Cells and Materials

Volume 3 Number 4 Article 5

1993

Heating of Calcium Phosphate Crystals: Morphological Consequences and Biological Implications

W. Bohne Faculte de Chirurgie Dentaire, Nantes

J. A. Pouezat Faculte de Chirurgie Dentaire, Nantes

L. Peru Faculte de Chirurgie Dentaire, Nantes

G. Daculsi Faculte de Chirurgie Dentaire, Nantes

Follow this and additional works at: https://digitalcommons.usu.edu/cellsandmaterials

Part of the Biomedical Engineering and Bioengineering Commons

Recommended Citation Bohne, W.; Pouezat, J. A.; Peru, L.; and Daculsi, G. (1993) "Heating of Calcium Phosphate Crystals: Morphological Consequences and Biological Implications," Cells and Materials: Vol. 3 : No. 4 , Article 5. Available at: https://digitalcommons.usu.edu/cellsandmaterials/vol3/iss4/5

This Article is brought to you for free and open access by the Western Dairy Center at DigitalCommons@USU. It has been accepted for inclusion in Cells and Materials by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected]. Cells and Materials, Vol. 3, No. 4, 1993 (Pages 377-382) 1051-6794/93$5.00 +. 00 Scanning Microscopy International, Chicago (AMF O'Hare), IL 60666 USA HEATING OF CALCIUM PHOSPHATE CRYSTALS: MORPHOLOGICAL CONSEQUENCES AND BIOLOGICAL IMPLICATIONS

W. Bohne*, J.A. Pouezat, L. Peru, G. Daculsi

Laboratoire de Recherche sur les Tissus Calcifies et les Biomateriaux, Faculte de Chirurgie Dentaire, Place A. Ricordeau, F-44042 Nantes, France

(Received for publication October 27, 1993, and in revised form December 29, 1993)

Abstract Introduction

Sintering hydroxyapatite (HA) and 13-tricalcium Synthetic calcium phosphates are applied in ortho­ phosphate ((3- TCP) affects the chemical composition , the paedic, maxillofacial, otorhinoplastic, and in dental crystallinity, and the morphological features as demon­ preprosthetic and periodontal surgery in order to fill strated by means of X-ray diffraction (XRD) , infrared bone defects and to enhance bone repair (Nery and spectroscopy (IR), and scanning electron microscopy Lynch , 1978; de Groot, 1983; LeGeros, 1988; Passuti (SEM). When heated to 1230°C, 16.7% ofHA had de­ and Daculsi, 1989; Daculsi et al., 1990, 1992; Hamel, composed to IJ-TCP. SEM investigations showed homo­ 1992). Before use, many are sintered at high tempera­ geneous, sharp angular polyhedric blocks of 30 to 50 Jlffi tures. Besides the modification of the chemical compo­ with rare surface pores. On heating at 1230°C, IJ -TCP sition and crystallinity of these materials, sintering also had entirely transformed to a-TCP. During sintering, causes modifications of the size, shape and surface tex­ the size of the powder grains increased and progressive ture of the powder grains. These modifications are tem­ bridging between the grains was observed. At 1230°C, perature dependent. They affect the surfaces in contact a network within round-shaped polyhedric blocks of 50 with tissue fluids and bone cells, and are thought to in­ to 90 Jlm was formed. In both, HA and IJ-TCP, surfaces fluence the biological reactions of these materials and were smooth. The chemical composition and the their solubility properties. The morphological modifica­ crystallinity of calcium phosphate ceramics determine tions have not been taken in account sufficiently by cli­ their dissolution behavior and osteogenic properties. nicians in dental prosthetics, periodontology, and osteo­ Nevertheless, their temperature dependent morphological articular surgery in choosing an appropriate material. features, such as, particle shape and size, surface The aim of this study was to point out the modifi­ texture, and porosity, as demonstrated in the present cations of the chemical composition and the crystallinity study, also influence the resorption rates, tissue of hydroxyapatite (HA) and 13-tricalcium phosphate ({3- responses, and wound healing duration. This should be TCP) during calcination, to demonstrate the size, shape emphasized more by clinicians in choosing an appro­ and surface texture modifications which affect the pow­ priate material for bone substitution. der grains and to discuss possible biological conse­ quences of those modifications.

