Professional Documents
Culture Documents
Abstract
A calcium phosphate phase that is equivalent in composition (carbonate content) and crystallinity (nanocrystals) to the mineral
phase of bone was prepared by a continuous precipitation method. The powder was compacted by cold isostatic pressing into
desired shapes with high compressive strength (range of 20–50 MPa, depending on the geometry). It is concluded that such implant
materials can be prepared with a fine-tuned biodegradability in combination with a high mechanical strength. The high mechanical
strength of the objects also permits further mechanical shaping procedures like drilling or cutting.
r 2003 Elsevier Ltd. All rights reserved.
Keywords: Calcium phosphates; Cold isostatic pressing; Bone substitution materials; Mechanical stability; Porosity
0142-9612/03/$ - see front matter r 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0142-9612(03)00281-3
4566 D. Tadic, M. Epple / Biomaterials 24 (2003) 4565–4571 4566
produced by remodelling. Necessary conditions are a The dried carbonate apatite powders were ball-milled
sufficient mechanical integrity, a controlled degradation and sieved to 250–400 mm particle size before processing.
rate, and a bioactivity that stimulates a simultaneous Compact ceramic bodies of variable shape were
replacement by newly formed bone tissue. Implant prepared by CIP (4000 bar) in a Dieffenbacher Isostat
degradation and new bone formation must be balanced press at room temperature (ca. 25 C). Mechanical
in such a way that a mechanically weakened stage parameters were determined on an electromechanical
during the healing process is avoided. It was the aim of testing machine by Schenck Trebel RM 100 (100 kN)
our research to prepare a synthetic carbonated hydro- with 0.5 mm min 1 on cylinders of different size accord-
xyapatite with the chemical composition and crystal- ing to DIN EN 658-2. For each geometry, three objects
linity of bone mineral, taking into account recent work were investigated and the results were averaged.
on carbonated apatite by other groups [21,25–29].
Earlier [30] we have shown how to prepare bone-like
carbonated apatites by an easy continuous process in 3. Results and discussion
unlimited amounts with constant quality. Here we show
for the first time that such material from precipitation Fig. 1 shows X-ray diffraction patterns of carbonated
can be shaped into macroscopic objects of different size apatite (4–5 wt% carbonate content) before and after
and shape by cold isostatic pressing (CIP). These objects CIP in comparison with natural human bone [19]. The
have a high mechanical stability while their chemical diffractogram shows two very broad peaks at 26 2Y
and crystallographic properties remain unchanged. and 31–34 2Y that are typical for a poorly crystalline
Further mechanical shaping (like drilling or cutting) is (nanocrystalline) apatite. Note that CIP does not
possible, making available porous objects that are influence the crystal structure of the obtained carbo-
suitable for bone tissue ingrowth. nated apatite. This is supported by IR and TGA
experiments before and after pressing that showed no
change in the composition. In particular, the presence of
carbonate is still evident from the IR absorption bands
2. Materials and methods at 870, 1420, 1480 and 1540 cm 1 (Fig. 2). Typical
objects of different size and shape prepared by CIP are
Biomimetic carbonated apatite comparable in chemi- shown in Fig. 3.
cal composition and crystallinity to bone mineral was After precipitation from solution, the material con-
obtained by continuous fast precipitation in a special sists of almost uniform spheres with a diameter of about
crystallisation device [30]. Briefly, aqueous solutions of 20–40 nm (Fig. 4a). These are agglomerated to larger
(NH4 )2 HPO4 and (NH4)2CO3 with a molar ratio of particles of irregular shape in the micrometre-size range
CO23 to PO34 of 0.5:1 were continuously mixed under (Fig. 4b). After CIP, macroscopic objects are obtained
nitrogen atmosphere at 37 C with an aqueous solution that consist of the ‘‘fused’’ primary particles in un-
3
of Ca(NO3)2 to result in a Ca2+ to PO4 molar ratio of changed geometry (Figs. 4c and d). Again, we conclude
1.67:1 after mixing. Both solutions were adjusted 100–400 C). This corresponds to a molar Ca/P ratio of
previously to pH 10 with ammonia solution beforehand. 1.59:1.
