- Email: [email protected]
Investigation of calcium phosphate formation from calcium propionate and triethyl phosphate
Ceramics International 42 (2016) 14061–14065
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Investigation of calcium phosphate formation from calcium propionate and triethyl phosphate Per Kjellin a,n, Anand Kumar Rajasekharan b, Fredrik Currie a, Paul Handa a a b
Promimic AB, AZ BioVentureHub, HB4, c/o AstraZeneca R&D, SE-431 83 Mölndal, Sweden Department of Chemistry and Chemical Engineering, Chalmers University of Technology, SE-41296 Göteborg, Sweden
art ic l e i nf o
a b s t r a c t
Article history: Received 24 March 2016 Received in revised form 26 May 2016 Accepted 3 June 2016 Available online 4 June 2016
Synthetic calcium phosphates are used in for example bone cements and implant coatings to increase biocompatibility. The common method to produce tricalcium phosphate (TCP) uses high temperatures, which creates large crystals with low specific surface areas. In order to investigate new methods to produce TCP at lower temperatures, the reaction between calcium propionate and triethyl phosphate conducted at 220 °C was studied. The method had a near 100% conversion rate, the main synthesis products were calcium phosphate and ethyl propionate. The formed calcium phosphate polymorph could be controlled depending on the water content of the precursor mixture. Anhydrous conditions created amorphous calcium phosphate. As the concentration of water increased, β-TCP was formed, followed by calcium deficient hydroxyapatite and monetite. The particle size increased with the water content, from 20 to 40 nm for amorphous calcium phosphate to tenths of micrometers for monetite. The specific surface areas varied between 209 m2/g for the amorphous product to 3.6 m2/g for the monetite product. & 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: TCP Calcium phosphates Nanocrystals Bone cements
1. Introduction Calcium phosphates are well known for their ability to stimulate the growth of natural bone. Hydroxyapatite (HA), tricalcium phosphate (TCP) and tetracalcium phosphate (TTCP) are some examples of calcium phosphate compounds which have a beneficial effect on the growth of bone tissue. Among these, hydroxyapatite is chemically most similar to natural bone [1], but has a slow resorption time when used as a bone substitute material. TCP, with the chemical formula Ca3(PO4)2, is a common ingredient in bone substitutes since it stimulates the growth of new bone, but also because it resorbs faster in the human body than HA. The effect of TCP on osteogenesis was studied as early as 1920 [2]. TCP exists in several polymorphs, where α-TCP and β-TCP are the most intensely studied. β-TCP is stable at room temperature whereas αTCP is the high temperature polymorph and is stable between 1125 °C and 1430 °C [3]. β-TCP is for example used as blocks or granules for bone defects or spinal fusions [4,5]. α-TCP forms a self-setting cement with water and is therefore used in calcium phosphate cements [6,7]. Due to its ability to react with water, it is generally considered that TCP has to be synthesized under anhydrous conditions. One common method to produce TCP is to react a calcium compound n
Corresponding author. E-mail address: [email protected] (P. Kjellin).
http://dx.doi.org/10.1016/j.ceramint.2016.06.013 0272-8842/& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
such as CaCO3 with a phosphorous containing precursor such as CaHPO4 at temperatures of above 800 °C [8,9]: 2CaHPO4 2H2O þCaCO3-Ca3(PO4)2 þ5H2O þCO2 Another method works by calcining amorphous calcium phosphate (ACP) or calcium deficient HA (CDHA), heat treatment at 700–1300 °C is used to induce a phase transformation from ACP to TCP or from CDHA to TCP [10–12]. A more recent synthesis method for TCP is flame spray pyrolysis, where calcium phosphates are produced in a flame, and the products are subsequently crystallized to TCP at 900 °C [13,14]. Synthesis of TCP at high temperatures may not be desirable in some cases. Calcination at high temperatures cause sintering effects which increase particle size and a decrease in the specific surface area. A small particle size creates stronger bone cements and small particles also resorb faster in the human body. A common method of decreasing the size of the particles and crystallites is by ball milling, either of the calcined product or before the heat treatment [15,16]. This is a slow process and may be expensive to scale up. Methods which create small particle sizes of TCP directly and at low temperatures are therefore of interest. There are a few reports in the literature which show that β-TCP can form at lower temperatures, these include precipitation in methanol [17], formation under hydrothermal conditions [18,19], and precipitation in ethylene glycol [20,21]. In this study we investigate the reaction between triethyl phosphate and calcium propionate. In theory, this synthesis should
14062
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
proceed as: 2(C2H5)3PO4 þ 3 Ca(C2H5COO)2 -Ca3(PO4)2 þ 6C2H5(C2H5COO) Thus, the synthesis products should be calcium phosphate and ethyl propionate. Using triethyl phosphate as a phosphorous precursor has the advantage that it is easy to control the water content of the reaction mixture and perform the synthesis under anhydrous conditions, as water is more easily removed from triethyl phosphate compared to other phosphorous compounds, such as for example phosphoric acid. Another purpose of the study was to investigate the formed calcium products from a reaction between a phosphate ester and a calcium carboxylate, since this reaction is a possible candidate to produce in situ polymerized calcium phosphate/polyester composites [22].
