sulfur oxides (SO,) and particulate matter iLondon. ronora) Site-specific Dioxins (Seveso. Love Canal. and Times Beach) Methyl isocyanate (Bhopal) Methyl mercury (Minamata) Cadmium ((tai -(tai) Indoor air (Formaldehyde. asbestos. NO.. CO)

What is a substantial risk? How safe is safe logical leadership and international competi- enough? These are questions that trouble the tiveness i-i.major segments of industry. public, industry, and regulators alike. Scientific Some of the economic and social costs related understanding and the available data are inad- to major environmental issues are discussed in equate to evaluate the true risk to individual the following sections. safety, or the true risk of damage to numan health or the environment, from exposure to Chemical Industry Safety most chemicals or to chemical plant or disposal The U.S. chemical and petrochemical indus- operations. Yet, legislators and regulators can- try safety record is generally good. The National not wait for all the data to come in before they Safety Council's 1985 data show that the chem- start to provide the public with the protection ical industry worker is only one-fourth as likely it is demanding. Laws have been passed and to hay..: a fatal on-the-job accident as the average regulations have been developed that require U.S. employee.' The Bureau of Labor Statistics government approval for production of new indicates that the 1984 chemical industry rate chemicals, design and operation of chemical for lost work days resulting from occupational plants, workplace exposures, certain product injury was 2.4 per 200,000 man-hours, compared uses, quantities and concentrations of chemicals to 4.7 for manufacturing as a whole and 3.7 for in effluent streams, and disposal of waste and all private sector employment.' Over the past by-product streams. 3ut because of the uncer- decade, annual fatalities from hazardous chem- tainties, some regulations have been written to ical accidents have numbered around 40, while protect against the effects of extremely unlikely highway fatalities have lumbered around ,000.3 worst-case scenarios, resulting in a misalloca- Nevertheless,accidentsandunintended tion of resources, reduceu technical innovation. chemical releases pose serious financial risks to and excessive costs. At the same time, other, the chemical and petrochemical industry. In less visible hazards that might be the focus of 1984 there were five major accidents in the appropriate regulation have been overlooked. hydrocarbon-chemical industries, totaling an This situation must be corrected. Society estimated loss of $268 million.' Hundreds of needs a clean and safe environment. It also lesser accidents occur yearly. The total annual needs to capitalize fully on new developments cost to the industry of accidents and unintended in chemistry. biotechnology, and materials sci- chemical releases is difficult to quantify.It ence, if the United States is to retain techno- includes significant costs owing to interruption

1 i C L.VIIRONSILV Ill. ?ROI 11 IPA, PRO( I. SS 1111 II, II1/ %KIWI S II 111 1:1 109

standing and important problems inenvironmentalprote( 'ion. Fossil fuels are used in such magnitude that emissions from cc.nbustion sources have a major impact on urban, regional, and 3Io'cal air quality. Combustion- generated pollutants are derived from contaminants in the fuel such as sulfur, nitrogen, and inorganic compounds, from incomplete combustion of the fuel, and from the high-teperature reac- t:on of nitrogen with oxygen in the air heated by combustion FIGURE 7..)In 1984, 14 years after the passage of the Clean Air Act. processes. These pollutants are significant areas of t:,e United States were still in violation of National Ambient emitted both in gaseous form Air Quality Standards (NAAQS) for ozone. Courtesy. Environmental Protec- (e.g., the oxides of nitrogenthe tion Agency. sum of NO and NO2, denoted as NOxsulfur dioxide, and un- burned hydrocarbons) and in of business as well as major liability an,1 liti- particulate form (e.g., fly ash and soot). The gation costs associated with injuries, deaths, 1970 Clean Air Act and its amendments are property damage, and insurance premiums. It ;I;rectedatreducing combustion-generated also includes losses oc product and feedstock emissions through the establishment of National that are direct profit losses for the manufacturer. Ambient Air Qra:ity Standards (NAAQS) for One estimate is that U.S. industry spent $7.7 oxides of suifur and nitrogen, carbon monoxide, billion in :985 for protecting worker safety and particulate matter, iead, and ozone. While sub- health;' the total annual cost of accidents and stantial reductions in urban levels of carbon unintended chemical releases by the U.S. chem- monoxide, sulfur oxides, and particulate matter ical and petrochemical industries is surely many have been achieved over the past 15 years, billions of dollars. ozone levels, controlled by an intricate chem- Costs associated with increased government istry involving organic gases and oxides of regulation are also difficult to quantify. Public nitrogen, have proved to be more resistant to concern in response to chemical release acci- control (Figure 7.3). Large-scale air quality dents affects regulators and community policy problems arising from combustion-generated groups. It is evident that the U.S. chemical emissions, such as acid rain (Figure 7.4), re- industry is already spending large an.Junts of gional hazes, and volatile toxic compounds, money to avoid accidents and to deal with their have also assumed prominence and will likely consequences when they occur; these costs are be the targets of future legislation. borne in part by the consumers. Continued Adding fluidized gas desulfurization to coal- expenditures are likely as industry strives to fired generating plants is estimated to add 15- achieve an "acceptable" level of public safety 25 percent to their total capital costs, or up to throughout all chemical industry operations. $125 million on a typical 500-megawatt unit.' Since the cost of reducing emissions by modi- Combustion of Fuels for Power Generation fying the combustion process i. usually an order and Transportation of magnitude lower than that of cleaning the fuel before burning or removing the pollutants The burr ig of Aid for power generation and from the exhaust gases, there are signi cant transportation presents some of the most long- challenges to develop clean, fuel-efficient com- 110 FRONlIERS IN CHEMICAL ENGINEERING

FIGURE 7.4Annual mean value of pH (acidity) in precipitation, weighted by the amount of precipitation in the United States and Canada for 1980. The low pH values seen in the eastern United States and Canada reflects the high acidity of the rainfall and other precipitation in this region. This geographic pattern of acid rain correlates with the emission and transport of sulfur dioxide from coal-burning plants. From U.S./Canada Work Group #2, Atmospheric Science and Analysis. U.S. Environmental Protection Agency. Washington. D.C.. 1982. bustion processes as well as to design more while around 500 million gallons are incinerated. economical processes for fuel cleaning prior to Of the total quantity of hazardous waste gen- combustion and for destroying or removing the erated, manufacturers account for 92 percent. residuals from postcombustion gases. It has been estimated that the chemical industry alone generates 71 percent of the manufacturer's total.' The Congressional Budget Office has Hazardous Waste Management estimated that the 1984 amendments to the The disposal of hazardous waste may well Resource Conservation and Recovery Act could have become the Achilles' heel of the American increase industrial compliance costs from be- manufacturing industry. More than 300 million tween $4.2 billion and $5.8 billion in 1983 to tons of hazardous waste are generated annually between $8.4 billion and $11.2 billion in 1990, by about 14,000 wstallations in the United depending on the level of waste reduction States. About 14.7 billion gallons of hazardous achieved by industry .9 waste is disposed of in or on the land each year, A variety of methods have been used over 12[, ENVIRONMENTAL PROTECTION, PROCESS SA1 LT1. If tZ.1RDOUSli AS TES 1 1 1 the past 100 years to bury hazardous waste. waters. After a few years of pursuing such Many of these burial sites now pose a threat to methods, it is clear that they do not provide a the health of nearby residents and, more broadly, solution to the problem of containment of waste to the nation's underground water supply (Fig- in existing landfills; slurry walls leak and clay ure 7.5). For example, recent studies by the caps crack. Itis also becoming increasingly state of California have shown that there are clear that it will require decades to accomplish widespread threats to groundwater in Califor- an adequate cleanup of hazardous waste sites nia's Silicon Valley; Santa Clara County leads nationwide. Thus, when the Superfund act was the nation in the number of sites on the National reauthorized in 1986, a significant focus was on Priority List, most of which are associated with the use of new technologies to decontaminate the electronics industry.'° The U.S. Congress soil and groundwater and to provide for long- Office of Technology Assessment has recently term containment of wastes (i.e., through en- projected that there are 10,000 sites nationwide capsulation). The level of expenditure could be that beiong on the National Priority I ist of toxic as high as $10 billion over the next 5 years. waste dumps." When one takes the cost of industrial com- In 1980, Congress appropriated $1.6 billion pliance with RCRA to handle currently gener- for a 5-year Superfund program. The original ated wastes and adds the cost of Superfund to Superfund legislation viewed cleanup of haz- clean up the wastes of the past, it becomes ardous waste sites as a relatively short-term obvious that there are strong incentives for program and anticipated that waste could be technology development in the area of waste contained for several decades by methods such minimization and treatment, and many oppor- as building slurry walls and clay caps to elimi- tunities for research and employment for chem- nate diffusion of buried waste into subsurface ical engineers.

Well Deep-Well Injection Spills and Buried Leaks Wastes Well

FIGURE 7.5Past methods of waste disposal threaten water supplies today. Reprinted from Opportunities in Chemistry. National Academy Press. 1985. 121 112 FRONTIERS /IV CHEM/C ENGINEERING

DESIGN OF INHERENTLY SAFER AND decision on whether to allow their construction LESS POLLUTING PLANTS AND and operation. PROCESSES The chemical reaction pathways chosen for a manufacturing process profoundly influence Few basic decisions affect hazard potential chemical plant safety because they determine or have more of an impact on environment than the nature and amounts of all substrates, prod- the initial choice of technology. Thus, when ucts, and reagents and implicitly govern the designing chemical manufacturing processes, it design and operation of all hardware. The ulti- is important to select sequences of chemical mate goal of chemical engineering research to reactions that avoid the use of hazardous feed- provide inherently safer plants should be to stocks and the generation of hazardous chemical elucidate the connection between process path- intermediates. It is necessary to find reactions ways and plant safety and to translate this conditions tolerant of transient excursions in connection into a quantitative form amenable temperature, pressure, or concentration of to engineering design calculations. The array of chemicals attd to use safe solvents whet. ex- chemicals, reactions, processes, and types of tracting reaction products during purification physical equipment used in industry is exceed- steps. Finally, it3 important to minimize stor- ingly diverse and constantly evolving. Tc effec- age and in-process inventories of hazardous tively address the need for inherently safer substances. The term "inherently sager plants" plants, chemical engineering research must be has been used to describe this approach.'2 One focused on fundamental issues that span the further consideration in the design of a new entire range of processing activities, from elu- process or plant is whether it is going to generate cidating detailee reaction mechanisms to un- polluting effluents or hazardous wastes. Good derstanding and predicting the gross response design should result in waste minimization in a of coupled equipment. The goal of such fun- manufacturing press or plant. damental research would ne to develop the tools Traditional analyses of process economics needed to define, discern, and assess the safety might show that inherently safer and less pol- issues associated with a given process design luting plants are lest efficient in terms of energy and its alternatives. o1 raw materials usage. Indeed, chemical plants New approaches to the design of commercial e been designed in the past principaii, 'o chemical syntheses should be pursued. A chem- naximize reliability, product quality, and prof- ical synthesis tree graph with a high-value prod- itabili!y. Such issues as chronic emissions, waste uct at its apex, lower-value raw materials at the disposal, and process safety have often been base, and reaction steps as nodes connecting treated as secondary factors. It has become all branches offers a basis for quantitative as- clear, however, that these considerations are as sessment of feasible and economic process al- important as the others and must be addressed ternatives. It could also serve to define the during the earliest design stages of the plant. safety and environmental impact of a pathway This is in part due to a more real' tic calculation and offer a basis for safe designs that produce of the economics of building and operating a minimal wastes. Tree graphs could be particu- plant. When potential savings from reduced larly helpful in evaluating the process safety accident frequency, avoidance of generating implications of highly selective synthesis routes. hazardous waste that must 'Je disposed of, and Near the apex of the chemical synthesis tree decreased potential liability are taken into con- graph materials that might be used in chemical sideration, inherently safer and less polluting reactions are of highest value. Overall raw plants may prove to cost less overall to build material and energy costs are lowest for those and operate. And in any case, if tilt American pathways that use the most selective reactions public is not convinced that chemical plants are to achieve the highest yield of the desired designed to be safe and environmentally benign, product. However, these selective reactions then the fact that they operate er.onomically often require the handling and storage of more will be of little consequence to the public's reactive, and hence more hazardous, chemicals.

122 EMIRONMENTAL PROTECTION, PROCESS ,S,AFE1Y,// IZARDO/ S t1 ISMS 113 For example, in the synthesis of Carbaryl at be improved if designers had the means to bhopal, a process that required storage of large envision the complete reaction topography and quantities of the reactive intermediate methyl to assess the consequences of straying from isocyanate was used, rather than a less flexible normal operations. This would involve devel- and more expensive straight-through reaction oping design tools that would incorporate chem- scheme with a minimal inventory of methyl ical pathway information more systematically isocyanate. Developing the methodology of us- into classical engineering design methods for ing process safety and environmental factors in reactors and associated equipment. synthesis tree graphs could provide a better framework for future plant design. COMBUSTION Most accidents in chemical plants occur when the plant is not operating at a steady statefor Many of the environmental issues listed in example, wh,mr it is starting up or shutting down Table 7.1 are intimately related to combustion. or when aansient of temperature, pressure, Combustion contributes significantly to emis- or reactant concentration occurs. Fundamental sions of pollutants into the environment, with research in non - steady -s: ate process control and effects ranging from thos, pertaining to indoor the management of process transients is there- air pollution to those affecting global climate. fore warranted. Design methodology poses a For this reason, combustion has been singled related research issue. It is obviously easier for out to illustrate the progress that can be made the designer of a plant or individual reactor to in resolving environmental issues through a envision how the equipment will operate during sustain.i fundamental research program and to the normal production mode than to envision demonstrate the potential added benefits of how it will operate under a host of potential continued in-depth study of the physical and scenarios that derive from process transients. chemical processes underlying combustion. The safety of chemical plants and reactors could Hydrocarbons and Fuel- Bound Nitrogen CH4 C2H6 The burning of fuel in a prac- H2O,OH HI H2O,OH tical combustion system, such as a power plant boiler or the cyl- 0 H' 0' OH CH, C H, C 113040 CH3C0 C113 inder of an internal combustion H engine, is at first glance very 0 H1 02 simple: a mixture of hydrocarbon 0, OH C4,0 C H,, CH,O, CH() and air is ignited and burned to C2H4 carbon dioxide and water. On H2O,OH H clescr examination this burning turns out to be one of the most CHO C2113 complex processes in all engi- 0 02,H neering. For example, the corn- H11H

OH H busiion of the simplest hydro- CO C2H2 CH,C0 CH, carbon fuel, methane, involves

IO 1OH more than 50 chemical reactions (Figure 7.6). During the past four 02 CO, CH, CH2O, CHO decades, major progress has been 0 made in developing a mechanistic understanding of the combustion of methane and C-2 hydrocarbons and their derivatives. FIGURE 7.6Mechanism of methane combustion. Rate constants of many individ-

123 114 FRONTIERS IN CHEMIC-11. ENGIATERING

INTERIOR WTI STAGE ual free-radical reactions have BURNER WALL been measured, and a good number of those not measured can be estimated from thermochem- ical kinetics and unimolecular reaction theory. Major unknowns in the mechanism by which a hydrocarbon fuel burns concern the pyrosynthesis reactions that lead to the formation of polycyclic aromatic hydrocarbons (PAHs) and soot and the oxidation chemistry of atoms other than carbon and hydrogen (heteroatoms) in the fuel, particularly nitrogen, sulfur, and halogens. Nitrogen oxide emissions from furnaces and boilers come mostly FIGURE 7.7Schematic of a low-NO,/SO, pulverized coal burner. The addition from oxidation of the nitrogen of oxygen is staged to produce an initial fuel-rich zone in the burner that atoms in the fuel, whereas in results in reduced emissions of nitrogen oxides. internal combustion engines these emissions are derived largely from oxidation of atmospheric nitrogen. Burners of tinued experimental and theoretical study of the advanced design currently reduce the emissions combustion chemistry of higher molecular weight of nitrogen oxides by a factor of 2 from uncon- hydrocarbons should provide the understanding trollecombustionsr!entsby staging the ad- needed for the design of combustors in which dition of oxygen to produce an initial fuel-rich the fuel/oxygen regime is selected to minimize regime in which the bound nitrogen is partially emissions of PAHs and nitrogen oxides. converted toN2(Figure 7.7). Potentially greater reduction in nitrogen oxides can be attained by Soot adding hydrocarbons downstream of the fuel. This is called reburning (Figure 7.8). To deter- Combustion processes are a major source of mine the optimal sequence of air and fuel particles emitted to the atmosphere. Particles addition requires detailed knowledge of both formed in combustion systems fall roughly into the fuel nitrogen chemistry and the !iydrocarbon chemistry. CHn The development of staged combustors for the control of nitrogen oxides is constrained partly by the formation of PAHs and soot COMBUSTION (Figure 7.9). The PAHs are potential carcinogens whose biological activity depends strongly on their molecular structure. It is postulated that they are formed under locally fuel-rich FUEL MOLECULE conditions by the succt sive addition of C-2 CONTAINING FUEL through C-5 hydrocarbons to aromatic rings NITROGEN ATOMS followed by ring closure. On the other hand, a FIGURE 7.8NO, control in combusion by reburning. Ad- staged combustor cannot be operated on too dition of hydrocarbons (CH,,) late in the combustion process lean a fuel mixture because formation of nitro- leads to the reduction of nitrous oxide (NO) to nitrogen gas gen oxides is favored under this regime. Con- (N;)

12--'A ENVIRONMENTAL. PROTECTION, PROCESS S %PET), IIAZARDO('S IVASTLS 115

Hydrocarbon C4I-15 + C2H2_,...O compounds are formed (Figure Fuel Butadienyl Acetylene 7.10). An understanding of the Benzeneenzene mechanism of soot formation has + C4 become more important because + C4 + C2 soot hampers the use of such important technologies as staged Napdthalene Acenapthylene combustion and diesel engines. Phenanthrene + C4 In addition, the sooting tendency + C4 of aromatic compounds is higher than that of aliphatics, and the aromatic content of fuels is expected to increase in the future

Cyclopenta(cd)pyrene Benzo(a)pyrene as petroleum resource availabil- Fluoranthene ity forces refiners to use fuel feedstocks with lower ratios of FIGURE 7.9Mechanism of formation of polycyclic aromatic hydrocarbons hydrogen to carbon. (PAHs) dunng combustion. Soot is objectionable not only because of its opacity but also because soot particles are car- two categories. The first, referred to as soot, riers of toxic compounds. When combustion consists of carbonaceous particles formed by products cool, soot particles provide conden- pyrolysis of the fuel molecules. The second, sation sites for hydrocarbon vapors, particularly referred to as ash, is composed of particles PAHs. Soot particles are agglomerates of small, derived from noncombustible constituents in roughly spherical units. The small vary in di- the fuel and from heteroatoms in the organic ameter from 0.005 to 0.2 p.m, with most in the structure of the fuel. range of 0.01 to 0.05 p.m, while the size and Soot can be produced in the combustion of morphology of the clusters can range from gaseous fueis and from the volatilized compo- aggregates of several particles to large contrails nents of liquid or solid fuels. Soot formation is several micrometers in diameter and hundreds a complex process involving the chemistry of of micrometers in length. Soot particles art not fuel destruction under fuel-rich conditions where pure carbon. The atomic ratio of hydrogen to hundreds of aromatics and other intermediate carbon decreases from around 1.0 at the point

H 114 14 C C C NI Ni th C C C.C/ H + C2H2(-H) +H(-H2) +C2H2 C-, H f- .1111----

+ H(-H2) +C2H2(-H) +C2H2(-H)

: SOOT

FIGURE 7.10 A possible mechanism for soot formation.

12 116 I. RON% II.RS IN ( III.III(' II, I. \Yd.\ LER/ Vf,

NATURAL GAS UOUID HYDROCARBONS of first formation in the flame to 0.1 to 0.2 in 30% 92% 211 QUADS 652 QUADS 205 4 TRILL CU FT the cooled exhaust. 114 9 BILL BBL

There is considerable uncertainty about the 1964 US CONSUMP'ON - I% mechanisms of soot particle inception and growth. FOSSIL FUEL Oi RECOVERABLE P RESERVES To evaluate potential measures to suppress soot RECOVERABLE RESERVES formation or to accelerate its postflame oxida- 1914 tion,itis important to understand how fuel COAL structure and combustion conditions influence 87 ea 6 234 QUADS the physical and chemical nature of the soot 267 3 RILL SHORT TONS particle. Particle inception in the flame occurs COAL by some sort of nucleation process not yet 260% 17 OUADS understood. Once formed, the soot particles PETROLEUM 781 8 MILL SNORT TONS 46 6% 3, OUADS grow by surface reactions of hydrocarbon rad- 5 3 BILL BBL I. S icals. It is generally accepted that acetylene is FOSSIL .UEL CONSUMPTION an important intermediate for growth of soot 1964

particles, but the precise pathway by which the NATURAL GAS 272% acetylene contributes to soot growth is not I II OUADS known. Although significant progress has been 175 TRILL CUR FIGURE 7.11U.S fossil fuel reserves and consumption in made in simulating the dynamics of multicom- 1984. Courtesy. Department of Energy. ponent aerosols, basic information on the fundamental chemistry and physics of soot formation and growth is needed to enable researchers tive as particle size decreases to the 0.1-0.5 Jim to predict the size and chemical composition of range, so that particles in this size range that soot particles as a function of fuel type and escape contain disproportionately high concen- combustion conditions. trations of toxic substances. The processes tha! govern the formation of Ash ash particles are complex and only partially understood (Figure 7.12). The mineral matter The United States uses a disproportionate in pulverized coal is distributed in various forms; amount of gas and oil compared to coal (Figure some is essentially carbon-free and is designated 7.11). It has long been known that the country at extraneous; some occurs as mineral inclu- mt'st increase the fraction of its energy that is sions, typically 2-5 gm in size, dispersed in the derived from coal if it is to be independent of coal matrix; and some is atomically dispersed foreign energy sources. The major obstacles to in the coal either as cations on carboxylic acid the wider use of coal are environmental. The side chains or in porphyrin-type structures. The mineral content of U.S. coals averages about behavior of the mineral matter during combus- 10 percent by weight, and the sulfur content tion depends strongly on the chemical and varies widely around an average of approxi- physical state of the mineral inclusions. mately 2.5 pe;nt. The minerals in coal lead During combustion the mineral inclusions to the formation of ash particles, and the sulfur decompose and fuse. Most of the mineral matter leads to sulfur dioxide in the flue gas. adheres to the char surface, but some is released Ash particles produced in coal combustion as micrometer-sized particles. As the char sur- are controlled by passing the flue gases through face recedes during burning, the ash inclusions electrostatic precipitators. Since most of the are drawn together and coalesce to form larger mass of particulate matter is removed by these ash particles. Char fragments are released and devices, ash received relatively little attention take with thew ash inclusions and ash adhered as an air pollutant until it was shown that the to the surface. As each char fragment bons concentrations of many toxic species in the ash out, an ash particle is produced of a size and particles increase as particle size decreases. composition determined by the evolution of the Particle removal techniques become le' s effec- char pore structure during combustion.

12C ENVIRONMENTAL PROTECTION, PROCESS SAFETY, ILIZARDOU,S0 4.STES 117

INTERNAL VOLATITE 9D° T REDUCING INORGANIC ENVIRONMENT VAPORS (Na, V, As )

METAL VAPORS CHAR PARTICLE SUBOXIDE SP. GR. < 2.0 PARTICLES AGGLOMERATION aOXIDATION (Fe, WO, Mg) NUCLEATION 0 0 0 00 00 0 °o MINERAL INCLUSION

1 -50 um EXTRANEOUS ASH 0 SP. GR. > 2.0 0 O

FIGURE 7.12Fly ash formation during coal combustion.

The high temperatures of coal char oxidation Sulfur Oxides lead to a partial vaporization of the mineralor ash inclusions. Compounds of the alkali metals, Current strategies for reducing sulfur oxide the alkaline earth metals, silicon, and ironare emissions from coal-fired combustorsare based volatilized during char combustion. The vola- on the addition of calcium sorbents that retain tilization of silicon, magnesium, calcium, and the sulfur either as a sulfide ina reducing iron can be greatly enhanced by reduction of slagging combustor or as a sulfate formed in their refractory oxides to more volatile forms entrained flow in a pulverized coal boileror (e.g., metal suboxides or elemental metals) in fluidized bed (Figure 7.13). The utilization of the locally reducing environment of the coal the calcium is currently as low as 20percent, particle. The volatilized suboxides and elemen- which resells in a large volume ofspent sorbent. tal metals are then reoxidized in the boundary The problems preventing better sorbent utili- layer around the burning particle, where they zation are the sintering of pores at hightem- subsequently nucleate to form a submicron peratures near the flame zone, the low diffusivity aerosol. of sulfur dioxide through the layer of calcium There is a general understanding that the size sulfate that forms on grains of the calcium oxide of ash particles produced during coal combus- sorbent, and pore plt.gging. There is opportunity tion decreases with decreasing coal particle size for major innovation in the design of sorbents and with decreasing mineral content of the for sulfur capture in combustors by tailoring parent coal particles. There are, however, no thL a- physical and chemical properties. The key fundamental models that allow the researchers characteristics of an ideal sorbentare large to predict the change in the size of ash particles surface area, mechanical strength, and fast and when coal is finely ground or beneficiatedor how complete utilization. Used sorbent should be ash size is affected by combustion conditions. regenerable or usable as a by-product.

12 7 118 FRONTIERS IX CHEMICAL ENGINEERING

is complicated by the strong link between flame propagation and the configuration of and ventilation in different enclosures, with high-rise Cyclone atriums in buildings being of special concern. Particle Separator Examples of the pressing problems to which the chemical engineer can contribute follow. 2 At present there is no small-scale test for Combustor predicting whether or how fast a fire will spread on a wall made of flammable or semiflammable (fire-retardant) material. The principal elements of the problem include pyrolysis of solids; char- External layer buildup; buoyant, convective, turbulent- Heat Exchanger boundary-layer heat transfer; soot formation in the flame; radiative emission from the sooty Steem flame; and the transient nature of the process Coe41 /Treated (char buildup, fuel burnout, preheating of areas Nftedwater not yet ignited). Efforts are needed to develop computer models for these effects and to de- 1 velop appropriate small-scale tests. Fluidized Bed Most fire deaths are caused by smoke inha- Crushed lation rather than by burns. Buildings now Limestone contain many synthetic polymeric materials that Air can burn to yield such toxic compounds as hydrogen cyanide and hydrogen chloride in addition to common combustion products such FIGURE 7.13Fluidized bed combustion. In fluidized bed combustion, air is blown into a bed -f burning coal, as carbon monoxide. For this reason, consid- limestone, ash, and gravel, swirling the mixture like a fluid era:ion is being given to banning certain mate- and greatly improving combustion (I). The hot stream of rials, at least in public buildings. Realistic hazard gases generated carries small particles up the combustor analyses for materials would be facilitated by (2). The particles of ash, gypsum, anu limestone (called computer models that could interrelate such "sand") in the gas stream are separated and collected by significant factors as the identity and amount of large cyclone separators (3). The hot gases (4) are ducted to preheat boiler water. The hot sand transfers heat to toxic products formed by combustion of these boiler tubes (5), generating steam. Cooled sand particles materials, the rate at which the materials burn, (6) are recycled to the combustor. Courtesy. E.I du Pont and the ease of ignition and smoke-forming de Nemours and Company. Inc. tendency of the materials. Furthermore, as combustion products are transported away from the flame (e.g., down a corridor), smoke parti- Fires and Explosions cles agglomerate and hydrogen chloride under- Fires and explosions cause major property goes mass transfer to and adsorption on a variety loss within the chemical process industry; more of surfaces. Any interdisciplinary effort to un- significantly, they account for an annual loss in derstand the hazards of ;fires involving synthetic this country of thousa1.ds of lives and the materials would benefit from chemical engi- destruction of billions of dollars of property. neering research expertise in reaction modeling, The chemical engineer can contribute to the chemical kinetics, and heat and mass transfer. solution of the overall fire problem by providing A small fire in a computer room, a telephone means of estimating the flammability, flame exchange, or an assembly plant for communi- propagation rates, and products of incomplete cation satellites can cause enormous damage combustion for the increasing diversity of in- because of minute amounts of corrosion on dustrial and manufacturing materials, including circuit elements. Furthermore, if either water polymer and ceramic composites. The problem or a halogenated agent is used to control the

120 ENVIRONMENTAL PROTECTION, PROCESS SAFETY'. HAZARDOUS WASTES 119

complex compounds and to mixtures of the chemicals found in chemical plants is a challenging problem.

HAZARDOUS WASTE MAN 4GEMENT There are three distinct prob- Trurment I i 4 lems in hazardous waste man- Disposal i Dispersion agement: (1) reduction in the gen-

_Hamm/ sad atelsdel eration of waste, already recovery roc..... mentioned in the section "De- ..L...... sign of Inherently Safer and Less Polluting Plants and Processes"; --ilimirdmOdleallon (2) disposal of generated waste (Figure 7.14); and (3) remediation of old, abandoned waste dis- L __, posal sites. The problems of handling 2 id disposing of radioactive FIGURE 7.14Strategies for dealing with continued generation of hazardous waste. Management paths for hazardous waste include temporary storage, waste are largely the concern of treatment. disposal. and dispersion. Courtesy, Office of Technology Assess- nuclear engineers, often working ment. with chemical engineers to develop separation and encapsulation technologies for radioactive fire, the agent itself or its decomposition prod- nuclides, and are not discussed here. The most ucts may cause damage to sensitive materials basic way to deal with the continuing generation or devices. As automated and robotic systems of hazardous waste is to accumulate,encapsu- become more preva!ent in manufacturing plants, late, and store only as a temporarymeasure vulnerability of plants to small fires will in- and to develop new approaches to reduce the crease. Chemical engineering research relevant volumes generated and to concentrate hazard- to this challenge includes the development of ous components or convert them into nonha- sensors for more sophisticated fire detection, zardous materials. Abandoned waste sitesare the design and development of materials and remediated by cleaning them upor containing techniques for encapsulation of sensitive device them before they contaminate groundwatersup- elements, research on surfaces and interfaces plies. The establishment of priorities for site to facilitate more effective equipment salvage, cleanup and the development of appropriate and research to develop a better understanding detoxification technologies requirean under- of corrosion mechanisms so that optimal strat- standing of the processes by which the waste egies for fighting small fires can be developed. can migrate or be transformed in the natural One of the great hazards in a chemical plant environment. is the potential for any deflagration (a fire in The development of a fundamental under- which the flame front moves through thecom- standing of the behavior of toxic chemicals in bustible mix at subsonic velocities) to turn into atmospheric, soil, and aquatic environments a detonation, in which the flame front propagates (Figure 7.15) and of possible mechanisms for at supersonic velocities, generating a blast wave. destroying toxic chemicals has lagged far behind Considerable basic research has been conducted tit?. rapidly unfolding problems surrounding dis- to understand this transition in the combustion posal. Nevertheless, the ability of American of hydrogen and significant progress has been manufacturing industries to remain internation- made. Extrapolating this understanding to more ally competitive depends on this. Engineers in

12 120 t RO VIERSIN (IIEMICAL ENGINEERING

FIGURE 7.151 ransport and transformation of toxic chemicals in soil environments (left) and water environments (right). Courtesy. Office of Technology Assessment. industry did not anticipate that the problems compose. Heating methods include resistive associated with hazardous waste disposal would electrical heating, the use of radio frequencies emerge so rapidly or that destruction and dis- or microwaves, radiative heating, and the use posal processes would be so difficult to develop. of het combustion products as a heat source. A major effort must be mounted to conduct Heating can be done in the absence of oxygen, advanced research and to educate engineers to in which case the process is known as pyrolysis. solve the problems associated with the disposal Pyrolysis yields different products than does and environmental behavior of toxic chemicals. combustion of waste in the presence of oxygen. The most important current technique for the Detoxification of Currently Generated Waste thermal destruction of waste is incineration, where the energy required for destruction is Many technologies have been proposed for provided by oxidation of the waste, sometimes detoxifying waste by processes that destroy supplemented with a fossil fuel. The major chemical bonds: pyrolytic; N-slogical; and cat- questitm about all thermal destruction tech- alyzed and uncatalyzed reactions with oxygen, niques is whether products hem the process hydrogen, and ozone. The following sections either traces of unreacted parent compound or deal only with research opportunities in the compounds synthesized from the parent com- areas of thermal destruction, biodegradation, pound at high temperaturewill pose a health separation processes, and wet oxidation. hazard. Concerns have been expressed about incin- Thermal Destruction eration on land and in the water. EPA's Science Advisory Board, in a 1984 report entitled Incin- Most organic molecules are not stable at eration of Hazardous Liquid Waste, stated, temperatures above 400°C for any significant "The concept of destruction efficiency used by length of time Therefore, many processes for the EPA was found to be incomplete and not detoxifying waste 'will heat the toxic compounds useful for subsequent exposure assessments." to temperatures at which they will rapidly de- It was recommended that the emissions and 130 ENVIRONMENTAL PROTECTION, PROCESS SAFETY, HAZARDOUSWASTES 121 effluents of hazardous waste incinerators be performance in terms of the destruction of analyzed in such a way that the identity,quan- selected components of the anticipatedwaste tity, and physical characteristics of the chemi- stream. These compounds, labeled principal cals released into the environment could be org:nic hazardous components (POHCs),are estimated. The Intern..tional Maritime Organi- currently ranked on the basis of their difficulty zation Scientific Group on Ocean Dumping, of incineration and their concentration inthe convened in London in March 1985, was unable anticipated waste stream. The destruction effi- to reach consensus on the following questions: ciency is expressed in terms of elimination of the test species, with greater than 99.99percent What is the relationship between destruc- removal typically judged acceptable provided tion and combust:on efficiencies overa wide that toxic by-products are not generated in the range of operation conditions? process. What sampling procedures should be used The effectiveness of incineration hasmost to obtain a gas sample representative of the commonly been estimated from the heating entire stack? value of the fuel, a parameter that has littleto What methodology should be used for col- do with the rate or mechanism of destruction. lecting particulate matter in the stack? Alternative ways to assess the effectiveness of What new organic compounds can be syn- incineration destruction of various constituents thesized during the incineration process? of a hazardous waste stream have beenpro- posed, such as assessment methods basedon Factors that influence the destruction effi- the kinetics of thermal decomposition of the ciency in incineration include constituents or on the susceptibility of individual constituents to free-radical attack. Laboratory local temperatures and gas composition, studies of waste incineration have demonstrated residence time, that no single ranking procedure is appropriate extent of atomization of liquid wastes, for all incinerator conditions. For example, dispersion of solid wastes, acceptably low levels of some test compounds, fluctuations in the waste stream composi- such as methylene chloride, have proved diffi- tion and heating value, cult to achieve because these compoundsare combustion aerodynamics, and formed in the flame from other chemical species. turbulent mixing rates. Rather than focus on specific incineration technologies, one must address the fundamental Mere destruction of the original hazardous physical and chemical processescommon to material is not, however, an adequatemeasure many of the possible incineration systems through of the performance of an incinerator. Products studies of (I) reaction kinetics of selectedwaste of incomplete combustion can beas toxic as, materials and (2) behavior of waste solutions, or even more toxic than, the materials from slurries, and solids in the incineration environ- which they evolve. Indeed, highly mutagenic ment. PAHs are readily generated along with soot in The combustion chemistry of methane and fuel-rich regions of most hydrocarbon flames. C-2 hydrocarbons is reasonably well under- Formation of dioxins in the combustion of stood. Progress is being made in addressing the chlorinated hydrocarbons has also beenre- pyrosynthesis reactions that lead to the for- ported. We need to understand the entire se- mation of toxic PAHs. Much of the literature quence of reactions involved in incineration in on combustion, though, is devoted to the flame order to assess the effectiveness and risks of zone, where heat release rates and free-radical hazardous waste incineration. concentrations are high. A key problem in The routine monitoring of every hazardous incineration chemistry involves understanding constituent of the effluent gases of operatinr the late stages of degradation of waste materials, incinerators is not now possible. EPA hases- where temperatures and free-radicalconcentra- tablished procedures to characterize incinerator tions are lower than in the flamezone. More- 131 122 FRONTIERS IN ('III 1N('.11. ENGINTERING

over, it has long been known that the introduc- Biodegradation tion of halogen atoms into -. flame can interfere with the combustion process by removing free Recent developments in molecular biology, radicals. The effect of such reactions on the particularly recombinant DNA technology, of- incineration of hazardous substances containing fer many new concepts for hazardous waste halogen atoms needs to be determined. We are treatment that were unthinkable a decade ago. concerned both with the destruction of the For example, the insertion of foreign genes from original compounds and with the production of one microorganism into another has become trace quantities of other hazardous species dur- relatively routine. Progress in achieving high ing the reaction. Thermal pyrolysis and reac- expression of a foreign inserted gene has also tions of the waste with common radicals such been impressive. A combination of molecular as OH, 0, H, and the halogens are most im- biology and chemical engineering could lead to portant 'n the flame environment. Both r.:ducing the design of new processes for waste treatment. and oxidizing atmospheres are encountered in The controlled use of biological systems or turbulent diffusion flames; therefore, an under- their products to bring about chemical or phys- standing is needed of the chemistry over the ical change is particularly attractive when deal- entire range of combustimi stoichiometries. ing with dilute waste streams. Biological sys- Studies of the incineration of liquid and solid tems thrive in dilute aqueous media, where they wastes must e'etermine the rates at which haz- can effectively degrade organic pollutants, ab- ardous comps nds are released into the vapor sorb heavy metal ions, or change the valence phase or are transformed in the condensed state of heavy metal ions (Table 7.2). Micro- phase, particularly when the hazardous mate- organisms and biological agents etcarry out rials make up a small fraction of the liquid reactions with great chemical specificity and burned. We must be particularly concerned with efficiency, and genetic engineering provides understanding the effects of the major compo- means for developing strains of organisms and sition and property variations that might be classes of enzymes with nearly unlimited ca- encountered in waste incinerator operations pabilities for effecting desired chemical changes. for example, fluctuations in heating value and In addition, many microbial systems have high water content, as well as phase separations. affinity for metal ions, and meta! ions are often Evidence of the importance of variations in moved from an aqueous solution into the cell waste properties on incinerator performance through active transport. Accordingly, such has been demonstrated by the observation of reactions as the biological reduction of a heavy major surges in emissions from rotary-kiln in- metal ion can be carried out at relatively fast cinerators as a consequence of the rapid release rates, although at millimolar concentrations. of volatiles during the feeding of unstable ma- There are significant research opportunities terials into the incinerator. for the mical engineers in the design and opti-

TABLE 7.2Examples of Biodegradation for Waste Management Major Contaminants Industry Effluent Stream Removed by Biodegradation Steel Coke-oven gas scrut bing operation Ammonia, sulfides, cyanides, phenols Petroleum refining Pnmary distillation process Sludges containing hydrocarbons Organic chemical Intermediate organic chemicals and Phenols, halogenated hydrocarbons, manufacture by-products polymers, tars, cyanide, sulfated hydrocarbons, ammonium compounds Pharmaceutical Recovery and punfication solvent Alcohols, ketones, benzene, xylene, manufacture streams toluene, organic residues Pulp and paper Washing operations Phenols, arganic sulfur compounds, oils, lignins, cellulose Textile Wash waters, dep discharges Dyes, surfactants, solvents

SOURCE: Office of Technology Assessment."