Key Words: Calcium phosphate crystals, sintering, size Materials and Methods and shape modifications, surface texture, chemical com­ position, crystallinity, biological implications, dental Sample preparation preprosthetic surgery, periodontal surgery, bone defect Carbonated HA and 13- TCP powders prepared by filling. Dr. R.Z. LeGeros (University of New York) were used. Both HA and 13- TCP were passed through a 100 /lm sieve. In free atmospherical conditions, the samples *Address for correspondence: were heated from room temperature to respectively W. Bohne 900°C and 1230°C with a Vectra furnace. Each heating Laboratoire de Recherche sur les Tissus Calcifies et les period was for one hour followed by maintaining the Biomateriaux, Faculte de Chirurgie Dentaire, specified temperature for 90 minutes. Cooling was per­ Place A. Ricordeau, formed for one hour. Then , all heated and unheated 44042 Nantes, powder samples underwent X-ray diffraction (XRD), in­ France frared (I R) and scanning electron microscope (SEM) ex­ Telephone number: 33-40 41 29 16 aminations. FAX number: 33-40 08 37 12

377 W. Bohne, J.A. Pouezat, L. Peru, G. Daculsi

Table 1. FWHM and D-value variations at different Results temperatures. Hydroxyapatite (HA) Temp. FWHM D (A) The FWHM-values at hkl reflections 002 and 310 oc (002) (310) (002) (310) had increased from room temperature to 900°C and had strongly decreased at 1230°C. Consequently, the D­ HA 20 0.1436 0.1409 614.7 626.5 value variations, which do not give the crystal sizes HA 900 0.1449 0.1575 609.2 560.4 directly but represent crystal size variations (Vetter et HA 1230 0.1113 0.1246 793.1 708.4 al., 1991), indicated crystal size decrease at 900°C and high crystal growth at 1230°C (Table 1). Infrared scans showed C03 peaks at 670 and 878 Table 2. FWHM and crystallinity index (Cl) value cm-1. Furthermore, when heated to 1230°C, a t3-TCP variations at different temperatures. phase appeared at 944 cm- 1. X-ray diffraction investi­ gations confirmed that HA had partly decomposed to Temp. CCC) FWHM D (A) 13-TCP. The degree of decomposition to /3-TCP, esti­ mated by the relative peak heights of the strongest apa­ t3-TCP 20 0.1457 0.620 tite (211) and t3-TCP (0210) hkl reflections (Apfelbaum t3-TCP 900 0.1470 0.625 et al., 1990), was 16.7%. {3- Tricalcium phosphate ({3- TCP) X-ray diffraction The FWHM-values calculated at the hkl reflection The powders were examined using an INEL XRG 2010 and the crystallinity indices (CI) measured on IR­ 3000 CPS diffractometer at 40 kV and 30 rnA connected scans exhibited little differences from 20 to 900°C to a CATO Multianalyser. The samples were exposed which do not seem to be significant (Table 2). for about 30 minutes to a Cu-radiation (A = 1.5406 A). At 1230°C, X-ray diffraction investigations The 20 reflections, relative peak intensities, and full showed that t3 -TCP had transformed to a-TCP. width at half maximum (FWHM) values were calculated Scanning electron microscopy (SEM) observations using a computer program developed at the "Institut des Materiaux", Nantes, France. Peak broadening is in­ Scanning electron microscopy investigations of versely related to both crystal size and internal lattice HA, for room temperature and 900°C samples, showed perfection (Vetter et al., 1991). So, using Scherrer's loosely packed clusters of round-shaped powder grains equation (Posner et al., 1963; Bonar et al., 1983; of about 0.2 Jlffi in diameter. Spaces between clusters Chantraineetal., 1988; Vetteretal., 1991) were about 1 to 3 Jlffi. All surfaces appeared rough, bridging between powder grains was not observed. D = Al(t3 cosO) [1] When sintered at 1230°C (Figs. 1 and 2), the particles D (the mean crystallite size in A) values could be deter­ and clusters had coalesced to form sharp angular poly­ mined; these are not absolute values and used only for hedric blocks of about 30 to 50 Jlffi, without or with rare comparison (Chantraine et al., 1988). A is the X-ray surface pores of 0.2 to 2 Jlffi. The boundaries of the wavelength (A) employed; t3 the FWHM measured; and original particles were either visible or had disappeared. 0 the reflectance angle. As t3 and e were measured in The surfaces seemed to be smooth. degrees, not in radians, these values were divided by At room temperature, t3-TCP showed loose bridg­ 57.3, the radian-degree conversion factor, before apply­ ing between round-shaped particles of about 2 Jlffi in di­ ing the above equation for D (Vetter et al., 1991). ameter. At 900°C, bridging between powder grains had progressed. The diameter of the bridged particles had Infrared spectroscopy increased to 4 to 6 Jlffi. The grain boundaries could be From unheated and heated HA and t3-TCP sam­ seen. The width of the free spaces was about 1 to 4 Jlffi. ples, pellets of 300 mg KBr/ 1 mg HA and 300 mg KBr/ 1 When sintered at 1230°C (Figs. 3 and 4), general bridg­ mg t3- TCP were prepared and examined by means of a ing was observed forming a true network within polyhe­ Nicolet 20 SXC FTIR spectrometer (32 scans) with dric blocks of about 50 to 90 Jlffi without any sharp an­ 1 gles. The spaces between the powder grains were about 4 cm- resolution. On HA scans, C03 peaks were identi­ fied. On t3 -TCP scans, the crystallinity indices were 1 to 6 Jlffi. The boundaries between bridged powder calculated (Rey et al., 1991). grains persisted; all surfaces seemed to be smooth (Fig. 5). Scanning electron microscopy Ali powder samples were carbon coated and ex­ Discussion amined at 7 kV in a JEOL JSM 6300 scanning electron microscope to assess the size, shape, and surface texture With regard to HA, the variability of the chemical modifications of the powder grains. and crystal structures of the used hydroxyapatites makes