The obtained powder was immediately filtered to avoid
crystal growth and dried at 70 C. Further details on the
experimental set-up can be found in Ref. [30]. All
solutions were prepared with compounds of p.a. quality
from Merck (Darmstadt, Germany).
The precipitated and dried calcium phosphate powder
was analysed by X-ray diffraction (XRD; Bruker AXS
D8 Advance; CuKa), infrared spectroscopy (IR; KBr;
Perkin-Elmer 1720X), thermogravimetric analysis
(TGA; TG/DTA-S II, Seiko Instruments Exstar 6000,
25–1000 C; 10 K min 1) and scanning electron micro-
scopy (SEM; LEO 1530; gold-sputtered samples).
Elemental analysis of the samples gave a calcium
content of 34.7 wt% (by atomic absorption spectro-
scopy), a phosphate content of 51.6 wt% (by photo-
metry), a carbonate content of 4–5 wt% (by TGA;
400–1000 C) and a water content of 7–8 wt% (by TGA;
4567 D. Tadic, M. Epple / Biomaterials 24 (2003) 4565–4571 4567
intensity / a.u.
natural bone
after CIP
before CIP
20 25 30 35 40
diffraction angle / 2
Preliminary attempts to prepare macroscopic objects strengths of 21.8–47.6 MPa (Table 1). Although the
by uniaxial hot pressing (4 kbar, 240 C) [30] showed geometry is not fully comparable with the earlier results,
that the mechanical stability of such objects was limited it is evident that the mechanical strength has improved
due to insufficient compaction (Table 1). In addition, by a factor of about 5–10. Apart from these numbers,
the objects were rather brittle, particularly at the edges. this is also demonstrated by an overall improved
This was decisively improved by the CIP technique. For mechanical stability towards manual scratching, torsion
cylinders of different diameters and heights, we obtained and bending, including the edges. Although compressive
Young’s moduli from 0.8 to 2.2 GPa and compression strength and Young’s modulus of the natural biomate-
rials bone and teeth are not achieved, the values are
within the range of other porous or compact calcium
P-O P-O phosphate-based biomaterials (Table 1). Isostatic press-
natural bone
C-O
ing of hydroxyapatite and sintered carbonated apatite at
OH 600 MPa (6 kbar) was also reported by Doi et al. [31],
H2O
after CIP P-O P-O but no mechanical properties were given.
H2O C-O
transmittance
(a) (b)
(c) (d)
Fig. 4. Scanning electron micrographs of synthetic bone mineral. (a) Spherical nanoparticles as obtained after precipitation; (b) Agglomerates of the
precipitated nanoparticles; (c) Surface of a cold-isostatically pressed material. The nanoparticles are preserved. (d) Fracture surface of a cold-
isostatically pressed material.
Table 1
Mechanical parameters of cold-isostatically pressed samples in comparison with other biominerals (bone and teeth) and biomaterials (selection from
the literature)
13 mm
7 mm
18 mm
16 mm
500-1500 m
objects with macropores of 500, 1000, and 1500 mm dissolved within a reasonable time, i.e. the implant is
diameter, depending on the drill size used (Fig. 5). Such finally replaced by new bone tissue. This is not the case
objects should be suitable as bone implants as osteo- for sintered hydroxyapatite ceramics as they have a
blasts are able to enter pores of this size. In vivo much lower solubility than the nanocrystalline carbo-
experiments of HAP implanted in rabbits showed a nated bone mineral that is dissolved by osteoclasts.