1 ri r>hr ENVIRONMENTAL PROTECTION, PROCESS SAFET}, IIAZARDOC.i tt ASI ES 123

mization of bioreictors for dilute waste stream ecosystem interactions, treatment, including the design of efficient con- exposure pathways, and tactors, the use of immobilized cells in reactors, probable short- and long-term impacts. and the elucidation of mass transport processes and reaction kinetics. Related research oppor- A substantial base of scientific information, tunities for chemical engineers include the for- monitoring methods, and predictive models is mulation of biocatalysts, the deve.4ment of required. Chemical engineers can assist biolo- bioseparations, and the use of chemical engi- gists in developing this base. For example, the neering expertise in process control and opti- chemical engineering tools used to analyze mization to better understand the behavior of chemical processes in industrial reactors would large microbial populations. be useful in analyzing a situation where genet- The promise of biological treatment of heavy ically engineered microorganisms were released metal ions has already been illustrated by strains into an underground waste site, with the earth of microorganisms that tolerate mercury, chro- itself being the chemical reactor. mium, and nickel--heavy metals that are generally toxic to microorganisms. Tolerant microorganisms have been isolated through classical Separation Processes adaptation and strain-selection studies rather A large fraction of the hazardous wastegen- than by recombinant DNA techniques. Mer- erated in industry is in the form of diluteaqueous cury-tolerant microorganisms have been shown solutions. The special challenges of separation to possess an enzyme that is not present in in highly dilute solutions may be met by the nonresistant strains. This enzyme, mercuric development of new, possiblyliquid-filled, reductase, is able to catalyze the reduction of membranes; by processes involving selective mercury(II) ions to metallic mercury. Since the concentration of toxic chemicals on the surfaces mercury(II) ion is the toxic species and the of particles; or by the use of reversed micelles. insoluble metal is chemically inactive, the mi- Liquid-filled,porous,hollow-fiber mem- croorganism is able to detoxify a solution that branes hold promise for improving the efficiency contains mercury(II) ions. and economics of extraction processes. Incon- Microorganisms hold tremendous promise for ventional liquid-liquid extraction, process de- improvements in the treatment of hazardous sign, hardware, and economics are dictated waste, but genetically altered microorganisms primarily by the relative densities of the two present both regulatory bodies and industry the liquid phases. Energy must be expendedto complex task of identifying, managing, and create a large surface area for diffusion to take controlling their use. While organisms that have place, while contact time may be shorter than been highly modified by classical strain selection desired because of high relative velocities of have been used safely in industry for years, the two phases. Hollow fibers containingpores there is a critical lack of data to either support filled with the extracting agent permit the waste or to allay concerns about the release into the and stripping fluids to flow in a countercurrent open environment of organisms that have been fashion on opposite sides of the membrane. In modified by recombinant DNA techniques. For this way, a high interfacial area can be main- example, to understand the potential conse- tained, regardless of the relative flow rate of quences of the release of genetically engineered fluid to extracting agent. In addition, theex- organisms, it is also necessary to know tracting agent can be renewed by continuous desorption into the stripping fluid. This tech- alternative approaches to meeting the need, nique is but one example of many newprocesses alternative systems for handling the orga- evolving in the field of membrane separations. nism, Another process for the separation of toxic how much material will be involved, chemicals from waste streams species involves how the organisms will move or be trans- adsorption from solution onto particles, fol- norted, lowed by sedimentation to remove the toxic- what chemical substances will be produced, laden particles. Solutes bound to the surface of

133 124 FRONTIERS IX CHEMICAL ENGINEERING

aqueous particles may participate in oxidation- acterization of reversed micelles by small-angle reduction reactions with the particles, undergo neutron and x-ray scattering. chemical transformations in whia the particle surface serves as a catalyst, or participate in heterogeneous photochemical processes. The Wet Oxidation design of effective engineering processes for the Incineration achieves high destruction effi- treatment of water supplies to remove toxic ciencies by fast free-radical reactions in the compounds uy adsorption/reaction/particle-re- presence of water vapor at high temperatures moval sequences demands fundamental data on (1,500-2,300°C)and 1 atm. Waste can also be the kinetics of the individual steps and the destroyed by oxidation at much lower temper- incorporation of the data into process models. atures by operating at high pressures, including A major challenge is to describe all relevant conditions above the critical point for water. chemical influences on the efficiency of removal For example, high destruction efficiencies (greater of specific toxic compounds. Among these are than 99.99 percent) of toxic organic compounds the physical and chemical properties of the can be achieved in2seconds at moderate absorbing particle surface, the alteration of temperatures(1,000-1,200°C)at250atm. The these properties by reactions or dissolution- lower reaction temperature permits the destruc- precipitation processes, and the stability of tion of waste of much lower heating value than aqueous particles to coagulation. In contrast to can be incinerated, at least without the use of the chemical conditions of conventional munic- auxiliary fuels. The chemistry of such reactions ipal water and wastewater processing, the con- at high pressures and moderate temperatures ditions selected or imposed by the special cir- needs to be further elucidated before wet oxi- cumstances for control of hazardous substances dation processes can be more widely used in may include extremes of pH, redox potential, hazardous waste management. ionic content, and organic content. These factors may become critical in the design of optimal processes combining adsorption, reaction, and Remediation of Toxic Waste Sites coagulation steps. Only two processes, high-temperature pyrol- Recent development of the use of reversed ysis and mobile incineration, have proved ef- micelles (aqueous surfactant aggregates in or- fective for soil decontamination and are consid- ganic solvents) to solubilize significant quan- ered to be commercially viable. Both involve tities of nonpolar materials within their polar heating the contaminated soil to a high temper- cores can be exploited in the development of ature, which is costly in terms of energy use new concepts for the continuous selective con- and materials handling. There are substantial centration and recovery of heavy metal ions opportunities fcr innovation and development from dilute aqueous streams. The ability of of processes for the separation of contaminants reversed micelle solutions to extract proteins from soils a--d the in-situ treatment of contam- and amino acids selectively from aqueous media inated soils. Examples of each are given in the has been recently demonstrated; the results following subsections. indicate that strong electrostatic interactions are the primary basis for selectivity. The high charge-to-surface ratio of the valuable heavy Separation Processes metal ions suggests that they too should be One generic problem in site remediation is extractaule from dilute aqueous solutions. the removal or deactivation of small quantities The potential of reversed micelles needs to of toxic organics from highly porous and sur- be evaluated by theoretical analysis of the metal face-active media such as soil. Alternative pro- ion distribution within micelles, by evaluation cesses to pyrolysis and high-temperature oxi- of the free energy of the solvated ions in the dation of soil, such as thermal desorption, steam reversed micelle organic solution and the bulk stripping, and supercritical extraztion, require aqueous water, and by the experimental char- less energy and thus should be investigated ENVIRONMENTAL PROTECTION, PROCESS SAFETY, HAZARDOUSWASTES 125

further. Fundamental research on the nature of piing strategies responsive to this needare the adsorbed state of organics in soil could have discussed in the following section. as a significant payoff the identification of alternative process paths. Basic measurements of desorption kinetics and pore diffusion in clas- BEHAVIOR OF EFFLUENTS iN THE sified fractions of soil components (e.g., clays, ENVIRONMENT silts, and sand) can provide the basis for de- It has been recognized for some time that veloping accurate models of such processesas fluids in motion, such as the atmosphereor the soil desorption and migration of contaminated ocean, disperse added materials. This property plumes. This information could be used to has been exploited by engineers ina variety of determine the conditions necessary for thermal ways, such as the use of smoke stacks for boiler desorption and steam stripping. furnaces and ocean outfalls for the release of Extraction with supercritical fluids, suchas treated wastewaters.Itis now known that carbon dioxide or methanol in carbon dioxide, dilution is seldom the solution toan environ- offers the potential for combining the highmass- mental problem; the dispersed pollutantsmay transfer coefficients of gases with the moder- accumulate to undesirable levels in certain niches ately high absorption capacities of liquid sol- in an ecosystem, be transformed by biological vents. In addition, the solubility characteristics and photochemical processes to other pollut- are highly sensitive to relatively small changes ants, or have unanticipated health or ecological in temperature and pressure. Thus, thecontam- effects even at highly dilute concentrations. It inants can be recovered from the supercritical is therefore necessary to understand thetrans- fluid after extraction and the supercritical fluid port and transformation of chemicals in the recycled at moderate cost. The method is being natural environment and through the trophic applied to tertiary oil recovery by the petroleum chain culminating in man. industry, a process somewhat akin to there- moval of organics from soil. Fundamentalre- search on the solubilities of organic compounds The Atmospheric Environment in supercritical fluids would expedite the eval- Over the last two decades, significantprogress uation and application of this promising tech- has been made in understanding the mechanisms nology. of transport and transformation of pollutants in the atmosphere. :lathematical models have been Biodegradation developed to describe the spatial and temporal distributions of sulfur dioxide, carbonmonox- The use of biodegradation for the treatment ide, nitrogen oxides, hydrocarbons, andozone. of dilute waste streams has already been dis- These models now serve av the backbone for cussed; it also has potentia! for in-situ treatment. the developmei t of state plIns for implementing The critical need is to learn how to select and the 1977 Clean Air Act amendments. Region- control microorganisms in a soil environment wide air pollution and acid rainare current to achieve the desired degradation of organics. subjects of intensive mathematical modeling efforts. But in spite of the strides that have Monitoring been taken, a number of important research problems remain in understanding the behavior One of the most important elements in the of atmospheric contaminants. remediation of existing waste sitesis early Organic compounds constitute about 25-30 detection and action. As an example, the cost percent of the fine aerosol mass (themass of cleanup at Stringfellow, California, increased contained in particles smaller than 2.5p,m di- from an estimated $3.4 million to $65 million ameter) in urban areas. They are of considerable because of pollutant dispersal during a decade interest because some of them, suc:1as PAHs, of inaction after the first identification of the are either suspected carcinogens or known mu- problem. The opportunities for innovativesam- tagens.Still,little headway has been made

135 126 FRONI MRS IN (11E.111C41. ENGIAEERING toward engineering their systematic reduction their time indoors, the quality of the indoor in the atmosphere. atmospheric environment is now receiving greater The problem is complex because many dif- attention from researchers and regulators. There ferent sources contribute to atmospheric load- has been a reported increase in both the con- ings of organic compounds. Not only do toxic centration and diversity of pollutants in indoor waste incinerators have to be considered, but environments; formaldehyde, nitrogen dioxide, so do more than 50 classes of mobile and carbon monoxide, and a diverse range of organic stationary combustion sources and industrial compounds have been identified. It is not certain processes that release small amounts of toxic whether this should be attributed to the use of organics mixed with other exhausts. In addition, new building materials and to changes in build- reliable aerosol source samples of PAHs and ing ventilation resulting from increased insula- their oxygenated or nitrated derivatives are tion or to the use of more sophisticated analyt- difficult to collect because these compounds are ical techniques. Since the principal indoor air present in both gas and aerosol phases. Special pollutants are known or suspected to adversely stack-sampling equipment must be designed to affect health (Figure 7.16), there is a need to acquire meaningful samples. Emissions undergo engineer systems that can reduce their genera- transport and chemical transformation in the tion. Chemical engineers can assist in devel- atmosphere. For example, mutagenic nitro com- oping such systems, including pounds can be created by the reaction of PAHs with HNO3, NO2, or N205. One way to analyze home heating and cooking burners that these atmospheric transformations is to com- minimize the generation of oxides of nitrogen; pare the chemical composition of primary source improved heat transfer devices that will effluents with that of ambient aerosol samples. allow for air exchange with the outside envi- However, source and ambient samplings now ronment while avoiding excessive loss of heat; vary in methodology and analysis so that dif- and ferences between them may be due to laboratory resins, binders, coatings, and glues for procedures. A comprehensive study, in which building materials that do not emit hazardous source and ambient measurements are made compounds, such as formaldehyde. and analyzed the same ways, is needed. Source emission data could then be correlated with Finally, there is a need for simple instrumen- atmospheric transport calculations, and the rel- tation that can be used to quantify occupant ative importance of source contributions to exposure to air pollution. ambient organics could be identified. When spills and releases of hazardous gases The Aquatic and Soil Environments or liquids occur, the concentration of the hazardous material in the vicinity of the release is Disturbingly little is known about the mech- often the greatest concern, since potential health anisms of groundwater contamination, including effects on those nearby will be determined by not only those for transport and dispersion but the concentration of the substance at the time also those for chemical transformation. Under- of the acute exposure. There are many models ground pollutant transport is often represented of routine continuous discharges (e.g.,dis- with rather simplistic plume models, in much charges arising from leaky valves in chemical the same way as traditionally done for the plants), but these cannot be applied to single atmosphere. These models do not take into episodic events. Research on the ambient be- account the fact that, for the underground trans- havior of short-term environmental releases and port process alone, the detailed mechanisms of the development of models for concentration flow through inhomogeneous porous media rep- profiles in episodic releases are crucial if we are resent a major source of added complexity that to plan appropriate safety and abatement mea- cannot be ignored (Figure 7.17). Chemical en- sures. gineering expertise in petroleum reservoir mod- Because most people spend the majority of eling can be applied to this area.

13C ENVIRONMENTAL. PROTECTION, PROCFCS 8,1FEll, II 1Z 1R001 'S Vi 1.SIE S 127

SINUSES NASAL CAVITY

FORMALDEHYDE

ORAL CAVITY TRACHEA NITROGEN DIOXIDE BRONCHUS AMMONIA, SULFUR DIOXIDE P,',RTICLES UNDER 3 MICRONS

CARBON MONOXIDE NITROGEN DIOXIDE (FROM ALVEOLI INTO BLOODSTREAM)

CARBON DIOXIDE (FROM BLOODSTREAM INTO ALVEOLI)

ALVEOLI FIGURE 7.16Health effects of indoor air pollution Gaseous and particulate contaminants frequently found in indoor air pollution affect different parts of the respiratory system Some. 'itch as carbon monoxide and nitrogen dioxide, move from the lungs into the bloodstream

Models of chemical reactions of trace pollut- Considerable work has been done on the ants in groundwater must be based on experi- behavior of pollutant species at air-water and mental analysis of the kinetics of possible pol- air-soil interfaces. For example, wet and dry lutant interactions with earth materials, much deposition measurements of variousgaseous the same as smog chamber studies considered and -%articulate species have been made overa atmospheric photochemistry. Fundamental re- wide range of atmospheric and land-covercon- search could determine the -urface chemistry ditions. Still, the problem is of such complexity of soil components and processes such as ad- thatspecies-dependent andparticle-size-desorption and desorption, pore diffusion, and pendent rates of transfer from the atmosphere biodegradation of contaminants. Hydrodynamic to water and soil surfaces are not completely pollutant transport models should be upgraded understood. There is int...11 to be learned about to take into account chemical reactions at sur- pollutant transfer at water-soil interfaces. Con, faces. cern about groundwater contamination by min-

_1 128 FRO,N7 /A (1fL,WC E,VGIATERING

Compliance point Aquifer

Waste constituent source Undetected fracture Aquifer Waste constituent source Ground water flow

Plume of contaminants

O Sampling wells, downgradlent Compliance Upgradlent well point

FIGURE 7.17Oversimplified and molz realistic views of plume migration in underground water. Plume migration is affected by inhomogeneities in the aquifer The diagram on the right shows that gravitational influences or fractures on the aquifer might cause plumes to flow in directions different from the direction of groundwater flow. Courtesy, Office of Technology Assessment.

eral processing leachates, by rainwater leaching requires rugged and reliable instrumentation of landfills, and by runoff of contaminated that can be transported and used in the field. surface water has heightened the opportunities Remote sensing technology should be explored for further work in this area. as a means of characterizing regional pollutant distributions. Ambient Monitoring The extent of groundwater contamination from landfills and storage tank leakage is often Advances in understanding the transport and unknown (Figur. /.18). It is important to devise fate of chemicals in the environment will depend measurement strategies to characterize the spa- on substantial improvements in measurement tial, chemical, and temporal nature of this prob- capabilities. Attention should be directed to- lem. Chemical engineers have been at the fore- ward instrumental techniques that can deter- front in using advanced mathematical tools and mine the oxidation state of inorganic species. instrumentation to characterize the size and which often has a marked influence on reactiv- extent of petroleum reservoirs. This technology ity,-nsport properties, and toxicity of the ion. should be transferred to the groundwater prob- r, Jicals and other highly reactive trace lem and, in particular, to the task of designing specks lay important roles in the chemistry of cost-effective sampling strategies. Other options the environment. Detection of these species include probing potential subsurface sources of

13C ENVIRONMENTAL PROTECTION, PROCESS SAT ETV, HAZARDOUS OA STLS 129

Multimedia Approach to Integrated Chemical Management Current laws and programs fo-

Fuel cus on the removal of pollutants Storage from the mediumair, water, or Tank landin which they are found, Sod Contaminated often with little regard for chem- By Residual Gasoline re421:7',r3"?/r ical management of the environ- .Aammulated ment as a whole. Because mod- Dianne, Gasoline/Water Inter/ace ern analytical techniques have Water Table revealed trace amounts of many Dissolved Hydrocarbons toxic chemicals throughout the environment, however, the me- FIGURE 7.18Leakage f:om underground storage tanks Some 2 million underground steel storage tanks are buned beneath service stations throughout dium-specific approach to pol- the country. Another 1.5 million steel gasoline tanks are buned on farms. lution control is now questiona- Thousands of "orphan" tanks are believed to have oeen left behind when ble." The diverse effects of acid service stations were razed for redevelopment. Leaks from these tanks could : in and of leachates from haz- allow hazardous organic compounds to migrate beneath the surface, polluting dous waste sites illustrate the sod and aquifers and releasing fum:s to the surface. Courtesy, E.I. du Pont de Nemours and Company, Inc. Nobility of chemicals in the en- virenment (Figure 7.19). The folio Ning list gives many areas pollutants by nondestructive methods such as of research opportunity in a multimedia ap- acoustic probing, eddy current techniques to proach to chemical management of the environ- assess tank corrosion, magnetometers for lo- meat: cation of buried drums, and electrical resistivity measurements. characterization of background levels of Improvements in the monitoring and opera- chemicals and geochemical cycles; tion of incinerators could minimize the acciden- basic kinetic studies (1' chemical degrada- tal r6ease of hazardous effluents. In particular, tion; fast-acting, continuous, on-line monitors are field experiments to measure pollutant fluxes needed to detect excursions in operating con- among media; ditions that could lead to toxic emissions. fundamental studies of interfacial dynam- Development of methodologies for charac- ics; terizing and measuring human exposure to better determination of Henry's Law con- chemicals is a challenging scientific and engi- stants for volatile species; neering undertaking. Data are needed for studies study of chemical speciation, including of risk assessment and health effects. During binding of organic compounds in soil and surface the past decade, rudimentary monitors have waters (Figure 7.20); become available to determine a person's ex- studies of sorption and desorption of heavy posure by measuring concentrations of a given n -tals on suspended organic matter; and pollutant in the air breathed. Efforts should be formation of precipitates in response to directed to lowering the cost and increasing the changes in pH. sensitivity, chemical selectivity, and accuracy of these monitors. Widesprevi ise of personal exposure monitors offers thi.itential for im- ASSESSMENT AND MANAGEMENT OF proving epidemiological studies and for devel- HEALTH, SAFETY, AND oping a more rigorous scientific basis for setting ENVIRONMENTAL RISKS standards in the workplace and the general Two of the biggest challenges facing chemical environment. engineering in the near future are (1) the iden- FRONTIERS IV CHEMICAL ENGINEERING

Prevailing Winds

Photochemistry

Cloud Processes Natural Sources \ NSW&

III

Marsh IndustryTransportation Aquatic Ecosystem Ocean Man-made Sources

Soil Minerals k Ground Water

FIGURE 7.19Mobility of acid raina multimedia problem. Reprinted from Opporntr.ttu'to Chennstrr. National Academy Press. 1985.

tification and evaluation of the lisksboth real and pen Avedto human health and to the Risk Assessment environment from exposure to chemical, (risk Risk assessment, an obvious precursor to risk assessment) and (2) the adequate control of management, first identifies a hazard and then these risks (risk management). The di/nate quantifies the likelihood of occurrence (hazard objective in meeting these challenges3 to en- assessment) and the impact (exposure assess- sure that risks in the :'hemical and processine, ment) associated with each hazard event. industries are viewed as accep:Phle by the public, regulatory bodie,,. and thr courts. while maintaining worldwide technological leadership Hazard Identification and Assessment and cost competitiveness in these industries Two main hazards associated with chemicals by capitalizing fully on advances in chem- are toxicity and flammability. Toxicity mea- istry, biotechnology, materials, and microelec- surements in model species and their interpre- tronics. tation are largely the province of life scientists. These challenges are critical k .'ne profession Chemical engineers can provide assistance in of chemical engineering, the chemical industry, helping life scientists extrapolate their results and our country. Risk assessment and manage- in the assessment of chemical hazards. Chemical ment involve input from a mu.titude of differ- engineers have the theoretical tools to make ent disciplines. The methodology is rapidly important contributions to modeling the trans- changing and extremely complex and requires port and transformation of chemical species in both technical input a id input from profes- the bodyfrom the entry of ecies into the sionals with expertise in legal, e^onomic. judi- body to their action at the ultimate site where cial, medical, regulatory, and pt "lic perception they exert their toxic effect. Chemical engineers issues. are also more likely than life scientists to ap-

140 ENVIRONMENTAL PROTECTION, PROCESS .SAFETY, HAZARDOUS HASTE'S 131 preciate realistic conditions and exposure scen- standing of the full spectrum of operations. It arios for the use of hazardous chemicals in is highly interdisciplinary and potentially re- industrial settings. Their assistance in interdis- quires manipulation of massive amounts of in- ciplinary efforts is needed to relate toxicity fulnation, some of which may be missing or of measurements to actual practice. uncertain validity. While the techniques and Identification of hazards of unexpected, epi- governing rules for risk assessment are gene' ally sodic events such as transportation accidents, straightforward, much creative work needs to equipment failure, fires, and explosions is largely be done before the methodology can be used the responsibility of the chemical engineer. The efficiently and effectively to anticipate and cor- most difficult problem in the quantification of rect safety problems or to analyze operating toxicity, explosion, and lire hazards of unex- abnormalities for precursors of danger. pected and episodic natures is the estimation of New techniques employing expert systems the probability of occurrence. Risk analysts for analysis of complex chemical processes can draw on both analysis and experience to gen- be used to anticipate safety problems associated erate sequences of component and subsystem with various design decisions. In many major failures that might lead to significant accidents. accidents, the relevant fundamental phenomena Methods such as fault-tree analysis can be used involved were totally unanticipated and were to display failure sequences in a logical format. understood only after considerable investiga- Mechanical failures, operator en ors, and man- tion. Some of these abnormal conditions might agement system deficiencies must all be consid- have been identified by a priori research guided ered in identifying and quantifying risks. by techniques for anticipating interactions that Risk assessment for complex systems involv- might be overlooked in conventional approaches ing hazardous chemicals requires deep under- to design. Other serious accidents were pri-:-

FIGURE 7.20 The binding properties of chemicals can affect their distnbution and retention in various environmental media. For example, the hypothetical environmental fates of two chemicals are contrasted. The profile on the left shcws the loading in various media over time for a chemical that binds strongly with lipid material. The profile on the nght shows the loading in vanous media o "er time for a chemical that binds strongly with organic material. Courtesy, Office of Technology Assessment.

1 4.;.

of - 132 FRONTIERS IN CHEMICAL ENGINEERING

ceded by abnormalities in operation that were society, itis becoming increasingly essential ignored until it was too late. Again, techniques that risk assessments be conducted and ade- for identifying and investigating such warning quately and carefully documented for all existing signals might have avoided disaster. industrial plants and transport systems that handle or store significant quantities of toxic or Exposure Assessment flammable chemicals. The same must be done insiting new facilities, selecting processes, There is a dearth of information and meth- designing processes and equipment, developing odology in the area of human and environmental operating and maintenance procedures, and de- exposure assessment. Standardized methodol- signing transport systems. ogies have not been developed; monitoring of Policies and procedures for risk management personnel exposure is rare; and suitably sensi- decisions must be established and be clear and tive, rapid-response, portable analytical equip- simple if the massive, but necessary, workload ment is limited. Fortunately, the exposure factor of risk assessment and management is not to is a controllable variable, at least theoretically, cripple the chemical industry's worldwide com- and is largely within the province of the chemical petitive position and consume inordinate re- engineer, who selects and designs processes, sources through inefficiency. sites and expands plants, and develops plant Managing chemical risk must proceed in the operating procedures and transport systems. To absence of perfect information on risks and how determine the likely dispersion of flammable to avoid them. The lack of critical information materials, as well as the transport of toxic or good alternatives is no excuse for inaction. chemicals in the environment, dispersion/reaction models (both near field and far field) for realistic accident scenarios (e.g., heavy gases, IMPLICATIONS OF RESEARCH dusts, aerosols) must be developed and verified FRONTIERS experimentally. This chapter has made clear the challenges to chemical engineers in research related to Risk Management environmental protection, process safety, and hazardous waste management. Chemical engi- Once the hazard and exposure assessments neering education must become strongly ori- are complete for any specific hazard,itis ented to these topics, as well. For example, relatively simple to determine how many people what might be characterized as the traditional will be affected and the severity of the effect approach to environmental concerns in process (i.e., the risk). It is considerably more difficult designestablishing the process and then pro- to decide whether these risks are warranted viding the necessary safety and environmental compared to the benefits. This is particularly controlsmust give way to a new approach true if the risks are uncertain, involuntary, or that considers at the earliest stages of des:gn not understood by those at risk; if those at risk such factors as are not primarily those who benefit; or if alternatives are unknown, uncertain, or impractical. process resilience to changes in inputs, The process is complex because the goals are minimization of toxic intermediates and multiple and frequently contraindicating. products, and Economic, liability, public image, and opinion safe response to upset conditions, startup, considerations are involved. Catastrophic haz- and shutdown. ards are less acceptable than smaller ones even if the absolute risk is identical. Voluntary risks As chemical engineering research develops new are a way of life for most people, but there is design and control tools to deal with these minimal tolerance for involuntary risks, partic- factors, these tools should be integrated into ularly if they are unknown or not understood. the curriculum. Process safety research is gen- In today's heavily regulated and litigious erally more advanced in industry than it is in

1 42 ENVIRONMEP FAL PROTECTION, PROCESS S41-E IT,HAZARDOUS WASTES 133

academia. Closer interaction between industrial of responding to unforeseen environmental researchers and academic researchers and ed- problems have certainly been great.Signifi- ucators is needed to disseminate insights and cantly increased support for funuarnentalre- knowledge gained by industry in this area. search is vital if universities are to preserve Other problems in environmental science and their role in long-term environmental research technologyas well as an introduction to the and in the education of tomorrow's researchers, social, economic, and political aspects of en- process designers, and regulators. vironmental issuesshould receive broad exposure in the curriculum. They should be in the content of existing courses wherever possible. NOTES ladustry has strong developmental research I. J. L. Makes.Natural Hazards Observer,10(3). programs in areas such as process safety, but January 1986, I. more fundamental research on process design 2. J. Mct.oughlin. "Risk and Legal Liability" in tools, on emerging environmental problems, or Deaiing with Risk,R. F. Griffiths. ed. New York: on general topics linked to public health and John Wiley8:.Sons, 1982. environmental protection requires stable, long- 3. National Safety Council.Accident Facts.Chi- term research support from the federal govern- cago: National Safety Council. 1985. ment. The chemical engineering profession stands 4. U.S. Department of Commerce. Bureauofthe ready to tackle these issues aggressively; does Census.Statistical Abstract of the United States: the federal government? 1987.107th ed. Washington, D.C.: U.S. Govern- The principal federal agencies that have sup- ment Printing Office, 1986 Table 697. ported environmental research generally have 5. One Hundred Largest LossesA Thirty-Year been the Environmental Protection Agency, the Review of Property Damage Losses in the Hy- National Science Foundation, the National drocarbon-Chemical Industries(OHL-9-86-7). Oceanic and Atmospheric Administration, the Chicago: Marsh and McLennan. 1986. National Institute of Environmental Health Sci- 6. McGraw-Hill Economics.Survey of Investment ences, and the Department of Energy. In recent in Employee Safety and Health,13th ed. New York: McGraw-Hill. 1985. years, many of these agencies have experienced 7. James L. Regens. The Regulatory Environment budget cuts that threaten their ability to maintain for Coal Development.- in vital research programs that anticipate environ- Costs of Coal Pol- lution Abatement: Results of an International mental problems, instead of reacting to the latest Symposium. Paris:Organisation for Economic crisis. Cutting back on basic research on envi- Cooperation and Development, 1983. ronmental problems is a false economy. Small 8.The National Survey of Hazardous Waste Gen- savings today on anticipatory research may erators and Treatment, Storage, and Disposal result in very large costs to society in the future, Facilities Regulated Under RCRA in/98/. since dealing with the consequences of envi- WESTAC. Inc.. 1984. ronmental problems is invariably more expen- 9. U.S. Congress, Congressional Budget Office. sive than research to anticipate and prevent Hazardous Waste Management: Recent Changes these problems. and Policy Alternatives.Washington. D.C.: Because of the critical importance of main- Congressional Budget Office, May 1985. taining our environmental quality and improving 10. U.S. Congress. Committee on Public Works and process safety and hazardous waste manage- Transportation. Subcommittee on Investigations ment, the committee recommends that these and Oversight.Hazardous Waste Contamination federal agencies undertake new initiatives in of Water Resources (Concerning Groundwater Contamination in Santa Clara Valley, chemical engineering research. The details of CA) (99- 321 Washington. D.C.: U.S. Government Print- proposed initiativ-s for EPA and NSF are spelled ing Office. out in more detail in Chapter 10 and Appendix II. U.S. Congress. Office of Technology Assess- A of this report. An investment in research to ment.SuperfundStrategy(OTA-ITE-252). anticipate and prevent environmental problems Washington. D.C.: U.S. Government Printing is likely to be highly cost-effective. The costs Office. 1985.

143 134 I ROA /MRSIN (11EtlICAL ENGINEER!VG

12. T. A. Kletz. Simpler, Cheaper Plants or Wealth 14. U.S. Congress. Office of Technology Assess- and Safety at WorkNotes on Inherently Safer ment. Technologies for Hazardous Waste Man- and Simpler Plants.London:Institution of agement. Washington. D.C.: U.S. Government Chemicai Engineers. 1984. Printing Office, 1981. 13. U.S. Environmental Protection Agency. Science 15. For a detailed discussion of this topic. see Y. Advisory Board. hu inei ation of Hazardou.s Liq- Cohen. ed. Pollutants in a Multimedia Environ- uid Waste. Washington. D.C.: U.S. Environ- ment. New York: Plenum Press. 1986. mental Protection Agency. 1984.

144 EIGHT Computer-Assisted Process and Control Engineering

Computers and computational methods have advanced to the point where they are beginning to have significant impact on the way in which chemical engineers can approach problems in design, control, and operations. The computer's ability to handle more complex mathematics and to permit the exhaustive solution of detailed models will allow chemical engineers to model process physics and chemistry from the molecular scale to the plant scale, to construct models that incorporate all relevant phenomena of a process, and to design, control, and optimize more on the basis of computed theoretical predictions and less on empiricism. This chapter explores the implications of these changes for process design, control, operations, and engineering information management. Future developments and opportunities in process sensors are also covered.

135

145 136 FRONTIERS IN CHEMICAL ENGINEERING

MAGINE A ROOM 50 feet long and 30 management of process information. Advanced feet wide, filled with cabinets contain- engineering development will be based more ing18,000 vacuum tubes intercon- than ever on mathematical modeling and sci- nected by miles of copper wire. This was the entific computation. Reliable modeling at the ENIAC (Figurc 8.1), a 30-ton behemoth that microscale, the individual pro,ess unit scale, was the world's first electronic computer. Now and the plant process scale will improve our imagine that ENIAC had been given the task ability to scale up processes in a few large steps, of solving a set of simultaneous linear equations possibly bypassing the need for a pilot plant that embodied 30,000 independent variables. and saving the 2 or 3 years required to build With a team of people v..arking around the clock and operate it. Process models capable of pre- to record intermediate solutions and feed them dicting dynamic behavior, operability, flexibil- back into the computer (ENIAC had a limited ity, and potential safety problems will permit amount of memory), ENIAC would still be these aspects of a process to be considered chugging away more than 40 years later to solve more fully earlier in the design stage. Because the problem! A supercomputer using modern improved computers can perform the extensive algorithms could solve the problem in an hour computations required by such models, it will or so. be possible to test alternative designs more The speed and capability of the modern com- quickly. puter have tremendous implications for the A chemical process must be designed to practice of chemical engineering. In the future, operate under a chosen set of conditions, each computer programs incorporating artificial in- of which must be controlled within specified telligence or expert systems will help engineers limits if the process is to operate reliably and design improved chemical processes more ef- yield a product of specified quality. Accurate, ficiently. Complex computations based on fun- complex, computer-solvable models of chemical damental engineering knowledge will allow en- processes will incorporate features of the con- gineers to design reactors that can virtually trols that are needed to maintain the desired eliminate undesired by-products, making pro- process conditions. Such models will be able to cesses less complex and less pol- lutinz. New sensors, many of which will be miniature analytical laboratories tied to miniature electronics, will allow rapid and accurate measurements for control that are currently impossible. New chemical products that today are discovered predomi- nantly through laboratory work for instance, reinforced plastics that are as strong as steel and weigh less than aluminum or drugs with miraculous propertiesmay be discovered in the future by computer calculations based on models that predict the detailed behavior of molecules. Chemical engineers will lead this revolution. They will need to be trained to use advanced FIGURE 11.1The Electronic Numerical Integrator and Calculator (ENIAC) computer techniques for process and As inventor, 3. Presper Eckert, circa 1946. ENIAC was the world's first design,processcontrol,and electronic computer. Courtesy, UNISYS Corporation.