2. Experimental 2.1. Calcium phosphate synthesis In a typical experiment, 9.31 g (0.05 mol) of calcium propionate (Fluka, 99%) was thoroughly mixed with 6.07 g (0.033 mol) of triethyl phosphate (Aldrich, 99.8%), and, for some of the experiments, with various amounts of water. Prior to the mixing, the calcium propionate was dried at 120 °C for 12 h to remove absorbed water. The triethyl phosphate was dried by using 4 Å molecular sieves (2–3 mm diameter, Scharlau). The mixture was placed in a glass vial in a PARR stainless steel autoclave with a PTFE inner vessel. The autoclave was placed in an oven at 220 °C for 24 h. The autoclave was allowed to cool to room temperature and the product was collected. The product consisted of a white solid substance and a liquid with a fruity odor. The liquid was collected for further analysis. The solid was split in two parts, one was calcined in an oven at 350 °C and the other part was washed repeatedly with a total volume of 500 ml of isopropanol and dried at 120 °C for 1 h. 2.2. Characterization For the X-ray diffraction (XRD) measurements, a Bruker D8 powder diffractometer with CuKα radiation (1.54 Å) was used. The scanning electron microscopy (SEM) analysis was done with a Zeiss Ultra 55 equipped with an Oxford INCA energy dispersive x-ray spectroscopy (EDX) system. Gas chromatography (GC) was performed with a Thermoquest Trace 2000 equipped with a Fisons MD800 mass spectrometer. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) measurements were done with a Perkin Elmer Optima 5300DV. The specific surface area was measured with a Micromeritics Tristar instrument, using nitrogen adsorption and the BET algorithm for specific surface area calculation.
3. Results and discussion In a series of experiments, the water content in the precursor mixture was successively increased, from anhydrous to 60 mmol. Fig. 1 shows XRD diffractograms for the synthesis with different concentrations of H2O. Diffractogram (A) is for the product synthesized under anhydrous conditions. As seen, this diffractogram has a large hump around 2θ ¼ 30° and no distinct diffraction peaks, this indicates that the sample consisted of ACP. Above 15 mmol, crystalline β-TCP was formed (B). A water content of above 40 mmol yielded CDHA (D). A water content of 60 mmol gave monetite, with minor CDHA content (E). The XRD measurements
Fig. 1. XRD diffractogram of calcium phosphate formed with different quantities of water. A) anhydrous conditions; B) Ca/H2O¼ 3.33; C) Ca/H2O¼ 1.67; D) Ca/ H2O¼ 1.25; E) Ca/H2O ¼ 0.83.