maximum in osteoconductivity at pore diameters of We have shown previously that under acidic condi-
about 130–260 mm [32] and 300 mm [33]. Of course, the tions (pH 4.4, pure water), our synthesised carbonated
mechanical strength of porous implants will be smaller apatites showed a much higher solubility than sintered
than that of compact implants. Like all ceramics, these calcium phosphate ceramics, like b-tricalcium phosphate
synthetic bone materials are brittle and have a limited (b-TCP) or highly crystalline sintered hydroxyapatite
elasticity (much smaller than that of natural bone). [30]. The surface of cold-isostatically pressed objects of
It must be noted that it is not clear at the moment why bone mineral-like apatite was investigated after the
such mechanically strong objects are obtained by CIP. simulated osteoclastic resorption (after 72 h at pH 4.4)
Definitely, the temperature is too low for any true [30]. It shows ‘‘resorption pits’’ of about 2 mm diameter
sintering to occur. Currently, we assume that the high that incorporate sub-micrometre-sized needles (Fig. 6).
specific surface area of the nanoparticles and the These probably result from a recrystallisation in
residual water content lead to some ‘‘chemical sinter- solution at that pH where dicalcium phosphate dihy-
ing/surface melting/recrystallisation’’ under this high drate (DCPD) has precipitated. However, the amount of
pressure. CIP experiments with completely dry powders foreign material is too small to be detected by X-ray
did not give mechanically stable objects (i.e. they could diffraction.
be crushed between the fingers), supporting this We therefore assume that macroscopic implants of
assumption. this synthetic bone mineral would participate in the
If the biological performance of different bioceramics general remodelling process of the body and thus be
is compared, it must be noted that the biological resorbed within a reasonable time. A macroporous
incorporation of a bone graft is a physiological process implant should also be degraded from within when bone
controlled by the cells (mainly osteoblasts and osteo- cells have grown into the macroporous material (500–
clasts) of the surrounding host bone. This biochemical 1500 mm).
response is influenced by the geometry, the chemical
composition and the morphological and mechanical
properties of a given biomaterial. While a compact 4. Conclusions
synthetic bioceramic shows good mechanical properties
(Table 1), only the surface of ceramic will only be in By a continuous crystallisation method, it is possible
contact with tissue and there will usually be no ingrowth to prepare bone mineral-like calcium phosphates with
of bone but only a degradation starting from the nanocrystallinity and a carbonate content as in bone
surface. Consequently, compact materials achieve only mineral in unlimited amounts with reproducible proper-
a geometrical fixation while porous materials provide a ties. By CIP, the precipitated powders can be processed
mechanical interlock caused by ingrowth of bone tissue into objects of a desired shape with considerable
into the pores. Due to this reason, porous bioceramics mechanical strength. Crystallinity and composition of
are preferred in clinical practice. Clinical practice also the material are not affected by this processing step.
demands that a bone graft material is resorbed or This offers the possibility to prepare biomimetic
4570 D. Tadic, M. Epple / Biomaterials 24 (2003) 4565–4571 4570
(a) (b)
Fig. 6. Surface of cold-isostatically pressed material after 72 h at pH 4.4 resembling osteoclastic resorption). In low magnification (a), resorption pits
are seen. In higher magnification (b), the surface is covered by small needles, probably of DCPD that was reprecipitated under these conditions.
materials for optimal bone substitution. Further shaping [4] Bohner M. Physical and chemical aspects of calcium phosphates
can be done by cutting and drilling, thereby creating used in spinal surgery. Eur Spine J 2001;10:S114–21.
macroporous implants or geometrically designed im- [5] Dorozhkin SV, Epple M. Biological and medical significance of
calcium phosphates. Angew Chem Int Ed Engl 2002;41:3130–46.
plants to meet individual implantation sites. Based on [6] Liu DM. Fabrication of hydroxyapatite ceramic with controlled
earlier in vitro resorption experiments at pH 4.4 (see porosity. J Mater Sci Mater Med 1997;8:227–32.
above), we propose such materials as biomimetic bone [7] Layrolle P, Ito A, Tateishi T. Sol–gel synthesis of amorphous
substitution materials that can be more easily resorbed calcium phosphate and sintering into microporous hydroxyapatite
during bone formation in the defect than sintered bioceramics. J Am Ceram Soc 1998;81:1421–8.