146 COMPUTER-ASSISTED PROCESS AND CONTROL ENGINEERING 137 predict the effects of process excursions and guage, using both pictures and voice. This the control measures needed to correct them. setting, which will address the need fornew Computer management of the process operation chemical processes and products by harnessing will rapidly initiate control correction of process almost unimaginable computing power, will pro- excursions. The development of new types of vide significant new research opportunities in process sensors will be essential to this degree chemical engineering. of process control and will eliminate the time- consuming withdrawal of process stream samples for analysis. USING THE COMPUTER'S POTENTIAL The design of a commercial process can Each decade over the last 35 years has seen generate an almost uncountable number of pos- the processing speed of newly designed com- sible solutions for seemingly simple problems. puters increase by a factor of about 100 owing Even a decade's 10,000-fold increase in com- to advances in the design of electronic micro- putational capability in terms of faster computer circuits and other computer hardware. On top speeds and better algorithms does not permit a of this has been another 100-fold increase per person to search among these alternatives, nor decade in computer speed, thanks to more would such a search make strategic sense. Give efficient methods of carrying out computations engineers a design problem for which the best (algorithms). It is not widely appreciated that solution is obvious from today's technology, new algorithms have been as valuable as hard- and they will quickly write down the correct ware design in improving computer perform- solution without searching. If the solution is not ance. With the combination of improved com- obvious, they will often home in on V' e infor- puter hardware and better algorithms, effective mation needed and perform the computations computer speeds have more than doubled on that will expose the right solution quickly and average each year. with minimal effort. By using intuition and The availability of computing resources is experience, they eliminate the need for testing also increasing rapidly. The actual and projected every possible alternative. We need to under- availability of high-speed supercomputers is stand how to use computer technology in much shown in Table 8.1. The projection for 1990 the same way, to solve complex problems where at least 700 computers of Cray I classrepre- many of the decisions are based on qualitative sents 35 times more available computing power information and insights that develop as the than that available in 1980. While continued problem is attacked. substantial investment in supercomputers is Encoding this activity in the computer in- needed, support for better ways of using them volves a type of modeling in which the capa- should not be neglected. It is conceivable that bilities of the designer and his tools, the alter- a new algorithm could effectively increase the native procedures by which complex problem power of supercomputing for a specific problem solving can be performed, and effective methods by a factor of 35 overnight. During the last two of information management are all incorporated. dem:as, many developments in numerical anal- Advances in artificial intelligence, expert systems, and information management will revolutionize the automation of this activity. giving us computers that can display encyclopedic TABLE8.1Actual and Estimated recall of relevant information and nearly human Supercomputing Resources Available reasoning capabilities. A HAL 9000 of 2001: A to Researchers in the United States, Space Odyssey fame may indeed exist in the 1980-1990 future. Number of Cray I-Class Computer-generated visual information, for Year Supercomputers example, three-dimensional portrayal of pro- 1980 21 posed new designs, will be commonplace in the 1985 142 future. Communication will be ;n natural Ian- 1990 700-1,000

147' 11U 1 00 R(1,\ I ILRS ( E\61,\TERAG ysis have had a profound impact on scientific computation Clearly computer technology Process Design in the Twenty-First Century has improved rapidly; there is little reason to doubt that it will It Is April 200e. A process engineer fora medium-size chemical continue to do so. The problem oompwr; is designing a process for a new series of products her is now and will continue to be company Is deveippinwet talldne to her computer and pointing the lack of people trained to ap- to the screen; she Indicates. that the process needs a reactor and ply computer technology to sci- KIMseperetersote *Woe reCyde ;Th000lnixt-s4 entific and engineenng tasks. The usingwiArtM00011401#000trereikicHirnethernettal models, improvements suggested in determineebtollharelioltriteMPirahie000050***1014isnse Chapter 7 in terms of our ability the PeldeetYiekt* *441#0041000-WHieue*Ii, 16041!isunt todesign andcontrolbetter tyofpec chemical products and processes will be made by chemical engineers who understand comput- minutia, ersnot by computer scientists onaGROI, or by software engineers. The prooseadesfigit countries that unJerstand this during Itrifigi-minitti distinction will lea: the world in ThsittifflPhiu0,=9*-4-- the process chemical technology. flowsheet tvg.kiiii**. 441,91olfariulisnent and operating oasts: .that kt 'undertatri MATHEMATICAL MODELS because it had to use:, .data sirtrapotrded from OF FUNDAMENTAL measured valtle&Attainglnee0request; the centpueitehovis PHENOMENA several alternative gaitratieettaidid carisklArak _indicates their projected costs, and toga -why it eliminated rs* them. Some Chemical engineers have irz.- of the ilovishests were eliminated- Weise. of N. ceet, others ditionzily used mati..-matical because the/ were toxic** unsafe, others because the startup models to characterize the phys procedures would be difficult; and still others because they were Ical and chemical interactions oc- based on uncertain extrapolation of experimental data. curring in chemical processes. After perusal of these process options, the engineer asks the Many of these models either have computer to select five designs for further study, and the computer been entirely empirical or have produces a paper copy of the flowsheet and design parameters relied on crude approximations of the basic physics or chemistry of the process. This is because a typical chemical by the available methous for solving its equa- process comprises an assemblage of interacting tions flows, transports. and chemical reactions Ac- Before the advent of modern computer-aided curate analysis and prediction of the behavior mathematics, most mathematical models of real of such a complex system require detailed chemical processes were so idealized that they portrayal of the physics of tra-;sport and the had severely limited utilitybeing reduced to chemistry of reactions, which calls for complex one dimension and a few variables, or linearized, equations that do not yield to traditional math- or limited to simplified variability of parameters. ematics. Nonlinear partial differential and in- The increased availability of supercomputers tegral equations in two and sometimes three along with progress in computational mathe- spatial variable, must he solved for regions with matics and numerical functional analysis is rev- complicit -d shape, that often have at least some olutionizing the way in which chemical engi- free boundary The more accurate the model. neers approach the theory and engineering of the more mathematically complex it becomes, chemical processes The ,neans arc at hand to but it cannot he mote complex than allowed fur model process physics and chemistry from the

14u COMP1 TERAS SI S /ED PROCESS 1:1 D ('01 / ROE E.11,1 \URI NI, 139

models that more fully incorpo- ratethebasic chemistry and for each. She then tells the computer to prepare more rigorous physics of a system provides a designs from each of the five flowsheets. The computer qubtions mechanism for assessing under- the third design because it involves extensive extrapolation of standinl of fundamental phe- known thermodynamic data However, since the engineer feels nomena in a system by compar- that this may be we of the better designs, she asks the computer ing predictions made by the model to notify the laboratory management computer, which is connected with experimental data. to a Monday teehnHan's terminal, to begin experimental deter- Better models can replace mination of the mirong thermodynamic data so that the design laboratory or field tests that are can be prepared. The computer reminds the engineer that precise difficult or costly to perform or contra of process conditions will be essential for another of the identify crucial experiments that designs and that a new sensor may be necessary. The engineer should be carried out. In either consults the artificial intelligence program, which tells her that case, they will significantly en- either of two sensors being developed in another company de- hance the scope and productivity partment may IN her need. The engineer uses her computer to ofchemicalengineeringre- schedule a meeting with the manager of sensor development searchers in academia and in- TWO years- liter, the process Is producing one of the new dustry. products. -The tkne from design to operation was short enough In process design, it is fre- that the convoy was able to capture a new market, and it plans quently discovered that many of to *stead the business with related products of the series. The the basic data needed to under- design engineer has developed a real-time dynamic simulation of stand a prccess are lacking. Be- the process and is running it on the company's parallel supercom- cause most current mathematical puter to test strategies for producing the related products from the models are not sufficiently ac- process equipment. She queries another computer to search the curate to permit direct scale-up historical moords of plant operation. She wants to know how well of the process from laboratory the dynamic model mirrors the actual plant operation and when data to full plant size, a pilot the process parameters were last adjusted to give a better fit plant must be constructed. A, between the plant and the model. She feeds this information to models are improved, it may be- her computer, which devises a comprehensive plan for scheduling come possible to evaluate design production among four of the new products, with provision for decisions with more confidence, revising the schedule as inventories and sale of the products and bypass the pilot plant stage. change. A new business has been created. Process technologies for which the use of more comprehensive molecular scale to the plant scale; to construct mathematical models can result in major im- models that incorporate all relevant phenomena provements include those for biochemical re- of a process; and to design, control, and opti- action processing, high-performance polymers. mize more on the basis of computed theoretical plastics, composites, and ceramics; chemical predictions and less on empiricism. Chemical reaction processing (e.g.,reaction injection engineers, using advanced computational meth- molding, reaction coating. chemical vapor dep- ods and supercomputers, can now readily iden- osition); microelectronic circuits; optical fibers tify the important phenomena in a complex and disks; magnetic memory systems; high- chemical process over the entire range of ap- speed coating; photovoltaic and semiconductor plicable conditions by exhaustive solution of materials; coal gasification, enhanced petroleum detailed models. The benefits of investing in recovery; solution mining; and hazardous waste less empirical, more fundamental mathematical disposal. To date the most extensive use of models are becoming clear: supercomputer modeling has been in space age weapons technologies, where objectives, eco- The capability t) construct mathematical nomics, and time frames differ from those in

14, 140 FRONTIERS IN CIIE.VICAL ENGINEERING

thechemicalprocessindus- tries. It is clearly in the national interest to stimulate the more Hypercubes extensive use of advanced computational methods and super- One thing that has not Changed sine the earliest days of the computers in other industries computer age &s ells: Pe fastest available oomputer is never fast critical to our worldwide com- enough $0 391ye fantod; 4101001000blem ef the day. Because of petitive position. A program to the raoltptaitolcurterit computers simpafying assumptions about encourage the greater dissemi- motet iit*elitniliata, Odd° for theprobternto be 8oIved nation of advanced computa- by 8.1001h. ;"!Or otept* petroleum tional techniques and hardware rese*****.. , , Il computer powsfilictates how many will offer challenges and oppor- the mew* and how complex tunities to computational math- ematicians and numerical analysts, to engineering scientists, toapplicationsandsofty are expertsinfirmsthat develop and manufacture :,upercompu- gime ters, and, above all, to perceptive ofMligill0140k leaders in high-technology pro- vaitiblaltd.A. cess industries. Corripllefehttee` The followingsectionsde- -cnbe in more detail a number ance =004111etlit? of areas in chemical engineering nettsbOkLerahrleteOLMOt %be overall architecture of °amputate in which the ability to develop has not chanaiM-Sintlithe UM* andthe use offiatet components and apply detailed mathematical is a strategylhatrifnowsiaming bib cilmlnlaNng felons. New models should yield substantial =Muter aithitecturea,tesed orupsiallel processing a>f subele- rewards merge of the same problerii, w( be-needed to achieve significant advances in speed to solve the complex problems of today and Hydrodynamic Systems tomorrow. Much of the current computational modeling research in chemical engi- bounded and the solution must extend to infin- neering is concerned wth the behavior of flow- ity, as in atmospheric systems; and instill ing fluids. The general system of equat:ons that others, such as the flow of blood in vessels, the describe fluid mechanics, called the Navier- boundary is deformable. Solution of the Navier- Stokes equations, has been known for more Stokes equations for systems of technological than 100 years, but for complex phenomena the interest rema,ns an exceedingly challenging task; equations are exceedingly difficultto solve. supercomputers are needed to treat those sys- Only recently have methods been devised to tems that can he solved. treat such phenomena as shock waves and turbulence. Further difficulties anse when dis- parate temporal and spatial scales are present Polymer Processing and wh.., chemical reactions occur in the fluid. The development of polymers and polymer Solutions of the Navier-Stoke:, equations can composites will benefit greatly from the avail- be smooth and steadyor they can exhibit ability of better computers and better algo- regular oscillations or even chaos. In some cases rithms. The inherent properties of a polymer the fluid flow is enclosed by a rigid boundary are governed by the chemical structure of its with a complex shape, as in the extrusion of molecules, but the property's of a finished poly- polymers: in others the flow is effectively un- mer product are affected by the interactions

1 5 t) COMPI/TER-ASNM1 D PROCIAS D CONTROL EA (1/NEER/N6 141

physicists, petroleum engineers, ;Ind chemical engineers. As more :01*APSIVO 01 SUCK r anew epOacts is the hypercube arch'. sophisticated techn ,ues are de- lediee'rn ,abrit we Wald together as a team veloped for locating and recover- 'iik:**.iniegratee teeof cheap ingpetroleum,mathematical lbeinoleedyfilatloated modeling is playing an ever-ex- aticoolor can panding role. ifftlije*teni two both Once regions that may contain etieftAbimonicsty in petroleum are located, local geological features must be sought *ler pieces to run that might have trapped the hy- :!*****0011sIblIV for drocarbons. The discovery pro- -40)mtelile, cess is based on a 'and of seismic f"filifred prospecting in which geologic aCoMplex maps are constructed from re- piece flected seismic signals generated other by explosions or vibrations at *AtellieeMeombiors can the surface of the earth. These 4ionfors of the signals are reflected or refracted in varying oevees by different sys- rock strata and are recorded by tem imtwodeltepreaeot exciting a set of receivers. Thus, the prob- newt of applications lem of interpreting signals can be cerdritio the donMereidttheilklet140.11rtfle4."Old dYnan'tcs likened to that of analyzing light and let flaw)bent Sk00400004.7'dkileited to tun on these beams reflected by an array of sin g, The new coineiller duke* promise* move the variously curved plates of glass field of chemical ennineering`substentially away from its depend- of different reflectivities sepa- ence on -simpre models toward- computer simulations and rated by liquids of dif..-rent re- calculations that more closely represent the Incredible complexity fractive indexes. The inverse of the ro world. mathematical problem of determining the earth structure and the rroperties of the strata from among these molecules, which are strongly influ- the recorded signals is extremeb, difficult. enced by the qv in which the material has After a hydrocarbon reseryr has been lo- been processed. While itis now possible to cated, the flow of oil, water, gas, --id possibly predict certain properties of polymers from their injected chemicals in the reservoir must be molecular structure, the ability to predict the modeled. This challenge is particularlyappro- effect of polymer processing steps on polymer priate for chemical engineers working withpe- properties is just being developed. Ideally it troleum engineers because of the important role would be desirable to model all steps from the played by molecular level interactions between formation of the polymer through its processing oil, subsurfaceater, and rock. Models for fluid and then predict the final properties of tne flow in porous media comprise coupledsystems material from Aructure-property relationships of nonlinear partial differential equations for Although such modeling is a formidable prob- conservation of mass and energy, equations of lem, is becoming feasible with the of state, and other constraining relationships. These supercomputers and improve4 algoriti, .is moiels are usually defined on irregular domains with complex boundary conditions. Their Petrol'urn Production numerical solution, with attendant d icuioes such Computation is widely used in petroleum as -,!-ioice of discret'ition methous and grid or- exploration and production by explorationgeo- ..;ntation ,is a ch, engmg intellectual problem

I 142 FRONTIERS IN CHEMICAL ENGINEERING

Once wells have been drilled into the for- into the important chemical and physical phe- mation, the local properties of the reservoir nomena found in combustors. Chemical engi- rocks and fluids can be determined. To construct neers have the mix of expertise necessary to a realistic model of the reservoir, its properties accomplish this. over its total extentnot just at the well sites must be known. One way of estimating these Environmental Systems properties is to match production histories at the wells with those predicted by the reservoir The environment can be likened to a giant model. This is a classic ill-posed Myers.; prob- chemical reactor. Gases and particles are emit- lem that is very difficult to solve. ted it .o the atmosphere by industrial and other When or if the reservoir is successfully sim- man-made processes, as well as by a variety of ulated, the engineer can turn to optimizing natural processes such as photosynthesis, vul- petroleum recovery, and theoretical ideas can canism, wildfires, and decay processes. hese be applied to models for various enhanced gases and particles can undergo chemical re- recovery methods to select optimal procedures actions, and they or their reaction products can and schedules (see Chapter 7). be transported by the wind, mixed by atmospheric turbulence, and absorbed by water droplets. Ultimately, they either remain in the at- Combustion Systems mosphere indefinitely or reach the earth's surface. Combustion is one of the oldest and most For example, the hazes of polluted atmospheres basic chemical processes (Plate 6, Its accurate consist of submicron aerosols of inorganic and mathematical modeling can help avoid explo- organic compounds, which are formed by chem- sions and catastrophic fires, promote more ef- ical reaction, homogeneous nucleation, or con- ficient fuel use, minimize pollutant formation, densation of gases (Plate 7). and design systems for the incineration of toxic Models of atmospheric phenomena are similar mrterials (see Chapter 8). For example, mod- to those of combustion and involve the coupling eling the initiation and propagation of fires, of exceedingly complex chemistry and physics explosions, and detonations requires the abrity with three-dimensional hydrodynamics. The to model combustion phenomena. Models of distribution and transport of chem'cals intro- the internal combustion engine can shed light duced into groundwater also involve a coupling on the influence of combustion chamber shape of chemical reactions and transports through or valve and spark plug placement on engine porous solid media. The development of ground- performance. Models at the molecular level can water models is critical to understanding the provide a fundamental understanding of how effects of land disposal of toxic waste (see Awls are burned and how gaseous and particu- Chapter 7). late pollutants are formed. This can lead to ways to improve the design of combustion systems. PROCESS DESIGN Mathematical models of combustion must incorporate intricate fluid mechanics coupled The primary goal of process design is to with the kinetics of many chemical reactions identify the optimal equipment units, the optimal among a multitude of compounds and fre2 connections between them, and the optimal radicals. They must also consider that those conditions for operating them to deliver desired reactions are taking place in turbulent flows product yields at the lowest cost, using safe inside chambers with complex shapes. Because process paths, and with minimal adverse ampact complete models of real combustors, incorpo- to the environment. Design is a complex prob- rating accurate treatment of both fluid mechan- lem that involves not only the quantitative ics and chemistry, are still too large for present computing depicted in the previous section, but computers, the challengeio construct simpler, also the effective handling of massive amounts yet still valid, models by using critical insik,ht of information and qualitative reasoning.

1 5 4 COMPUTER-ASSP;TED PROCESS AN!) CONTROL ENGINEERING 143 Computer-Assisted Design of New Processes the identification of the most viable process Designs for new processes proceed through option in a relatively short amount of time. at least three stages: As the designer moves from conceptual design toward final design, he or she must analyzea Conceptual design- -,he generation of ideas number of alternatives for the final design. The for new processes (process synthesis) and their development of large, computer-aided design translation into an initial design. This stage programs (so-called process simulators) such as includes preliminary cost estimates to assess FLOWTRAN, PROCESS, DESIGN 2000, and the potential prottability of the process, as well ASPEN (or other equivalent programs used in as analyses of process safety and environmental various companies) has significantly automated co,lerations. the detailed computations needed to analyze Final designa rigorous set of design cal- these various process designs. The availability culations to specify all the significant details of of process simulators has probably been the a process. most important development in the design of Detailed designpreparation of engineer- petrochemical plants in the past 20 years, cutting ing drawings and equipment lists needed for design times drastically anti resulting in better construction. designed plants. Although the available simulators have done The key step in the conceptual design of a much to achieve superior design of petrochem- new chemical manufacturing process is gener- ical processes, there is considerableroom for ating the process flowsheet (Figure 8.2). All improvement. For example, better modelsare other elements of computer-aided design (e.g., needed for complex reactors and for solids process simulation, design of control systems, processing operations such as crystallization, and plantwide integration of processes) come filtration, and drying. Thermodynamic models into play after the flowsheet has been estab- are needed for polar compounds. Moreover, the lished. In current practice, the pressure to enter current process simulators are limited to steady- the market quickly often allows for the explo- state operations and are capable of analyzing ration of only a few of the process alternatives only isolated parts of a chemicai plant atany that should be considered. To be fair to today's iven time. This z ampartmentalization is due designers, it is possible to generate a very large to the limitations on computer memory that r'imber of alternative process paths at the prevailed when these programs were first de- conceptual stage of design, and yet experience veloped. This memory limitation resulted ina indicates that less than I percent of the ideas computational strategy that divided the plant for new designs become commercial. Thus, the into "boxes" and simulated static conditions challenge in computer-aided process synthesis within each box, iteratively merging the results is to develop systematic procedures for the to simulate the entire plant. With today'F su- generation and quick screening of many process percomputers, itis possible to simulate the alternatives. The goal is to simplify the synthe- dynamics of the entire chemical plant. This sis/analysis activity in conceptual design and opens the way for dramatic advances in mod- give the designer confidence that the initial eling and analysis of alternative process designs, universe of potential process paths contained because the chemical reactions that occur in all the pathways with reasonable chances for manufacturing ;.1 ocesses are usually nonlinear commercial success. The advances in computer- and interd.pendent, and random disturbances aided process synthesis that are possible over in the process -an propagate quickly and threaten the next decade are dramatic. They include both the operation of the entire plant. To nullify the an increasing level of sophistication (e.g., the effects of such disturbances, the designer must synthesis of heat exchanger networks, se- know the dynamics of the entire plant, so that quences of separation processes, networks of control failure in any one unit does not radiate reactors, and process control systems) and com- quickly to other units.Itis now within our putational procedures that should make possible reach to integrate this sophisticated level of

1`)0 4J Quit Design file TT_PL_FIGe

Handle Files Operations

Add Unit

Add Stream TEAR AST: P&L Figb Solution is 512,519,539 Solution is 512,535,540 or other equivalents

S8 S5

Si S4

S2 S3 S6 511 S12 11 14 S9 513 510 S16 515 514 S23 519 S27 S17 S24 ;13U- 525 526 -D-0 0 0 oS?8 v S22 541

S20 S36 518 S29 521 S39 Describe S34 S35 S3 S38 -DO o 0542 D S37 S40 S32

S33

Select Solvent 15 -;

FIGURE 8.2Example of a flowsheet generated by computer-aided pro ess on the left side of the figure shows options available in the particular design synthesis, as would be seen on the screen of a computer terminal. The column program being run Courtesy, Peter Piela, Cariegie-Mellon University. COMPUTER-ASSISTED PROC'ESS AND CONTROL ENGINEERING 145 design ,.nd analysis on a plantwide scale (in- size of units that are undertaken to improve cluding design and performance modeling of operating costs, plant flexibility,or safety. plantwide control systems) into the computational tools used to analyze and optimicthe performance of individual processes in the plant. Research Opportunities in Process Design In the detailed design stage fora chemical The overall goal of process design research manufacturing process, a chosen process &sign is to develop a systematic procedure, probabiy must be converted into a list of equipment items in the form of an interactivecomputer program, to be purchased and a set of blueprints to guide that contains design heuristics and interchange- their assembly. The design is presentedas a able approximate and rigorous models thatcan detailed process flow diagram (PFD), from which lead an engineer from an initial conceptto a is constructed a list of all needed items of final design as quickly as possible. The final equipment, and piping and instrumentation dia- design must include zonsiderations ofecc com- grams (PIDs) that show the equipment and its ics, controllability, safety, and envircnmental interconnections. The next task is to establish protection. We need to extend the conceptual the phys'cal layout for the entire plant. Ad- and final design procedures that have been vanced computer-based drafting tools aid in all developed for petrochemicalprocesses to pro- these activities cesses for producing polymers, biochemicals, and electronic devices. We also must develop systematic synthesis/analysis procedures for Computer-Assisted Process Retrofitting studying batch processes analogous to thepro- The preceding section focused largelyon the cedures that have been developed for studying design of new plants. However, theseproce- continuous petrochemical processes. dures can also be adapted to the retrofitting of There are come aspects ofprocess design in existing plants. Retrofitting is generally under- wNich decisions arc based primarilyon past taken to increase the capacity ofa plant; to experience rather than on quantitative perform- make use of a new technolog! suchas an ance models. Problems of this type include the improved catalyst, a new material ofconstruc- selection of construction materials, the selection tion, or a new unit operation; or to respond to of appropriate models for evaluating the phys- a significant change in the cost of energy or raw ical properties of homogeneous and heteroge- materials. A fair amount of retrofitting in the neous mixtures of components, and the selec- chemical process industries in recent years has tion of safety systems. Advances inexpert been undertaken to improve energy efficiency systems technology and information manage through plantwide energy integrationretrofit- ment will have a profound impact on expressiig ting of about 50 processes to incorporate modern the solutions to these problems. heat-exchanger network synthesis concepts has In summary, systematic procedures must be reduced energy requirements in the chemical developed for the following: industry by 30 to 50 percent. Retrofitting will generation of process alternatives; continue to play a major role in the design of quick screening of process alternativesus- chemical plants as new procedures forcom- ing both rule-of-thumb and short-cut calcula- puter-aided synthesis of separation systemsare tions; applied and research on process synthesis be- inclus n of controllability, safety, anden- gins to yield large savings by helping existing vironmental factors in the initial design; petrochemical plants produce the same mix of more detailed screening of a small number products through more economical chemical of promising design alternatives; reaction pathways. design of the process and managementof We need to develop a systematic approach its construction; to analyzing the impact of making changes in use and extension of expert systems con- the connections fetween process units or in the cepts to handle aspects of design that deal with

155 146 I. KONTIERS IN CHEMICAL ENGINEEREVG a mixtin : of qualitative and quantitative infor- contained in a process model, and measure- mation; and ments are used to evaluate the degree to which retrofitting of existing plants. the process conditions deviate from those of the model. When a mismatch occurs between PROCESS OPERATIONS AND CONTROL actual process conditions and those postulated by the model, a control and operating strategy Process operations and control have a tre- is invoked to correct the process conditions. A mendous impact on the profitability of a man- critical interrelationship exists between mea- ufacturing operation. In some cases, they can surements and operating/control strategy, one determine the economic viability of a manufac- that is too often neglected. The control strategy turing facility. For example, Du Pont's Process depends on what information is or can be Control Technology Panel has estimated that if available, even while it dictates which mea- Du Pont were to extend the degree of computer surements are needed. This is perhap., best process control that has been achieved at a few i:lustrated in a number of manufacturing proof its plants across the entire corporation, it cesses in cutting-edge technologies, where con- would save as much as half a billion dollars a trol and operating strategy is circumscribed by year in manufacturing costs. If Du Pont's num- the lack of appropriate sensors for may critical bers are representative, the entire chemical processvariables.Conventionalestimation industry could save billions of dollars each year techniques are used that infer the values of through more widespread application of the best unmeasured variables from measured variables, available process control. This could be the bu, these provide imperfect guidance for process single most effective step that the U.S. chemical control. process industry could t, ice to improve its global Even something as empirical as process mea- competitive position in manufacturing. surement cannot be divorced from the need fcr Why are such savings suddenly possible? good process modeling. In the absence of a Because of the explosive developments in com- good model, it may not be known what variables puter technology, research on operations and affect process operation or product quality and control is no longer constrained by lack of should therefore be measured or estimated. computing power. In particular, the traditional Process measurements are subject to errors. boundaries between design, control, optimiza- Random (stochastic) disturbances are ubiqui- tion, simulation, and operation are disappearing. tous, and gross or systematic errors can be Control is becoming a part of process design; caused by malfunctioning sensors or instru- simulation and optimization are becoming comments. The detection and elimination of these ponents of control design. errors are essential if the data are to be used Research opportunities in process operations for process operations and control. The success and control lie in three areas: of this screening depends on the measurements collection of information through process themselves, the failure data available, and the measurements- process control strategy. At the present time, interpretation and communication of infor- diagnostic programs are not applied to most mation by use of process models; and sensor failure data. The detection and remedia- utilization of intormation through control tion of significant errors in measurements for algorithms and control strategies for both nor- process control pose interesting research op- mal and abnormal operation. portunities.

Measurements Interpretation of Process Information The essence of process control is to take The quality of an operation and control strat- appropriate and quick corrective action based egy depends on the quality of the model on on measured information about the behavior of which itis based (Figure 8.3). We are only the process. The concept of the process is beginning to understand this relationship quan-

156 COMPUTER-ASSISTED PROCESS.4 A P CONTROL E. N(d.V.b.RI.N6 147

Alarm management also requires research. Modern chemical plants usually have audible and visual annunciators to warn operators when key variables deviate from acceptableor safe values. A process upset in a plant that has several interconnected Lnits with many feedback controls can set off multiple alarms, and the consequences of misinterpreting the alarms can range from inefficient process operation to outright d' :aster. When the alarm sounds, the operator must decide quickly what action to take. A hybridization of expert systems and process control systems can assist the operator in interpreting process status after an abnormal event. The need for better handling of abnormal events makes research is artificial in'elligence of great importance to the chemical industry.

Integration of Process Design with Control '. and in 1/10,000 of a second It can corrnound Most continuous plants are now designed for the process model's error 87,500 times'. steady-state operation with little regard for the FIGURE 8.3The importance of accurate process models. ease (or difficulty) with which the steady state Copynght 1988 by Sidney L. Harris. can be maintained tarough control. Such a plant can be difficult to control once it deviates from the steady state. titatively, even in the relatively simple context Design and control have traditionally been A the feedback control of linear systems. Even treated separately for the following i: it is assumed that the structure of the process reasons: model is correct, we do not yet know how to The problems in each area alone areex- translate uncertainties in model parameters into ceedingly complex. uncertainty in the performtAce of the control The interactions between design, control, system. A more difficult problem is to assess and optimization are poorly understood. the effect of an incorrect model structure, such The computational requirements of an in- as a wrong set of basic equations, on the tegrated approach to design and control have performance of the control strategy on which it been beyond the capability of available hard- is based. Understanding the effect of model- ware and software. process mismatch on control system perform- For example, a chemical plant might be de- ance provides a critical research opportunity. signed to achieve high efficiency by integrating In the context of operations and control, the operation of many individual process units simulations can be used to test new process across the plant (e.g., by using waste heat from strategies as well as to train operating personnel one unit as an energy source in another unit). to control the process and to respond to emer- However, the tight coupling of process units gencies. The increasing use of simulation in generally makes the entire plant more difficult process control requires that the cost of dynamic to control. Therefore, this is a factor that must simulation be brought down. This could be done be considered at the design stage. No method- by taking advantage of new developments in ology currently exists for including this consid- computers, such as new user interfaces, com- eration in plant design; its development consti- puter architectures, and languages, and by de- tutes a significant research opportunity. veloping faster numerical integration algorithms The supercomputer power that is becoming for ordinary differential equations. available will provide the opportunity tocom-

15"4 ' 7I 148 1.110\11E11.i.V LAGIVEERIVG b'ne process and control system designincluding optimiza- tionInto one large problem that A Revolution in Process Control can be solved in a way that accounts for then intera:tions The Until recently, the chemical process industries have largely tried success of such a consolidation to solve crifficuft control problems by using simulation, case studies, willdepend on the develop- and ad hoc on-line adjustment of controller knobs a procedure ment of approximate compatible that can require personal intuition end C011tilVJOUS human super- models and of techniques to revision of the controls. Theta industries did not use modern control late model quality to perform- techniques because available coned theory did not address ance. practical issues. The contref t that WM *minim( inthe 1960s and 1970s assumed that the *aligner had a linear mathwomlical model of the process * be controlled OS- well as a substantial Robust and Adaptive Control body of fixed knowledge on *e$ disturbances and the in Control systems are designed "noise in the nessurernedi made by process seneom Stich a from mathematical models that knew process control theorybut on fatal shortamie2s: there are generally imperfect descrip- are no t. uly Sneer processes and there are no fixed, kirovm spectra tions of the real process. Itis of information on disturbincesteximeasurernent :dee. Controllers essential that control systems op- designed on the bads dines( males theory will fail in the real erate satisfactonly over a wide world of nonlinear* and uncertainty. range of process conditicis. Thus, New research advances In -control theory that we brkiging it the control algorithm must pro- dose to practical problems' are promising dramatic new devel- vide for control of the process opments and attracting widespread bilabial inte,:adt °noddies:. even when the dynamic oehavior advances is the development of "robust!' systerne. A robust control of the process differssignifi- system is a stable, dosed-loop system that can operate success- cantly from that prefiicted by the kdly even if the model on which it is based does not adequately model. A control system with describe the plant. A second advance is drz use of powerful this characteristic is sometimes sernivmpkical formalisms in contrt problems, particularly %:-.are called robust. In fact, a tradi- the range of possible process variables is coma *road. tional disregard for the model The development of In model control," a design technique error problem is one of the main that *Wass traditional and robust techniques for designing control reasons for the frequently cited system has provided the framework for unilykig and extending gap between theory and practice these sivances. It is now evadable in commercially evadable in process control. Industry needs design software. algorithms that are robust rather The combination of these advances is revolutionizing process than ones that "get that last half control, apawning unprecedented research activity in both aca- percent performance." Control demia and industry. strategies that work all the time within reasonable limits are better than those that work optimally some of the Importance in the design of ....laptive controllers time but that frequently require reversion to for processes since it is that very mismatch that manual control. drives changes in the controller parameters. Because over time a process often changes, The engineering theory and methodology for its model parameters must he continually up- designing reliable adaptive controllers for chem- dated: in extreme cases. the basic model must ical processes are inthe earlieststages of be reformulated. An adaptive system is a control development. system that automatically adjusts its controller Finally, there are always process operations settings or even its structure to accommodate in which neither classical nor modern control changes in the process or its environment The is effective. Such operations may require qual- problem of model-plant mismatch is of crucial itative decision making or the use of past knowl-

15E; FRONTIERS (lIEWICAL ENGIAFFP'S(.

(1) (2)

(3) (4)

PLATE5Stages in the retorting of an oil shale particle in a hot inert gas are shown. [l, 2, 31 The retorting zone moves inward as products diffuse out of the particle. A coke layer on the particle is formed as a final step of the retorting process. [4, 5, 61 Retorting goes ic com- pletion in the center of the r -tide, and a fully coked particle remairs. Courtesy. Amoco Corporation. (5) (6)

(IL

PLATE 6 The turbulent environment in which combustion, a chemical process. takes place is dramatized by this Schli- eren photograph of a propane diffusion flame. Courtesy, Norman A. Chigier, Carnegie Mellon University.

159 I RONTIERS IN CIIE111Cil, ENGINEER/AG

PLATE 7Chemical engineers develop models to understand the formation. transport, and environmental fate of airborne pollutants sui.h as ozone. Thi, photograph shows a graphic display of a chemical engineering model for ozone concentrations in the Los Angeles basin. Courtesy. John Seinfeld. California Institute of Technology.

PLATE SCorrosion is a chemical process whose result., are easy to see in the world around us. In this picture. corrosion of reinforcing steel has caused the concrete pillars to wall weakening the bridge and forcing the installation of wooden joists to temporarily support the bndge deck structure. The results of corrosion impose significant eccnomic costs on societyIn 1982, these costs were estimated at about $120 Courtesy. Robert Baboian. Texas Instru- ments. Inc

1 6 1) COMPUTER-ASSISTED PROCESS AND COATROI. ENG ISEERIN l, 149

edge.Artificialintelligence techniques offer PROCESS SENSORS promise for control system design in thesecases. Ifwe had a completely accurate model of a Batch Process Engineering process and accurate measurements of process disturbances at their inception, then corrective The production of filie and specialty chemi- action could be taken directly without the need cals, which are usually made Ly batchpro- to measure the output streams from the process cesses, is becoming increasingly important and after the disturbance has propagated through it. competitive. The efficient operation of multi- But because we generally do not have adequate product and multipurpose batch plants offersa models, the output streams of processes must variety of challenging research problems for be measured for the purpose of feedbackcon- chemical engineers. Most industrial batch chem- trol. ical operations are now scheduled by intuitive, The sensor is the "fingertip" of the process ad hoc methods that consist of modest variations control system. The principal challenge inpro- around historical operating patterns and that cess sensing is the development of analytical make little or no use of computer technology. sensors, particularly for determining process It is now widely recognized that the scheduling stream composition. Such sensors eliminate the problems associated with batch processesare need to withdraw samples to determineprocess immensely comp!ex and, in fact, are among the and product parameters, a practice that should most difficult combinatoria'_ problems known. be minimized because of inherent problems Limited progress has been made in using math- (e.g., samples of reactive intermediates may be ematical models in the simplest types of batch toxic or otherwise dangerous,or the interven- process scheduling. Current algorithms are too tion represented by withdrawing a sample may computationally demanding and complex for affect process operation). Since it is important industrial use. An important intellectual chal- for process control not to disturb the normal lenge is to generate a unified field of batch operation of the process, sensorsare needed process engineering theory and to put it into a that can operate in the environment of th' practical context by using case studies. process stream The key to meeting this ch' Linear control theory will be of limited use lenge is a fundamental understanding of for operational transitions from one batchre- physical and chemical ;nteractions at the gime to the next and for the control of batch sor-environment interface and, in pa. tict plants. Too many of the processes are unstable the transport and kinetic processes thatoccur and exhibit nonlinear behavior, such as multiple there. steady states or limit cycles. Such problems often arise in the batch production of polymers. The feasibility of precisely controllingmany Future Sensor Developments batch processes will depend on the development Thetechniques used in the chemical proof an ppropr4ate nonlinear control theory with cessing of electronic microcircuits (see Chapter a high level of robustness. 4) are being adapted to the microfabrication of While startup and shutdown occur relatively two- and three-dimensional structures for solid- infrequently in lar.le continuous plants, theyare state sensors. These techniques will permit the inherent in batch plant operation. Most startup integration of transducers, optcelectronics, sig- and shutdown procedures, whether devisedem- nal-conditioning and data-processing devices, pirically or theoretically, are designed to follow and micromechanical devices into extremely a recipe ofactions with no feedback. Thus, if up- small packages Reduced size offers advantages sets occur, there is often no way to change the in thermal uniformity and response speed; shock startup or shutdown in time to _void unwanted and acceleration resistance; and reduced weight, process excursions. Procedures are needed that volume, power, and cost. incorporate feedback and adaptive techniques Solid-state sensors may be developed that to the problem of plant startup and shutdown. will he responsive to a broad range of acoustic

161. 150 / RIATII.RS (7 /L 'II( U. /..N6II..1 RI N'l,

TRANSMITTER inputs, electromagnetic radiation, ionizing ra- FIBER diation, and electrochemical stimuli. Response LILY elements may be tailored for high selectivity a among ions, free radicals, or specific compounds. Alternatively, elements with low selec- RECEIVER SENS01.' OPTO ELECTRONIC P.TWOF,K tivity are also useful because information from NUB EXTRINSIC an array of such sensors, each with a different PROCESS OPTICAL SENSOR OR but known broad response, can be processed OUTPUT (OPTROOE, to provide quantitative analysis of a complex mixture. Complex mixtures also lend them- (23 selves to chromatographic analysis. It has been tea.. INTRINSIC FIBER SENSOR NETWORK shown that gas chromatographic data can be analyzed on a silicon chip, although with some FIGURE 8.4Configurations for sev,dt different kii.. 3 of loss in recognition reliability. Combinations of optical fiber sensing systems are shown. The common factor gas or liquid chromatography or capillary elec- in all these systems is the use of an optical fiber as an integral elemei ;' in the system, either to carry light to and trophoresis of microsamples with mass spec- from discrete sensors (often referred to as optrodes), or as trometry may be developed to provide superior sensitr:e elements themselves (intrinsicfiber sensors) performance. Courtesy, AT&T Bell Laboratories. The development of biological sensors is taking place at a rapid pace. Biological sensors analyze chemical mixtures using biological re- the development of instrumentation with no agents of exquisite specificityfor example, electrical components in the sensor (Figure 8.4). enzymes, immunoproteins, monoclonal anti These devices will operate by transmitting probe bodies, and recombinant nucleic acids. Such light from a remote source to the process sensor sensors may permit the analysis of fast reactions with an optical fiber light guide. In the sensor, of species in very dilute media. Multicomponent the light signal will be altered by the sensed biological sensors may be able to perform com- environment (e.g., by absorption of certain plex analyses that involve multiple reactions, wavelengths, fluorescence, or scattering) and with automatic -egeneration of the biochemical will thus be "encoded" with information. The reagents or removal of interfering species. encoded signal is transmitted through the optical Unfortunately, current biological sensors are fiber to a transducer that produces an electronic extremely delicate. Even when the biological signal. The advantages of suc.! systems include reagents are immobilized on a solid carrier, inherent safety, low signal attenuation, and the such sensors require careful construction and ability to multiplex signals ir, the optical fiber. frequent recalibration. are not always amenable Such instrumentation can incorporate additional to automation or unattended operation, and chemical, biological, and electronic component. sometimes have inconsistent dynamic response and is likely to play a major role in many future and limited life. Although biological reagents sensor systems. are ideally suited for some atmlications, partic- Future sensors and their associated data pro- ularly those in relatively mild environments, cessing elements will need capabilities beyond they may not survive the harsher conditions those required for the simple measurement of often found in process industries. Here again, process variables, sty as periodic self-calibra- miniaturization of the biological sensor and its tion against known standards, automatic com- direct integration into an optoelectronic trans- pensation for environmental or other interfer- ducer, potentiometric electrode, or membrane ences, signal conditioning including linearization are promising approaches. With the current or other variable calculation, and fault recog- worldwide interest in biotechnology, major in- nition and diagnosis Some of these capabilities novations in biological sensors can be antici- are now available to a limited extent. Others pated. will become available with the continued de- Further advances in optoelectronics will allow velopment of integrated sensors and data pro-

162 COMPL TER- -ISMS TED PRO(TSS 1.\U ( .0 N 1 ROI. t A 6/ V/ I R/N G 151

cessing instrumentation. Arrays of sensors have mation management and decision making will already been mentioned in connection with be essential, and advanced capabilities inuser complex mixture analysis, but they may also interfaces and networking will bring new di- be used to provide redundancy, fault detection. mensions to this application. For example,cur- and data reconciliation. rent on-line literature search systems allow data sharing among many users, significantly increas- Research Opportunities ing individual productivity. However, this technology is generally only used to manage well- The availability of high-quality, real-time in- organized data; basic research is needed to apply formation on the conditions and composition of it to engineering data, which are notas well he process stream will permit engineers to organized. develop a completely new generation of process A process engineer will need to be able. to control strategies. The physical, chemical, and store and access relevant data rapidly in order biological phenomena at the sensor/process in- to carry out process development and design in terface must be understood and translated into less time and to solve problems arising from sensor technology. Chemical engineers are well new and complex design requirements (e.g., positioned to contribute to the development of designing for multiple objectives of profitability, improved process sensors in a variety ofways, safety, reliability, and controllability). Topro- including vide for rapid data storage, access, and transfer, new generations of computer hardware (bulk work in interdisciplinary collaborations with storage, network, and work stations) and soft- electronicengineers,biologists,analytical ware (data bases) will be needed. Some specific chemists, and others to elucidate the bic::.-Tical, research challenges for chemical engineers, chemical, and physical interactions to be meas- working together with computer scientists and ured; information specialists, follow. application of fundamental principles of reaction engineering and transport phenomena Most chemical manufacturingprocesses in to the design of sensor surfaces; the future will be monitored and controlled by development of new process control sys- computer. Process data will be collected con- tems and operation strategies in response to tinuously and stored either locally or ina central improved capabilities for measurement and aata base. Research is needed to develop mu- determination of the implications tor pro- tually compatible, efficient algorithms for stor- cess design of wholly new types of process ing and retrieving process data. In addition, sensors. computerized procedures will be needed for sifting through voluminous process data for PROCESS ENGINEERING INFORMATION information to use in process improvement and MANAGEMENT in the generation of new processes. Methods must be developed for storing In thenext decade, competition among in- judgments, assumptions, and logical informa- dustrialized countries will Se influenced by the tion used in the design and development of way in which information and knowledge are processes and models. managed in industry. The challenge is to be the Procedures and methods will be needed for first to find relevant information, to recognize retrieving and operating on other types of im- the key elements of that information, and to precise data. apply those elements in the manufacture of Efficient transfer of information amongen- desired products. gineers and designers will dependon how easily Computer technology will continue to provide these data can be accessed. The special needs new generations of hardware and software for of chemical engineers in this areathe partic- fast information processing and low-cost storage ular ways in which they generate, manage, and and retrieval The use of computers for infor- use informationmerit study. 152 tRoAllERS ry cia.mi(':tL ENGINEERING

produce them, chemical engineers must be IMPLICATIONS OF RESEARCH broadly versed in advanced computer technol- FRONTIERS ogy. This can only happen if they have access The speed and capability of the modern com- to state-of-the-art computational tools through- puter, as well as the developing sophistication out their educational careers, not in an isolated of chemical engineering design and process course or two. control tools, have tremendous implications for Making this broad access and utilization of the practice of chemical engineering. Chemical the computer in education possible will require engineers of the future will conceptualize and substantial government, academic, and indus- solve problems in entirely new ways. There are trial funding to provide both hardware and two bottlenecks to the application of these software. In some cases, groups may need powerful resources, though. First, there are not emote access to networks of supercomputers. enough active research groups at the frontiers In other cases, dedicated array proce. sors and of computer-assisted process design and con- other computational hardware may be required trol. A larger effort is needed in order for the in the chemical engineering department. If this field to keep pace with the expanding power of country is to maintain a leadership role in available computers. Second, many chemical chemical technology, critical needs for both engineering departments lack the computational research support and facilities acquisition must resources needed to fully integrate advances in be addressed. The status of funding in this field design and control into their curricula. For the and a specific initiative to achieve the goals full potential of the computer to be realized in outlined above are discussed in Chapter 10 and improved design of chemical products and the Appendix A. improved design and operation of processes to

... 1 ()`4 NINE Surfaces, Interfaces, and Microstructures

urfaces, interfaces, and microstructuresare key to an improved understanding of catalysis, electrochemistry and corrosion, processes for the manufacture of microcircuits, colloids and surfactants, advanced ceramics and cements, and memoranes. Chemical engineers canuse their knowledge of thermodynamics, transport phenomena, kinetics, anu process modeling to explore a variety of research frontiers. These include the development of molecular-level structure-property relations for guiding the production of mate-4als with specified physical and chemical surface properties; the development ofan improved understanding of elementary chemical and physical transformations occurring at phase boundaries; the application of modern theoretical methods for predicting atomic and molecular bonding at surfaces, the energetics and kinetics of surface processes, and the thermc4v- namics governing the formation of two- and three-dimensional microstructures; and the integration of fundamental knowledge to achieve realistic models of process operation that can be used for process design and evaluation. Chemical engineers will have to integrate their insights with those of researchers in other disciplines (e.g., chemistry and physics) t" advance these fields. Access to advanced instrumentation wo' be critical to the success of chemical engineers in addressing these frontiers.