Table 1 Overview of the calcium phosphates formed with different water contents. Water amount
Molar ratio Ca/H2O
Polymorph
0 5.5 mmol 10 mmol 15 mmol 30 mmol 40 mmol 60 mmol
– 9 5 3.33 1.67 1.25 0.83
Amorphous Amorphous Amorphous β-TCP β-TCP þ CDHA CDHA Monetite
are summarized in Table 1. From the XRD analysis, no difference was seen between the samples which were calcined at 350 °C and the samples which were washed with isopropanol. The only difference between these cleaning methods was the color; calcined samples had a slight brownish tint whereas the isopropanol-washed were clear white, which indicates that the calcination procedure left small amounts of carbon in the sample. SEM analysis was done on samples of the amorphous powder, crystalline β-TCP and the monetite sample. Fig. 2(a) and (b) shows images from the amorphous samples. At low magnification it can be seen that the amorphous samples formed irregular shaped particles with sharp edges and smooth surfaces, the size of the particles were approximately 100–200 mm. Higher magnifications reveal that these particles consisted of granules with sizes of 20–40 nm. The crystalline β-TCP samples formed similar structures on the micrometer level, and at higher magnifications it can be seen that these agglomerates also were composed of spherical granules (Fig. 3(a) and (b). The main difference is that these granules are smaller compared with the amorphous samples. EDX measurements for the amorphous and β-TCP samples revealed a Ca/P ratio of 1.46, close to the theoretical ratio of 1.5. At a H2O content of above 40 mmol, monetite was formed instead of CDHA. SEM analysis of these samples show flake-like particles at low magnification (Fig. 4(a)), while higher magnifications reveal sharp-edged crystals with clearly visible growth steps (Fig. 4(b)). Each crystal was covered with a thin layer of irregular shaped particles, most probably this phase is CDHA, as seen in the diffractogram at around 2θ ¼32°. The nitrogen adsorption measurements showed a specific surface area for the amorphous sample of 209 m2/g while the β-TCP sample had a surface area of 65 m2/g. For the monetite samples, the specific surface area was 3.6 m2/g.
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
Fig. 2. a and b. SEM images of samples consisting of amorphous calcium phosphate.
The liquid ester from the β-TCP sample with 0.27 g water added in the precursors was analyzed with Karl Fischer titration, in order to measure the total amount of water remaining in the liquid part. This analysis showed a water content of approximately 0.1 g. Accordingly, of the 0.27 g added water approximately 0.17 g was absorbed in the TCP powder and 0.1 g in the ester. In order to assess the stability of the β-TCP at higher temperatures, a sample was heated to 1000 °C for 12 h. The XRD diffractograms for the sample before and after heat treatment is shown in Fig. 5. Vertical bars indicate reflections for β-TCP (JCPDS 09-0169). As seen from the diffractograms, both samples consist of β-TCP with nearly identical peak positions. The heat treated sample has slightly more narrow peaks which indicates sintering and crystal growth of the particles. Also, small peaks at 2θ ¼31.8° and 49.5° seen for the heat treated sample coincide with the XRD pattern for hydroxyapatite (JCPDS 09-0432), with reflections (2 1 1) and (2 1 3), respectively. Due to the low intensity of these peaks, the type of apatite which was formed is hard to determine. The liquid part from the synthesis was investigated with gas chromatography. A mixture of equal weights of triethyl phosphate and ethyl propionate was prepared, this chromatogram is seen in Fig. 6, lower image (A). Triethyl phosphate had a retention time of 15.82 min, whereas ethyl propionate had a retention time of 7.04 min. Samples from the synthesis were also analyzed, shown in the upper image (B) of Fig. 6. As seen, this sample consists mainly of ethyl propionate, with a minor content of propanoic acid (approx. 1%, seen directly to the right of the ethyl propionate
14063
Fig. 3. a and b. SEM images of samples consisting of crystalline TCP.
peak). This shows that the triethyl phosphate is completely decomposed during the synthesis. However, there was the possibility that heavier phosphorous containing compounds were dissolved in the mixture without showing up on GC, so ester samples were further investigated with ICP-AES. These measurements revealed a total phosphorous content of 0.25 wt%, which further corroborates that a very high amount of the phosphate in the triethyl phosphate is converted to TCP. In contrast to the calcium phosphate, the ester formation was not affected by the water content of the precursors, ethyl propionate was the main product, regardless of the water amount. The amount of water in the precursors had a direct effect on the formed calcium phosphate polymorph. As seen from Table 1, ACP was formed at anhydrous conditions and up to a Ca/H2O molar ratio of 3.33, above this water content β-TCP was the synthesis product. It can also be seen from this Table that the formation of β-TCP occurred in an interval between Ca/H2O¼3.33–1.67. Multiple experiments in this range revealed that the border between ACP and βTCP formation was quite sharp, and no diffractogram showed a coexistence between ACP and β-TCP. In contrast, the crystalline phases were mixed in certain intervals, at Ca/H2O41.67, CDHA together with β-TCP was formed. The effect of water on phase transformations of ACP is described in the literature, but this phase transformation is normally direct from ACP to CDHA [23]. The reason to the phase transformation at certain Ca/H2O ratios is unclear, and will be subject to further investigation in the future. The ester formation on the other hand was not affected by the amount of water, ethyl propionate was the main product
14064
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
Fig. 6. GC diagrams of a 50/50 mixture of ethyl propionate and triethyl phosphate (diagram A) and sample from the synthesis (diagram B).