378 Heating of Calcium Phosphate Crystals

Figure 4. Detail of a-TCP formed by bridging of 13- TCP powder grains sintered at 1230°C, showing a network structure. Bar = 2 Jlm.

Figure 5. Smooth surface texture and grain boundaries of coalesced a-TCP grains formed from 13- TCP powder grains sintered at 1230°C. Bar = 200 nm.

the comparison between different laboratory studies dif­ ficult (de Groot, 1983). Moreover, HA shows variable chemical stoichiometry and crystal structure according to the conditions used in its synthesis (Sakae, 1988). Beside these factors, the calcination of HA and 13- TCP powders also affects the stoichiometry of the powder crystals, and modifies their solid phase constituents. On one hand, HA partially decomposes to 13-TCP at about Figure 1. Sharp angulated polyhedric blocks of 1000°C, while on the other hand, rhombohedric 13-TCP coalesced HA powder grains sintered at 1230°C. transforms to monoclinic a-TCP between 1120 and Bar = 10 J.tm. 1180°C (Fowler and Kuroda, 1986; Binder-Royer, 1990). During cooling, Binder-Royer (1990) observed Figure 2. Surface of coalesced HA powder grains sin­ a limited transformation of a-TCP to 13-TCP, a finding tered at 1230°C showing some pores of 0.2 to 2 J.tm. not reproduced by our study. Therefore, both synthesis Bar = 2 JLffi. and manufacturing conditions are important, because the employed methods and the sintering temperatures affect Figure 3. At 1230°C sintered 13 -TCP powder grains the chemical composition and the crystallinity of HA and forming polyhedric blocks of a - TCP showing bridging 13-TCP powders and consequently their biological behav­ phenomenon. Bar = 10 J.tm. ior (Daculsi and LeGeros, 1989; Ducheyne et al., 1993).