153 165 154 I. RO\ill:R'. IV ( III:111(,11, LCINELRIM,

ICROELECTRONIC CIRCUITSfor com- croci ystallites, thin films, micelles, and micro- munications. Controlled permeability composites that are assembled into more com- films for drug delivery systems. Pro- plex structures ot. scales from microsccpic to tein-specific sensors for the monitoring of bio- larger. Tne ultimate products formed from mi- chemical processes. Catalysts for the produc- crostructures may include tailored forms of tion of fuels and cherr, cals. Optical coatings for particles, fibers, sheets, porous and sponge-like window glass. Electrodes for batteries and fuel structures, and a vast range of composites and cells. Corrosion- resistant coatings for the pro- as3emblages. So it is that microstructures, in- tection of metals and ceramics. Surface active terfaces, and surfaces represent an expanding agents, or surfactants, for use in tertiary oil and exciting frontier of untold potential. The re 'very and the production of polymers, paper, frontier extends from humalay designed mate- textiles, agricultural chemicals, and cement. rials and products all across energy and natural What do then products have in common? resources processing, environmental engineer- They all are based on materials that have pre- ing, and biochemical and biomedical technolo- cisely defined microstructures and/or surface gies. and near-surface properties. In fact, surfaces, interfaces, and microscale structures are important in virtually every aspect of modern The Nature of Structure technology; they influence the quality and value Figure 9.1 illustrates a variety of different of a broad range 4 products. Modern high- structures. This selection is by no means all- technology materials and products can be thought inclusive; a host of related structures such as of as populations of molecules that are distin- colloids, microstrands, thin films, microporous guish-d by the ways in which they are spatially solids, microemulsions, and gels could also have organized to provide useful, often unique prop- been shown. The parts of each of these struc- erties and pe ' mance. Organizational forms tu. ate distinguished by the zonesinter- include microstructures such as domains, mi- fae.:sbetween them, which often seem to be

(2)

P - SUBSTRATE ENHANCEMENT t...?LETION CHANNEL CHANNEL IMPLANT

(3)

FIGURE 9.1Examples of different types of surfaces, interfaces, and micro- structuies. (I) A biological membrane composed of phospholipid re-'lcules In which protein molecules are embedded (2) An NMOS logic circuit. (3) A section of ZSM-5 zeolite.

166 tit REM LS, Li a RI.,1t ES, t Nil 'OK ROSIRt el 1 RI S 155

so sharply defined that they are callcd surfaces. chemical engineering. The ability to produce Well-defined surfaces occur between solids and microstructures with the desired properties is either gases or liquids and thus are commonly leading to a wide variety of new materials and found in catalytic and electrode reactions. More products that promise to improveour quality diffuse interfaces may occur between solids,as of life and to provide new business opportunities in microelectronic devices, and between fluids for U.S. industry. Five specific examples of the or semifluids, as in many polymeric and colloidal societal and technological impacts of micro- systems. structural engineering, corresponding to Chap- Whether they are called surfaces or inter- ters 3 through 7 in this report, are given in the faces, when the zones between parts of a struc- following subsections. ture are "thin"from a fraction of a micrometer (the limit of the ordinary microscope) down to molecular dimensionsthe matter in themas- Biochemical an! Biomedical Engineering sumes a character that is somewhat different Mi :yr biological processes dependon cellular from that seen when the same matter is in the microstructure. These include selective and re- form of a bulk solid. This special character of action-coupled transport, antibody-antigen in- a molecular population arranged as an interfacial teractions, enzyme catalysis and the synthesis zone is manifested in such phenomena as surface of proteins and membranes. Surface and inter- tension, surface electronic states, surface reac- facial phenomena affect cell growth through tivity, and the ubiquitous phenomena of surface their influence on cell immobilization, cell-cell adsorption and segregation. And the structuring interactions, and cell disrupticn and sepal ation of multiple interfaces may be so fine thatno of coastituents. A wide variety of therapeutic part of the resulting material has properties products, then, exert their influence on living characteristic of any bulk material; the whole systems by influencing molecular events occur- is exclusively transition zones of one kindor ring at biological interfaces. In addition, the another. practical implementation of cell culture for the The finer the scale of structuring in a material, commercial production of biochemicals (bio- the more the material is taken up by interfacial technology) is heavily dependent on advances zones. Asthe proportionof material in intel-- in understanding how cell-surface interactions facial zones increases, the special character of mediate important cellular events. Finally, sur- those zones begins to dominate the properties face and interfacial phenomena are important of the total structure. This is why the perform- to the function of a variety of biomedical de- ance of highly microstructured materials is often vices, including artificial tissues andorgans, marked by superior physical properties (e.g., sustained-release drug delivery systems, and mechanical strength, toughness, and elasticity) future generations of biosensors. or chemical properties (e.g., resistance to oxidation or corrosion, selective permeability, and catalytic activity). Further, since "the perfor- Electronic, Photonic, and Recording mance is in the interface," there are strong Materials and Devices economic incentives to develop high-perform- Very-large-scale integrated (VLSI) micro- ance products from ;nexpensive bulk materials, electronic circuits epitomize intricately de- by modifying their surface and interfacial prop- signed solid microstructures thatare painstak- erties, to compete with existing products com- inglybuiltundermeticulouslycontrolled posed of expensive, homogeneous materials. conditions. Structure sc Ales in microelectronic devices are now at theI µm -level and are Relationship to Applications of Chemical expected to reach the I -nm level in the next Engineering decade. The production of such devices depends on the carefully controlled deposition. patt..cn- Surface and interfacial properties and pro- ing, and etching of thin layers of metals, semi- cesses affect virtually every aspect of modern conductors, pol:mters and ceramics. The grow-

1 6 7 . F. 156 ROA /MRS /A CM, 1/// /1,N (//14:/:///,VG

ing field of optoelectronics depends on the sired catalytic activity and separation selectiv- formation of carefully tailored glass films and ity. fibers to serve as light guides. Recording materials also depend on the careful generation Environmental Protection, Process Safety, and control of microstructure. Thus, magnetic and Hazardous Waste Management memories require ultrafine particulate coatings, whereas the new field of optical storage requires Theproduction of aerosols, soot, ash, and polymeric and inorganic thin films that undergo fines during combustion, calcining, and incin- finely tuned structural changes upon illumina- eration depends on a large number of physical tion. The production cf these and many other and chemical l rocesses at gas-solid and/or gas- materials will present a growing set of challenges liquid interfaces. Similar phenomena are im- to chemical engineers in the future. portant in the formation of suspensions, slimes, sludges, slurries, and other waterborne particulates from natural resource processing and all Polymers, Ceramics, and Composites sorts of chemical manufacture. The separation of particulates from air or water requires highly Microstructures, surfaces, and interfaces play microstructured separators (e.g., membranes, a central role in the production of ceramics, colloidal adsorbents, porous adsorbents, and glasses, and organic aid inorganic polymers. micellar and reverse-micelle scavengers). Like- Themechanical and chemical properties of wise, the understanding of soil contamination cement and concrete are also highly depende.:: and &contamination by hazardous wastes de- on the formation of the proper microstructure. pends c knowing soil and r.'neral microstruc- Interfacial chemistry is critical in determining ture as weli as the interactior of waste materials the strengt and durability of fiber- and fabric- reinforced composites and laminuted high- with these porous matrices. performance polymer composites.Exciting opportunities are low emerging for molecular composites of rod- and coil-type polymers. INTELLECTUAL FRONTIERS Crucial to the future of chemical engineering aretwo profound questions that define the field's Processing of Energy and Natural Resources frontiers: The recovery of pen oleum from sandstone How do the properties of a system or and the release of kerogen from oil shale and productand thus its processing characteris- tar sands both depend strongly on the micro- ticsdepend on the local structure of the system structure and surface properties of these porous (i.e., the size and shape of its parts, their media. The interfacial properties of complex liquid agentsmixtures of polymers and sur- contacts and connectivity, and their composi- factantsare critical to viscosity control in tion)? How does local structure depend on the tertiary oil recovery and to the comminution of starting materials and processing conditions tv minerals and coal. The corrosion and wear of which the system or product was cleated? How mechanical parts are influenced by the com- should the process be designed and controlled position and structure of metal surfaces, as well to achieve reliably the desired structuring? as by ti.e interaction of lubricants with these surfaces. Microstructure and surface properties The answers to both nuestions can come only are vitally important to both the performance from interdisciplinary research. The first ques- of electrodes in electrochemical processes and tion impels physicists, chemists, materials sci- the effectiveness of catalysts. Advances in syn- entists, geologists, biologists, and engineers thetic chemistry are opening the door to the alike. So does the second, but it serves as a design of zeolites and layered compounds with more potent driving force to materials engineers tightly specified properties to provide the de- and process engineers. If chemical engineers do

16E, SURFACES, INTERFA.ES, AND MICROS !RUC I ORES 157

connected porespace filled with fluid). But as the dimensions of the supporting matrix become smaller, ac,-ess to the catalyst on the matrix surface becomes more strongly controlled by diffusion, a process quite slow compared to convection, which is favored by larger interstices and pores. Moreover, the activity and specificity of the catalyst itself are often influenced by the way the catalyst is deployed on the supporting surface, by the nature of that surface, and by the catalyst's interactions with the sur- FIGURE 9.2This high-resolution electron micrograph shows the unique pore face. structure of the ZSM-5 zeolite catalyst. Molecules such as methanol and hydrocarbons can be catalytically converted within the pores to valuable fuels The development of new and and lubricant products. Courtesy. Mobil Research and Development Corpo- improved catalysts requires ad- ration. vances in our understanding of how to make catalysts with specified propertizs: the relationships not take up the research opportunities that come betweensurface structure, composition, and to them in the area of structure, researchers in catalytic performance; the dynamics of chemical other disciplines will. But chemical engineering reactions occurring at a catalyst surface; the has much to contribute to interdisciplinary at- deployment of catalytic surface within support- tacks on structure- function ritionships and ing microstructure; and the dynamics of trans- process-structure connections. These potential port to and from that surface. Research oppor- contributions, and the research opportunities tunities for chemical engineers are evident in associated with them, are discussed in the four areas: catalyst synthesis, characterization remainder of this section. of surface structure, surface chemistr,and design. Catalysis Catalysts are most often used to promote Catalyst Synthesis reactions of fluid reactants. They are, with some The introduction of new types of catalytic exceptions, colloidal, amorphous, or micro- ma.erials has often led to the development of crystalline states that, to be accessible by the nev or improved chemical processes. Examples reactants, are deployed on supporting matrices are zeolite catalysts for petroleum cracking and with a large ratio of surface area to volume. organic synthesis (Figure 9.2), platinum-based The largest possible ratios are achieved, though, reforming catalysts, Ziegler-Natta catalysts for by suspending fine partic':s containing theca- olefin polymerization, and catalysts for control talyst in the fluid of reactants, thus creating the of automobile exhaust emissions. Synthetic in- problem of removing the fine catalytic particles organic chemistry is currently opening upways from the fluid after the reaction is complete. of preparing new multicomponent compounds, Alternatively. the fine particles may be packed many of which have compositional and geo- together in one of several ways (e.g., as a metrical characteristics that suggest they !might porous 'ocrl; as the internal surface of a fine, have potential as catalysts. Such materials in- consolidated porous medium; as an inte -sper- clude the molecular sieves based on silicon siou of connected solid; or as a semisolid and aluminum phosphates, pillared clays and other

16c. 158 FRONTIERS IN CHEMICAL ENGINEERING layeredmaterials,supported transition metal clusters, and metal carbides and nitrides. There Zeolltes and Shape-Selective Catalysis is a growing interest in studying the influence of promoters on the catalytic properties of nonnoble The technological progress of the pardon reeling indusby metals, which could lead to a has beer snarled by, the discovery and development of awe to reduction in the demand for cat- turn more of a given bent of-gore di into rocks* such as alysts based on metals such as geedinsoileoel WO* platinum, palladium, and rho- collections of- listable-0- diummetals forwhichthe !red edict* wihlylttF li United States is almost totally containing dependent on foreign sources. useful nags*to Catalysts that do not contain these known es essioblik metals but possess many of their Clad**1111* catalyticproperties havere- sucharc cently been developed. adayste leen ,Aft kr A challenge particularly suited mold* die --11000:140#001,. to chemical engineers is the de- not change_ teksbelaelk.lt. velopment of process models for selecevity tor- girders predicting the microstructure and of key fridrotferilltelefir, surface structure of catalysts as nation'esseters amide twit** a function of the conditions of crude oil, emir* the *non* their preparation. Such models Yews. could be used not only to guidl AncivarsatotFuiorhMNustiincrudead the preparation of existing main polisherprodUc.1e1110-10,111);te -01.11001$ terials, but also to explore pos- hydrocarboaefinYtineyterialit sibilities for making novel cata- fluiditySraOin-Weifisfebt. lysts. and hes** *AO be' wax01941.00411b00111. the teskbatetileof *Pk* Characterization of Catalyst bons. A dillerook1004 Structure solosorinee281*Ondits*, Characterizationofcatalyst decrimineleinerenlameasitn, structure and composition is es- ProPoolYkixo*OknO0000,0111011001*, 1#1,011ieteilles sential to achieving a fundamen- t h a t a r e a m e l l a n t a . l i t -100114111- tal understanding of the factors dodo hydrootibmitt i be controlling catalyst activity, se- used to 40.4sceidy*ii*iiiiivieset fitteedsiiinaleteing. *is lectivity, an i stability. During anionic* dent Mil Oita** POSIIOOlt$046ofotif mode the past 15 years, the application offby IV b 10,1000,010ftiOn. 411160:1***1011 by of surface science techniques shapeidedityciaddagatitmesnwntoectilet Tlfi illest lNd (e.g., low-energy electron dif- inexpeniaitiprolotteydielidegiedgedicedetopiesi*aorW- fraction [LEED], Auger electron wide. spectroscopy [AES], x-ray pho- Stoompoisiodhsoolllesisin also Wm. Ibilliestieineleleong toelectron spectroscopy [XPS], poterdelprodittecif diondoetwOrloe, speolowtyegkehirockst andultravioletphotoelectron d oll* NW*. M dde alio, dM 000.10 Protium* le the_ one spectroscopy [UPS]) has led to capable of seceping van the mei poi* stmolum. This is the a very rapid advance in under- basis of the NIKO* cOlWerdefl of methanol to gookNie standing how metallic catalysts function.Knowledgeof the

170 SURFACES, INTERFACES, AND MICROSTRUCTURES 159

structure of supported metal catalysts has been advanced through the use of high-resolution trans- missionelectronmicroscopy (TEM), extended x-ray absorp- FAUJAINTE - tion fine structure spectroscopy TYPE ROUTE (EXAFS), and, more recenti. solid-state nuclear magnetic resonance spectrometry (NMR). Major advances in understanding zeolite structure have resulted from combining information obtained from x-ray diffraction, silicon-29 and aluminum-27 NMR, infrared spectrosc ')py, and neu- ZSM-5 ZEOIJTE tron diffraction. Unfortunately, many of the currently known techniques must be usej ex situ, making it difficult to observe catalyst structure and composition The kernewmi *as ass and pore mars-seabons of two types of sedatesare sham). Mid A reale has a dime-dimensional channel system during use or to examine the *lb pores of at hot 7.4in diameter. A pore k byroad by inyeen atoms dynamics of the changes in these in a rime. akirome 75144 zeolite has intemonneded channels running Ii one pro- Therefore, it ;s es- direcIfon, with pores lb its ht tbarneler. ZSM-5 pores are ton.;%1 by 10 rhygen Mat greater attentithi be Moms in a Ong. Reprinted wish permission from Chemical Access, gi 84(2), February 190, 32. to developing in-situ characterization techniques based on infrared spectroscopy, Raman spectroscopy, EXAFS, NMR, ZSPA45, the pray a Warble+, orsolhied of New Zealand's g000line and neutron diffraction. isproduct-0d. P Adod promises for lie produclon a. ultrahigh quay** &ad IA stank and payleno are bona on the sainnamilliind. fovea's* prkelPfill. Surface Chemistry ifrorq. Of*** reaolkino, lapookily *ow Involving the cow Knowledge of the structure bkialloit Orlon inaleolifii, elleugh WAY Irmo" stays and bonding of molecules to sur- on that aily irony aridonit to product. Carrying ad such faces has been obtained from rosalions in lha callus ofillonnall Itibukir pores c milescan such techniques as LEED, elec- madradly hdluonoolhair .aadonpollhostrys. Title balled trandtlon- tron energy-loss spectroscopy aNMetrolooffoty. Trarnilionalalo Mad* is the WNW phonon). (EELS),secondary-ionmass won in th ionhanood Walla* oboinad for nom catalysts In spectrometry (SIMS), infrared xylem loortmodzallon, a process praollosd conanerclally one large spectroscopy(IRS),Raman sash spectr, scopy, and NMR spec- cherfahry le an ended wimple of has a throe- tro-ietry. The scope of such dImOnsionot surfs* ow illsr. the cons of dishiest reacdons, stu;!;,9 needs to be greatly ex- asiladiktglor are product out of a hod of paint's! candidates. In pandr,e. to include the effects of addilloniolhanar:nononarcialopplosiOns that they have found, coadsorbates, pmmeters, and ahapninidellua resoMlis lisle proPridod the bolo for a rich new poisons. Greater emphasis should area of Mal* Winos and tedviokigy, sit; expected lo yawn be given to developing new pho- yin mars Inds** knowledgo, and applications. ton spectrlscopies that would peroit observation of adsorbed species in the presen..e of a gas

. 1 7 160 I RUMEX', IV ( 111-111C1I, 1,1GIALLKI1G

or liquid phase. Attention also needs to be given boundary layer near the electrode surface. In to studies of surface reaction dynamics to obtain certain applications, the pore size and connec- a fundamental understanding of the :lementary tivity of the electrode can also be important. reaction steps involved in the decomposition or rearrangement of an adsorbed species and of its reaction with coadsorbed species to form Charge Transfer eitherdesirableorundesirableproducts. There are two issues of fundamental impor- Knowledge gained from such studies combined tance to the kinetics of electrochemical pro- with information on the structure of the catalyst cesses. The first is the dependence on distance surface will lead to an improved understanding of electron transfer between sites Shat are not of what types of centers are critica! for achieving in contact. An understanding of this is critical hinss activity and selectivity and the role of for creating three-dimensional catalytic stric- p flS and other substances in causing catalyst ture:: through which charge percolates to fixed deactivation. The information gained from such sites. The second is the kinetics of electron studies will provide vital input to large-scale transfer at well-defined sites such as individual scientific computations of molecular dynamics defects on single crystals or on selected planes aimed at predicting the influence of surface of carbon. An ..aproved understanding of the composition and structure on catalyst perform- physical processes governing charge percolation ance. and conduction through an electrode and the factors influencing electron transfer at the electrolyte-electrode Catalyst Design interfaceisneeded. Such knowledge would make it possible so choose Recent theoretical studies have demonstrated electrode materials and tailor their microstruc- that it is possible to calculate accurately adsor- ture to suit particular applications. The identi- bate structure and energy levels, to exp!ain fication of cheap electrode materials to replace trends with variations in metal composition, and platinum would be a very significant accom- to interpret and predict the influence of pro- plishment. moters and poisons on the adsorption of reactants. Additional efforts along these lines will Molecular Dynamics contribute greatly to understanding how catalyst structure and composition influence catalyst- Mechanistic studies are needed on a select adsorbate interactions and the reactions of ad- number of electrochemical reactions, particu- sorbed species on a ,atalyst surface. With larly those involving oxygen. These studies are sufficient development of theoretical methods, far from routine and require advances in knowl- it should be possible to predict the des.red edge of molecular interactions at electrode sur- catalyst composition and structure to catalyze faces in the presence of an electrolyte. Recent specific reactions prior to formulation and test- achievements in surface science under ultrahigh ing of new catalysts. vacuum conditions suggest that a comparable effortin electrochemical systems would be equally fruitful. Electrochemistry and Corrosion There is no fundamental theory for electro- Electrochemical processes (e.g., electrolysis, crystallLation, owing in part to the complexity electroplating, electromachining, current gen- of the process of lattice formation in the pres- eration, and corrosion [Plate 81) are distin- ence of solvent, surfactants, and ionic solutes. guished by their occurrence in a boundary region Investigations at the atomic level in parallel between an electrolyte (liquid or solid) and an with studies on nonelectrochemical crystalli- electrode. The course of these processes is zation would be rewarding a id may lead to a strongly dependent on the potential at the elec- theory for predicting the evolution of metal trode surface, the composition and structure of morphologies, which range from dense deposits the electrode, the composition of the electrolyte, to crystalline particles and powders. and the microstructure of the electrolyte in 'he Many electrochemical reactions consist of SURFACES, INTERFACES, AND MICROSTRUCTURES 161

rosic electrodeposition,or etching. Mechanistic details re- Fuel Calls for Transportation main essentially unknown. Im- provedinsightwouldbenefit Coithitreltact signilkardi more useful energy out technologies that depend on the of our piesibui-bill meourthierfhink how remarkable it would be formation and stabilization of to caw oiltribMfigklbitzproeseiss at elliciencies not possible in controlled surfaces. troth heitenakies. These are not empty drilirl,tilkeriessibst sew_ thstdamonstrated in 1830. Called a Supramolecular fusicillithittrkkaboelthmiosteleviearney evertor* reshape major Microstructures firwrgto 1104001001*. *My 44:110sattoiliria ettbemistry were fascinated by the As noted earlier, the kinetics pokilikel4 fief Ode. Their *any attempts to use fuel cells to of electrochemical processes are *thm of materiels problems and influenced by themicrostructure °P104:10060* 1 lioply_riesnlythiit advances in science of the electrolyte in the electrode atien*Wreillittwitokraullid In pribliaitand reliable fuel cells. boundary layer. This zone is pop- TothibitHABOkparat. Mhite Mil an high - pressure 'hydrogen- ulated by a large number of spe- okyopit-fuir(tellab rams- MOS* power while In orbit. But cies, in luding the solvent, re- how ean,wathlrig the : pollen*, 1.4 luelvells back to bath? , actants, intermediates, ions, in- klust dell 'bir-used- in transportation vehicles, it could hibitors, promoters, and impur- have a fasjiiillitheollit worldwide consumption of petroleum. ities. The way in which these kisjortslOrtriemsnla that are needed for this to happen include species interact with each other incremektallmalliciencyotfusiatk,inCreathigtheirpowardc alty, is poorly understood. Major im- reducklgitgir Manufacturing cost, and developing fuel cell designs provements in the performance capable.), avid ankle. of batteries,electrodeposition Sokol andirderfacial engineering are the key to addresskig systems, and electroorganic syn- these teelmolOgical needk The conversion of chemical energy to thesis cells, as well as other elec- electrical 4ilergy WI fuel nil takes Owe at the interface of an trochemical processes, could be electrode Atli electrolyte. iwzglinolve misdate are needed achieved through a detailed un- for althIbTodsw-thateriall that combine losiptems stability with the derstanding of boundary layer ablikylecateffasili caiVenslon reaction. inexpenelve electrotees structure. are neadsort molester strudures liwit favor the transport of Sinceelectrochemicalpro- protons-and ortygsn lareackve ekes atthe interface. The prospects cesses involve coupled complex for gnaw superior materiels a* he taproot, by developing new phenomena, their behavioris techniques for in situ characterization of eolidl p id interfaces. complex. Mathematical model- There is exciting new engineering and science in the old field ing of such processes improves of fuel cells. Even more important, the payoff of this research may our scientific understanding of be a truly stunning advince in our use of energy. them and provides a basis for design scale-up and optimization. The validity and utility of such large-scale models is ex- complex sequences of steps, such as iii elec- pected to improve as physically correct descrip- troorganic synthesis. In these, the key to high tions of elementary processes are tLied. yield is knowledge of the sequence so that adroit choices of electrode and solution materials can be made. More thorough documentation of rate Electronic, Photonic, and Recording Materials and Devices and equilibrium constants is mandatory to transfer such scientific understanding into engineer- The creation of microstructure with well- ing practice. defined electrical or optical properties is critical The influence of added agents and inhibitors to the production of integrated circuits and is important in processes that involve cor- recording materials. The processes used to de-

173 162 I RUN! II.RS IN (///.111( I/ / N6/\//RIV,

fine microstructure in this context are described in Chapter 4, Common to the fabrication of all Photoresi.st Processing electronic and photonic materials are the dep- Polymer films that are sensitive to light, x- osition, patterning, and etching of thin films. rays, or electronsknown as photoresistsare The preparation of magnetic recording materials used extensively to transfer the pattern of an also involves the creation of very small magnetic electronic circuit onto a semiconductor surface. particles and their distribution in a thin layer of Such films must adhere to the semiconductor binder on a substrate. Virtually all aspects of snrface, cross-link or decompose on exposure these processes involve surface and interfacial to radiation, and und(sgo development in a phenomena. The challenge to chc -'-.:al engi- solvent to achieve pattern definition. Virtually neers is to understand the fundamental elements all aspects of photoresist processing involve of each processing step at a level where this surface and interfacial phenomena, and there knowledge can be used to guide the design and are many outstanding problems where these fabrication of high-density, superfast circuits phenomena must be controlled. For example, and storage devices. the fabrication of multilayer circuits requires The scientific problems that must be ad- that photoresist films of about 1-gm thickness dressed to meet the chal'enges posed by the be laid down over a semiconductor surface that decreasing feature and domain size include the has already been patterned in preceding steps. following: A planar resist surface is essential to the successful execution of subsequent steps, but characterization of microstructures; it is as yet difficult to attain. A knowledge of identification of the factors affecting the the ways in which polymer viscoelasticity, sur- controlled application and development of pho- face tension. and surface adhesion affect the toresists: theology of resist flow is needed. Another area determination of the elementary processes requiring research is the solvent development involved in chemical vapor deposition, plasma of resist films after exposure of the films to deposition, and etching of thin films: and radiation through a mask. In this step, itis mathematical modeling of all aspects of essential to remove only those parts of the microstructures formation (e.g., in photoresist polymer that have been degraded by the radia- spincoating, resist patterning, and thin film dep- tion. Research is needed to understand how osition and etching. solvent composition, residual polymer stress, polymer adhesion, and the swelling of an nnir- r-diated polymer affect the geometric definition of the developed well. Such problems become Characterization of Microstructure increasingly important as resolution is pushed to 1 p,m and then to 0.1 p,m. Advances in integrated circuit technology and in tne production of high-density storage devices depend on making ever-smaller microstruc- Chemical Vapor Deposition and Plasma tures. An essential aspect of this problem is the Deposition /Etching of Thin Films ability to characterize the physical and chemical Both thermal and plasma-assisted chemical properties of domains having dimensions be- vapor deposition techniques are used routinely tween 0.1 and 1 gm. Visualizatimi and elemental to deposit thin films, and plasma etching is used mapping of microstructural elements at this to define fine features in the firms. Understand- scale have been accomplished by use of scan- ing the fundaments: reactions involved in these ning electron microscopes and, more recently, processes is essential to developing an under- high - resolution scanning Auger and x-ray pho- standing of how best to control the deposition toelectron spectrometers. If chemical engineers or etching of thin films and the design of equip- are to play an effective part in the future of the ment to carry out such steps. To make progress electronics and phntonics industries, they must in this area, chemical engineers need to identify be familiar with such modern analytical devices. the chemical species present in the gas phase S( RI: WES, !Nil la tNI) Uk ROSI RII I RI 163 of both thermal and plasma reactors through terfacial properties play a dominant role in the use of such techniques as emission spec- determining the behavior of the overall system. troscopy, laser-induced fluorescence spectros- Such systems are often characterized by large copy, electron spin resonance spectroscopy, surface-to-volume ratios (e.g., thin films, so's, and mass spectrometry. The dynamics of gas- and foams) and by the formation of macroscopic phase chemical reactions need to be understood assemblies of molecules (e.g., colloids, micelles, for processes involving not only neutral species vesicles, and Langmuir-Blodgett films). The but also electrons and ions. Another task of peculiar properties of the interfaces in such equal importance is understanding how reactive media give rise to these otherwise unlikely (and gas-phase species interact with solid surfaces often inherently unstable) structures. to achieve film growth or etching. While some The formation of ordered two- and three- of the elementary processes are similar to those dimensional microstructures in dispersions and occurring at the surface of catalysts, others, in liquid systems has an influence on a broad such as ion bombardment and photon-assisted range of products and processes. For example, etching, are specific to systems found in the microcapsules, vesicles, and liposomes can be electronics and photonics industries. Because used for controlled drug delivery, for the con- of their knowledge of transport phenomena, tainment of inks and adhesives, and for the chemical engineers are expected to contribute isolation of toxic wastes. In addition, surf ictants significantly to an understanding of how local continue to be important for en ._ced oil re- electrical fields and concentration gradients in- covery, ore beneficiation, and lubricat'on. Ce- teract to influence such processes as the ani- ramic processing and sol-gel techniques foi the sotropic etching of semiconductors. fabrication of amorphous or ordered materials with special properties involve a rich variety of Mathematical Modeling colloidal phenomena, ranging from the production of monodispersed particles with controlled The systems involved inmicroelectronic surface chemistry to the thermodynamics and processing are usually so complex that they can dynamics of formation of aggregates and micro- rarely be described by simple conceptual models. crystallites. It is therefore necessary to develop mathemat- The current status and the emerging oppor- ical models that incorporate fundamental infor- tunities in the science of colloids, surfactants, mation in order to understand such processes and fluid interfaces can be addressed conveni- adequately. The advantage of mathematical ently by considering a threefold hierarchy of modeling has been demonstrated for simple systems as follows: systems, and more detailed models will continue to appear with the growing access to large-scale individual molecules (e.g., surfactants), self-assembling(associated)microstruc- computing.Theconventionalmacroscopic models will have to be augmented with micro- tures of surfactants and oil molecules in the colloidal size range; and scopic treatments of interface formation so that macroscopic (often structured) systems made process conditions and interface properties can eventually be related. A close collaboration up of associated microstructures and bulk phases. between experimentalists and theoreticians will This last category may also include fluid lead to detailed models for simulating such interface systems vith unstructured bulk phases processes as chemical vapor deposition, plasma and/or moderate surface-to-volume ratios. etching, photoresist spinning, and photoresist Significant breakthroughs have been made in development. recent years in the identification, preparation, characterization, and understanding of entities Colloids, Surfactants, and Fluid Interfaces at all these levels, creating new opportunities for the successful use of colloidal and interfacial The area of colloids, surfactants, and fluid phenomena in chemical engineering and pre- interfaces is large in scope. It encompasses all senting new challenges as described below. fluid-fluid and fluid-solid systems in which in- New surfactant molecules ar : being designed

115 164 IRO:111.k \ 1%, i tit 1,1(17.1. %GIi i

with novel and special properties, particularly with the inclu- Flat -Panel Image Displays: sion of fluorine atoms and sili- An Emerging Application for con-containing substituents (both Chemical Engineering Principles yielding surface activityin or- ganicmedia).Multifunctional The most common imaging device that we encounter is probably surfactants are being designed as the picture tube in a television set. It's a bulky device and it coupling agents, release agents, requires a lot of electrical power to operate. These two eisadvan- rewetting agents, and steric sta- tages have sparked the search for a better alternativea display bilizers for wci colloids. Other with the imaging power of the picture tube, but one that is more recent developments include the portable and that has lower power requirements. synthesis if di-tail (and even tri- One candidate for the "picture tuba of the tutu._is the electro- tail) surfactants for use in the phoretic image display, or EPID. EPIDs are flat-panel displays tha preparation of thinfilms and contain submicron particles of pigment suspended ina liquid along membranes. Despite these ad- wilt a dye that provides contrast. When an electrical potential is vances, a htof possibiaties for applied to the system, pigment particles are driven to the interface structural modifications of hy- between the suspending liquid and a viewing plate (usually glass). drophilic and hydrophobic groups There they can be seen under normal illumination. EPIDs have the insurfactantsremainsunex- potential of providing an image that has extremely high optical plored. contrast under normal light onditions, that is legibleover a wide Self-assembling structures in- range of viewing angles, that is inherently retained on the display clude 7.-!onolayers and micelles. (as opposed to an image needing constant refreshingas on a TV both of which have received much picture tube), and that requires low voltage and power. study. However, new ap- The EPID concept car be combined with other developments proaches andpossibilitiesfor in imaging technology 10 produce optics: devices such as light these structures ace emerging. valves and xAy imagers. The accompanying illustration shows For instance, chrornophores are an example of an electrophoretic x-ray imager. being IncorporateJ in monolayer The most crucial part of this chemical display technology is the assembliestoProduce Lang- liquid dispersion of pigment particles and dye. The dispersion must muir-Plodgett flits with a vari- 1,9 stable, even when Ito particles are compressed to form an ety of unique optical properties image where the concentration is 10 times higher than in the bulk There has been great interest in of the liquid. The pigments must be able to retain their charges incorporating chemical function- after numerous whaling operations. The fluid must allow fast ality into monolayer- forming sur- response time. Unwanted migration of particles when an electrical factants to permit lateral poly- field is applied and other electrohydrodynamic effects must be merization, either in monolayers controlled. on liquid substrates or in Lang- Many of the crucial problems for researchers in this area are muir-BlAgett films on solids, thus the same as the ones encountered in other areas of surface and yielding exceedingly thin films or interfacial science. The -search of chemical engineerson high- membranes with structural integ- performance ceramic materials, field-induced bioseparations, and rity. For micelles. greater refine- fouling also addresses phenomena such as agglomeration and ment in the determination of mi- clustering in dispersions and rheology of dispersions. For EPIDs, cellar shapes, structures, and properties, as well as the investigation of the kinetics of micelleformation and other microstructural assemblies. and new en- disintegracon have become possible thanks to tities have been discovered and identified. These recent advances :n the use of photon correlation include inverse micelles, vesicles, liposomes, spectroscopy, small-angle neutron scattering. Mayers. microemulsions, liquid ci vstallites. and and neutron spin-echo spectroscopy. Notable a variety of as yet unnamed entities formed by advances have also been made in the study of the interaction of dissolved polv mers (often