decrease in specific surface area (and an increase in crystal size). An interesting possibility of this synthesis method would be to use a dicarboxylate such as calcium succinate as a calcium source and use phosphorylated glycerol as the phosphorous source, to produce an in situ polymerized, biodegradable polymer/TCP scaffold.
4. Conclusions
Fig. 4. a and b. SEM images of samples consisting of calcium phosphate (Monetite).
The calcium phosphate formation from the reaction between calcium propionate and triethyl phosphate was investigated. The inorganic part of the synthesis products showed to be highly sensitive to the water content of the precursors. Amorphous calcium phosphate was formed at anhydrous conditions, whereas addition of a small amounts of water created β-TCP. Higher amounts of water created increasingly higher amounts of CDHA and monetite. The formed ACP and β-TCP were nanosized with high specific surface areas, which makes this method potentially useful to create in situ polymerized ACP/polymer or β-TCP/polymer scaffolds.
References Fig. 5. XRD diffractograms of a β-TCP sample, as synthesized (lower diffractogram) and the same sample heat treated at 1000 °C (upper diffractogram). Vertical bars show reflections for β-TCP from literature data. (2 1 1) and (2 1 3) are hydroxyapatite reflections.
regardless of water content. As Karl Fischer titration showed, parts of the water was present in the formed ester, and some parts were present in the TCP powder. The fact that the amorphous composition did not form CDHA directly with water may be due to the presence of the ester compounds. These evidently bind water and may slow down the ripening process from ACP-TCP-CDHA. Formation of monetite at higher water concentrations is unexpected, excess water should produce mainly CDHA. Monetite formation is mainly favored at lower pH [24], so the reason may be that a low pH is created by the formation of small amounts of propanoic acid. The transformation from ACP to monetite was also accompanied by a significant
[1] H.A. Lowenstam, S. Weiner, On Biomineralization, Oxford University Press, New York, 1989. [2] F.H. Albee, H.F. Morrison, Studies in bone growth, triple calcium phosphate as a stimulus to osteogenesis, Ann. Surg. 71 (1920) 32–39. [3] M. Mathew, L.W. Schroeder, B. Dickens, W.E. Brown, The Crystal Structure of α-Ca3(PO4)2, Acta Crystallogr. B 33 (1977) 1325–1333. [4] E.W.H. Bodde, J.G.C. Wolke, R.S.Z. Kowalski, J.A. Jansen, Bone regeneration of porous β-tricalcium phosphate (ConduitTM TCP) and of biphasic calcium phosphate ceramic (Biosel1) in trabecular defects in sheep, J. Biomed. Mater. Res. A 82A (2007) 521–776. [5] G. Russmueller, D. Moser, E. Spassova, R. Plasenzotti, P.W. Poeschl, R. Seemann, S. Becker, K. Pirklbauer, C. Eder-Czembirek, C. Czembirek, C. Perisanidis, R. Ewers, C. Schopper, Tricalcium phosphate-based biocomposites for mandibular bone regeneration – a histological study in sheep, J. Cranio Maxillofac. Surg. 43 (2015) 696–704. [6] K.S. TenHuisen, P.W. Brown, Formation of calcium-deficient hydroxyapatite from α-tricalcium phosphate, Biomaterials 19 (1998) 2209–2217. [7] H. Monma, S. Ueno, T. Kanazawa, Properties of hydroxyapatite prepared by the hydrolysis of tricalcium phosphate, J. Chem. Tech. Biotechnol. 31 (1981) 15–24. [8] X. Yang, Z. Wang, Synthesis of biphasic ceramics of hydroxyapatite and β-tricalcium phosphate with controlled phase content and porosity, J. Mater. Chem. 8 (1998) 2233–2237.