[8] Linhart W, Peters F, Lehmann W, Schilling AF, Schwarz K,
hydroxyapatite ceramics. As chemical composition, Amling M, Rueger JM, Epple M. Biologically and chemically
crystallinity, geometry, and macroporosity can be optimized composites of carbonated apatite and polyglycolide as
adjusted by the preparation process, we believe that bone substitution materials. J Biomed Mater Res 2001;54:162–71.
the biological requirements for different applications [9] Verdonschot N, van Hal CTH, Schreurs BW, Buma P, Huiskes R,
(orthopaedic surgery, maxillofacial surgery) can be met Slooff TJJH. Time-dependent mechanical properties of HA/TCP
particles in relation to morsellized bone grafts for use in impaction
by choice of the appropriate material. grafting. J Biomed Mater Res Appl Biomater 2001;58:599–604.
[10] Zyman ZZ, Ivanov I, Rochmistrov D, Glushko V, Tkachenko N,
Kijko S. Sintering peculiarities for hydroxyapatite with different
degrees of crystallinity. J Biomed Mater Res 2001;54:256–63.
Acknowledgements [11] Wen HB, de Wijn JR, van Blitterswijk CA, de Groot K.
Incorporation of bovine serum albumin in calcium phosphate
This project was supported by the Deutsche For- coating on titanium. J Biomed Mater Res 1999;46:245–52.
schungsgemeinschaft (Bonn) and the Fonds der Che- [12] Esenwein SA, Esenwein S, Herr G, Muhr G, Ku.sswetter W,
mischen Industrie (Frankfurt). We are grateful to Prof. Hartwig CH. Osteogene Aktivita.t BMP-3-beschichteter Titanpr-
o.fko.rper unterschiedlicher Oberfla.chenbeschaffenheit im ortho-
D. Sto.ver, Dr. H.P. Buchkremer, Dr. E. Schu.ller and topen Lager des Riesenkaninchens. Chirurg 2001;72:1360–8.
Dr. M. Bram (Research Center Ju.lich, Institute for [13] Kokubo T. Apatite formation on surfaces of ceramics, metals and
Materials and Processes in Energy Systems) and to polymers in body environment. Acta Mater 1998;46:2519–27.
Matthias Matthes for help with the cold isostatic [14] Willmann G. Coating of implants with hydroxapatite—material
pressing, and to Prof. G. Eggeler and Dr. B. Skrotzki connections between bone and metal. Adv Eng Mater 1999;1:
95–105.
(Materials Science, University of Bochum) for help with [15] Epple M, Choi J, Ko.ller M, Bogdanski D, Muhr G, Mu.ller D.
mechanical testing. Calcium phosphate coating of Nitinol shape memory alloys for
surgical application. Eur J Trauma 2002;28:139–41.
[16] Habibovic P, Barre"re F, van Blitterswijk CA, de Groot K,
Layrolle P. Biomimetic hydroxyapatite coating on metal implants.
References J Am Ceram Soc 2002;85:517–22.
[17] den Boer FC, Wippermann BW, Blokhuis TJ, Patka P, Bakker
[1] LeGeros RZ. Calcium phosphates in oral biology and medicine. FC, Haarman HJTM. Healing of segmental bone defects with
In: Myers HM, editor. Monographs in oral science. Basel: Karger; granular porous hydroxyapatite augmented with recombinant
1991. human osteogenic protein-1 or autologous bone marrow.
[2] Elliot JC. Structure, chemistry of the Apatites and Other Calcium J Orthopaed Res 2003;21:521–8.
Orthophosphates. Studies in Inorganic Chemistry, vol. 18. [18] Weiner S, Wagner HD. The material bone: structure–mechanical
Amsterdam: Elsevier; 1994. function relations. Annu Rev Mater Sci 1998;28:271–98.
[3] Suchanek W, Yoshimura M. Processing and properties of [19] Peters F, Schwarz K, Epple M. The structure of bone studied with
hydroxyapatite-based biomaterials for use as hard tissue replace- synchrotron X-ray diffraction, X-ray absorption spectroscopy
ment implants. J Mater Res 1998;13:94–117. and thermal analysis. Thermochim Acta 2000;361:131–8.
4571 D. Tadic, M. Epple / Biomaterials 24 (2003) 4565–4571 4571