1'It SURFACES, INTERFACES, AND MICROSTRUCTURES 165

Methodstodetermine and V control the properties of individual surfactant molecules and to determine the conditions needed to produce well-defined molecular assemblies are just beginning to emerge. We are at the threshold of being able to produce deliberately structured supramolecular entities with properties tailored to meet special applications.Some additional examples of problems that will have significant impacts over the next one or two decat...s follow: ltemopsomt serallog Mastrada New methods to produce large quantities of mono-sized Kai-0* *fay rixePor with an electroPhoretic image display. When a-seitage is applied Across the image cell, pigment particles and particles of nearly any inorganic eounlierions *the electrophotegc liquid separate. Most a the applied voltage material desired (e.g., metals, dreqsUcrairs sisessdeiSe jayer. /key isiinsure under this condition leads to oxides,silicates,sulfides) are thegreadOneta cAigielnerat the Pc-l*id interface dueto the generation needed for the processing of ce- of x-mpindui' dsharges-inthe Se. After the x-ray exposure, the applied voltage ramics, electronic materials, and is reduced to Zero, and the Apse* particles se driven to the viewing plate. The image becomes vigil* upon illumination. other engineered materials. New methods of emulsion polymerization, particularly the some Temarsh questions where chemical engineering principles use of swelling agents, to pro- are particuliny miewAnt Include the following: duce monodisperse latexes of any bow do Wass interact _4, they are repeatedly and desired size and surface chem- forcskOkiacked and unpsalcod inelk cal fields? istry are also needed. Perfect HOW aro ,ps Mee, counterim* polymers, and surfactants spheres as large as 100 Rni can transported acmes an EPID cell? now be produced in the zero- How dose fluid motion in an EPID device affect ihis Overall gravity environment provided by transport? the space shuttle. These spheres How fast do structures famed by charged particles dissipate and other mono-sized particles in applied fields? of various shapes can bz. used as model colloids to study two- and Questions similar to these are being researched by chemical three-dimensional many-body enginems who are active in the braid field of colloid and interface systems of very high complexity. miens. Future cant teations by chemical engineers may be the Refinements in the theory key to the success of the EPID concept in visual display applica- of interparticle long-range van tions. der Waals fences (the Laitdau- I theory) are within reach. New techniques are now proteins) or other macromolecules with the available for measuring the complex dielectric above structures. Many of these exist in the constants of various media required for the cellular makeup of living tissues (their study is implementation of that theory. called membrane mimetic chemistry), and host- Recognition and description of new inter- guest systems or artificial enzymes may also be particle interaction forces such as those owing produced. to magnetic dipoles, steric and electrosteric repul-

177 166 / RO\ Ill R' /\ ( III VI( V. I.V,I\II.KI1(,

mon, and long-range solvent ordering offeroppor- and cements. The chemical engineer's knowl- tunitics to study interfacial molecular pheno- edge of reaction kinetics, surface phenomena, mena that were previously difficult tc describe. and transport phenomena could contribute ef- New experimental techniques for the direct fectively to the development of new materials. measurement of interparticle forces are now Bulk ceramics are produced conventionally available and can be used to unders:and the by the sintering of powders. The strength, physicochemical factors that control adhesion, toughness, thermalstability,anddielectric coating phenomena, tribology, and others. properties of the fired ceramic depend strongly New optical (static as well as dynamic) on the size and uniformity of the precursor techniques for the study of long-range order in powder and on the chemical properties of the structured continua are beginning to appear and powder surface. can be used to understand the constitutive Examples of the need for improved ceramics properties and relations in complex (polymeric, technology, either to produce ceramics more nematic, and other structured) fluids. economically or to produce ceramics with im- New application of modern statistical me- proved performance, abound in both structural chanical methods to the description of structured and electronic applications. They include au- continua and supramolecular fluids have made tomobile engine components, armor, welding it possible to treat many-body problems and nozzles, artificial hip joints, wear elements of cooperative phenomena in such systems. The valves and pumps, cutting tools, electronic increasing availability of high-speed computa- packages, and a host of other current or future tion and the development of vector and parallel applications of this exciting new area of mate- processing techniques for its implementation are rials science. Among new developments in ce- making it possible to develop more refined de- ramic materials are ceramic glasses, micropo- scriptions of the complex many-body systems. rous ceramic filtering media, ceramic-ceramic Because of the increasing level of control that composites and microcracked composite ce- is now possible in the preparation of model colloids ramics for catalyst supports, ceramic fibers, and surfactants, model many-body systems can ceramic thin films and coatings, permselective be created in the laboratory and studied by non- membranes for applicati' 'i separations and intrusive instrumental techniques in parallel with sensors, and a variety of high-performance ce- computational and theoretical sophistication. ments. Essential to these improved ceramics is the In view of the above developments, it is now control of particle size and uniformity through possible to formulate theories ofthe complex phase well-characterized chemical reactions. Chemi- behavior and critical phenomena that one observes cal engineers have rich opportunities for con- in structured continua. Furthermore, there is cur- tribution through surface and interfacial engi- rently little data on the transport properties, : heo- neering of preceramic particles and powders. logicalcharacteristics, and thermomechanical Specific research areas include the study of properties of stich materials, but the thermody- chemical reactions affecting powder particle namics and dynamics of these materials subject nucleation, precipitation, surface structure and to long-range interparticle interactions (e.g., dis- composition, size distribution, shape, shape joining pressure effects, phase separation, and distribution, surface charges, agglomeration, viscoelastic behavior) can now be approached deagglomeration, tribological characteristics, and systematically. Such studies will lead to significant rheology. intellectual and practical advances. Important research opportunities in surface and interfacial engineering also exist with re- Ceramics, Cements, and Structural spect to the properties of finished ceramic bodies, such as surface energy and susceptibility Composites to crack propagation. Sintering mechanisms and The development and control of microstruc- kinetics represent a very important area for ture are critical in the processing of ceramics scientific investigation. Progress it- essing SI RI 1(1.5. I\II RI it i1,I ROW RI11 RI 1 167

steps and can he influenced by a ho,t of low-concentration addi- High-Tech Concr to tivesExamples include the superplasticizers, which not only Wo are surrounded by concrete, from our basement floors to reduce the ,,iscosity of freshly the support structures of many office buildir -s, from the sidewalks mixed concrete but also affect outside to dams and bridges all across Am. .a. Because concrete its fin41 rrner,y, eture. Collab- is so inexpensive and commonplace, it is not often thought of as orative research air.zing chemical a high - technology material. But research into chemical and miners engineers, civil engineers, and additives is yielding new insights into the complex chemical cilloid and surface chemists can reactions that occur when concrete sets. These insights, in turn, accelerate progress toward are leading tcs Improvements in the inner structure of concrete. By achieving suneaof formulations finding ways of decreasing concrete pc Jsity and increasing the for cement and concrete. strength of its interparticle bonds, researchers have pushed the maximum compressive strength of cone rute upwards from 4,000 6,000-1 to more than 40,000 psi. Membranes The secret of the new high-perfurmance concrete he use of Membranes are thin two-di- chemicals to modify surfaces and in:arfaces the bulk mensional structrres designed to -aterial. The best-known chemical additives are potent surfac- pass preferentially certain com- tarts, the so-called superplasticizers and high-range water re- ponents. -hly eficient sepa- ducers, that reduce overall porosity in fresh concrete while main- rations of gaseous or liquid com- taining Its workability prior to hardening. The, ddition of highly ponents car be achieved with divided minerals, such as microsilicon, in combination with a such technologies as reverse os- superplasticizer greatly improves concrete properties such as mosis, ultrafiltration, gas sepa- z:wasion resistance ari,i resistance to concrete breaks'e in by ration, microfiltration, dialysis, deicing seas, seawater, and frseze-thaw cycles. and electrodialysis. In these sys- The payoff to society from rater attention to the surface and tems, separations aririven either interfacial engineering of concrete is potentially immense: high- by pressure -Jr by a gradient in tech concretes that will prolong the life of public works and reduce chemical or electrochemical po- their maintenance costs as well as dramatic new applications for tential. Membranes are also find- this old reliable material. ing increasing use in controlled drug release devices and biosensors. Traditional .pplications these issues may permit the application of ce- of membrane technology have barely scratched ramics of known high-performance character- the surf-ce ofan exciting and rapidly developing istics to areas in which their use is now unec- ar-a. onomical. There are two major frontiers in membrane Finally, there is a need for chemical enginee research, one technological and the other sci- to bring their expertise in surface and interfacial entific. At the technological frontier, chemical engineering to the problems of developing better engineers can make important contributions to varieticF of Portland cement and concrete Both the development of r.ew materials, the engi- these commodities are produced in vast quan- neering of structure or morphology into me, - .ities each year, and major improvements in branes, and the identification of new ways of their properties (e.g., hve7e-thaw durability. using perm.olective membranes. corrosion resistance, and compressive strength) On the materials side, there is c,,nsiderable tremendously benefit society. The prop- ii,terest in developing novel membrane mate- ...es of Portland cement and u,.-..-ete are rials that are fu, ctionalized to selectively adsorb controlled by the microstructure of the mate- a specific component from a fluid phase. Mem- rials. The microstructure of concrete is devel- brane materials thv are enviromentally stable oped through a remarkably complex series of and resist. to foulint al : also needed. Since

17( 168 IRON. !!LRS ( ULM( U. L6LVLLKI\( higher fluxes of permeates can be achie,..i by decreasing m,r Modern Membranes: Process Technology brane thickness, there is increa3- Enables New Product Design ing emphasis on building stmc- tural integrity into the membrane. How do new materials and processing technologies affect Possibilities include the use of product design? Recent developments in membrane-based prod- laminated polymer membranes ucts show how design strategy has been profoundly changed and porous ceramic substrates the development, by chemical engineers, of new materials aid for ultrathin polymer layers. processes for spinning hollow fibers. On the applications s;de, in- Ten years ago, most membranes were made in flat rhea's or tegrated membrane processes by assembling numerous sheets into sandwiches. This quasi-two- represent an attractive area to be dimensional geometry imposed significant restrictions on mem- de velooed, in which a membrane brane product design. For exunple, sandwiches of planar mem- separa:.;ln is combined with a branes weredifficultto engimier because of the problem of sealing conventional separation to ac a multiplicity of thiflat components around their edg The complist. a job neither process advent of material,.nd processing technology for spinning hollow by itself could do. An example fibers, a chemtca engineering achievement, has changed the way is the distillation of azeonopic that membranes find use in product design. mixture. where a pervaporation Hollow tubular membranes have important physical advantages. module can be used to get around If the tube is small in diameter (and about 1 mm is typical), most the azeotropic composition. An- polymers can withstand considerable "hoop stress," i.e., pressure other example is the use of hol- differential across the wall. A tube with a small diameter also has low fiber membi anes in bioreac- a high ratio of surface area to volume. This provides a large tors. Here the membrane acts as amount of interfacial area, a desirable characteristic in a membrane a support for an immobilized cell that operates by passing substances through its skin selectively. or enzyme and at the same time By tailoring the composition of the hollow fiber surface, one can facilitates the supr'if oxygen vary the kind of separation ft will perform. If fie wall of the fiber is and/or nutrients. In a particularly made from a microporous material, ultrafilters can be made that elegant extension, a permselec- will separate cells from a fermentation broth or a single size- tive membrane may be combined fraction of proteins from a mixture. If the wall of the fiber is wade with a catalytic membrane to selectively remc ve a dilute reactant from a stream containing inerts and to '.:nological transport models already exist, tt-o- generate a product stream in which the produc lccular-scale moo for describing t:ie transport concentration is many-fold higher than that of of organic permeants and the transport of con- the reactant going in. Finally, there are some densil,le vapors through glassy or nonequilib very exciting opportunities for the development rium matrices have yet to be devehped. The of "smart" membranes that respond to the types application of structural probes, such as carbon- or concentrations of specir present in the fluids 13 N:vIR spectroscopy and XPS, could contrib- contacting them. One example is a membrane ute to the development of structure-permeability that regulates the delivery of insulin from a probe Likewise, elucidation of the physical reser, oir into patient's bloodstream in re- and chemical processes involved in membrane sponse to blood sugar level. synthesis could aid in producing membranes The s2ntific base for rationally designing with the desired microstructures. membrane polymers for specific applications is very limited, aP,1 hence there is an immense fron, ier to be c. .iered. Work is also needed RESEARCH NEEDS on transport flam.mtals, structure-permep To understand how the properties and per- bility relationships, a: el elucidation of how to formance mate. ial are tied to .ts microstruc- control membrane morphology. While phenom- ture and how microstructure depends on pro-

160

Pirli 111111111MIP St Rt V LS, ATER!' 1(1. S.t NI) III( ROSIN! ( II RI S 169

ture-related research One reason that surfaces, interfaces, and to alt wit only one kind of molecule to pass through, one can make more complicatedmicrostruc- membranes such as Monsanto's Prism'', membranes that separate tures are a frontier of chemical hydrogen from other gasesorGE's oxygen enrichment meri- engineering and procesing re- t, _nes. search isthat modern science The future is bright for devices based on hollow-fiber membranes. has recently spawned a number By combining, in one device, membrane layers whose properties of mizrostructural probes of un- change in response to their environment, one will be able to make precedented resolution and util- s.ilf-regulating or "smart" .,anes. An example might be a ity. For the first time we have hypothetical artificial pancreas that would regulate the passage of the proper tools to attack the insulin to a diabetic.Atwo-layer hollov. tube, capped at both ends, molectar and chemical basis of would encapsulate a reservoir of insuNn. The outer layer of this microst. uct ures . device would have the enzyme glucose oxidase immobilized within Of course, our understanding t. When glucose in the bloodstream reacts with this enzyme, of microstructures will be ad- hydrogen peroxide is released. The hydrogen neroxide would vanced through an interplay of diffuse into the second layer, a redox membrant whose ability to observation, conceptualization. pass insulin out from the central reservoir is controlled by the experiment, and theory. But in amount of hydrogen peroxide diffusing into and out of this layer. 'Ins area of engineering science, The overa3 system would release insulin in proportion to the advances will come most oft.. amount of glucose in the bloodstream, just like the par. when 411eddydeveloped instru- Techniques such as enzyme immobilization and th, menta; probes are adapted to of layered structures of redox polymers are already use° in me new systems or new probes are deve:opment of diagnostic sensors for clinical and biological perfected to answer questions application' o make dwices lik -, the artificial pancreas a reality, arising from practical problems. these techniques must be combined with hollow fiber technol'..gy The adaptation -,,e levelopment to engineer membrar with the appropriate interfaces. It is an of instrumental probes for sys- area where chemistry, physics, biology, and engineering interact. tems of interest to chemical en- Chemical engineers have the holistic perspective needed to gather gineers demand cooperative ef- these strands together. forts with the originating scientific disciplines and with instrument manufacturers. In such efforts, cessing, researcher' must be able to detect chemical engineers can bring inportant refine- microstructure, ciiai actenze it, resolve its shape ments or innovations to 'instrumental practice. and connectivity, and measure itssize and This has already occurred, for example, in the composition. They need to visualize the mi ro- development of video-enhanced optical micros- structure, whether directly through some ,.ort cop;', rand- freezing cryo-electron microscopy. cf microscopy or indirectly by means of theory the nalysis of solid catalytic surfaces. and the based on model-dependent synthesis trom me4- probing of solid-liquid interfaces important in surements. The challenges are enormous be- electrochemical catalysis. cause of the small size and complexity of microstructures, the fluidity and thermal fluctuations of liq,lid and semiliquid systems, and the rap- Mk ras«)py and Mk roto.nography idity of many physical transformations a.id The directisualization of microstructure chemical reactions. may be accomplished by vi.r,us forms of microscopy. Recent refinements in microscopy instrumentation t:Arik,ues are epitomized by video-enhanced interference phase-contrast microscopy, which Instrumentation for experimental observatic, is emerging as a workhorse pr be for colloidal and :neasuremert is paramount in microstruc- suspension', and other i iicrost ucttd liquids.

1 8 ;. 170 FRONTIER.S IN ";NGINEERING This technique is capable of resolving structures at distances approaching the wavelength of visible light (350 to 800 nr Another useful tool, ana ar- guably the most powerful probe of surfactopography on scales from those of the light microscope down to 5 nm, is scanning electron microscopy'ith x-ray microanal; sisTrim technique combinemag-"evingpower, depth of field, anu ability to analyze local composition. It may also be used to study the internal microstructure of specimens by fracturing them (sometimes 'fter FIGURE 9.3Measuring less than 1/100,000,000th of an inch, the hills in this freezing). Scanning electron mi- micrograph are individual atoms on r silicon crystal that have been enlarged more than I cros;,opy is certain to become a billion times using a scanning tunneling microscope. The microscope collects digital information that is plotted by a computer. Bands very useful tool in the hands of on the hills are contours assigned ,y the computer to help researchers see chemical engineers, particularly how crystalline strictures are formed. Copyright AT&T. Microscapes. as they apply the principles of chemicalengineeringscience (e.g., sophisticatedunder- standing of heat and mass transfer, phase change, "environmental stages" that function only with and chemical reactions) to interpreting images severely reduced effectiveness). However, the and developing ancillary techniques. inteL3e electrical fields of the scanning tip can Even greater magnifying power is provided strongly affect the specimen locally. Equipment by transmission electron microscopy, and in and techniques are rapidly being refined, and it some instances this technique can be comple- appears that scanning tunneling microscopy will mented with energy loss spectroscopy. Trans- be playing an important role as a probe of active mission electron microscopy can resolve micro- sn"d-gas and solid-liquid interfaces. structure down to atomic scale., (0.1 to 0.5 nm) X-ray microtomography is a new develop- and requires the skillful application of special- ment of great promise for re. instructing, dis- ized techniques to extremely thin, solid, or playing, and analyzing three-dimensional mi- solidified specimens (or replicas of specimen crostructures.Resolution around 1 p.m surfaces such as in' er.al surfaces of fractured has been demonstrated with currently available samples). Correct interpretation of images re- synchrotron sourcesof x-rays,x-rayde- quires not only considerable experiencebat tectors, algorithms, and large-scale computers. also a fundamental understanding of sample The potential for microstructural research in behavior during preparation and under the elec- composites, porous materials, and suspensions tron beam and of the contrast mechanisms atthisandfinerscalesappearstobe urm:erlying an image. tremendous. Scanning tunneling microscopy is a recent Magnetic resonance imaging, or microtomog- invention of great potential (Figure 9.3). Capable raphy by multinuclear magnetic resonance, is of resolving surface topography down to atomic another new development that is c ven more dimensions, it operates perfectly well on sur- exciting because it provides three-dimensional faces immersed in gas or liquid, whereas elec- mapping of th(: abundance of a variety of atoms. tron microscopy requires that the specimen be Compositional aspects of microstructure can studied under a vacuum (except for special thereby be resolved. However, the resolution

182 SURFACES, INFLRI ACES, AAD IIICROS1RI C I I RI S 171 of currently available instruments does not yet at wwch the absorption occurs and the shape approach I Rm. of the absorption feature can provide detailed information about electronic structure, molec- Scattering Methods ular bonding, and the association of molecules into microstructural units. Beam. of electromagnetic radiation of appro- Nuclear magnetic resonance (NMR) spec- priate wavelength are scattered when they in- troscopy is an enormously powerful toot that, teract with the gradients inherent in structured in chemistry, has become a mainstay for ana- materials. By measuring the ways in which the lyzing molecular structure and environment. In intensity of scattered radiation varies as a func- recent years, NMR spectros-opy has proved tion of the angle at which the radiation initially us -ful for studying catalysts, amorphous semi- strikes the sample, the wavelength of the radia- conductors, and colloidal-scale microstructure tion, and the time, many aspects of the structure and molecular aggregates. Examples of the of materials; ran be inferred. application of NMR spectroscopy to problems Bulk het( ogeneities arid surface topography of interest in chemical engimering include iden- are both marked by electron distributions that tification of the secondary building units in- vary in thei polarizability. These variations are volved in zeolite synthesis, analysis of the cap: t ...1! of scattering photons. In liquid and development of bicontinuous "liquid micro- semiliquid materials, where the variations them- sponge" in surfactant-oil-water systems, the selves fluctuate over time and space, static light clustering of hydrogen in amorphous silicon scattering and its dynamic complement, .oton photovoltaic devices, and the structural char- correlation spectroscopyare important probes acterir ltion of ea, bonaceous deposits that lead of larger colloidal-scale microstructures and to for, ation of coke on catalysts. In addition their thermal motions, which are often finger- to r . aiding time - avenged information, NMR prints of structure. For solids, the scattering of spectroscopy can be used to probe the dynamics x-ray radiation can be used to characterize the of molecular motion on time sea- s ranging from structure of both crystalline and amorphous 106 to I second. Thus, for example, time-re- materials. Of particular interest in terms of solved NMR techniques have made it possible amorphous materials is the technique of ex- to characterize the dynamics of forming the tended x-ray absorption fine structure, which precursors to zeolite synthesis. provides information on atomic( oordination The vibrations of molecular bonds provide number and local bond distances. insight into bonding and structure. This infor- Generally, tl.e more intense the available mation can be obtained by infrared spectroscopy beam source, .!. shorter the time scales, the (IRC), laser Raman spectroscopy, or electron weaker the heterog( !lefties, and the longer the energy loss spectroscopy (EELS). IRS and distances that can be probed by a scattering EELS have provided a weal ii of data about the method. Hence, there is a strong drive to utilize structure of tatalysts and the bonding of adsor- high-powered lasers, synchrotrons, and intense bates. IRS has also been used under reaction neutron sources in research on surfaces, inter- conditions to follow the dynamics of adsorbed faces, and microstructures. This is particularly reactants, intermediates, and products. Raman true in the study of liquid materials and of spectroscopy has provided exciting information systems that undergo rapid physical transfor- about the precursors involved in the synthesis mations or chemical reactions. of catalysts and the structure of adsorbates present on catalyst and electrode surfaces. Resonance Spectroscopies Molecular-level :haracterization of surface composition and structure can be obtained The interaction of radiation with a material through a variety of electron and ion spcctros- can lead to an absorption of energy when the copies. The two-dimensional structure of sur- radiation frequency matches one of the resonant faces and ordered arrays formed by adsorbates frequencies of the material. Tice exact frequency is revealed by low-energy electron diffraction

1Q el . t 'a.kjfj 172 FRO% MKS IA (111.111( if. E\GIAI;ERING

(LEED). This technique can also be used to follow phase changes and surface reconstruction in real light to time. The atomic composition of spersrorweter surfaces can be determined by microscope objective Augerelectronspectroscopy upper rod (AES), x-ray photoelectron 'We spectroscopy (XPS), and sec- piezoelectric tube ondary ion mass spectrometry mica sheets on silica ...ovatie clamp (SIMS). While SIMS provides disks the highest elemental sensitivity, man support AES and XPS can resolve spatial variations in composition down clamp adjusting stiff double- to 0.1parr. XPS, in addition, rod cantdever spring giveinformation on the valence state of individual atoms, from variable helical spring whichdetails stiffness ofinteratomic force- bonding can often be inferred. measuring The density of electrons in bond- spnng ing orbitals can be obtained from lower rod ultraviolet photoelectron spec- 0 cm 5 troscopy (UPS). When carried white fight out with monochromatic beams TGURE 9.4 The direct force measurement apparatus shown here can mt asure of synchrotron radiation,this the forces between two curved molecularly smooth surfaces in liquids. Mica technique can also identify the surfaces. either raw or coated. are the p. mary surfaces used in this apparatus. orientation of individual atomic The separation between the surfaces is measured by optical techniques to and molecular orbitals at a solid better than 10 nm. The distance between the two surfaces is controlled by a three-stage mechanism that includes a voltage-driven piezoelectric crystal surface. With the exception of tube supporting the upper mica surface: this crystal tube can be dispLced L: D, each of these techniques less than 10 nm in a controlled fashion. A force-measuring spring is attached can be used to characterize poly- to the lower mica surface and its stiffness can be varied by a factor of 1.000 crystallinefilms, amorphous by shifting the position of a movable clamp. Reprinted with pc-mission from materials, and powders, as well Prot. Natl. Acad. Sri. USA. 84. July 1987. 4722. as single crystals. Its use will continue to expand as its cost falls, Other Important Methods its complexity decreases, and its capabilities multiply. The statics and dynamics of microstructures The eleeiron tunneling microscope tip is cur- are governed by the forces that create or main- rently gaining recognition as the most exquisite tain them. Rarely can the forces be cured micromanipulator for measuring local deform- directly. But forces between special surfaces rbility of solid surfaces, down to nanometei s immersed in fluid can now be accurately gauged and smaller. For microstructures on scales of at separations down to 0.1 nm with the direct micrometers and larger, micromanipulation ap- force measurement apparatus, an ingenious paratus from biology and biophysics is turning combination of a differential spring, a piezo- up in probes of deformability and force; mi- e'lctric crystal, an interferometer, and crossed croelectronic devices are in the offing. Micro- cylindrical surfaces covered by atomically smooth electrade probes continue to evolve. Laser- layers of cleaved mica (Figure 9.4). This recent doppler motion probes capable of micrometer development is finding more and more appli- resalution and birefringence and dichroism mea- cations in research on liquid and semiliquid surements are becoming important in the char- microstructures, thin films, and adsorbed layers. acterization of many surfaces. SURFACES, INTERFACES, AND MICROSTRIVTL RES 173

Particulate microstructures. as well as the widespread acquisition, or while the req .mite fragments obtained by disrupting more viten- financial resources cannot be marshaied, an sive structures, are separated by equipment of effective strategy may be for funuing agencies varying cost and sophistication: ultrafiliers, ul- or institutions to provide reasonably complete tracentrifuges, gel permeation and size exclu- subsets of the sophisticated instruments needed sion chromatographs, and electrophoretic sep- for surface and microstructural engineering to arators. The ultimate goal is rapid, automatic selected groups that would focus on a coherent sorting of individuals from a population of theme. Theesult would be to create a small particles upon sensing one or more properties communitj of users (e.g., three to four faculty of each. In the laser spectroscopy cell sorter, members) with similar interests in terms of use this capability has reached down to the scale of of the equipment. While some needs of the living cells through application of several tech- group may go unmet, requiring the use of nologies, including ink-jet printing. The cell equipment in other locations, the investigators sorter has already opened up research into the shoed generally be able to carry out the major population structure of cell cultures, a basic portion of their preparative and analytical work narch problem important in biotechnology. in close proximity to their laboratories.

Cost and Availability Theory Instruments for probing microstructures and Significant advances are needed in our current their changes typically follow a rule that the understanding of how molecules interact with costs of purchase, installation, operator exper- surfaces and with each °the: to form micro- tise, and equipment maintenance become higher structural units. Theoretical efforts along these as the dimensions of the structure to be meas- lines should be carried out starting at themo- ured become smaller, as the connectivity and lecular level and extending to the level of bulk shape to be examined become more complex, materials. The development of a hierarchy of and as the time between events to be resolved theoretical methods for predicting the behavior becomes shorter. Management of such research of increasingly complex ensembles of molecules gets complicated as its scale moves into the will be invaluable in understanding how best to gray area between small science and big science. orocess materials. Examples of specific areas The need arises to share instruments within a requiring develvpment were discussed earlier department, an institution, or a regional center, in this chapter. or, in sor rases, a national or international Development of the necessary theoretical facility, i -I activity that can become cumber- models will involve a careful intsgration of some when the instruments being shared are insights from different disciplines. Concepts central to an investigation. C:itting-edge re- new to chemical engineers (e.g., fractals, Monte search often calls for improvement, adaptation, Carlo methods, and pe,lation theory) will and augmentation of equipment. Research can have to be introduced to provide more accurate be stultified when scheduling a shared instru- and/or computationally efficient means for for- ment inhibits hot pursuit of a finding or an idea, mulating process descriptions. Chemical engi- when the needs of others prevent modification neers will need to become more familiar with of an instrument, or when a key experiment recent advances in applied mathematics and faces the risk of a !emporary shutdown of an computer science in order to work productively instrument. with researcliers from these disciplines. In par- Over the long run, the costs of sophisticaed ticular, collaborative efforts between theoreti- equipment fall to the level where the equipment cians and experimentalists should be encour- can be P.:quired for use and adaptation by aged as a means to new theoretical approaches individual research groups. But in the interim, and insights. while costs are sufficiently high that the level The need for access to supercomputers, dis- of usage of advanced equipment does not justify cussed in detail in Chapter 8, cannot be over-

X85 174 ; Rai I II,RS li ( Min( I.\61\1.1.1a emphasized. In the past, many of the major with time constants of less than a nanosecond. problems in the processing of structured ma- It will also be essential that chemical engineers terials have yielded to analysis once sufficient become familiar ss ith modern theoretical con- compEtation,.! power was "provided to permit cepts of surface physics and chemistry. colloid the utilization of very detailed physical models. physical chemistry, and rheology, particularly Supercomputers have made possible s' iificant as it applies to free surface flow and flow near advances in the modeling of plasma r. actors, solid boundaries. The application of theoretical complex electrochemical systems, coating lows, concepts to ierstanding the factors ctrol- and stress fracturing of polynrrs and ceramics. ling surface woperties and the evaluautAt of Advanced computational tools will become even complex process models will require access to more important as chemical engineers attack supercomputers. the important and highly complex problems now Funding must be provided to support rcst:arch on the cutting edge of research on surfaces, at academic and industrial institutions. Re- interfaces, and microstructures. searchers in universities will require funds for research assistants, instrumentation, computer time, and travel to use special facilities such as IMPLICATIONS OF RESEARCH synchrotron radiation sources, neutron sources, FRONTIERS and atomic resolution microscopes. The primary There is an increasing societal need for ma- support fo- these efforts should come from terials with surtace and interfacial properties federal agencies, with additional support pro- tailored to meet specific application. This spec. vided by industry. industry will also need to trurn of materials is extremely broad; it ranges finance its own research and development ef- from thin films for microelectronic circuits, to forts. One should anticip2:1 that generic long- high-strength concrete for roads and building-, range work will be carried out at universities, to membrales for food protection. The devel- whereas research leading to specific products opment and production of cch advanced ma- and processes will be conducted primarily in terials., and of sIrface active agents, will be rich industrial laboratories. Collaborative investi- ill technical challenges for chemical engineers. gations between university and industry scien- To address these challenges, chemical engi- tists should be strongly encouraged, since such neers will need state-of-the-art analytical in- efforts will help define the goals and objectives struments, particularly those thf.tt can provide of intermediate- and long-range research and information about microstructures for sizes down facilitate the transfer of new ideas and tech- to atomic dimensions, surface properties in the niques into practice. presence of bulk fluids, ane dynamic processes

186 T E N

Recommendations

HIS REPORT HAS been about the future though, is needed. Students must be exposed contributions of chemical engineering to both traditional and novel applications of research to the national well-being, and chemical engineering. Emerging applications about emerging res-arch ken iers of special should be highlighted and an expanded range importance. Some of the implications of these of process scales and chemistries should be frontiers have already been explored in individ- taught. ual cLapters. In this chapter the committee Chemical engineers of the future will not only draws together its principal recommendations be interested in the transport and chemical for action by policymakers in four sectors: reactions of small molecules, they will also be academia, industry, government, and profes- interested in large organic m,Aecules, proteins, sional societies. polymers (both organic 2.nd in )rgarlic), interfaces, surfaces, solids, and composite materials. Each of these new arcas rA interestto the RECOMMENDATIONS FOR ACADEMIA extent that they are ne t iow treated in depth in the core courses of the curriculumwill Education and Training of Chemical require new emphasis it teaching. Colecourses Engineers such as separations (unit operations) should be modified to cover new t..idlenges. For ex Ample, Chemical engineering undergraduate curri- a separations course might be modified to ad- cula have traditionally been designed to train dress challenges in bioprocessing and ultrapur- students for employment in the conventional ification by placing less eraphasis on t:istillation, chemical processing industries. The current core absorption, and conventional extraction, and curriculum is remarkably successful in this ef- more emphasis oa examples of generating high hrt. Chemical engineers will continue to play selectivity 'n separations.' Similarly, unitop- a major role in the chemical and petroleum erations laboratories and design courses in most industries, but new areas of application as well institutions are in need of rethinking and revi- as new emphases on environments: protection; talization to accommodate new needs and op- process safety; and advanced computation, de- portunities. sign, and process control will require some Another important need in the curriculum is modifications of the curriculum. for a far greater emphasis on design and control These modifications will not r itail a massive for process safety, waste minimization, and revision of the curriculum. Continued emphasis minimal adverse environmental impart. These is needed on basic principles (e.g., thermody- themes need i a be woven into the curriculum namics, transpon, separations, reaction engi- wherever possible. The AIChE Center for neering, and design) that cut across many appli- Chemical Process Safety is attempting to pro- cations. A new way of teaching these principles, vide curricular material in this area, but a larger

175

1878 7 176 I. ROAIILICS IN CIILVICIL L '1 UNLERE1 6 effort than this project is needed. Several large There will be a greater variety of products chemical companies have significant expertise and processes, increased demands for quality, in this area. They should do more to share their and a necessity to shift rapidly from one product insights and methods with academic researchers to another. and educators. (See "Recommendations for Chemical engineers will be working with a Industry" below.) broader :-., ectrum of colleagues; not just chem- Certain emerging specialties of chemical en- ists, but physicists, biologists, material scien- gineering will require a deeper exposure of tists, and electrical engineers. engineering students to other disci-Nnes. For The chemical processing industries will example: increasingly have a global outlook. They will require engineers with a greater understanding Students interested in biochemical engi- of the global economy, knowledge of engineer- neering will need to become familiar with the ing research and advance,occurring in other language and concepts of the biosciences, par- countries, and appreciation for other cultures. ticularly molecular biology. These industries will be operating in a For those interested in advanced materials, complex regulatory environment. They will re- more in-depth knowledge of statistical and con- quire engineers with a broad perspective on tinuum mechanics than that provided in trans- economic and environmental policy. port and mechanics courses seems essential. With the current emphasis on streamlined Chemical engineering students interested management and lean staffing in the processing in electronic materials need to understand the industries, chemical engineers will have fewer element! of electrical engineering and solid- peers and fewer superiors; their ability to make state physics in order to work productively with decisions will be tested early and often. colleagues in these disciplines. A greater emphasis on surface and inter- Chemical engineering faculty need to consider facial phenomena is needed for all chemical these challenges in planning for tomorrow's engineers interested in materials engineering. more broadly based curriculum.