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065 [9] M. Ermrich, F. Peters, X-ray powder diffraction data of synthetic β-tricalcium phosphate, Mat.-Wiss.Werkstofftech 37 (2006) 526–529. [10] S. Somrani, C. Rey, M. Jemal, Thermal evolution of amorphous tricalcium phosphate, J. Mater. Chem. 13 (2003) 888–892. [11] E.D. Eanes, Thermochemical studies on amorphous calcium phosphate, Calc. Tiss. Res. 5 (1970) 133–145. [12] S. Raynaud, E. Champion, D. Bernache-Assollant, P. Thomas, Calcium phosphate apatites with variable Ca/P atomic ratio I. Synthesis, characterisation and thermal stability of powders, Biomaterials 23 (2002) 1065–1072. [13] W.J. Stark, S.-E. Pratsinis, M. Maciejewski, S.F. Loher, A. Baiker, Flame synthesis of metal salt nanoparticles, in particular calcium and phosphate comprising nanoparticles, US Patent 7,879,303. [14] S. Loher, W.J. Stark, M. Maciejewski, A. Baiker, S.E. Pratsinis, D. Reichardt, F. Maspero, F. Krumeich, D. Günther, Fluoro-apatite and calcium phosphate nanoparticles by flame synthesis, Chem. Mater. 17 (2005) 36–42. [15] J. Bae, Y. Ida, K. Sekine, F. Kawano, K. Hamada, Effects of high-energy ballmilling on injectability and strength of β-tricalcium-phosphate cement, J. Mech. Behav. Biomed. Mater. 47 (2015) 77–86. [16] B. Nasiri-Tabrizi, A. Fahami, Mechanochemical synthesis and structural characterization of nano-sized amorphous tricalcium phosphate, Ceram. Int. 39 (2013) 8657–8666.
14065
[17] J.-S. Bowa, S.-C. Lioub, S.-Y. Chen, Structural characterization of room-temperature synthesized nano-sized β-tricalcium phosphate, Biomaterials 25 (2004) 3155–3161. [18] T. Toyama, K. Nakashima, T. Yasue, Hydrothermal synthesis of β-Tricalcium phosphate from amorphous calcium phosphate, J. Ceram. Soc. Jpn. 110 (2002) 716–721. [19] R. Vani, E.K. Girija, K. Elayaraja, S.P. Parthiban, R. Kesavamoorthy, S.N. Kalkura, Hydrothermal synthesis of porous triphasic hydroxyapatite/(α and β) tricalcium phosphate, J. Mater. Sci.: Mater. Med. 20 (2009) S43–S48. [20] J. Tao, W. Jiang, H. Zhai, H. Pan, X. Xu, R. Tang, Structural components and anisotropic dissolution behaviors in one hexagonal single crystal of β-Tricalcium phosphate, Cryst. Growth Des. 8 (2008) 2227–2234. [21] L. Galea, M. Bohner, J. Thuering, N. Doebelin, C.G. Aneziris, T. Graule, Control of the size, shape and composition of highly uniform, non-agglomerated, submicrometer β-tricalcium phosphate and dicalcium phosphate platelets, Biomaterials 34 (2013) 6388–6401. [22] K. Pielichowska, S. Blazewicz, Bioactive polymer/hydroxyapatite (nano)composites for bone tissue regeneration, Adv. Polym. Sci. 232 (2010) 97–207. [23] S.V. Dorozhkin, Amorphous calcium orthophosphates: nature, chemistry and biomedical applications, Int. J. Mater. Chem. 2 (2012) 19–46. [24] B. Jokic, M. Mitric, V. Radmilovic, S. Drmanic, R. Petrovic, D. Janac’kovic´, Synthesis and characterization of monetite and hydroxyapatite whiskers obtained by a hydrothermal method, Ceram. Int. 37 (2011) 167–173.