379 W. Bohne, J. A. Pc It zat, L. Peru, G. Daculsi

With regard to the crystal size modifications (D phates (BCP) of various {3- TCP/HA ratios have also been value variations) found in our study, they are in accord used for bone defect filling, but systemic studies in ani­ ance with findings in the literature (Young and Brown mal models where HA/ {3- TCP ratios were varied and se­ 1982; Sakae et al., 1989). These facts are well known quential time points studied, are lacking. BCP resulted But the morphological modifications of calcium pho in rapid bone replacement after implantation into perio­ phate powder grains during sintering, as shown in ou dontal defects in monkey (Weng et al., 1991) and in man SEM study, seem not to have been taken sufficiently int•J (Nery et al., 1990). BCP containing 80% {3-TCP and account by clinicians. Indeed, the size and shape o f 20% HA (80/20 BCP) formed dense bone much more these particles, their surface texture, and their porosity , than 50/50 BCP (Weng et al., 1991; Hamel, 1992). Six are assumed to be important factors for dissolution be ­ months after implantation, most of the implanted gran­ havior, and hence, for the rate of biodegradation (Klei ules had disappeared in the defects (Weng et al., 1991). et al., 1983; LeGeros, 1988; LeGeros et al., 1988; va An in vivo study of root socket filling after tooth extrac­ Blitterswijk and Grote, 1989). Therefore, sintered HA tion in beagle dogs by means of 85/15 BCP (Hamel, blocks were estimated not to be appropriate for bone de ­ 1992), showed active resorption and advanced dissolu­ fect filling because of their lack of porosities (Dehen an d tion of the powder grains and bone apposition, 7 months Niederdellmann, 1989). Studies on dry HA powders after implantation. A complementary radiographic study (Thomann et al., 1990) and laser irradiated HA of human root sockets filled by 40/60 BCP exhibited (Meurman et al., 1992) showed that their surfaces pro­ densified root sockets, 10 months later, which was inter­ vided significant amounts of calcium and phosphate ion s preted as new-bone formation (Hamel, 1992). In addi­ adsorbed onto the apatite surface. The presence of these tion, the alveolar process importance in BCP filled sites ions slows the dissolution rate (Ducheyne et al., 1993 ) was about twice that in control sites (Hamel, 1992). because the interface is charged (Meurman et al., 1992) . Furthermore, the solubility rates of calcium phosphates in decreasing order are: a-TCP, {3- TCP, HA (Fowler Conclusions and Kuroda, 1986; LeGeros et al., 1988; Meurman et al., 1992; Ducheyne et al., 1993). Calcium phosphate ceramics such as HA, {3- TCP, Plasma spraying of calcium phosphates revealed or BCP of various {3- TCP/HA ratios are frequently used that {3-TCP was more readily resorbable than HA which , in bone filling. In the present study, the SEM investi­ in vivo, underwent minor surface dissolution (Ellies et gations were of particular interest. On one hand we al. , 1992). {3-TCP is an extremely unstable form of cal­ found homogeneous angular blocks of HA/ {3- TCP pre­ cium phosphate (Okazaki and Sato, 1990) and its dis­ senting a high volume/surface ratio at 1230°C, while on solution rate is at least 10 to 20 times higher than HA the other hand, during sintering, progressive bridging (Klein et al., 1983). HA ceramics possessed a higher between powder grains of {3- TCP followed by the forma­ osteogenic potential than {3- TCP (van Blitterswijk and tion of round shaped a-TCP blocks with a low volume/ Grote, 1989), but {3- TCP seemed biodegradable in con­ surface ratio, which were composed of a network of trast to HA which showed no detectable resorption after round shaped structures. In both sintered HA and 9 months implantation (Klein et al., 1983). {3- TCP, surfaces were smooth. Besides the chemical All implanted biomaterials encounter inflamma­ composition and the crystallinity of the employed materi­ tion caused by the wound inflicted during healing (van als, the temperature dependent morphological features, Blitterswijk and Grote, 1989). The presence of foreign such as, granule size, surface texture, and porosity, bodies leads to the so-called foreign-body reaction pre­ which determine the dissolution behavior, tissue re­ senting macrophages and multinucleated cells which ex­ sponses and wound healing duration should also be con­ hibit signs of phagocytosis, while the exudate cells show sidered by clinicians in odontology and in osteoarticular proliferative activity 4 weeks after implantation (van surgery to choose an appropriate material. It seems to Blitterswijk and Grote, 1989). However, in the case of be of great interest to study the morphological modifica­ HA implants, there is little proliferative activity of tions of BCP powders during sintering to determine the exudate cells. As related by van Blitterswijk and Grote optimal sintering temperatures. As systemic studies in ( 1989), the implant shape and the surface structure have animal models where HA/TCP ratios were varied and se­ a direct influence on the biological performance of bio­ quential time points studied seem not to have been made, materials. Size is an important parameter: above a cer­ such studies are needed to assess the behavior of synthe­ tain size, whole particles cannot be phagocytosed. Sur­ sized calcium phosphates in bone sites. face texture affects tissue responses, making them more severe for rough than for smooth surfaces. Further­ Acknowledgement more, round particles induce a less unfavorable reaction than angular material, and particles seem to cause a more severe inflammatory response than porous bodies. We wish to thank Prof. Raquel LeGeros, New Similarly, round particles lead to faster resorption of in­ York University, for supplying the calcium phosphate flammation than sharp ones. To combine the advantages samples. This work was sponsored by INSERM CSF of both HA and {3- TCP, so-called biphasic calcium phos- 93-05 and CNRS EP59.