It will not be easy to shoehorn all of these elements into a 4-year program, and the initia- The Future Size and Composition of tion of carefully delimited electives and options Academic Departments for students is probably the best approach. The committee applauds the recent action of the How can chemical engineering departments AIChE to allow more flexibility in the choice implement this broader curriculum as well as of science electives by undergraduates in ac- aggress'vely respond to new research challenges credited departments. The development of such and opportunities? A bold step is needed. The electives and new course sequences is not a committee recommends than universities con- casual undertaking. It will require greater in- duct a one-time expansion of chemical engi- volvement by faculty in providing advice and neering departments over the next 5 years, counsel to students at the sophomore and junior exercising a preference for new faculty capable level, who may not be prepared to make defin- of research at interdisciplinary frontiers. itive choices about specialization in their major. This expansion will require a major commit- The mix of industries into which chemical ment of resource.on the part of universities, engineers are being introduced is changing rap- government, private foundations, and industry. idly. Chemical engineers have always be Can such a preferential commitment to one proud of their flexibility and willingness to nse discipline be justified, particularly nt a time of to new challenges. This characteristic will be severe budgetary austtaity? Each agency re- needed more than ever in the future, as diversity sponsible for funding chemical engineering re- within the processing industries greatly in- search will have to answer th.% question for creases. For example: itself, but the cotr.aittee believes that the fol- RECOMIENDATION 5 177

lowing arguments for expansion are worthy of other disciplines should be encouraged. Cross- consideration. disciplinaryresearchpartnershipsbetween chemical engineers and colleagues in other dis- U.S. leadership in technology is under ciplines should be stimulated and rewarded. challenge today as never before. Future markets `,See sections on "Cross-disciplinary pioneer for biotechnology products, advanced mate- awards" and "Cross-disciplinary partnership rials, and advanced information devices will be awards" under "Recommendations for Gov- won by those who are best able to integrate ernment.") innovative product design with efficient process Academic institutions should make sub- design. Chemica: engii.cerc, the uniquely "mo- stantial commitments (e.g.. cost sharing and lecular" engineers, have powerful tools that can maintenance for instrumentation and facilities) be refined and applied to these challenges. They to make additional outside investments in chem- can make significant contributions to U.S. ca- ical engineerin8 as effective as possible. This pabilities in these high-technology areas. will be particularly needed to justify additional The existing chemical processing industries funding for advanced instrumentation andcom- are high-technology enterprises too. Moreover, puters needed for research and education. the chemical industry is one of the most suc- Academic institute -rs should takesteps to cessful U.S. businesses on world markets, with strengthen even further the !ong-standing ties a $7.8 billion trade surplus in 1986. The chemical between chemical engineering departments and processing industries require a continuing sup- the chemical process adustries, with a partic- ply of highly trained engineers armed with the ular emphasis on the flow of personnel between latest insights from the core researchareas of academia and industry. Mechanisms for inchemical engineering. Redirecting resources from creased interchange include appointing indus- these established and vital research programs trial researchers as adjunct professors inaca- to the emerging applications of the discipline demia; providing industrialsabbaticals for risks reducing the U.S. leadership roiition ina university professors; inviting industrialre- large and very successful industrial sector. Is searchers as seminar speakers on campus; plac- this an appropriate Wurse to take at a time ing student researchers in industrial laborato- when industrial competitiveness and the large ries; and expanding the number of masters U.S. trade deficit are such pressing national degree programs, evening continuing education concerns? programs, and other short courses targeted to industrial research personnel. In^reasing the number of resear^h groups and Chemical engineering departments should the resource levels in chemical engineering use additional re-aurces to promote greater seems to be the most practical way of (1) moving interaction with small process technology firms chemical engineers agg .ssively into the new in electronics and biotechnology. These firms areas among this report's research priorities play an unheralded but key role in developing while (2) maintaining the discipline's current industrial technology in these areas. (See section strength and excellence. The following re. om- on "Improvment of Links Between Chemical meneations for universities are ;ntended to en- Engineering Departments and Small Process sure that such an additional investment would Technology Firms" under "Recommendations be used most effectively. for Governme

Chemical engineering departments should use additional resources to appoint and en- RECt.MMENDATIONS FOR INDUSTRY courage taculty to address the frontier areas The committee encourages decision makers identifi..d in this report. Many of these frontiers in the chemical processing industries to increase will require more emphasis on interdisciplinary their commitment to research in their companies research. The selected appointment of chemical both to retain their world leadership in chemical engineering faculty whose backgrounds are from technology and to deliver to American society

1St)

1 1 /8 I ROIll RS I\ I.(dALI.RIG the maximum be, _fit from its investment in basic research. These industries should take RECOMMENDATIONS FOR GOVERNMENT advantage of opportunities to leverage their Balanced Portfolios funds by working cooperatively in generic non- propriet ..ry areas, such as in-situ processing and Maintaining thc ,alth of any research cntcr- -olids processing (see Chapter 6) and health, prise is a difficult undertaking. It requires the safety, and environmental protection (see Chap- recognition that future research advances are ter 71. unpredictable and may substantially reorder the Industry should also continue to commit re- initial priorities assigned to particular areas of sources to academic research, for reasons that investigation. It demands an appreciation that go far deeper than the desirability of additional researchers in different areas of science and funds. The development of any engineering technology, or at different stages in their profes- field, and particularly one as closely linked to sional careers, have different requirements foi manufacturing as chemical engineering, needs support. To maintain the vitality of chemical the intellectual guidance that can only come engineering, and to enable its researchers to from an industry with a stake in research out- make the greatest possible contribution to solv- comes. Also, industry has to be linked aca- ing the problems facing society, a philosophy demia so that new laboratory results can be of support for the field is needed that can be rapidly transferred to product and process de- described .n terms of three "balanced portfo- sign. An industry committed to financial spon- lios": one of priority research areas, one of sorship and personnel e changes with academia funding sources, and one of support mecha- will make sure that the crucial industrial intel- nisms. lectual involvement needed for success exists. The committee's balanced portfolif "fierity Thus, the committee urges that: research areas has already been discussed in the Executive Summary and the preceding chap- industry continue to expand its support of ters of the report. A discussion of the portfolio academic chemical engineering research in each of funding sources is presented later in this of the priority research areas identified in this chapter and, more comprehensively, in Appen- report, and dix A. The following section discusses the policymakers in industry facilitate the ex- committee's views on a balanced portfolio of cha:: 4e of people across the interface between support mechanisms. ry and academia. A number of possible mechanisms for such exchanges are listed under Support Mechanisms in Perspective "Recommendations for Academia." They will be most effective if corporate culture and in- Different research frontiers require different centive structures promote their use. mixes of support mechanisms. The appropriate mix for a particular area depends on several of One area where intellectual involvement from factors, including the nature of the scientific industry is essential is in the area of process area; its requirements for expensive equipment, safety. Several large chemical companies have instruments, or facilities; and tine need for trained committed substantial resources to advancing personnel from that area in the broader econ- their understanding of this area. The generic omy. insights that have emerged from their work need Table 10.1 summarizes a range of possible to be shared with academic researchers and funding mechanisms for research. Many of these educators as well as with smaller chemical mechanisms are well known and need no further companies. The AIChE Center for Chemical elaboration. Some of them. ho: 'ever, are new Process Safety is one mechanism for such trans- mechanisms being proposed by the committee fer of knowledge. Others, including direct con- and are briefly described below. The great tacts between academic and industrial research- number of pos("ble support mechanisms for ers in this area, are needed. academic research should be placed in the

190 2E( (JUME N D4 /70.1 S i79

TABLE 10.1 Possible Support Mechanisms for Chemical Engineering Research Support Mechanism Liks.iy Applicants Single Investigator Awards

Starter grants New investigators Presidential Young Investigator Extremely promising new investigators awards Solicited project grants Younger investigato. Career development awarus Extremely meritorious younger investigators Unsolicited project grants Meritonous midstream investigators Research excellence awards" Extremely meritonous midstream and senior investigators Cross- Disciplinary Awards

Cross-disciplinary pioneer Promising new investigators moving into awards° chemical engineenng from other disciplines Cross-disciplinary partnership Two or three co-pnncipal investigators from awards° different disciplines Equipment and Facibues

Equipment and instrumentation One or more research groups with special grants requirements Regional and national equipment Researchers needint, unique equipment or facilities facilities (e.g.. synchrotrons or neutron- scattenng facilities) Centers and Academic - Industrial Consortia

NSF Science and Technology Collaborative groups of principal investigators Centers Engineering Research Centers Cross-disciplinary engineenng research attacking several aspects of a common problem area Other centers and consortia" Large research centers funded by industry at academic campuses Improving links to small high- Small process technology companies working technology firms° with academic investigators ° New mechanism discussed in the text.

following perspective. Academic chemical en- Because of these advantages, and because gineering research has traditionally taken place many of the most exciting research problems in the milieu of the small research group. Led in chemical engineering lend themselves to the by a single principal investigator, the small scale and environment of the small research research group enjoys the advantages of decen- group, grants to individual academic invecti- tralization and freedom to move in new direc- gators should continue to remain the mainstay tions as opportunities unfold. The environment of the portfolio of support mechanisms for of a small group, where individual students take research. responsibility for significant parts of the group's In this context, how can interdisciplinary work, also facilitates the training of independ- research on the frontiers of the discipline be ent-minded researchers. Much of the vitality of best conducted? There is an obvious role for chemical engineering over its history can be large assemblages of researchers in centers of traced to innovative and exploratory research various types. The committee also believes that groups led by individual principal investigators. the special needs of interdisciplinary research

191 180 1 ROA MRS IV ( III IIll II. LAGIALLRIA6

areas can be met by creating two new funding and compete with other researchers on an eq- mechanisms modeled on the best features of uitable basis in the peer-review system. the small group model. Another type of individual award that belongs Cross-disciplinary partnership awards. The in the balanced portfolio is the type of "research first mechanism is designed to encourage two excellence award' exemplified in the industrial or three small research groups to submit joint world by the IBM Fellows program, These applications. Such a partnership would allow awards would choose extremely meritorious researchers to hegin to communicate across investigators for research support on the basis disciplinary lines without necessitating the cre- of the creativity, productivity, and significance ation of large and formal entities, Cross-disci- of their recent work, rather than on the basis plinary partnership awards should be considered of a project proposal. They would be provided in a variety of research areas where interdisci- funding for themselves and one or two associ- plinary cooperation is vital, and particularly for ates to pursue topics entirely of their own research areas associated with the emerging choosing for a few years. The result is that a technologies described in Chapters 3 through 5 select group of the very best researchers would of this report. A case in point would be bio- receive the opportunity to explore more spec- chemical engineering where more effective links ulative and high-potential research ideas. An in research and training are needed between experimental program of this type, with no more chemical engineers and life scientists. than 15 such awards active at any given time, Cross-disciplinarypioneer awards. The might leaven the entire field of chemical engi- secondproposedinterdisciplinaryfunding neering as well as provide an award that would mechanism would encourage top-quality young represent the pinnacle of individual personal researchers trained in other disciplines (e.g., achievement in chemical engineering research. molecular biology, chemistry, materials sci- Like other fields where the small research ence, and solid-state physics) to accept tenure- team is still a vital and appropriate funding track faculty positions in chemical engineering mode (e.g., chemistry, solid-state physics, bi- denartments at leading universities. The com- ology, and materials science), the cost of con- mittee believes that these departments would ducting cutting-edge research in chemical en- be enriched by the judicious appointment of gineeringis growing rapidly. The complex young faculty who are at the cutting edge of problems of the future will require groups that disciplines important to chemical engineering are larger than today s. The cost of state-of- frontiers. Some appointments of this type are the-art instrumentation and computational fa- already being made, but these "pioneers" face cilities will continue to grow. It is important co .derable obstacles in making the transition that agencies funding chemical engineering re- from the area in which they were trained to search ensure that their investment is made another discipline. These include the difliculties most effectil by realistically estimating the of immersing oneself in a new intellectual area research costs of groups at the frontiers of the while trying to stmt. a new research program field. In Appendix A (Table A.2), the committee that, absent a track record of success, might provides its own estimates of appropriate levels not appeal to proposal reviewers (whether from of support for groups of different , zes.Partic- the pioneer's initial or newfound field) who ularly in a time of budgetary stringency, re- have the more traditional perspectives of estab- search effectiveness is maximized by funding a lished disciplines. The proposed award would smaller number of excellent projects ade- provide start-up research funding for up to 5 quately, rather than funding a large number of years. It would allow cross-disciplinary pioneers projects inadequately. the opportunity to engage themselves fully in Funding agencies with vital programs of sup- chemical engineering research so that they can, port for small research groups may also want by the end of the award, contribute to the to consider implementing one of the following intellectual life and teaching of their department t.vo mechanisms:

192 RECOMMENDATIONS 181

Improvement of links between chemical researchers from academic, industrial, and gov- engineering departments and small process ernment laboratories. The most ambitious pro- technology firms. There is an important problem gram of support for cross-disciplinary centers in some key emerging technologies that is not is 'Lie NSF Engineering Research Centers (ERC) addressed by existing programs in funding agen- Program. The cross-disciplinary character of -. It deals with the generation and transfer many of the frontiers discussed in this report of expertise and ideas from the research labo- makes them fruitful areas for center-based re- ratory to the production line in biotechnology search, a reality that has already been recog- and process technologies for electronic, pho- nized by the establishment of ERCs in biopro- tonic, and recording materials and devices. In cess engineering, compound semiconductors, these areas, a key role in generating new process composite materials, hazardous waste manage- concepts and equipment is played by a large ment, and computer-assisted design. Chemical number of relatively small firms. These firms engineering plays an important role in nearly are capital-poor but rich in problems that would all these centers. benefit markedly from the insights of academic chemical engineers. The United States could Federal agencies should try to stimulate in- significantly boost its competitive position in dustrial interest in consortia to bolster U.S. these areas by facilitating information transfer industrial competitiveness in technology. DOE between academia and this segment of industry. should foster interest in consortia to address The problem for funding agencies with an in- macroscale energy and natural resource pro- terest in promotirg U.S. capabilities in thin area cessing research. Here, the experiments that is how to create incentives for academic and need to be carried out are large in size and industrial researchers to seek out and forge costly. Generic research could be carried out links with one another. in such a way as to (I) allow for fruitful inter- Since the partnerships that the committee change of ideas between industrial and academic would like to see fostered will be individualized researchers, and (2) improve the research effi- to 1?le particular needs and interests of the ciency of the energy processing industries while participants, it recommends that any program leaving each individual energy company free to allow for a wide range of proposed activities. develop its own proprietary processes and tech- Examples of the sort of initiatives that might nologies. be funded include the following: (1) Grants to Another area where the large-scale approach provide special instrumentation and facilities to might be appropriate is in research related to be shared by researchers from chemical engi- environmental protection and process safety. neering departments and high-technology firms. Again, this is an area where cooperationon The facilities or instrumentation would be lo- generic research among companies, as well as cated at a university, but availaNe for use by between industry and academia, focused on ref earchers at the high-technology firms. (2) generic research problems, cou'd lead to sig- Sa ,batical awards for academics at smaller firms nificant research advances. EPA should stim- that are not likely to be plugged into a large ulate such arrangements to expand the amount university-based center. (3) Starter grants for of research needed in this area. industrial liaison programs in chemical engineering departments. A key feature of any of Recommendations for Specific Federal these programs would have to be significant, Agencies ongoing person-to-person contact between academic and industrial research groups National Science Foundation

Large research centers and consortia. Cen- The NSF has a logical role in each of the ters are a common mechanism of support to following frontier research areas. stimulate cross-disciplinary interactions among Biotechnology. NSF should sustain the researchers, or to facilitate cooperation among growth and quality of its existing research sup-

1 9 3 182 FRO\ HERS IN ('ILL-.111C IL IAGINEERING port. This area should be targeted by NSF for ana interface studies. The need for such dedi- new support to cross-disciplinary pioneers, cated instrumentation can be met at a funci,Ig partnerships, instrumentation and facilities, and level of about $i million per year. incentives to improve links between academia and small high-technology firms. To make its most effective contribution to Electronic, photonic, and recording mate- these areas, NSF should consider the following rials and devices. The committee recommends new support mechanisms for chemical engineers. a 5-year pattern of budgetary growth to achieve 30 groups funded at an aggregate level of $6 Cross-disciplinary pioneer awards. The million per year. This area should also be committee recommends that a steady-state pro- targeted for new support to cross-disciplinary gram of 25 awards in the three areas identified pioneers and improved links between academia above be achieved over 5 years, at an average and small high-technology firms. level of $100,000 per award per year. Polymers, ceramics, and composite mate- Research excellence awards. The commit- rials. In addition to continued growth in support tee recommends that a steady-state program of for research on polymers and polymeric com- 15 awards be achieved over 3 years, at an posites, a new thrust is recommended to estab- average level of $100,000 per year per award. lish six to eight centers for the chemical engi- Improvment of links between chemical en- neering of ceramic materials and composites gineering departments and high-technology firms. over the next 5 years, funded at a total annual NSF should create a program to provide incen- level of $4 million per year. Cross-disciplinary tives for partnerships between academic chemical pioneers should also be supported in this area. engineers and researchers in small process tech- Energy and natural resources processing. nology firms in biotechnology and electronics. NSF should sustain its support of basic research in complex behavior in multiphase systems, Department of Energy catalysis, separations, dynamics of solids transport and handling, and new scale-up and design The committee proposes major new initiatives methodologies. for the Office of Fossil Energy and the Office Environmental protection, process safety, of Energy Research (OER) to support in-situ and hazardous waste management. NSF should processing of resources and the development strongly support continued growth in this area, of tomorrow's liquid fuels for transportation. focusing on engineering design methodology for New initiatives are also proposed for these two process safety and waste minimization. In FY offices for the chemical engineerug of advanced 1986, only three chemical engineering groups materials and the development of computational working in this area were funded by NSF, with methods and process controlthe Division of total support of less than $250,000. Engineering and Geosciences in OER should Computer-assisted process and control en- continue and expand its support of fundamental gineering. The committee recommends a 5-year research while the Office of Fossil Energy starts pattern of growth to a program supperting 35 a new program on applications to design and groups with access to state-of-the-art worksta- scale-up of large-scale technologies. An initia- tions, software, and computer networks, in tive is recommended for the Office of Energy addition to the existing ERCs related to this Utilization in chemical and process engineering topic. These smaller groups should receive total for biotechnology applied to renewable sources NSF support of $8 million per year as a target, for chemical feedstocks. Surface and interfacial keyed to additional industry support. engineering should be a strong focus for initia- Surface and interfacial engineering. NSF tives in the OER (e.g., catalysis and colloid should expand its support in this area, with science and technology) and in Office of Energy emphasis on acquisition by chemical engineers Utilization (e.g., corrosion science and electro- of state-of-the-art instrumentation for surface chemical engineering).

19.4 ._ RECOMMENDATuns 183

National Institutes of Health national "Center for Engineering Research on Environmental Protection and Process Safety" The committee recommends that the National that would provide both unique state-of-the-art Institutes of Health undertake an initiative to laboratoryfacilities and computationalre- support basic research in chemical and process sources to chemical and process engineering engineering science for ultimate application to researchers from academia, federal laborato- biotechnology and biomedical devices. Such an ries, and industry. initiative should be targeted towards (1) en- couraging the submission of proposals from cross-disciplinary partnerships between chem- National Bureau of Standards ical engineers and life scientists and (2) shaping The small NBS program in chemical engi- the biochemical engineers of the future by using neering should receive substantially greater Institutional National Research Service Awards funding to fulfill critical needs for evaluated to facilitate the expansion of graduate course data and predictive models. The committee requirements in biochemical engineering to in- supports NBS plans to focus on data needs in clude greater exposure to the life sciences. emerging technology areas such as biotechnology and advanced materials. Department of Defense Bureau of Mines For the Department of Defense (DOD), the main relevant thrust of this report is in the area The committee recommends that the Bureau of materials. The committee recommends that fund a modest number of university-based cen- DOD formulate integrated initiatives (i.e., from ters focused on in-situ processing of dilute basic research to testing and evaluation) on resources. This initiative would complement the problems where improvements in chemical pro- one proposed for DOE. cessing is the key to enhanced performance (e.g., polymer-based optical fiber and components; processing for high-strength, high-mod- RECOMMENDATIONS FOR ulus fibers; manufacturing process technology PROFESSIONAL SOCIETIES for composite materials; and joining and repair science for complex materials systems based American Institute of Chemical Engineers on polymers, ceramics, and composites). At the Every constituent part of the American In- basic end of the research spectrum, these would stitute of Chemical Engineers (AIChE) jour- translate into major initiatives for support of nals, committees, local sections, divisions, and molecular science and engineering. student chaptersshould take up the challenge of examining the frontiers presented in this Environmental Protection Agency report and, in the context of their mission and purpose, seek out ways of rejuvenating the The Environmental Protection Agency should profession of chemical engineering. The AIChE revitalize its research grant program in its Office should set the goal of attracting to its meetings of Exploratory Research and substantially in- and journals significant numbers of researchers crease its support to chemical engineers inves- from other disciplines who are working on tigating important challenges to environmental problems closely related to those at the cutting quality. Stability in the EPA research program edge of chemical engineering. The AIChE should aver several years is needed to attract the best provide awards that recognize the achievements scientific and engineering research talents to of cheri.ical engineering researchers in frontier these problems and to allow them to work areas. The Institute should continue to promote efficiently on their solution. reforms in chemical engineering education that The EPA should also consider creating a would ensure that students would be prepared

1 95 184 FRONTIERS IN CHEMICAL ENGINEERING

for new challenges. Finally, the AIChE should undertake cooperative efforts with other orga- CONCLUSION nizations to advance the interests of the disci- Chemical engineering is r. discipline that in- pline. tegrates the rese rch advances of a number of scientific areas. Paramount among these is chemistry, but fields such as applied mathe- American Chemical Society matics, biology, computer science, condensed- Over 12,000 chemical engineers are members matter physics, environmental science and en- of the American Chemical Society (ACS), and gineering, and materials science also provide ACS meetings, journals, and abstracting ser- important insights that chemical engineers use. vices support chemical engineering research in This report has highlighted chemical engineering important ways. The findings and recommen- accomplishments and frontiers in a number of dations of this report should be of interest and areas touched on by these disciplines. It is concern to the ACS. In fact, many of the important that these other disciplines remain recommendations for the AIChE could be prof- vital, too. itably implemented by the ACS Divisions of Two years ago, the National Research Coun- Industrial and Engineering Chemistry, Colloid cil published a report on research frontiers and and Surface Chemistry, Environmental Chem- needs in chemistry entitled Opportunities in istry, Fuel Chemistry Microbial and Biochem- Chemistry, and known colloquially as the "Pi- ical Technology, Polymer Chemistry, and Pol- mentel Report" after its chairman, George C. ymeric Materials: Science and Technology. The Pimentel. That report identified many of the ACS is a natural partner for the AIChE in joint same areas discussed in this report and focused undertakings to benefit chemical engineering, on how chemists contribute to them. "inis com- and the committee recommends that both or- mittee endorses the recommendations contained ganizations pursue opportunities for future joint in Opportunities in Chemistry, and urges their undertakings. implementation in addition to the recommendations contained in this volume. The two reports, like the two disciplines, should be seen Council for Chemical Research as complementary, not competing. A vital base The Council for Chemical Research i: not a of chemical science is needed to stimulate future professional society, but it deserves mention in progress in chemical engineering, just as a vital this reNrt because it provides a unique forum base in chemical engineering is need( i to cap- for interactions between chemists and chemical italize on advances in chemistry. engineers and between academic and iddustrial researchers. It should be credit:-.! for building effective bridges among these groups and high- NOTE lighting the complementary character of chem- 1. More detailed suggestions for education and train- istry and chemical engineering. The committee ing in separations ate contained in Separations recommends that the Cou eil continue to pro- and Purification: Critical Needs and Opportunities mote these interactions, focusing on the fron- Washington, D.C.: National Academy Press, 1987. tiers in this report that are of mutual interest to chemists and chemical engineers.

19C APPENDIXA

Detailed Recommendations for Funding

CURRENT FUNDING PATTERNS Nearly 90 percent of federal supper( for academic basic and applied research came from More than 18 separate federal agencies funded only two agencies: NSF and DOE. This narrow chemica: engineering research in FY 1985, the funding base is not good for the health of most recent fiscal year for which.tual ex- academic research, which is best served by penditure data are available for all government pluralism among funding sources. Neither is it agencies (Table A.1). Their support to all per- in the best interest of applied and developmental formers of chemical engineering researchac- federal programs that depend on chemical en- ademic, private, and federalexceeded $254 gineering. As was observed at a recent confer- million. Much of this funding, though, was for ence on research and national priorities: research in federal and private laboratories. A good development manager inevitably runs into fundamental problems requir- $120 Industry Support Other Support ing a research solution. The $110 road to a solution is easier if the manager has close ties $100 to the research community. $90 The way to maintain such ties is through the mainte- a$80 nance of an ongoing basic $70 research program in fields underlying the development $60 activity.' $50 This principle underlies industry's support of academic chem- $40 cc ical engineering research. In the $30 last 6 years, industrial support has nearly quadrupled; it has been $20 the main engine for funding $10 growth in academic chemical engineering (Figure A.1). Chemical $0 engineering now leads all engi- 1980 19e2 1984 1986 neering disciplines in the pro- Fiscal Year portion of academic support FIGURE A.1Industrial support of academic chemical engineering nearly coming from industry and other quadrupled from 11:80 to 1986 and is the major factor behind growth in nonfederal sources (Figure A.2).2 academic funding in this period. Data from Council for Chemical Research. Industry is often stereotyped as

185 9 186 . t PP E .% DIA A

TABLEA.1Federal Support for Basic and Applied Chemical Engineering Research in FY 1985 (thousands of dollars) Sponsoring Agency and Subdivision Support to All Performers Support to Academia Department of Agriculture Agricultural Research Service 3.267 0 Cooperative State Research Service 2.353 2.353" Department of Defense Department of the Army 26.689 1,072 Department of the Navy 24.790 414 Department of the Air Force 1,673 332 Defense Agencies 126 0 Department of Commerce INBS) 1,660 250" Department of Energy 136.625 15.182 Department of Health and Human Services (N11-11 Department of the Interior Bureau of Mines Geological Survey° 4.264 0 Minerals Management Service 500 0 Department of Transportation Federal Highway Administration :00 0 Federal Railroad Administration 341 0 Research and Special Programs Administration 60 0 Environmental Protection Agency 18,026 520" Federal Emergency Management Agency 200 N/A National Aeronautics and Space Administration 674 674 National Scienc: Foundation 32,606 27,957 Arms Control and Disarmament Agency Su N/A

TOTAL 254.134" 48,754"

° Estimate. 6 Academic data includes development as well as research Data for chemical engineering not broken out from data for all engineering research or "engineenng. not elsewhere classified." ° Includes the Office of Water Research and Technology. SOURCE: National Science Foundation."

being more oriented toward applications and lowest rate of growth in federal support (Figure less interested in basic research than federal A.3) This is a puzzling pattern to encounter at agencies. Yet the chemical prozessing industries a time when government support for basic have invested in academia in a very enlightened research is widely seen as a way of promoting manner, even during a major recession in the the international competitiveness of U.S. in- early 1980s and several quarters of reduced dustry. profits or slow growth. Presumably, they are Balancing a portfolio of funding sources is supporting basic research because they believe always a good idea, but it assumes even more it will yield insights essential to the long-term importance as chemical engineering depart- profitability of their businesses. ments seek to expand into new areas. This Unfortunately, federal support of academic appendix outlines initiatives for growth in fed- chemical engineering has grown at only meager eral programs to respond to the opportunities rates in the last 6 years. Thus, among engi- facing chemical engineers. The initiatives at- neering disciplines, chemical engi:leering has tempt to match research priorities with the simultaneously experienced the second highest missions and purposes of each agency. rate of growth in industrial support and the

19C APPENDIXA 187

TABLEA.2Costs of Doing Frontier Research in Chemical Engineering Annual Level of Effon" A C NSF" Faculty summer salary (2 mot. 2 18.0 9.0 9.0 1 9.0 8.6 $4,500/mo

Postdoctoral. $27.000/yr 1 27.0 I 27.0 0.5 13.5 0.5 13.5 1.2 Graduate students. $11.000/yr 8 88.0 6 66.0 4 44.0 2 _ 22 0 12.6 Supplies. $6.000/yr/FTE (9) 54.0 (7) 42.0 (5) 30.0 (3) 18.0 8.5 Services or other personnel. (9) 31.5 (7) 24.5 (5) 17.5 (3) 10.5 5.1 $3.500/yr/FTE Equipment and small instru- 10.0 8.0 6.0 5.0 4.9 ments Indirect costs (34%) 77.7 60.0 40.0 26.5 13.9

TOTAL 306.2 236 5 160.7 104.5 54.8

Under each level of effort are two columns. The first gives multipliers for each budgetcategory and the second gives subtotals and totals in thousands of dollars. This column shows the breakdown of costs for the average grant in FY 1986 fromthe NSF Division of Chemical. Biochemical, and Thermal Engineering (CBTE). The 506 grantsmaue by CBTE supported 508 senior investigators for a total of 95 man-years (2.25 months per investigator) at a cost of $4.340,340: 43 postdoctorals(0.085 per grant) for a total of $622.322 of support: 722 graduate students (1.53 per grant) fora total of 56.396.492 of support: 52.567.871 of other personnel costs: $2.451.796 of equipment and instrumentation: and $4.294.378 of other direct costs. Theaverage indirect cost rate for rat CBTE grants was 34 percent. SOURCE: NSF Directorate of Engineering.

THE COST OF FRONTIER RESEARCH tween two research groups. Levels B and C show the costs of large and moderate-sized Before the detailed agency recommendations research groups led by single principal investi- are presented, some general comments will be gators. A group that wants to tackle important made about future research costs. research problems at the frontiers of the disci- 1Pu:suing the frontiers described in this report pline will probably need to be about the size of will require more resources than have teen Level D to maintain both vitality and continuity. needed in the past. Problems of greater com- Throughout the following sections, the cost plexity require larger research groups to make figures in Table A.2 will be used to derive target optimal progress. Access to more sophisticated budgets for proposed initiatives. It should he instrumentation and facilities is costly. Table stressed, though, that the levels of effort shown A.2 presents a range of grant sizes that will be in Table A.2 are not meant to imply that there required to perform frontier research efficiently are only four ways to organize a research effort, in the future. The estimates in this table for or to suggest that an explicitly multitiered sys- stipends and salaries are reasonable estimates tem of support should be introduced. These of current costs in chemical engineering de- levels are meant only to be illustrative of the partments at major universities. Estimates in resources needed today to conduct state-of-the- the table for costs of services, supplies, and art research in chemical engineering. ordinary equipment are close to current NSF averages in chemical engineering. Four different levels of effort are shown, NATIONAL SCrENCE FOUNDATION along with the level of effort supported by the The National Science Foundation is the larg- current average grant from the NSF Division est source of federal support for academic of Chemical, Biochemical, and Thermal Engi- chemical engineering research (see Table A. I). neering (CBTE). Level A shows the costs of a Itr support of the discipline comes through a substantial cross-disciplinary partnership be- variety of programs and divisions (Table A.3).

. 1 9 5,: 188 APPENDIX A

These programs have a logical Non-Federal Support role in each of the priority areas spelled out in this report, and All Engineering increased support from NSF is vital to the goal of expanding chemical engineering into new Aero/Astr areas. The committee proposes the following new initiatives for the Foundation.

Biotechnology and Biomedicine Since publication of a preliminary report by this committee in 1984,5 NSF has increased its support of biochemical engineering by putting in place a new program in biotechnology in its Division of Emerging and Criti- Mechanical cal Engineering Systems, by funding an Engineering Research Center focused on biotechnology Other. n.e.c processing, and by increasing support to the CBTE program 1 on biochemical and biomass en- 10% 20% 30% 140% 50% 60% gineering. The committee ap- Fraction of Support AVER plauds this progress and strongly encourages NSF to sustain the FIGURE A.2Chemical engineering currently leads all engineering disciplines growth and quality of its research in the fraction of its support coming from nonfederal sources. Data from National Science Foundation.' support in this area.

NSF Initiative I made to researchers with biological back- The committee recommends that NSF include grounds who are interested in chemical engi- biochemicai and biomedic I engineering in a neering. larger program of 5year cross-disciplinary pioneer awards (seeChapter 10 for a description). Materials The overall program would be open to candidates from any other discipline with an interest For materials-related priority areas, the com- in chemical engineering, and a steady-state mittee recommends small groups or cooperative program of 25 awards would be achieved over efforts among small groups as the preferred 5 years. The committee recommends that such mode of research organization. There are sev- awards be funded at least at Level D (see Table eral reasons for this: A.2) with thepportunity to grow to Level C if warranted. This would mean an initial award Many of the frontier research questions in the range of $100,000, and a program total of outlinedinChapters 4 and 5 can be profitably at least $2.5 million at steady state. All candi- attacked by adequately supported groups led dates for cross-disciplinary pioneer awards would by a single principal investigator or by multi- compete in one pool, regardless of their area of disciplinary collaborations between small re- interest within chemical engineering. It is likely, search groups. however, that some of the awards would be Some important typesofrest arch questions APPENDIX A 189

EDNon-Federal Support 0 Federal Support engineering community that the MRLs are more directed toward All Engineering physics and perhaps not open to significant participation by chemicalengineers. Itmay be less expensive for NSF te inves- Aero/Astro tigate ale reasons for this perception and to facilitate access by chemical engineers to existing facilities at their home institu- Chemical tions than to create new centers. There already is a significant demand in the materials and electronics industry for chemical en- CM gineers. These personnel needs are likely to grow as future international competition focuses on materials processing. The ex- Electrical isting and anticipated demand for materials-oriented chemical engineers will be most effectively met by a broad-based pattern of Mechanical M support, rather than one concentrated in a few large centers.

Other, n.e.ckNILIVL NSF Initiative 2 In FY 1986, there were only 0% 20% 40% 60% 80% 100% 120% 140% 15 NSF-supported chemical en- aver aver. gineering groups working on the Percentage Growth FY 80 - 85 problems of electronic,pho- FIGURE A.3Paradox. Among engineering disciplines, chemical engineenng tonic, and recording material3 enjoyed the second largest percentage growth in nonfederal support from FY and devices. Thirteen of these 1980 to FY 1985. During this same period, it also expenenced the lowest were supportedbythe Directo- percentage growth in federal support. Data from National Science Foundation., rate of Engineering and shared a total budget of $755,152. The other two were in the Divisior of Materials Research of the Direc- require expensive instrumentation and equip- torate of Mathematical and Physical Sciences. ment that mast be modified extensively by the Their budgets totaled $211,200. Six of these 15 research team in order to perform its experi- groups are led by Presidential Young Investi- ments. In such cases, sharing even the same gators. A 5-year initiative should be put in place type of equipment among groups with different to double the number of groups Norking in this experimental objectives becomes impossible. critical area. This is an achievable goal if the Creating small groups or collaborations best existing groups are allowed to expand in among groups at institutions that have Materials size to produce more faculty candidates, if Research Laboratories (MRLs) can be a cost- existing chemical engineering researchers with effective way to add chemical engineeringex- interests in this area are given the resources to pertise and insights to existing NSF- supported shift their programs, and if some researchers efforts. There is a perception in the chemical from related disciplines elect to becomecross- 190 APPENDIX 4

TABLE A.3 NSF Support of Chemical Engineering in FY 1986 (thousands of dollars). Directorate, Division, and Program Total Directorate of Engineering Office of the Assistant Director 247 Electrical, Communications, and Systems Engineering 54 Chemic il, Biochemical, and Thermal Engineering Kinetics and Catalysis 2.841 Biochemical and Biomass Engineering 2,731 Process and Reaction Engineering 3,248 Multiphase and Interfacial Phenomena 2,155 Separation and Purification Processes 2,870 Thermodynamics and Transport Phenomena 4.322 Thermal Systems and Engineering 2,381 Mechanics, Structures, and Materials Engineering 1,125 Design, Manufacturing, and Computer Engineering 377 Emerging and Critical Engineering Systems Biotechnology 2,667 Bioengineering 15 Cross-Disciplinary Research Engineering Research Centers 3,638 Industry-University Cooperative Research Projects 243 Industry-University Cooperative Research Centers 508 Directorate of Mathematical and Physicai Sciences Chemistry 425 Materials Research 3,133

TOTAL 32,981 ' NSF support of chemical engineering research by all performers. SOURCE: NSF Directorate of Engineering and NSF Directorate of Mathematical and Physical Sciences.

disciplinary pioneers in chemical engineering NSF Initiative 3 departments. An appropriate steady-state group size for research in this area would be some- The chemical engineering of polymers or where between Lev:Is B and C. This might composites was the subject of at least 34 grants result in an eventual proiuction of about 40 in the Directorate of Engineering in FY 1986 Ph.D. researchers per year with expertise in the (totaling $2.37 million), 12 in the Division of broad range of materials and devices for infor- Materials Research of MPS (totaling $2.83 mil- mation storage and handling. lich.), and an Engineering Research Center grant It is somewhat surprising that the average of $1.25 million. In contrast, there was virtually size of the 13 Engineering Directorate grants in no identifiable NSF support in FY 1986 for the this area is only about $58,000. The PYIs are chemical engineering of ceramics.In addition obviously getting industry co-funding, bu the to continued growth in support for research on total size of individual programs in this area polymers and polymeric composites, a major must still be below the optimum level. A min- new thrust is recommended in the chemical imum target for research support, apart from engineering of ceramics.An initial thrust might special instrumentation and facilities, should be be to solicit proposals to es:ablish a number of about $6 million by the and of the proposed university-based centers on the chemical engi- initiative. Industrial co-funding could be re- neering of ceramics that could then lay the quired to obtain state-of-the-art equipment or foundation for a more broadly based research to upgrade facilities. effort. Cross-disciplinary interaction between

2' i 2 APPENDIX t 191

chemists, chemical engineers, and ceramists of Maryland established in 1985. In FY1986, would have to be a key featu -e of these centers. NSF support of the chemical engineeringre- One could imagine such centers being about search at this center was reported to be $2.24 twice the size of a Level A researchgroup. If million. However, this is the totalsupport re- six to eight such centers were set upover the ceived b' that center in 1986, and only about next 5 years, their steady state cost (less special 25 percent of the work being carried out is in equipment and instruments) would be in the chemical engineering. In 1986,an Engineering range of $4 million per year. In addition to Research Center on design was established at centers, cross-disciplinary pioneers should be Carnegie M ion University withan initial grant supported in this aim from NSF of $2.0 million. Again, about 25 percent of the center's eff-,rt is in chemical Processing of Energy and Natural Resources engineering. Thus, NSF has committedto annually fund about $1 million in chemical engi- NSF Initiative 4 neering research in design methodologyover the next few years through these two ERC:,. National Science Foundation programs in Twenty-two other research groups incom- catalysis; multiphase systems; separations; dy- puter-assisted process and control engineering, namics of solids transport and handling; and six of which are led by PYIs, received $1.91 methodologies for design, scale-up, and control trillion in funding from NSF in FY 1986. While play a key role in supporting more applied the average grant size for thisgroup of inves- research on processing of energy and natural tigators (about $86,800) is much larger than the resources. These NSF programs must be sus- average grant size for the CBTE Division,a tained and nurtured. comparison with the levels of effort in Table A recent report from the National Research A.2 shows that in absolute terms thesegrants Council recommends that the NSF Separation still do not provide for a very substantialpro- and Purification Processes Program receivea gram. The committee recommends a major substantial increment in its budgetover the nex. initiative for NSF: a 5-year pattern of growth 5 years.6 The committee endorses thoserec- from the current 22 Levcl D grantsto 35 Level ommendations. B grants. These groups will also needaccess to state-of-the-artworkstations,software,and Environmental Protection, Process Safety, computer networks. Strong co-funding from and Hazardous Waste Management industry in addition to the NSF Level Bgrants will help to meet this need, as wellas the need NSF Initiative 5 for periodic upg ades. At the end of 5years, the NSF investment in this area, exclusive of The National Science Foundation should the ERCs, should be at a total level of about strongly support growth in this researcharea, $8 million. At steady state, this initiative will with a special focus on engineering design and produce about 50 new Ph.D.sper year with control methodology for process safety and expertise in computer-assisted design, control, waste minimization. In FY 1986, only three and operations. They will havean immense chemical engineering groups working in this impact on chemical engineering education and area were supported by NSF, with combined practice. support of less than $240,000.