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Investigation of calcium phosphate formation from calcium propionate and triethyl phosphate Per Kjellin a,n, Anand Kumar Rajasekharan b, Fredrik Currie a, Paul Handa a a b
Promimic AB, AZ BioVentureHub, HB4, c/o AstraZeneca R&D, SE-431 83 Mölndal, Sweden Department of Chemistry and Chemical Engineering, Chalmers University of Technology, SE-41296 Göteborg, Sweden
art ic l e i nf o
a b s t r a c t
Article history: Received 24 March 2016 Received in revised form 26 May 2016 Accepted 3 June 2016 Available online 4 June 2016
Synthetic calcium phosphates are used in for example bone cements and implant coatings to increase biocompatibility. The common method to produce tricalcium phosphate (TCP) uses high temperatures, which creates large crystals with low specific surface areas. In order to investigate new methods to produce TCP at lower temperatures, the reaction between calcium propionate and triethyl phosphate conducted at 220 °C was studied. The method had a near 100% conversion rate, the main synthesis products were calcium phosphate and ethyl propionate. The formed calcium phosphate polymorph could be controlled depending on the water content of the precursor mixture. Anhydrous conditions created amorphous calcium phosphate. As the concentration of water increased, β-TCP was formed, followed by calcium deficient hydroxyapatite and monetite. The particle size increased with the water content, from 20 to 40 nm for amorphous calcium phosphate to tenths of micrometers for monetite. The specific surface areas varied between 209 m2/g for the amorphous product to 3.6 m2/g for the monetite product. & 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: TCP Calcium phosphates Nanocrystals Bone cements
1. Introduction Calcium phosphates are well known for their ability to stimulate the growth of natural bone. Hydroxyapatite (HA), tricalcium phosphate (TCP) and tetracalcium phosphate (TTCP) are some examples of calcium phosphate compounds which have a beneficial effect on the growth of bone tissue. Among these, hydroxyapatite is chemically most similar to natural bone [1], but has a slow resorption time when used as a bone substitute material. TCP, with the chemical formula Ca3(PO4)2, is a common ingredient in bone substitutes since it stimulates the growth of new bone, but also because it resorbs faster in the human body than HA. The effect of TCP on osteogenesis was studied as early as 1920 [2]. TCP exists in several polymorphs, where α-TCP and β-TCP are the most intensely studied. β-TCP is stable at room temperature whereas αTCP is the high temperature polymorph and is stable between 1125 °C and 1430 °C [3]. β-TCP is for example used as blocks or granules for bone defects or spinal fusions [4,5]. α-TCP forms a self-setting cement with water and is therefore used in calcium phosphate cements [6,7]. Due to its ability to react with water, it is generally considered that TCP has to be synthesized under anhydrous conditions. One common method to produce TCP is to react a calcium compound n
Corresponding author. E-mail address: [email protected] (P. Kjellin).
http://dx.doi.org/10.1016/j.ceramint.2016.06.013 0272-8842/& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
such as CaCO3 with a phosphorous containing precursor such as CaHPO4 at temperatures of above 800 °C [8,9]: 2CaHPO4 2H2O þCaCO3-Ca3(PO4)2 þ5H2O þCO2 Another method works by calcining amorphous calcium phosphate (ACP) or calcium deficient HA (CDHA), heat treatment at 700–1300 °C is used to induce a phase transformation from ACP to TCP or from CDHA to TCP [10–12]. A more recent synthesis method for TCP is flame spray pyrolysis, where calcium phosphates are produced in a flame, and the products are subsequently crystallized to TCP at 900 °C [13,14]. Synthesis of TCP at high temperatures may not be desirable in some cases. Calcination at high temperatures cause sintering effects which increase particle size and a decrease in the specific surface area. A small particle size creates stronger bone cements and small particles also resorb faster in the human body. A common method of decreasing the size of the particles and crystallites is by ball milling, either of the calcined product or before the heat treatment [15,16]. This is a slow process and may be expensive to scale up. Methods which create small particle sizes of TCP directly and at low temperatures are therefore of interest. There are a few reports in the literature which show that β-TCP can form at lower temperatures, these include precipitation in methanol [17], formation under hydrothermal conditions [18,19], and precipitation in ethylene glycol [20,21]. In this study we investigate the reaction between triethyl phosphate and calcium propionate. In theory, this synthesis should
14062
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
proceed as: 2(C2H5)3PO4 þ 3 Ca(C2H5COO)2 -Ca3(PO4)2 þ 6C2H5(C2H5COO) Thus, the synthesis products should be calcium phosphate and ethyl propionate. Using triethyl phosphate as a phosphorous precursor has the advantage that it is easy to control the water content of the reaction mixture and perform the synthesis under anhydrous conditions, as water is more easily removed from triethyl phosphate compared to other phosphorous compounds, such as for example phosphoric acid. Another purpose of the study was to investigate the formed calcium products from a reaction between a phosphate ester and a calcium carboxylate, since this reaction is a possible candidate to produce in situ polymerized calcium phosphate/polyester composites [22].