380 Heating of Calcium Phosphate Crystals

References calcium phosphate materials in bone tissue. J Biomed Mat Res 7: 769-784. Apfelbaum F, Mayer I, Featherstone JDB (1990). LeGeros RZ (1988). Calcium phosphate materials The role of HPO/- and CO/- ions in the transformation in restorative dentistry: A review. Adv Dent Res 2: of synthetic apatites to ,S-Ca(P04h. J Inorg Biochem 38: 164-180. 1-8. LeGeros RZ, Parson JR, Daculsi G , Driessen F, Binder-Royer A (1990). Etude de !'influence de Lee D, Liu ST, Metsger S, Peterson D, Walker M la composition de 1'hydroxyapatite frittee sur ses propri­ ( 1988). Significance of the porosity and physical chemi­ etes mecaniques (Study on the influence of the composi­ stry of calcium phosphate ceramics. Biodegradation-Rio­ tion of sintered hydroxyapatite on its mechanical proper­ resorption. Bioceramics: Material characteristics versus ties). Doctoral Thesis, Universite de Grenoble, France: in vivo behavior. Ann NY Acad Sci 523: 268-271. 47-49. Meurman JH, Voegel JC, Rauhamaa-Makinen R, Bonar LC, Roufosse AH, Sabine WK, Grynpas Gasser P, Thomann JM, Hemmerle J, Luomanen M , MD, Glimcher MJ (1983). X-ray diffraction studies of Paunio I, Frank RM (1992) . Effects of carbon dioxide, the crystallinity of bone mineral in newly synthesized Nd:YAG and carbon dioxide-Nd:YAG combination and density fractionated bone. Calcif Tissue Int 35: lasers at high energy densities on synthetic hydroxy­ 202-209. apatite. Caries Res 26: 77-83. Chantraine A, Very JM, Baud CA (1988). A bio­ Nery EB, Lynch KL (1978): Preliminary clinical physical study of posttraumatic ectopic ossification. Clin studies of bioceramics in periodontal osseous defects. J Orthopaed Rel Res: 255: 289-292. Periodontal 49: 523-529. Daculsi G, LeGeros RZ (1989). Comparative dis­ Nery EB, Lee KK , Czajkowski S, Dooner JJ, solution of enamel apatites and sintered HA. In: Tooth Duggan M , Ellinger RF, Henkin JM, Hines R, Miller Enamel V. Fearnhead R (ed.). Florence Publishers, M , Olson JW, Rafferty M, SullivanT, Walters P, Welch Yokohama, Japan. pp. 482-488. D , Williams A (1990). A Veterans Administration coop­ Daculsi G, Passuti N , Deudon C, Martin S, erative study of biphasic calcium phosphate ceramic in LeGeros RZ (1990). Macroporous calcium phosphate ce­ periodontal osseous defects. J Periodontal 61: 737-744. ramic for long bone surgery in human and dogs. Clinical Okazaki M , Sato M (1990). Computer graphics of and histological study. J Biomed Mater Res 24: 379-396. hydroxyapatite and 13-tricalcium phosphate. Biomaterials Daculsi G, Bagot d'Arc M , Corlieu P, Gersdorff 11: 573-578. M (1992). Macroporous biphasic calcium phosphate effi­ Passuti N, Daculsi G (1989). Les ceramiques en ciency in mastoid cavity obliteration: experimental and phosphate de calcium en chirurgie orthopedique (Calci­ clinical findings. Ann Otol Rhinol Laryngol 8 : 669-674. um phosphate ceramics in dental surgery). La Presse de Groot K (1983). Ceramics of calcium phos­ Med 18: 28-31. phates: Preparation and properties. In: Bioceramics of Posner AS , Eanes ED, Harper RA, Zipkin I Calcium Phosphate. De Groot K (ed.). CRC Press, Boca (1963). X-ray diffraction analysis of the effect of fluo­ Raton, FL. 99-114. ride on human bone apatite. Arch oral Biol 