Surface and Interfacial Engineering Computer-Assisted Process and Control Engineering NSF Initiative 7 NSF Initiative 6 The National Science Foundation shouldex- pand its support to surface and interfacialen- Chemical process systems is one focus ofan gineering, focusing on surface chemistry,ca- Engineering Research Center at the University talysis, electrochemistry, colloid and interfacial

203 192 ,iNIA MX ,t

phenomena, and plasma chemistry. State -of- the -art research in these areas is very costly, DEPARTMENT OF ENERG Y because experimental apparatus must be tai- The Department of Energy has wide-ranging lored to individual experiments. Expensive in- programmatic interests to which chemical en- struments ($200,000 to $500,000) are often ex- gineering can make important contributions. tensively modified it, the course of studies and These include familiar areas of fossil resource become, for all purposes, instruments dedicated production and processing, catalysis, separa- to a particular group. For such studies, there tions, and nuclear energy. They also include are few financial savings to be realized from less familiar areas such as material, process assembling investigators into centers. The com- design and control, and molecular biology. mine.] recommends that the NSF provide funds for chemical engineering groups to acquire In-Situ Processing of Energy and Mineral sophisticated instrumentation for studying sur- Resources faces, interfaces, and microstructures. The committee estimates that, in a given year, DOE Initiative 1 somewhere between 10 and 25 of the active groups in surface and interfacial engineering The committee's prime initiative for DOE is will need to acquire a major instrument for the support of research on fundaniental phe- adaptation and use. A funding level of $5 million nomena important for in-situ processing of hy- per year for major dedicated instrumentation drocarbon resources. (A related initiative for can meet most of these needs. the Bureau of Mines is discussed later in this appendix.) The important fundamental problems in this area are outlined in Chapter 6. The Research Excellence Awards size and extreme complexity of the environments in which these phenomena occur will NSF Initiative 8 require expensive, large-scale, prolonged field The committee recommends that a steady- experiments. Such large-scale research, though, state program of 15 research excellence awards will be quite different from the demonstration in chemical engineering be achieved over 3 projects funded by DOE in the late 1970s and years. This new mechanism is described in early 1980s. Rather than demonstrating the Chapter 10. Because these awards are intended maturity of technologies and their readiness for to fund speculative high-potential research, they commercial application (an activity in which, it will probably work best in the milieu of a small has been argued, the federal government should research group. Thus, the committee recom- not be involved), the focus of large-scale fun- mends ththey be funded at about Level D. damental research proposed for this initiative The steady-state cost of the initiative would be would be to build a nonproprietary knowledge $1.5 million. base relating experimental results on the smaller scales of test systems and equipment to results obtained in the larger and more complex systems Conclusion found in the field. The committee's recommendations for NSF A variety of support mechanisms for carrying target about $22 million in growth over the next out such research could be envisioned that 5 years in six major initiatives, and a less would include sponsorship of individual re- determinate amount of growth in the other two search projects in academia or federal labora- initiatives. The six major initiatives would add tories, where appropriate; a DOE equivalent of 84 new research groups to chemical engineering the NSF Engineering Research Centers, but over 5 years, a 17 percent increase in the number with more cooperative involvement from in- of groups funded. In terms of dollars, the dustry; and stimulation by DOE of industrial initiatives would amount to a rough doubling of consortia both to carry out joint research among the amount that chemical engineering research companies on nonproprietary topics and to received from the CBTE Division in FY 1986. support relevant research in academia. The

2t,; APPENDIX A 193

importance of this research is such that sub- program has been in the areas of catalysis and stantial interest in cooperative research alight separations. Given the broad range ofenergy be generated in industry if DOE took a lead applications in which the structure and chem- role in providing stimulus and partial funding. istry of interfaces is important, the committee recommends that the Division undertake an Liquid Fuels r ". Future ;nitiative in the chemical control of surfaces, interfac-s, and microstructures. This would include support of work by both chemists and DOE Initiative 2 chemical engineers in the areas of surface chem- A second research initiative for DOE centers istry, plasma chemistry, and colloid and inter- on facilitating progress towards the trxt gen- facial chemistry. eration of liquid fuels. This is a very broad topic, encompassing many areas including ca. talyzis (see Chapter 9), solids processing, sep- Microstructured Materials arations, materials development, and advanced scale-up and design techniques. The Office of DOE Initiative 5 Energy Research (OER) and the Office of Fossil Materials in general, and ceramics in partic- Energy should work together to cc,ordinate ular, are heavily emphasized in the DOE Divi- research in these areas. Some research areas, sion of Material Sciences. Up to now, this notably solids processing, may require the same program has had relatively little involvement type of large-scale fundamental research called from chemical engineers. Given that chemical for in the previous research initiative. The processing approaches to ceramics have a bright mechanisms proposed there for stimulating large- future, both for structural applications and pos- scale research may be applical-le here, as well. sibly for ceramic superconductors, the OER should consider a major thrust in chemical Advanced Computational Methods and pro, essing of materials, with a view towards Process Control the more facile production of defect-free ceramics for energy and energy-saving applications. DOE Initiative 3 The Division of Engineering and Geosciences in OER supports cutting-edge research in fluid Biotechnology dynamics and process design and control. These programs should be sustained as a vital part of DOE !nitiative 6 the balanced portfolio of support for these areas Bioprocessing is of interest to the DrJE Di- within chemical engineering. Process design, vision of Energy Conversion and Utilization scale-up, and control have already been men- Technologies (ECUT), which is concerned with tioned as important keys to in-situ processing. increasing the efficiency of energy conversion The Office of Fossil Energy should consider and the use of renewable resources. A recent setting up a research program in this area that report of the National Research Council pro- would support fundamental process design and poses a comprehensive program in this area for control research that would be particularly ap- EC UT, involving both chemical engineering and plicable to large-scale projects. the life sciences, and funded for :0 years at an annual level of $10 miliion.7 The committee endorses these recommendations and urges their Surface and Interfacial Engineering implementation. DOE Initiative 4 The Division of Chemical Sciences in OER NATIONAL INSTITUTES OF HEALTH supports basic chemical research. The primary The National Institutes of Health is the pre- involvement of chemical engineers with this mier sponsor of health-related research in the

2f15 194 11TE ,1 Ill XI

United States. Its long-term support of the basic NIHInitiative 2 biosciences is responsible for the advances that The National Institutes of Health could also have made biotechnology possible. Sophisti- play a vital role in shaping the biological chem- cated engineering will be needed, though, if ical engineers of the future by using Institutional biotechnology is to make its full potential con- National Research Service Awards to i 3vide tribution to the nation's health. NIH supports biochemical engineering graduate students with a great deal of research aimed at elucidating a greater exposure to the life sciences. A sig- molecular processes in living systems; identi- nificant limiting factor in expanding the graduate fying molecules of potential therapedtic value; curriculum in this way is the problem of sup- and developing potential routes to them, whether porting students for an additional year prior to via synthetic chemistry or recombinant DNA their immersion into grant-supported research. organisms. It provides less support to the problem of turning these potential synthetic routes into practical, economic processes. In part this DEPARTMENT OF DEFENSE is because NIH has traditionally focused on The chemical engineering research frontiers basic science, leaving commercialization of dis- of most relevance to the Department of Defense coveries to others. But there is a knowledge are in materials. Faster electronic devices, more gap in the basic engineerir; science for bio- reliable communication systems, and stronger technology and biomedicine that is not being structural components are all needed by DOD filled by industry. This gap impedes the full in order to fulfill its mission. Chemical process- conversion of new biological knowledge into ing is a valuable tool to tailor these materials products and therapies for improving human for specific military uses. health. There is a role for NIH, consistent with A special strength of the DOD research in- its historical mission and philosophy, to (1) frastructure is its vertical integration from basic expand the base of fundamental knowledge in research, through applied and exploratory re- the chemical engineering of biological systems search, to advanced testing and evaluation of and (2) train a new generation of chemical technologies in the field.The committee rec- engineers to be more conversant with the bio- ommends thatDODexploit this strength by logical sciences. Both steps will allow chemical formulating integrated initiatives around topics engineers of the future to expertly apply engi- where advances in chemical processing can neering principles to biological problems. exert leverage.The development of more secure signal and communication systems might be one such topic. Chemical processing plays an im- NIHInitiative 1 portant role in the manufacture of both glass- The committee recommends that NIH un- based and polymer-based optical fibers. The dertake an initiative to bring chemical engi- latter are more easily fit to conrectors and neering researchers into more effective contact attached to one another. They might be most with biological and medical researchers. The appropriate for defense systems requiring trans- mechanism for accomplishing this would be mission of data over short distances (measured through cross-disciplinary partnerships in re- in meters) rather than long distances. An initi- search, with groups ranging upwards in size ative to develop practical polymer-based com- from a traditional research project led by two munications systems would require a significant co-principal investigatorsone from chemical basic research effort in polymer chemistry and engineering and one from the life sciencesto chemical engineering to resolve materials chal- program projects and perhaps even centers lenges to low-loss optical transmission in poly- involving many investigators from engineering, mers, a significant effort in chemical ergir.eering biology, and medicine. At the low end of this to provide the fundamental insights needed to spectrum, a level of funding similar to Level A fabricate optiLol-quality polymer lenses, split- in Table A.2 would allow for a substantial ters, and connectors, as well as research in interdisciplinary effort. related disciplines (e.g., electrical engineering)

2 ( G 1

APPENDIX A 195

to develop overall system concepts and design cidate joining and repair on the molecular level, and to produce a system that could he integrated and to integrate new insights from the molecular into existing hardware. The potential payoffs of level into the contributions that would be made success in this initiative would be enormous, on the systems level by other disciplines (e.g., both in terms of national security (e.g., secure materials, mechanical, and aerospace engineer- optically based communicatiGns systems cheap in:). enough to install nationwide) and in terms of society's need for rapid, efficient transmission of data (e.g., optically based local telephone ENVIRONMENTAL PROTECTION AGENCY systems and local area networks capable of simultaneously transmitting enormous quan- EPA Initiative I tities of Jigitized voice and data signals). The Environmental Protection Agency needs This is just one example of several such DOD a strong basic research program, especially in initiatives that might be built around scenarios chemical science and technology. The commit- that assume that basic chemical problems in tee urges EPA to revitalize its research grant materials could be solved by concentration of program in the Office of Exploratory Reseaich. resources on fundamental research. One can As part of this revitalization, EPA should seek just as easily imagine other initiatives, such as to fund the best chemical engineering research the following: groups investigating important ongoing challenges to environmental quality: The high strength of Kevlar® fibers is due to the way in .hich they are processed, rather fundamental chemical processes important than their intrinsic chemical composition. If in the generation and control of toxic substances processes could be found that were capable of by combustion, creating in other materials the highly ordered chemical processes involved in the trans- structure seen in Kevlaro, it might be possible port and fate of hazardous substances in the to fabricate extremely durable treads for use in environment, and modern infantry vehicles and tanks. Such a design methodologies that could result in practical focus might be used as an organizing waste and process hazard minimization in chem- focus for a substantial program of fundamental ical manufacturing plants. research on polymer processing for high strength These areas all nromise substantial advances in and toughness. improving our environment, but will not yield Most composites used in aircraft must be that promise in an atmosphere of on-again, off- "laid up" by hand, because a reliable manufac- again funding. Stability in the EPA research turing technology for composites has yet to be program over several years is needed to attract discovered. Chemical processing combined with the best scientific and engineering research textile engineering could be used to achieve talents to these problems and to allow them to major advances in the manufacture of reliable work efficiently on their solution. composites for major structural components of aircraft. Another composites problem is joining and EPA Initiative 2 repairing these materials. Unlike metals, where The EPA should consider establishing a na- patches can literally be bolted onto a system tional Center for Engineering Research on En- without substantially degrading performance, vironmental Protection and Process Safety perfonr.2nce it composites is very sensitive to (CERES), modeled on the National Center for the means t.,:, which composite components are Atmospileric Research. Chemical and process joined to each other, or repaired The practical engineering researchers would benefit froma problem of repair. the composite aircraft of special collection of state-of-the-art laboratory the future cculd he focus of a significant facilities and computational resources dedicated fundamental chemical engineering effort to elu- to research on environmental protection, pro-

fa 21 ) 7 1% ...1 196 iPPEND/A 4 cess safety, and hazardous waste management. mixtures found in bioprocessing systems. The As a centralized fatality, CERES could serve a committee endorses this effort and encourages coordinating role to enhance cooperation in the NBS to provide the needed ievel of funds research across institutional boundaries and to for an optimal effort. In the materials area, the diffuse rapidly into industry research advances need for international standards for advanced made in academic and other laboret7N :z.s. The materials, such as polymer blends and ceramics, specific tasks and possible organization of such is acute. Again, the NBS has started an effort a center, as well as is potential relationship in this area as part of the international Versailles with a new NSF center on hazardous waste Project on Advanced Materials and Systems management, should be 'he subject of an in- (VAMAS). Currently, about 100 U.S. research- depth study by EPA and any competition for ers are involved in VAMAS-related research, siting and 3perating this center should be open in both industry and academia. The amount of to the most meritorious proposal,'. Ether it federal funding for this effort, though, is less originates from a university, a federal 'abora- than $500,000. This type of project is extremely tory, the nonprofit sector, or some coloination important to the rapid worldwide development of the three. of advanced materials, and should be funded at a level more commensurate with that importance. NATIONAL BUREAU OF STANDARDS

The National Bureau of Standards has a BUREAU OF MINES unioue role to play in supporting the field of chemical engineering. It should be the focal The Bureau of Mines, within the Department point for providing evaluated data and predictive of the Interior, funds a substantial amount of models for data to facilitate the design, the chemical engineering research in its in-house scale-up, and even the selection of chemical laboratories, particularly in the area of hydro- processes for specific applications. metallurgical separation processes. The U.S. Despite the plethora of data in the scientific minerals industry is currently in a depressed literature on thermophysical quantities of sub- state typified by diminished research efforts stances and mixtures, many important data gaps within industrial laboratories and, in some cases, exist. Predictive capabilities have been devel- wholesale termination of research operations. oped for problems such as vapor-liquid equilib- As a result, new researchers have bleak pros- rium properties, gas-phase andless accu- pects for industrial employment. At the same ratelyliquid-phase diffusivities, and solubilities time, the United States cannot afford to !ose a of nonelectrolytes. Yet there are many areas professional generation of research per,onnel where improved predictive models would be of in an area that would be of critics' importance great value. An accurate and reliable predictive if foreign supplies of certain metals were inter- model can obviate the need for costly, extensive rupted. experimental measurements of properties that The committee recommends that the Bureau are critical in chemical manufacturing process- fund a modest number of university-based cen- es. ters focused on in-situ processing of dilute Particular attention should be given by NBS resources. This initiative would complement the to data needs in the emerging technology areas najor one proposed for DOE. Such centers served by chemical engineering (i.e., biotech- Jhould explicitly focus on generic themes, such nology and materials). In the area of biotech- as separations from highly dilute solutions, nology, the NBS is attempting to identify and multiphase flow though porous media, or the assign priority to the thermophysical properties development of sensors and other instrumen- of greatesimportance to scale-up and com- tation. The goal of the centers program would mercialization, and to identify promising theo- be to stimulate fresh ideas and insights in metals- retical approaches that could lead to generic related processing research and to train a new predictive models for the types of complex generation of research engineers flexible enough APPENDIX A 197 either to move into a revitalized minerals in- Washington, D.0 : National Science Foundation, dustry or to find employment in the broader 1986. sector of process industries. 4. National Science Foundation, Division of Science Resources Studies.Federal Support to Universi- ties, Colleges, and Selected Nonprofit Institu- tions, Fiscal Year 1985.Washington, D.C.: Na- NOTES tional Science Foundation, 1987. 5. National Academy of Sciences-National Academy I .National Academy of Sciences, Government-Uni- of Engineering-Institute of Medicine, Committee versity-Industry Research Roundtable.What Re- on Science, Engineering, and Public Policy. "Re- search Strategies Best Serve the National Interest port of the Research Briefing Panel on Chemical in a Time of Budgetary Stress? Report of a and Process Engineering for Biotechnology," in Conference.Washington, D.C.: National Acad- Research Briefings 1984.Washington, D.C.: Na- emy Press, 1986. tional Academy Press, 1984. 2. National Science Foundation, Division of Science 6. National Research Council, Committee on Sepa- Resources Studies.Academic Science/Engineer- ration Science and Technology.Separation and ing: R&D Funds, Fiscal Year 1985.Washington, Purification: Critical Needs and Opportunities. D.C.: National Science Foundation, 1986. Washington, D.C.: National Academy Press, 1987. 3. National Science Foundation, Division of Science 7. National Research Council, National Materials Resources Studies.Federal Funds for Research Advisory Board.Bioprocessing for the Energy- and Development, Fiscal Years /985, 1986, and Efficient Production of Chemicals.Washington, /987,Volume XXXV (Detailed Statistical Tables). D.C.: National Academy Press, 1986.

21 W APPENDIXB

Contributors

The Committee on Chemical Engineering Frontiers: Re- search Needs and Opportunities would like to express gratitude to the following individuals who provided suggestions, pres- entations, written submissions, or critiques that were used in the preparation of this report. In many cases, these inputs were solicited by one of the panels involved in the writing of this report. The organizational affiliation of contributors at the time Ciey provided their input is also listed. Contributors of figures and photographs for the report are acknowledged where their submissions appear. The committee, of course, is responsible for the final content of the report.

ABRAMOWITZ, P. H., St. Joe Minerals BOHRER, M., Ni &T Bell Laboratories Corporation BORTZ, S. A., IIT Research Institute AGARWAL, J., Charles River Associates BOSANQUET, L. Monsanto Company AHEARNE, J. H., Resources for the Future BOUDART, M., Stanford University ALKIRE, R. C., University of Illinois BRATON, R., IBM ALLRED, V. D., Marathon Oil Company BRIAN, P. L. 1., Air Products and Chemicals, ALPAN, F. F., Pennsylvania State University Inc. ALPERT, S. B., Electronic lower Research BRISTOL, E. H., Foxboro Company Institute BROWN, R. A., Massachusetts Institute of ANDERSON, H. R., IBM Technology ANDERSON, T. J., University of Florida BROWNAWELL,D., Exxon Chemical ANGUS, J. C., Case Western Reserve Company University BUILDER, S. E., Genentech, Inc. APPLEBY, A. J., Electric Power Research BURGER, R. M.,Semiconductor Research Institute Corp. ARKUN, Y., Rensselaer Polytecnnic Institute CHIEN, H. H., Monsanto Company ASTROM, K. J., I -Ind Institute of Technology COOPER, W., Eastman Kodak Company ATHERTON, R. W., Sentry Computer Cox, R. K., E.I. du Pont de Nemours & Co., Integrated Manufacturing Systems Inc. Aztz, K., Stanford University CROWE, C. M., McMaster University BAEDER, D. L., Consultant DAHLSTROM, D. A., University of Utah BAILEY, J. E., California Institute of DEEN, W. E., Massachusetts Institute of Technology Technology BAXTER, J., Illinois State Geological Survey DOHERTY, M. F., University of BENNION, D. N., Brigham Young University Massachusetts BLANCH, H. W., University of California, DORSCH, R. R., E.I. du Pont de Nemours & Berkeley Co., Inc.

198 210 APPEND! V B 199

DOSCHER, T. M., The Doschers Group KAHN, L., Illinois State Geological Survey Doss, J. E., Tennessee Eastman Company KALELKAR, A. S., Arthur D. Little, Inc. ECONOMY, J., IBM Almaden Research Center KALHAMMER, F., Electric Power Research EGGERT, R. G., Pennsylvania State Institute University KARDOS, J. L., Washington University, St. EIDEL, J., Illinois State Geological Survey Louis ERLINGER, H., Illinois State Geological KELLOG, H. J., Columbia University Survey KEMP, H. S., E.I. du Pont de Nemours & EVANS, J. W., University of California Co., Inc. FAULKNER, L., University of Illinois KING, A. G., Ferro Corporation FEFFERMAN, G. B., AT&T Bell Laboratories KING, C. J., University of California FISHER, D. G., University of Alberta KLEIN, L. C., Rutgers University GEORGAKIS, C., Lehigh University KNIGHTS, J., Xerox Palo Alto Research GORBATY, M. L., Exxon Research & Center Engineering Co. KRAMER, M. A., Massaschusetts Institute of GREEN, D. W., University of Kansas Technology GREGOLI, A. A., Cities Service Research LARRABEE, G. B., Texas Instruments, Inc. Company LEES, J. K., E.I. du Pont de Nemours & Co. GUTOFF, E. B., Polaroid Corp. LEESLEY, M. E., Austin, Tex. HALE, J. C., E.I. du Pont de Nemours & Co., LEIGHTON, M. W., Illinois State Geological Inc. Survey HAMMOND, G. S., Allied Corporation LEWIS, D., GAF Corporation HANISCH, W., Cetus Corporation LEWIS, I. C., Union Carbide Corporation HitSHIMOTO, I., Kyoto University LIN, CHENG-YIN, AT&T Bell Laboratories HELT, J. E., Argonne National Laboratory Liu, H. T, Union Carbide Corporation HENCH, L. L., University of Florida LYNCH, R. W., Sandia National Laboratory HENLEY, E. J., University of Houston- MACDONALD, S., IBM Almaden Research University Park Center HENRIE, T. H., U.S. Bureau of Mines MAH, R. S. H., Northwestern University HERBST, J., University of Utah MALLINSON, J. C., University of California, HIRASKI, G. J., Shell Development Company San Diego Hi.AVACEK, V., State University of New MANZIONE, L. T., AT&T Bell Laboratories York, Buffalo MARGOLIN, S. V., Consultant HOPFENBURG, H. L., North Carolina State MATHIS, J. F., NL Industries University MATTSON, J. F., Sepracor, Inc. HORWITZ, E. P., Argonne National MAZDIYASNI, K. S., Wright-Patterson Air Laboratory Force Base HOWES, M. A. H., IIT Research Institute MEMERING, M. N., Pfizer Pigments, Inc. HUGHES, R., Illinois State Geological Survey MERRIMAN, J. R., Union Carbide Corporation HUGHES, T. R., Chevron Research Company MERZ, J. L., University of California, Santa HUNT, A. J., Lawrence Berkeley Laboratory Barbara IRVING, J. P., Chevron Oil Field Research MILLER, J., University of Utah Co. MILLER, S. D., Ampex Corporation JACQUES, D., Exxon Chemical Company MORARI, M., California Institute of JEFCOAT, I. A., University of Alabama Technology JEZL, J. L., Amoco Chemicals Company MORRISON, D. L., IIT Rescaich Institute JEROME, J., Northwestern University MOTARD, R. L., Washington University JOHNSON, D. W., AT&T Bell Laboratories MUNTER, J. D. JONES, F. N., North Dakota State University MURATA, H., The Furukawa Electric Co., JUBA, M R., Eastman Kodak Company Ltd. 209 APPENDIX B

NAGEL, S. R., AT&T Bell Laboratories Scum J. W,Chevron Research Company NEWMAN, J. S., University of California (retired) NISHIMURA, A., 0 Konica Technology Inc. SHAH, S. L., University of Alberta O'BRYAN, H. M., JR., AT&T Bell SHAH, Y. T., University of Pittsburgh Laboratories SHAW, D., Texas Instruments, Inc. Orro, R. J., Lawrence Berkeley Laboratory SHAW, H., New Jecey Institute of OUBRE, C., Shell Development Company Technology OVERCASH, M. R., North Carolina State SIEG, R. P., Chevron Research Company University (retired) PALSSON, B., University of Michigan SKALNY, J. A., W. R. Grace and Company PEPPAS, N. A., Purdue University SMYRL, W. H., University of Minnesota PETRICK, M., Argonne National Laboratory SMITH, D., E. I. du Pont de Nemours & Co., PHILLIPS, J., Boeing Computer Services, Inc. Inc. PIMEVTAL, D., Cornell University SNEDIKER, D. K., Battelle Columbus PIPES, R. B., University of Delaware Laboratories POSNIK, A., Ferro Corporation SOHN, H. Y., University of Utah PRATER, D., California Institute of SOLOMON, P. R., Advanced Fuel Research Technology SQUIRES, A. M., Virginia Polytechnic PRETT, D. M., Shell Development Company Institute PROGELHOF, R. C., AT&T Bell Laboratories STADTHERR, M. A., University of Illinois PRUD'HOMME, R. K., Princeton University STEINDLER, M. J., Argonne National RAJAGOPALAN, R., Rensselaer Polytechnic Laboratory Institute STEPHANOPOULOS, G., Massachusetts RANKIN, J. D., Imperial Chemical Industries Institute of Technology RAWSON, N. E., IBM STEVENSON, F. D., U.S. Department of RAY, W. H., University of Wisconsin Energy REKLAITIS, G. V., Pur,!ue University TEBO, P. V., E. I. du Pont de Nemours & RICE, R. W., W. R. Grace and Company Co., Inc. RIPPIN, D., ETH-Zentrum THEMELIS, N. V., Columbia University THURSTON, C. W., Union Carbide ROBERTSON, J. L., Exxon Research and Engineering Co. Corporation TOBIAS, C. W., University of California RODRIGUEZ, F., Cornell University TRACHTENBERG, I., Texas Instruments, Inc. RoGERS, K., National Science Foundation ULMER, K., Genex Corporation ROLF, M. J., Owens/Corning Fiberglas UMEDA, T., Chiyoda Chemical Engineering & ROSENBERG, R. B., Gas Research Institute Construction Co., Ltd. ROSSEN, R. H., Exxon Production Research VALENTAS, K., Pillsbury Corporation Company WADSWORTH, M., University of Utah ROSTENBACH, R. E., National Science WANG, D. I. C., Massachusetts Institute of Foundation Technology RUSHAK, K., Eastman Kodak Co. WENDT, J. 0. L., University of Arizona SATHER, N. F., Argonne National Laboratory WHITE, J. R., Consultant SAV 'CKI, E., Microsafe, Inc. WHITELEY, R. L., Research and Innovation SAXENA, A. N., Gould AMI Semiconductors Management SCHOENBERGER, R. J., Roy E. Weston, Inc. WHITESIDES, G. M., Harvard University SCHWARTZ, B., IBM WYMER, a. G., Oak Ridge National Sam, C. D., Oak Ridge National Laboratory Laboratory

2 i 2 APPENDIXC

The Chemical Processing Industries

Any discussiln of the chemical industry must make clear how that industry is being defined. Chemistry, chemicalpro- cessing, and chmicals are ubiquitous in modern society and industry. In this, report, where the term "chemical industry" is used in connection with statistics or economic data, the data refers to the industry segment defined by Standard Industrial Classification (SIC) code 28. Where the term "chemical processing industries" or "CPI" is used in this report, a broader range of industries is meant. The definition of the CPI adopted by this report is the same one used by Data Resources, Inc., and Chemical Engineering magazine in their reporting on industry. The industry segments and SIC codes subsumed by this definition of the CPI have been supplied to the committee by Chemical Engineering and are reprinted here with permis- ,ion.

Industry Segment SIC Code Chemicals (including petrochemicals) Alkalies and chlorine 2812 Industrial gases 2813 Industrial inorganic chemicals, n.e.c.* 2819 Synthetic rubber (vulcanizable elastomers) 2822 Gum and wood chemicals 2861 Cyclic crudes and intermediates 2865 Industrial organic chemicals, n.e.c. 2869 Chemicals and chemical perparations, n.e.c. 2899

Plastics materials, synthetic resins, and nonvulcanizable elastomers 2821 Man-made fibers Cellulosic man-made fibers 2823 Synthetic organic fibers, except cellulosic 2824 Drugs and cosmetics Biological products 2831 Medicinal chemicals and botanicals 2833 Pharmaceutical preparations 2834 Perfumes, zosmetics, and other toilet preparations 2844

201

2i 3 202 1/11: ND/ I II

Industry Segment SIC Code Soap, glycerin, cleaning, polishing, and related products Soap and other detergents except specialty cleaners 2841 Specialty cleaning, polishing, and sanitation preparations, except soap and detergents 2842 Surface active agents, finishing agents, sulfonated oils, and assistants 2843

Paints, varnishes, pigments, and allied products Inorganic pigments 2816 Paints, varnishes, lacquers, enamels, and allied products 2851 Fertilizers and agricultural chemicals Nitrogenous fertilizers 2873 Phosphatic fertilizers 2874 Fertilizers, mixing only 2875 Pesticides and agricultural chemicals, n.e.c. 2879 Petroleum refining and coal products Petroleum refining 2911 Paving mixtures and blocks 2951 Asphalt felts and coatings 2952 Lubricating oils and greases 2992 Coke and by-products: part of SIC 3312"Blast furnaces (including coke ovens), steel works, and rolling mills" 3312 Rubber products Tires and inner tubes 3011 Rubber footwear 3021 Reclaimed rubber 3031 Fabricated rubber products, n.e.c. 3069 Leather tanning and finishing 3111 Foods and beverages Condensed and evaporated milk 2023 Wet corn milling 2046 Cane sugar, except refilling only 2061 Beet sugar 2063 Malt beverages 2082 Malt 2083 Wines, brandy, and brandy spirits 2084 Distilled, rectified, and Mended liquors 2085 Flavoring txtracts and syrups, n.e.c. 2087 Roasted coffee, includes instant coffee 2095 Food preparations, n.e.c. 2099 Fats and oils Cottonseed oil mills 2074 Soybean oil mills 2075 Vegetable oil mills, n.e.c. 2076 Animal and marine fats and oil 2077 Shortening, table oils, margarine, and other edible fats and oils, n.e.c. 2079

2i APPENDIX c 203

Industry Segment SIC Code Wood, pulp, paper, aid board Woodpreserving 2491 Pulpmills 2611 Paper mills, except building paper mills 2621 Paperboard mills 2631 Paper coating and glazing 2641 Building paper and building board mills 2661

Stone, clay, glass, and ceramics Flat glass 3211 Glass containers 3221 Pressed and blown glass and glassware,n.e.c. ?1,29 Brick and structural clay tile 3251 Ceramic wall and floor tile 3253 Clay refractories 3255 Structural clay products, n.e.c. 3259 Pottery and related products 3261 3262, 3263, 3264, 3269 Gypsum products 3275 Abrasive products 3291 Asbestos products 3'92 Steam and other packing, and pipe and boiler covering 3293 Minerals and earth, ground or otherwise treated 3295 Mineral wool 3296 Non-clay refractories 3297 Nonmetallic mineral products, n.e.c. 3299 Lime and cement Cement, hydraulic 3241 Lime 3274 Metallurgical and metal products Electrometallurgical products 3313 Primary smelting and refining of copper 3331 Primary smelting and refining of lead 3332 Primary smelting and refining of zinc 3333 Primary production of aluminum 3334 Primary smelting and refining of nonferrous metals, n.e.c. 3339 Secondary smelting and refining and alloying of nonferrous metals and alloys 3341 Enameled iron and metal sanitary ware 3431 Electroplating, plating, polishing, anodizing, and coloring 3471 Coating, engraving, and allied services, n.e.c. 3479 Explosives and ammunition Explosives 2892

215 204 APPENDIX C'

Industry Segment SIC Code

Small arms ammunition 3482 Ammuaition, except for sn.all arms n.e.c. 3483 Ordnance and accessories, n.e.c. 3489 Other chemically processed products Part of broad woven fabric mills, wool, including dyeing and finishing 2231 Finishers of broad woven fabrics of cotton 2261 Finishers of broad woven fabrics of man-made fiber and silk 2262 Dyeing and finishing textileF, n.e.c. 2269 Artificial leather, oilcloth, and other impregnated and coated fabrics, except rubberized 2295 Adhesives 2891 Printing ink 2893 Carbon black 2895 Carbon ano graphite products 3624 Semiconductors and related devices 3674 Storage batteries 3691 Primary batteries (dry and wet) 3692 Part of photographic equipment and supplies 3861 Lead pencils, crayons, and artists materials 3952 Carbon paper and inked ribbons 3955 Linoleum, asphalted felt base, and other hard surface floor covering, n.e.c. 3996 Manufacturing industries, n.e.c. 3999 Engineering design/construction Engineering firms, engineering and construction firms, consulting engineers, independent R&D central engineering, and others 891 *n.e.c. = not elsewhere classified

2iG Index

A surface and interfacial phenomena in, 155 synovial fluids in joints, 20 Adhesives Nee also Prostheses manufacturing problems, 69-70 Ash from combustion processes, 98, 115, 116-117, 156 research needs on, 166 Automobiles U.S. competitiveness in, 71 composites applications, 64, 71 uses, 69. 76 electric vehicle technology, 95 Adipic acid from microbial fermentation of glucose, 26 emissions, 108 Agriculture polymer applications, 14 biologically denved pesticides, 23 opportunities for chemical engineers in. 17, 22-23 pollution from, 108 B product markets, 18 Biochemical processes in humans, measurement of. 31 veterinary pharmaceuticals, 23 Biochemical synthesis Air pollution opportunities for chemical engineers in. 23-24 indoor, 126 Nee also Cell/tissue culture modeling, 14, 125, !7,6, 142 Bioengineenng monitonng, 125-126 curricula, 32 reduction strategies, 113-117 faculty needs, 33 Aircraft funding for research, 26, 32-33, 180 ceramic applications in, 65-66 instrumentation and facility needs, 33-34 composites applications, 62, 64, 71, 76 manpower needs, 34 Amencan Chemical Society, recommended role of, 184 surface and interfacial phenomena in, 2, 17. 27-28, 155 American Institute of Chemical Engineers Biological systems, complex Center for Chemical Process Safety, 175, 178 engineering aaalysis of, 2, 30-31 continuing education program, 77 interactions, modeling, 2, 17, 26-27 promotion of cross al ciplinary cooperation, 34 Biomedicine recommended role of, 183-184 clinical Implants and biomedical devices, 27, 31 Antibiotics, penicillin, 10, 11-12 contributions of chemical engineers to, 19-20 Antibodies, 20, 22, 27 diagnostics, 18-20, 169 Antigens, 14, 22 educational/training needs for engineers, 32-33, 176 Antthemophilic factors, 22 international competition in, 17, 25-26 Artificial organs, tissues, and fluids membrane technology applications, 20, 168-169 aqueous and vitreous humors, 20 opportunities for chemical engineers in, 2, 17, 18-31 chemical engineenng contnbutions to development of, product market, 18 10, 19-20 research recommendations, 27, 30-31 future targets, 19-20 therapeutics, 19, 21-23, 30-31, 154, 163 heart program, 10 see also Artificial organs, tissues, and fluids, Prostheses hybridization, 19 Bioprocessing kidney program, 19 batch, 20, 28- -29 pancreas, 20, 169 continuous, 2, 29 performance forecasts, 31 high-fructose corn syrup, 24-25 skin for burn patients, 20 monitoring and control, 2, 29; see also Sensors "smart membrane" device, 20 of monoclonal antibodies, 20

205 2i7 206 VDE1

proteins, 29-31 zeolites, 89, 156, 157-159, 171 research recommendations, 2, 17, 26-31 Ziegler-Nana, 157 scale-up, 29, 31 Cell/tissue culture separation technologies, 2, 29-31 animal, 27, 28 see alsoCell/tissue culture; Fermentation processes bioreactors, 2, 29-30 Bioreactors constraints on, 23-24, 30 analytical instrumentation, developmental processes, 30 facility costs, 33 animal ceil glowth in, 27, 28 interferon production, 29 for biomass energy production, 97 plant, 23, 28 contamination prevention, 28 recovery of products from, 30 design, 2, 14, 21, 27, 28 research recommendations, 2, 29-30 feedback control systems, 14 surface and interfacial phenomena in, 155 hollow-fiber, 20 tissue plasminogen activator, 21 for manufacturing processes, 28-29 -2ellular processes, modeling of, 26-27 for pharmaceutical production, 21 Cement and concrete, 156, 167 properties measured by sensors in, 30 -amics, advanced for waste treatment, 123 applications, 65-66, 71, 166 see alsoCell/tissue culture chemical additives in processing, 67-68 Biotechnology chemical synthesis and processing, 2, 56, 74, 165, 166 benefits of research, 18 colloidal phenomena in processing, 163 foreign accomplishments in, 25-26 composites, 68-69, 71 funding for research, 18, 33, 181-183, 188, 193, 194 conventional ceramics contrasted with, 65 international competition in, 17, 25-26 crack initiation and propagation in, 65, 68, 166 markets for products, 18, 24 cutting tools, 65 opportunities for chemical engineers in, 17, 18-25 economic impacts, 65 research recommendations, 26-31 functions, 66 see alsoClones/cloning; Recombinant DNA technology interconnection substrat-s from, 48 Bureau of Mines international competition in, 71 funding from, 104, 1% market forecasts, 65 recommended role, 183 materials for, 74 Butanol from microbial fermentation of glucose, 26 membrane applications, 168 microstructural control, 65-66, 166 nonoxide, 73 opportunities for chemical engineers in, 66-69 notential Importance, 71 Catalysis/catalysts powder processing, 74 activity and specificity, 157 properties, 66, 166 automobile exhaust emission control, 157 research needs on, 74 biological, 123 sol-gel processing, 49, 56, 66-67, 163 charactenzation of structure, 158-159, 171 superconducting properties, 49, 56 enzyme role in, 24 transformation toughening, 68 fuel production, 154 zirconia, 68 modeling microstructure and surface characteristics, Chemical effluents 158 ambient monitoring, 128-129 nonnoble metals, 158 aquatic and soil behavior, 126-128 olefin polymerization, 157 atmospheric behavior, 125-126 particle surface as, 124 multimedia approach to managing, 129 performance determinants, 19, 156 see alsoHazardous wastes petroleum cracking, 157-159 Chemical engineering history photosensitive, 102 contributions to technological needs, I, 10, 11-12, 19- platinum-based reforming. 157 20 potential, 157-158 emerging and enduring characteristics, 13 removal problems, 157 engineering science movement, 12 research opportunities, 157-160 first university program, 11 scale of research, 14 future directions, 12 -IS shape-selective, 158-159 paradigms, traditional, 11-17 silica-alumina, 158 Chemical engineering opportunities solar energy applications, 102 agriculture, 17, 22-23 surface chemistry, 159-160 biochemical synthesis, 23-24 three-dimensiwai structures, 160 biomedicine, 19-22

.21S 207

biotechnology, 2, 17, 18-31 design, 75-76, 112-113, 144-146 cut:lysis, 157-160 engineering science movement, 12 ceramics, 56, 65-69, 166-167 environmental protection, 59, 112-113 colloidal science, 163 information management, 151 composites, 64-65, 68-70 for integrated circuits, 14 deposition processes, 56-57 integration of,seeProcess integration electrochemistry, 160-161 for interconnection materials and devices, 46-48 environmental protection, 13, 24-25, 59 for light wave media and devices, 42-45 fluid interfaces, 163 for microcircuits, 40-42 health, 17, 19-22 for monoclonal antibodies, 20 liquids, complex, 69-70 for packaging materials for electronic devices, 39, 46- membrane technology, 16'-168 48 natural resource recovery, 25 for photovoltaics, 49 polymer science, 55-56, 63-65 for polymeric materials, 55-56, 74-75 process interactions, 52-53 reactor engineering and design, 53-54;see also purification proce TS, 54 Bioreactors reactor engineeri. g and design, 53-54 for recording media, 45-47 surfactants, 163-164 retrofitting, 145 thin films, 56-57 safety in, 59 Chemical engineering research sensors, 149-151 biological interaction., 26-27 for superconductors, 49 biological surfaces and interfaces, 27-28 types of operations, 11 bioprocessing, 28-30 ultrapunfication, 54 complex biological systems, 30-31 unit operations concept, 11-12, 14, 46, 51 core areas, I see also13-- irocessing; Materials synthesis and federal role, 178-183 processing fire hazards, 118 Chemical processing industries high-priority areas, 1-7 diversity, 176 industry role, 177-178 employment, 10-11, :3 industry/university collaboration, 34, 59, 174, 177 share of total manufacturing, 10 instrumentation and facilities, 33-34, 77, 173-174, 177 shipment values, 10 interdisciplinary cooperation, 4, 14-15, 22, 31, 32-33, standard industrial classification codes, 201-204 34, 38, 64, '6, 77, 100, 103-104, 118, 156, 179-180 U.S competitiveness, 3-4, 11, 13, 177 macroscale, 14 value added by manufacture, 10-11 mesoscaie, 14 Chemicals microscale, 14 acceptable risks, 107-108 natural resource recovery, 80-81, 99, 100 binding properties, 131 needs, 1-6, 31-34, 59 commodity, 13, 18, 26 small-group, 179 environmental impacts of, 5, 106, 107-111, 142 U.S. competitiveness in, 38 hazards associated with, 130 Chemical industry life cycle in environment, 106 accidents, 13, 107, 108-109, 113, 126, 131 markets for, 18 coo, rative research within, 178 microbial fermentation routes to, 26 costs of accidents, 108 public opt. ion on hazards from, 107 expenditures for safety, 109 regulation of, 108 fires and explosions, 118-119, 131 specialty, 13, 18, 149 new areas of application, 175 toxicity measurements, 130 -131 process safety research, 178 see alsoChemical effluent s; Petrochemicals regulation costs, 109 Citric acid from microbial fermentation of glucose, 26 safety record, 108-109 Clones/cloning Chemical manufacturing processes from microbial cells, 28 batch, 20, 28-29, 149 from plant or animal cells, 211 challenges in, 52-59 of plants, 23 computer-assisted, 135-152 protein production by, 30 contamination prevention in, 54 tissue plasminogen activator, 21 continuous, 2, 29, 45, 54, 147 Coal control,seeProcess control acetic anhydride from 86, 90-91 current, 38-49 combustion, 116-118 deposition of thin films,seeDeposition processes for consumption, 116 thin films co-processing with heavy crude oil, 89

,:2j 208

environmental constraints on use, 116 aircraft applications, 62, 64, 71, 76 liquefaction, 88, 89, 99 annotropic, 64 mineral content, 116 aramid fibers in, 65 power generation, 109 attachment to other materials, 3, 76 pyrolysis, in situ, 87-88 automotive applications, 64, 71 reserves, 86, 116 bicontinuous, 71-72 solids handling, 99 carbon fiber, 70 Coal gasification ceramic, 68-69, 71 environmental impacts, 87 crack inhibition in, 68 in situ, 87 curing, 75, 76 integrated gasification combined cycle, 88. 90 design problems, 76 processes, 25, 86-88, 99 fiberglass, 64 research needs, 99 flaw detection and repair, 76 see also Synfuels interfaces, role in synthesis, 72 Coatings international competition an, 70-71 corromon-resistant, 154 isotropic, 64 hermetic, 50, 57 laminates, 64, 156 magnetic tape, 46-47, 57, 156 liquids, 69-74; see also Liquids, complex optical, for window glass, 154 lubncants, 69. 74 optical fiber, 45, 50, 55, 57 manufactunng methods, 68-69 polymer resist, 41-44 mechanical strength, 64, 72 research needs on, 166 molecular design concepts, 71-72 solution, 39 opportunities for chemical engineers in, 64-65, 68-70 ultrapure glass fiber, 39 poly(ethylene terephthalate) fibers, 70 see also Thin films polymenc, 62, 64-65, 70-72, 75, 156 Colloids/colloidal phenomena properties. 68, 69 in ceramics processing, 163 silicon carbide whiskers, 68-69 microscopic examination of, 169-170 third-generation fibers in, 70 spectroscopic examination of, 171 three-dimensional, 71-72 Combustion trussworks, 64, 71 ash formation, 98, 115, 116-117, 156 two-dimensional reinforcement, 64 coal, 116-1" see also Coatings complexity of, 113 Computer-aided design and manufactunng dioxin formation, 121 availability, 137-138 efficiencies, improvement in. 161 development of, 76 environmental impacts of, 109-110, 113-119 education and training, 136. 140 fires and explosions, 118-119, 142 hydrodynamic systems, 140 fluidized-bed, I17, '18 for new processes, 143-146 free-radical concentrations, 121-122 petroleum production, 141-142 of fuels, 109-110, 115 polymer processing. 140-141 hazardous waste incineration, 120-122, 126, 129, 142 in process operations and control, 146-149 of hydrocarbons, 113-114, 121 programs, 143 in situ, for oil recovery, 83, 85-86 recommended research, 5-6 interfacial phenomena in, 156 for retrofitting of processes, 145 mathematical modeling of, 14, 142 see also Models/modeling; Supercomputers of methane, 113 Council for Chemical Research, recommended role of, monitoring of hazardous effluents, 121 184 of municipal wastes, 91-92 Cytomegalos ELISA test for. 20 nitrogen oxide formation from, 114 pollution control strategies, 116 D polycychc aromatic hydrocarbon formation, 115, 121, 125, 126 Department of Defense pyrolysis, 120 funding from, 194-195 recovery or metals from fly ash. 98 recommended research role, 183 soot formation, 113, 114-116. 121. 156 Department of Energy staged, 115 funding from. 192-193 sulfur oxide formation, 117 recommended research role, 181, 182 Competition, see International competition; U.S. Deposition processes for thin films competitiveness chemical vapor. 39, 43-45, 49, 50, 52. "1, 162 Composites in magnetic media production, 52 adhesion between fiber and matrix. 68, 72 in microcircuit manufacture, 2, 39, 41 INDEX 209

monitoring, 162-163 chemical engineers' role in, 37, 59 in optical fiber manufacturing, 43-45 employment of chemical engineers in, 59 in photovoltaics manufactunng, 39, 54 evolution, 38, 40 physical vapor, 39 small-firm role in generating new processes/equipment, plasma vapor, 39, 44, 162 59 thermal, 162 Electrorefining undesirable phenomena, 57 recovery of spent nuclear fuel, 93 U.S. competitiveness, 50 research needs, 93-94 vertical axial, 43 Employment of chemical engineers Drugs, see Pharmaceuticals biochemical and biomedical engineers, demand for, 34 electronics Industry, 59 oil industry, 10.I 1 E statistics, by industry, 10 Energy crisis, 11, 80 Education/traiiiing Energy industries access to computers, 152 chemical engineering contnbutions to, 80 bioengineering, 32-33, 176 Importance, 3-4, 80 in computer applications, 136, 138, 140, 152 research needs, 4, 13 curriculum, core, 12, 15, 77, 175 shipment values, 80 electronics, 176 Energy research environmental concerns in process design, 132-133 nuclear ener^v, 95 faculty needs, 33, 176-177 trends, 80 flexibility in science electives, 176 Energy resources processing graduate, 32-33, 77 separations processes, 100 industry role, 176 solids, 99-10e joint appointments with biological and medical technical prklems, 80 faculties, 33 See also specific energy resources life sciences, 18, 20, 31-32 Energy sources materials phenomena and processing, 77, 176 oatteries, 95, 1!)2 process design and safety, 175 electrochemical energy comers's:in and storage, 95, 102, recommendations, 6, 132-133, 175-177 161 sabbaticals for industrial researchers, 77 electrolysis cells, 95 separations courses, 175 foreign, U.S. dependence on, 80-81 size and composition of academic departments, 6, 176- fuel cells, 95, 161 177 geothermal, 96 surface and interfacial phenomena, 176 municipal solid wastes, 91-92 undergraduate, 32, 175 plant biomass, 96-97 workshops, 77 solar, 95-96, 102 Electrocheniml processes technologies for exploiting, 81-97 characteris Acs, 160 see also Coal; Fuels; Nuclear energy; Oil charge transfer, 160 Environmental impacts and issues electrode performance determinants, 154, 156 acid rain, 5, 107, 110, 125, 129 energy conversion and storage, 95, 102 accidents at chemical plants, 13, 107, 108-109 microstructure influences, 161 automotive emissions, 108, 109 modeling of, 161 bloaccumulation of chemicals, 107 molecular dynamics, 160-161 coal conversion to liquid and gaseous fuels, 87-88 research challenges, 95 chlorofluorocarbons, 108 'ectron-beam processing combustion processes, 14, 109-110, 113-119, 156 thography, 50 energy utilization, 107, 108 silicon-on-insulator structures, 50 greenhouse effect, 108 Electronic materials and devices human activities, 107 chemical engineering aspects of, 39, 40 mathematical modeling of, 142 commercial life cycle, 38 ore recovery processes, 100 energy to manufacture, 38 pesticides, 108 fabrication, 162 polymer processing, 63 integration of manufacturing processes, 2 semiconductor industry wastes, 59 transistor, 38, 40 see also Hazardous wastes value, 38 Environmental protection see also Integrated circuits; Microcircuits; cost considerations, 112 Semiconductors in electronics industry, 59 Electronics industry engineering employment needs, 34

22 210 I \DI:1

opportunities for chemical engineers in, 13, 24-25, 59, Films 105, 106, 132-133 LangmumBlodgett, 163, 164 recommended research, 4-5 see also Thin films regulations and statutes, 108-110, 123, 125 Fires and explosions, 118-119, 142 see also Pollution control technologies Fluid flow and dynamics Environmental Protectior Agency, recommended role in artificial organs, 19 183 of complex liquids, 69-70, 73-74, 140 Enzyme-linked immunosorbent assay, 20 in electrophoretic Image displays, 164-165 Enzymes in enhanced oil recovery, 83 artificial, 165 of film coatings, 57 in high-fructose corn syrup production, 24-25 at interfaces, 163-166 isomerase, 24-25 modeling, 31, 141-142 lifetime determinants, 25 optical techniques for study of, 166, 172 mercuric reductase, 123 in pollutant transport, 126-127 protease, 29 in porous media, 141-142 separation from complex mixtures, 25 in reactor engineering and design, 54 tissue plasminogen activator, 21 Freeze drying to stabilize penicillin solutions, 12 uses in synthetic chemistry, 24 Fuel cells for transportation, 95, 161 waste treatment applications, 24 Fuels Epitaxy combustion, environmental impacts of, 109-110 molecular beam, 43, 50 conversion of methanol to gasoline, 89-90, 93, 158-159; in optoelectronics manutactunng, 43 see also Synfuels solid-phase, 50 ethanol as a gasoline substitute, 96 vapor-phase, 43 gaseous and liquid, coal conversion to, 86-88 Etching nuclear, chemical process steps, 93-94, 100-101 anisotropic, 163 opportunities for chemical engineers in, 4 of microcircuits, 39, 41, 57 plant biomass sources, 96-97 of optoelectronics, 43 reactor materials, 4, 102 in photolithography, 44 U.S. supplies, 80 of photovoltaics, 39 Funding plasma, 39, 55, 58, 162 by academic-industrial consortia104, 179, 181 process monitoring, 162-163 biochemical engineering research, 26, 32-33, 180 rates, 58 biotechnology/biomedicine research, 18, 33, 181-183, reactive ion, 43 188, 193, 194 selectivity, 58 Bureau of Mines, 104, 196-197 of semiconductors, 163 computer-assisted process and control engineenng, 182, of silicon, 58 191, 193 wet chemical, 39 cross-disciplinary partnership awards, 33, 177, 179, 180, Ethanol 182, 183 from microbial fermentation of glucose, 26 cross-disciplinary pioneer award., 180, 182 Europe current patterns, 185-186 biotechnology research in, 25-26 Department of Defense, 194-195 composites technology, 71 Department of Energy, 104, 133, 192-193 see also specific countnes electronic, photonic, and recording matenals and devices, 182 energy/natural resources processing, 104, 182, 183, 191, F 192 Environmental Protection Agency, 133, 195-1% Federal Republic of Germany environmental research, 133, 182, 183, 191 biotechnology institutes, 25 equipment and facilities, 177, 179-181, 182 separation process applications, 25 foreign, 26 Fermentation processes hazardous waste management, 182, 191 batch, 29 IBM Fellows award, 180 beer, 29 from large research centers, 181 continuous, very large, 26 liquid fuels, 193 enzyme culturing, 21, 24 materials research, 182, 183, 188-191, 194-195 glucose, commodity chemicals from, 26 mechanisms, federal, 178-181 immobilized-cell, 26 microstructure research, 174, 193 penicillin recovery, 12 National Bureau of Standards, 1% problems, 24 National Institutes of Health, 133, 193-194 INDEX 211

National Science Foundation, 104, 133, 187-191 Instruments for studying, 170 nuclear energy research, 95 modeling, 39 process design, 182 in ocean thermal energy conversion, :15 process safety, 133, 182, 191 in retorting of shale oil, 85-86, 87 recommendations, 7, 104, 133, 178-181, 185-197 in semiconductor materials prepay' on, 41-42 research excellence awards, 180, 182, 192 Hemodialysis and he-tofiltration, 10, 19, 26 single investigator awards, 179 High-fructose corn syrup, bioprocesses, 24-25 surface and interfacial engineering, 182, 191-192, 193 Hormones, human growth, action, 22 tenure-tra:k faculty positions, 180 Human serum albumin, 22 Hydrocarbons, combustion of, 113-114 Hydrodynamic systems, mathematical modeling of, 140 G Gamma globulin, 22 Genetic engineering,seeBiotechnology; Clones/cloning I Germany,seeFederal Republic of Germany Imaging devices, electrophoretic 164-165 Glasses Implantation processes, U.S. competitiveness on, 50 halide and chalcogenide, 57 In situ processing in optical fibers, 57 coal, 87-88 stress corrosion crack growth inhibitors, 57 combustion, 83, 85-87 Glycerol environmental impacts, 98, 99 from microbial fermentation of glucose, 26 metals and minerals, 25, 97 oil shale, 99 petioleum, 83, 85-87, 98-99 H problems, 98-99 research needs, 98 Hazardous wastes Information management for process engineering, 137, amount generated annually, 110 151 biodegradation, 122-123, 125 Information technologies burial methods, 110-111 hazardous wastes generated by, 59 chemical engineering opportunities, 4-5, 105 international competition, 38 containment problems, 111 materials and devices, 37, 38, 40 detoxification, 120-124 product obsolescence, 52 from electronics industry, 59, III world markets for storage and handling devices, 38 groundwater contamination, 1 II, 126-129, 142 Injection molding, equipment design, 14 heavy metal ions, 122-123, 124 Instruments improper disposal, 107, 1 11, 128-129 cost and availability, 173 incineration of, 110, 120-122, 124, 126, 129, 142 direct force measurement apparatus, 112 land disposal, 5, 110, 1 I I, 128-129, 142 laser- doppler motion probes, !72 management of, 48, 59, 110-111, 119-12i microelectrode probes, 172 monitoring of sites, 125, 128-129 for microstructure studies, 169-173 multimedia approach to, 129 see alsoMicroscopy; Microtomography; Spectroscopic National Priority List of toxic waste dumps, 111 methods polycyclic aromatic hydrocarbons, 115, 121, 125, 126 Insulin, 19 separation processes, 123-126, 156 Integrated circuits site remediation, 124-125 complexity and capability, 40 spent nuclear fuel management and disposal, 93 etching processes, 58 from substrate formation for interconnection devices, manufactunng methods, 38, 58 a materials, 41 thermal destruction, 110, 120-122, 124 microstructure characienzation, 162 wet oxidation, 124 monolithic, 40 see alsoWaste management packaging, 58 Health plastic packaging, 48 contributions of chemical engineers to, 19 scale of research, 14 opportunities for chemical engineers in, 17, 19-22 thin films in, 2 see alsoArtificial organs, tissues, and fluids; very large scale, 155 Biomedicine; Pharmaceuticals; Prostheses see alsoMicrocircuits Heat transfer Interactions,seeBiological systems, complex in design of packaging, 39 Interconnection/packaging materials and devices in fermentation processes, 24 board composition, 47, 48, 52

2 212 INDEX

chemical manufactunng processes, 38. 39. 46-48. 57 silicon-on-Insulator structures. 50 IBM mult'layer ceramic interconnection package. 48 superconductors. 52 international competition in, 52 technology transfer from U.S. to. 70 for high-frequency data transmission. 57 materials. 39, 47, 48. 56 modeling applications in design of, 39. 58 K plastics, 48 Kidneys polymers in, 48, 55, 56 artificial. 19 process design challenges, 48 function, 19 substrate ormation, 48, 56 Korea thin film deposition on, 57 magnetic tape market. 50 transfer molding process, 48, 56 polymer processing. 63 ultrahigh-speed modules, 48 U.S. competitiveness in, 48 world market, 38 L Interfaces, see Surface and interfacial engineering Landau- Lifshitz theory. 165 Interferons Light wave media and devices action. 22 applications, 42-43 production, 31 chemical manufacturing processes for, 39. 42-45. 156 International competition commercial life cycle, 38, 42-43 access to prototypes and, 50 energy to manufacture, 38 biotechnology and biomedicine. 17. 25-26 future of, 43 ceramics. 71 International competition in, 50 composite materials, 70-71 local area networks. 55 fermentation processes. 26 microstructure control in. 156 interconnection and packaging. 52 packaging. 39 light wave media and devices, 50 polymer applications in, 55 microcircuits, 49-50 value, 38 photovoltaics, 50 world market. 38 polymers, 63, 70-71 see also Optical devices; Optical fibers product quality and performance factors, 13 Liquid crystals, 71, 72 recording media, 50-52 Liquids, complex separations technology, 26 applications, 74 superconductors, 52 *liocompatible polymer solutions. 20 see also U.S. competitiveness composites, 69 74 Ion microbeam technologies design of. 70 Japanese competitiveness in. 50 interfacial properties, 156 in optoelectronics manufacturing. 43 microstructured fluids, 14 Isomenzation, xylene, 159 modeling of, 74 molecular behavior. 14. 70, 72-73 ordered, fluid mechanics of. 74 J particles or micelles in. 74 Japan polymeric, 74 biotechnology research, 25 processing. 73-74 carbon fiber technology. 70 rheology, 74 ceramics technology, 71 scale of research. 14 composites technology. 71 suspensions, 14 deposition and processing technologies for thin films, Lithography 50. 52 electron beam. 50 fermentation processes. 26 equipment. 50 interconnection technologies, 52 high-resolution. 50 kidney dialyzers, 26 x-ray. 50 laser applications, 50 Little, Arthur. D.. 11 lithography equipment, 50 Lubricants magnetic media manufacturing, 50-52 components and performance charactenstics. 69 microcircuit processing technology, 50 dewaxing of, 158 optical fiber manufacturing, 50 interaction with surfaces, 156 polymer research and development, 63, 70 U.S. competitiveness in. 71 separation research and development, 25-26 Lymphokines, 22

22 ' INDEX 213

M depletion, 4, 24-25, 80, 98 reserves, 4, 80 Manufacturing processes,seeChemical manufacturint U.S. dependence on foreign sources, 80. 97 processes Metals and minerals, recovery and processing Mass spectrometry biological systems, 25 microstructure charactenzation, 159 and construction costs, 103 monitoring deposition/etching processes, 163 from fly ash from power plant combustors, 98 secondary ion, 159, 172 high-concentration ore deposits, 97 Mass transfer h!drothermal deposits, 103 in artificial organs, 19 in situ processing, 4, 25, 97, 103 in fermentation processes, 24 industnal waste streams, 25, 98, 123, 124 instrun.ents for studying, 170 low-concentration ore deposits, 97-98, 100, 102 of plant cells from bioreacton 28 research needs, 97 reactor engineering and design considerations, 54 solids handling, 99-100 in semiconductor materials preparation. 41-42 solvent extraction, 97, 103 Massachusetts Institute of Technology steel making, 97, 103 Biotechnology Process Center, 25 sulfide deposits, 101 chemical engineering program, 11 technologies, 25, 97-98 development of freeze drying process, 12 Methyl ethyl ketone from microbial fermentation of Materials, advanced glucose, 26 aircraft applications, 62 Micelles, 124, 163, 164 for membranes, 167-168 Microcircuits nondestructive testing of, 76 chemical engineering contnbutions to, 39 for platelet storage, 19 chemical manufacturing proc ,sses, 39, 40-42, 50, 53, process-related, needs, 102 56-57, 155, 162 see alsoCeramics, advanced; Composites; Polymenc component density, 40-41, 162 materials/polymers gallium arsenide, 41, 55 Materials research heat dissipation, 47 chemical engineering frontiers in, 71-76 International competition in, 49-50 instrumentation and facility needs, 77 photolithographic processes, 41, 56 microscale structures and processes. 71-73 polymers in, 55, 74 Materials synthesis and processing process integration, 50 combining, 3 research frontiers, 162 complex liquids, 73-75 silicon, 41-42 monitoring of, 75 size limits, 54, 56, 57, 155 powders, 74 substrates, 40 Materials systems theoretical research needs, 173-174 chemical processing in fabncation, 3, 76 three-dimensional, 41-42 design, 75-76 uses, 40, 149, 154 detection and lepair of flaws in, 76 see alsoIntegrated circuits; Semiconductors Membranes Microscopy applications, 20, 167-169 scanning electron, 162, 170 cellular, 26-27 scanning tunneling, 170, 172 ceramics in, 168 transmission electron, 170 design of, 168-169 video-enhanced interference phase - contrast, 169-170 energy and natural resource processing applications, l00 Microstructures/microstructured materials hollow-fiber, 123, 168-169 charactenzation, 124, 162, 168-173;see also laminated polymer, 168 Microscopy; Microtomography; Spectroscopic liquid-liquid extraction, 123 methods materials and research needs, 167-168 control in electronic, photonic, and recording materials mimetic chemistry, 165 and devices, 155-156 Monsanto's Prisms, 169 microscopic examination, 169-170 organic polymer, 100 microtomographic examination, 170-171 separation processes, 19, 20, 25-26. 100, 123. 167 nature of, 154-155 "smart," 20, 168-169 organizational forms, 154 surfactant applications in processing of, 164 probes, 172-173 synthetic, 27 research needs and opportunities, 2 -3, 156-174 transport dynamics, 27, 31 scattering methods for examining, 171 Metals and minerals supramolecular, 161

-22 214

technological impacts of, 155-156 National Science Foundation ultimate products from, 154 Engmeenng Research Centzrs, 181 see alsoCeramics, advanced; Composites, Polymenc funding from, 104, 187-191 materials/polymers Materials Research Laboratories, 77 Microtomography recommended role, 181-182 multinuclear magnetic resonance, 170-171 Natural resource recovery/processing x-ray, 170 cost determinants, 80, 97 Models/modeling design and scale-up, 102-103 accuracy A-147 energy resources, 81-97 adsorption ...,iion/pt nide-removal separation process, environmental constraints, 97 124 high-concentration raw materials, 97 air pollutant transport and chemical reactivity, 14, 125, in situ processing, 4, 25, 98-99 126, 142 interfacial phenomena in, 156 biological interactions, 2, 26-27 !ow-concentration law materials, 97-98 catalyst microstructure and surface structure, 158 media microstructure and surface properties, 156 cellular, 26-27 metal/mmeral resources, 25, 97-98 chemical dynamics in electronic device manufacture, 2, opportunities for chemical engineers in, 4, 24-25, 103 57-58 primary processes, 81 combustion systems. 14, 142 research frontiers, 98-104 complex liquid structure-property relationships, 74 secondary processes, 81 electrochemical processes, 161 separation processes, 100-102 environmental systems, 142 solids processing, 99-100 Escherichia coll.27 solvent extraction, 97 fluid flow in porous media, 141-142 technical problems, 80 heat transfer in design of packaging, 39 technologies for exploiting, 91-98 hydrodynamic systems, 140 see aisoEnergy sources; Metals and minerals; Oil; magnetic media manufacturing processes. 58 Petroleum refineries/refining mathematical, of fundamental phenomena, 138-142 Nitrogen oxides from combustion processes, 114 membrane transport phenomena, 168 Nondestructive testing of advanced materials, 76 microelectronics processing, 163 Nuclear energy of molecular events in unit operations, 12 development efforts, 93 Navier-Stokes equations, 140 fast breeder reactors, 92, 93 petroleum production, 14, 141-142 fission, 92-94 plasma processing. 58 fusion, 92, 94-95 polymer processing, 140-141 Integral Fast Reactor, 93 with supercomputers, 136, 173-174 light water reactors, 92 thin film processing, 58 Nuclear fuel cycle, 93, 100-101 underground pollutant transport, 126-127 viscous fluid flows, 58 Monitoring 0 air pollution, 125-126 Oil bioprocessec, 29 consumption, 116 chemical spills/releases, 126 co-processing of coal with, 89 incinerator effluents, 121 problem constituents in, 89, 158 instrumentation needs, 128-129 reserves, 4, 82, 84-86, 116 nondestructive probes, 128-129 reservoir simulation, 140-141 personal exposure, 129, 132 shales, 84-86, 87, 89, 99, 100, 103, 156 remote sensing technology, 128 synthetic base, 69 of toxic waste sites, 125 ultraheavy crudes, 82, 102 see alsoSpectroscopic methods see alsoLubricants; Petroleum refinenes /refining Mono layers, 164 Oil industry, employment of chemical engineers in, 11 Oil recovery processes N chemical flooding, 83 enhanced, 4, 14, 81-84, 86, 125, 154, 156 National Bureau of Standards hydrogen reaction, 102 funding from, 1% in situ combustion, 4, 83, 85-87, 100 recommended role, 183 miscible flooding, 83 National Institutes of Health modeling, 141-142 funding from, 193-194 pnmary, 81 recommended role, 183 secondary, 81-82

22 C INDEX 215

from shales, 99, 100, 103 chemical steps in, 44 steam injection, 83, 84 Photonics supercritical extraction, 125 chemical engineenng aspects of, 39 thermal, 8245 fabncation, 162 see also Petroleum refineries/refining integration of manufacturing processes, 2 Optical devices see also Light wave media and devices Rat-panel image displays, 164-165 Photovoltaics see also Light wave media and devices chemical engineering contributions to, 39 Optical fibers chemical manufacturing processes, 38, 49, 53, 54 characteristics, 43 costs of manufacturing, 49 coatings, 45, 50, 55, 57 focus of research, 49, 53 connectors and splice hardware, 55 gallium arsenide cells, 49 costs of manufacturing, 45, 55 hydrogen clustering in, 171 data transmitting capacity, 42. 44 international competition in, 2, 38, 52 drawing processes, 57 polycrystalline module efficiency and reliability, 49 glass, 44-45, 57 world markets, .13 hermetic coatings, 50 Platelet storage, 19 high-strength, 50 Pollution control technologies international competition in, 50 calcium sorbent addition, 117 manufacturing methods, 38, 40, 43-45, 57 in coal-fired generating plants, 109 low-cost components, 55-56 electrostatic precipitators, 116, 124 polarization-maintaining, 55 emission reduction strategies for combustion products, polymers in, 56, 57 113-117 Rayleigh scattering in, 44 for fly ash, 116 sensor applications, 43, 55 incinerators, 120-122, 124 thin film deposition on, 2, 57 through plant and process design, 112-113 ultrapurification of materials, 54-55 separation of power plant emissions, 100 Organs, see Artificial organs, tissues, and fluids for soot, 116 for sulfur oxides, 117 P see also Separation technologies/processes Polymeric materials/polymers Packaging materials for electronic devices, see acrylate, 55 Interconnection/packaging matenals and devices aramid, 63, 65 Peptides automotive applications, 14 neuroactive, 22 biocompatible, for human fluid replacements, 20 regulatory, 22, 23 chemical engineering contributions to, 63 Pesticides chemical synthesis and processing, 2, 55-56, 74-75, 156 biologically derived, 23 composites, 64-65, 70, 69-71, 72, 74, 140, 156 Petrochemicals in drug delivery systems, 22, 23 biological routes to, 25 elastomeric, 57 see also Oil electrically active, 39 Petroleum refineries/refining epoxy-Novolac prepolymers, 56 catalytic cracking, 156-159 fibers, high-strength, 63 dc Agn, 11, 103 glassy, 57, 74 flexicoking, 91 high-strength/high-modulus, 70 fluid-bed coking, 89 in information storage and handling devices, 55 hydrocarbon separations, 100 interaction with monolayers and micelles, 164 hydrotreating processes, 89, 92 interconnection substrate applications, 48, 58 mathematical modeling of, 141-142 international competition in, 70-71 new raw materials for, 88-91 Kevlar 0, 63 research challenges, 91, 103 low-molecular-weight, 69 scale-up, 103 in membranes, 100, 168 Pharmaceuticals methacrylate, 55 bioprocessing techniques, 21 microstructure, surfaces, and interfaces, 156 drug delivery modes, 21-23, 154, 155, 163, 167 multicomponent blends, 75 markets for, 18 molecular design, 75 therapeutic targets of opportunity, 22 opportunities for chemical engineers in, 63-65, 75 surface and interfacial phenomena in, 155 optoelectronic applications, 43, 45, 55, 56, 156 veterinary, 23 photoresists, 41-43, 55, 58, 74-75, 162 Photolithographic processes p,:ybenzothiazole, 63 216 /APEX

polyethylene, 63 environmental protection and safety considerations, 2, polyimides, 56 112-113, 131-132, 137 polypropylene, 63 goals, 142 pour point depressants, 69 integration with process control, 147-148 property determinants, 140-141 matenals manufactunng, 75 recording applications, 55, 56 research opportunities, 145-146 redox, layered structures of, 169 research recommendations, 5, 54, 75, 112 research needs on, 55 retrofitting, 145 resource recovery applications, 83 scale-up, 29, 31, 102-103, 139 rod- and coil-type, 156 stages, 143-144, 145 solids dispersants, 69 tree graphs, 112-113, 131 spinning fibers from anisollopic phases, 63 twenty-first century, 138-139 thermally stable, 56 Process integration thermosetting epoxy, 64 importance, 52-53 toxic emissions from burning of, 118 in magnetic tape manufactunng, 50-51 viscosity modifiers, 69, 83, 156 for microcircuit manufacture, 50, 53 Polymerization processes research needs, 2, 52-53 batch, 149 for semiconductor manufacturing, 53 emulsion, 165 Process safety fluid flow during extrusion, 140 cost considerations, 112 free radical influences on, 72 education, 175 mathematical modeling of, 75, 140-141 in electronics industry, 59 polyethylen', 63 fire prevention, 118 processing of complex liquids dunng, 73 reactor materials and, 54 reactive extrusion, 73 research challenges and recommendations, 5, 105, 132- reactive injection molding, 73 133, 178 solvent-polymer interactions, 74 Professional societies for thin films and membranes, 164 cooperation and communications among, 34 UNIPOL method, 63 recommended rolt of, 183-184 Powders, processing of, 74 Prostheses Power generation electrochemical signal transduction systems for, 20 coal-fired, 109-110 ....e alsoArtificial organs, tissues, aid fluids from combustion of solid wastes, 92 Proteins environmental impacts of fuel combustion for. 109-110 bioprocessing of, 29-31 geothermal, 96 interferon production, 29 see alsoEnergy s' urces; Nuclear energy Purification processes Process control for antibiotic preparation, 11-12 adaptive, 148-149 in high-fructose corn syrup production. 25 batch process engineering, 149 for optical fiber grade RICE, 55 bioprocessing operations, 29 for pharmaceutical production, 21 computer-assisted, 137, 146-149 for proteins, 30-31 in high-fructose corn syrup production, 25 se, alsoUltrapurification integration of process design with, 147-148 internal model control, 148 interpretation of information, 146-147 R materials manufacturing, 75 Reactor design and engineenng measurements for, 146 for electronic, photonic, and recording materials and monitoring of, 2, 29, 75 devices, 53-54, 57-58 non-steady-state, 113, 147 for pharmaceutical production, 21 problem-solving strategies, 148-149 see alsoBioreactors process transients, management of, 113 Recombinant DNA technology robust, 148-149 pharmaceutical production through, 21 scale of research on, 14 waste management applications, 122-123 sensors for, 149-151 see alsoClones/cloning Process design Recording and storage media blurring of product design 'nd, 14 chemical engineering aspects of, 38-40, 45-46 computer-assisted, 131, 136-139, 143-145, 147-148 coating processes, 57, 156 contamination prevention, 54 commercial life cycle, 38 education, 175 compact disks, 45, 52

228 INDEX 217

E-beam, 39 chromatographic, 29 energy to manufacture, 38 cyclone separators, 117 femte cores, 40 distillation of azeotropic mixtures, 168 formats, 45 economics, 100, 101 integration of manufacturing processes, 2 education/training on, 175 international competition in, 50-51, 57 energy consumption oy, 100, 102 magnetic, 39, 40, 46-47, 50-52, 54, 58, 156, 162 energy/natural resources, 100-102 manufacturing methods, 38-40, 45-47, 50-51, 5- 58 enzymes and amino acids from complex mixtures, 25 materials, 45, 51 equipment improvement, 14 mathematical modeling of manufactunng processes, 58 examples according to property differences, 101 microstructure characteriza'ion, 162 fixed-bed adsorption, 100 macrostructure control in, 156 foreign accomplishments, 25-26 optical, 39, 40, 45-46, 52, 156 hazardous wastes, 123-126, 156 polymers in, 55, 56, 58, 156 hemodial "sis /hemofiltration, 19 read-only optical disks, 45, 52 heterogeneous feeds, 100 read-write optical disks, 46, 52 high-temperature, 100-101 recording density determinants, 46. 162 homogeneous mixtures, 100, 102 surface and interfacial phenomena in fabncation of, 162 ion-exchange, 100 ultrapurification, 54 laser spectroscopy cell sorter, 173 value, 38 liquid-liquid extractions, 100 world market, 38 membrane, 19, 20, 25-26, 100, 123, 156. 167-168 see also Thin films pollution control, 100, 156 Risk assessment reactor materials, needs, 4, 102 construction materials, 118 research needs, 2, 30, 98, 100-102 exposure assessment, 129, 132 selectivity improvement, 54, 100, 156 hazard identification and assessment, 129, 130-132 steam from brine, 96 standard for, 107-108 steam stripping, 124-125 Risk management, 132 supercritical extraction, 124-125 Rubber, synthetic, 11 thermal desorption, 124-125 zeolite applications, 100 S see also Purification processes; Ultrapunfication Shell Development Company Safety, chemical industry. 108-109 contribution to penicillin production, 11-12 Selective ion transport across membranes, 27 polypropylene manufacturing process. 63 Self-assembling structures, 164 Silicon Semiconductors -on- insulator structures, 50 amorphous, 171 polycrystalline, chemical production steps, 43 chemical manufacturing processes, 40, 41, 57, 59 ultrapure single-crystal, 39, 41 Group III-V compound, 43 Solar power hazardous wastes from, 59 advantages and disadvantages, 95 integrated processing, 53 conversion costs, 95 in optoelectronic devices, 43 materials problems, 102 ultrapurification of matenals, 54 photovoltaics efficiency, 49, 95 world market, 38 research challenges, 95-96 Sensors storage systems, 102 arrays, 150, 151 thermal energy conversion, 95, 102 biological, 150, 155, 167, 168 Sol-gel processing computerized, 136 ceramic powder preparation by, 49, 66-67 73 future developments, 136, 149-151 hemical steps in, 45, 67 membranes in, 167 double-alkaloid systems, 73 noninvasive on-line, for bioprocessing. 2, 29-30. 154 optical fiber manufactunng, 45, 57 optical fiber, 43, 55, 150 preform manufacture, 67 process, 2, 14, 29-30, 136, 137, 149-151 problems in, 67 protein-specific, 154 research needs on, 73 research recommendations, 151 single-alkaloid systems, 73 solid-state, 149-150 Solids processing Separation techr.ologies/processes costs, 99 adsorption/reaction/particle removal sequence, 123-124 crushing, grinding, and milling, 99 bioproducts, 2, 29-31 efficiency, 99

229 218

equipment design and scale-up, 4, 99, 100, 143 in concrete and cement, 167 interdisciplinary cooperation, 100 in film deposition, 58 research needs, 99-100, 103 fluid, 163, 166 scale-up factors, 103 in fuel cell technology161 Soot from combustion processes, 114-116 Importance, 154-156 Soviet Union, fission research, 95 lubricant interaction with, 156 Spectroscopic methods in microcircuit processing, 162 Auger electron, 158, 162, 172 multiple, structuring of, 155 carbcn-13 NMR, '68 in natural resources recovery, 156 catalyst charactenzation, 158, 159, 171 properties and processes, 155 electron energy loss, 159, 171 research needs and opportunities, 6, 72, 156-171 electron spin resonance, 163 role in materials chemistry, 72 emission, 163 solvent/polymer, 55 extended x-ray absorption fine structure, 159, 171 in surfactants, 163 infrared, 159, 171 tissue-implant, 27 laser-induced fluorescence, 163 Surfactants laser spectroscopy cell sorter, 173 in cemert and concrete, 167 low-energy electror diffraction, 158, 159, 171-172 di -tail and tri-tall, 164 for micelle studies, 164 in enhanced oil recovery, 154, 163 for microstructure characterization, 158-159, 162,171- monolayer-forming, 164 172 multifunctional, 164 for monitoring deposition/etching processes, 163 property control measures, 165 neutron spin-echo, 164 research opportunities, 163-164 nuclear magnetic resonance, 159, 171 resource recovery applications, 83-84 photon correlation, 164 superplasticizers, 167 Raman, 159, 171 Synfuels small-angle neutron scattering, 124, 164 catalytic conversion to liquid fuels, 87 solid-state nuclear magnetic resonance, 159 Fischer-Tropsch process, 87 structure-permeability probes, 168 methanol to gasoline, 89-90, 93, 158-159 ultraviolet photoelectron, 158, 172 natural gas to gasoline, 90 x-ray photoelectron, 124, 158, 162, 168, 171, 172 production process, 86-88, 90-91 Storage media,seeRecording and storage media uses, 25 Sulfur oxides from combustion processes, 117 Supercomputers applications, 139-140, 143 T artificial intelligence, 136-139 availability, 137-138, 152 Thin films Cray I class, 137 controlled permeability, 154 expert systems, 136, 137 deposition processes, 2, 41, 50, 54, 56-58 hypercube architecture, 140-141 on interconnection devices, 2, 57 need for, 173-174 low-temperature methods, 56 speed and capabilities, 136-138, 152 mathematical modeling of processes, 58 Superconductors on optical fibers, 2, 57, 156 ceramics in, 49, 56 pharmaceutical applications, 154 chemical manufacturing processes for, 49 on recording/storage media, 2, 46, 57, 70 cooling, 49 organic, 41 high-temperature, 2, 49, 52 property determinants, 58, 156 international competition in, 52 research needs on, 58 materials, 49, 56 silicon dioxide, 41 metal oxide, 56 surfactant applications in processing of, 164 uses, 49 U.S.-Japanese competition, 50 see alsoSem'conductors Tissue culture,seeCell/tissue culture Surface and interfacial engineenng Tissue plasminogen activator, 21, 22 biological, 2, 17, 27-28, 155 Tissues,seeArtificial organs, tissues, and fluids; in catalytic and electrode reactions, 155, 159-160 Prostheses in ceramics, 166-167 Training,seeEducation/training characterization techniques, 158-159 Transmission electron microscopy, 159 in colloidal systems, 163-164 Transportation, environmental impacts of fuel combustion in composites, 64, 72, 156 for, 109-110

230 INDEX 219

U Uranium-235 scale-up of manufactunng process, 11 Ultrapurification for cell culture processes, 21 for electronic, photonic, and recording materials and V devices, 2, 41, 54 Vaccines, see Pharmaceuticals of silicon for semiconductors, 41 Vesicles, 163 Union Carbide. UNIPOL process, 63 United Kingdom biotechnology institutes, 25 W Unocal Corporation, 96 Waste management U.S. competitiveness biological treatment, 17, 24, 122-123, 125 adhesives, 71 multimedia approach, 5, 129 biotechnology, 25 nuclear waste repositones, 94 ceramics, 71 regulation of, 110, 123 chemical processing Industries, I I. 13, 38, 50 site remediation, 124-125 composite manufactunng and processing technology, 71 soil decontamination, 124-125 interconnection and packaging, 52 Superfund Program, 1 1 I liquid crystals, 71 see alsoHazardous wastes lubricants, 71 Wastes microelectronics, 2, 48, 50 metal/mineral recovery from, 98 optical technologies. 2, 45, 50 municipal solid, as an energy source, 91-92 photovoltaics, 2 polymers, 70-71 recording media, 2, 50-52 z superconductors, 52 Zeoldes, 156

23i