8: 549-570. Dehen M, Niederdellmann H (1989). Einlagerung Rey C, Renugopalakrishnan V, Collins B, von pyrolisiertem Knochenersatzmaterial im ersatz­ Glimcher MJ (1991). Fourier transform infrared spec­ starken Knochenlager (Insertion of pyrolyzed bone re­ troscopic study of the carbonate ions in bone mineral placement material in heavily substituted bone layers). during aging. Calcif Tissue Int 49: 251 -258. Dtsch Zahnarztl Z: 9: 695-697. Sakae T (1988). X-ray diffraction and thermal Ducheyne P, Radin S, King L (1993). The effect studies of crystals from the outer and inner layers of of calcium phosphate ceramic composition and structure enamel from human dental enamel. Arch oral Biol 33: on in vitro behavior. I. Dissolution. J Biomed Mater Res 707-713. 27: 25-34. Sakae T, Davies JE, Frank RM, Nagai N (1989). Ellies LG, Nelson DGA, Featherstone JDB Crystallographic properties of a series of synthetic (1992). Crystallographic changes in calcium phosphates hydroxyapatites. J Nihon Univ Sch Dent 31: 458-463. during plasma-spraying. Biomaterials 13: 313-316. Thomann JM, Voegel JC, Gramain Ph (1990). Fowler BO , Kuroda S (1986). Changes in heated Kinetics of dissolution of calcium hydroxyapatite pow­ and in laser-irradiated human tooth enamel and their der. III: pH and sample conditioning effects. Calcif probable effects on solubility. Calcif Tissue Int 38: Tissue Res 46: 121-129. 197-208. van Blitterswijk CA, Grote JJ (1989). Biological Hamel L (1992). Interet preprothetique et prothe­ performance of ceramics during inflammation and infec­ tique des phosphates de calcium en odontologie (Pre­ tion. Biomaterials 5: 13-43. prosthetic and prosthetic uses of calcium pho phates in Yetter U, Eanes ED, Kopp JB, Termine JD, dentistry). Doctoral Thesis, Universite de Nantes, Robey PG (1991). Changes in apatite crystal size in France: 61-82. bones of patients with osteogenesis imperfecta. Calcif Klein CPAT, Driessen AA, de Groot K , van den Tissue Int 49: 248-250. Hooff (1983). Biodegradation behavior of various Weng HT, Uoshima M , Lin CT, Kinoshita A,

381 W. Bohne, J.A. Pouezat, L. Peru, G. Daculsi

Borgese DA, Oda S, Ishikawa I (1991). Evidence of Discussion with Reviewers rapid bone replacement after implantation of biphasic calcium phosphate ceramics into periodontal defects in C. Rey: Please describe the starting materials more monkeys. Dent Jpn 28: 155- 159. precisely. Is HA stoichiometric? How was the Young RA, Brown WE (1982). Structures of bio­ stoichiometry measured? logical minerals. In: Biological Mineralization and Authors: /)-TCP was stoichiometric with a Ca/P ratio Demineralization. Nancollas GH (ed.). Springer, Berlin. of 1. 5. HA was prepared by precipitation. It was pp. 10 1-141. carbonated and non-stoichiometric.

C. Rey: The transformation of a-TCP into {J-TCP de­ pends essentially on the cooling rate of the sample. Two different experiments cannot be compared and the dis­ crepancy observed might not be due to "synthesis or manufacturing conditions" but simply to different cooling rates. Authors: In our study, we did not observe transforma­ tion of a-TCP to/)-TCP during cooling, but we mention­ ed this transformation because it was observed by Daculsi and LeGeros (1989) and by Ducheyne et al. (1993).

382