2. Experimental 2.1. Calcium phosphate synthesis In a typical experiment, 9.31 g (0.05 mol) of calcium propionate (Fluka, 99%) was thoroughly mixed with 6.07 g (0.033 mol) of triethyl phosphate (Aldrich, 99.8%), and, for some of the experiments, with various amounts of water. Prior to the mixing, the calcium propionate was dried at 120 °C for 12 h to remove absorbed water. The triethyl phosphate was dried by using 4 Å molecular sieves (2–3 mm diameter, Scharlau). The mixture was placed in a glass vial in a PARR stainless steel autoclave with a PTFE inner vessel. The autoclave was placed in an oven at 220 °C for 24 h. The autoclave was allowed to cool to room temperature and the product was collected. The product consisted of a white solid substance and a liquid with a fruity odor. The liquid was collected for further analysis. The solid was split in two parts, one was calcined in an oven at 350 °C and the other part was washed repeatedly with a total volume of 500 ml of isopropanol and dried at 120 °C for 1 h. 2.2. Characterization For the X-ray diffraction (XRD) measurements, a Bruker D8 powder diffractometer with CuKα radiation (1.54 Å) was used. The scanning electron microscopy (SEM) analysis was done with a Zeiss Ultra 55 equipped with an Oxford INCA energy dispersive x-ray spectroscopy (EDX) system. Gas chromatography (GC) was performed with a Thermoquest Trace 2000 equipped with a Fisons MD800 mass spectrometer. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) measurements were done with a Perkin Elmer Optima 5300DV. The specific surface area was measured with a Micromeritics Tristar instrument, using nitrogen adsorption and the BET algorithm for specific surface area calculation.
3. Results and discussion In a series of experiments, the water content in the precursor mixture was successively increased, from anhydrous to 60 mmol. Fig. 1 shows XRD diffractograms for the synthesis with different concentrations of H2O. Diffractogram (A) is for the product synthesized under anhydrous conditions. As seen, this diffractogram has a large hump around 2θ ¼ 30° and no distinct diffraction peaks, this indicates that the sample consisted of ACP. Above 15 mmol, crystalline β-TCP was formed (B). A water content of above 40 mmol yielded CDHA (D). A water content of 60 mmol gave monetite, with minor CDHA content (E). The XRD measurements
Fig. 1. XRD diffractogram of calcium phosphate formed with different quantities of water. A) anhydrous conditions; B) Ca/H2O¼ 3.33; C) Ca/H2O¼ 1.67; D) Ca/ H2O¼ 1.25; E) Ca/H2O ¼ 0.83.
Table 1 Overview of the calcium phosphates formed with different water contents. Water amount
Molar ratio Ca/H2O
Polymorph
0 5.5 mmol 10 mmol 15 mmol 30 mmol 40 mmol 60 mmol
– 9 5 3.33 1.67 1.25 0.83
Amorphous Amorphous Amorphous β-TCP β-TCP þ CDHA CDHA Monetite
are summarized in Table 1. From the XRD analysis, no difference was seen between the samples which were calcined at 350 °C and the samples which were washed with isopropanol. The only difference between these cleaning methods was the color; calcined samples had a slight brownish tint whereas the isopropanol-washed were clear white, which indicates that the calcination procedure left small amounts of carbon in the sample. SEM analysis was done on samples of the amorphous powder, crystalline β-TCP and the monetite sample. Fig. 2(a) and (b) shows images from the amorphous samples. At low magnification it can be seen that the amorphous samples formed irregular shaped particles with sharp edges and smooth surfaces, the size of the particles were approximately 100–200 mm. Higher magnifications reveal that these particles consisted of granules with sizes of 20–40 nm. The crystalline β-TCP samples formed similar structures on the micrometer level, and at higher magnifications it can be seen that these agglomerates also were composed of spherical granules (Fig. 3(a) and (b). The main difference is that these granules are smaller compared with the amorphous samples. EDX measurements for the amorphous and β-TCP samples revealed a Ca/P ratio of 1.46, close to the theoretical ratio of 1.5. At a H2O content of above 40 mmol, monetite was formed instead of CDHA. SEM analysis of these samples show flake-like particles at low magnification (Fig. 4(a)), while higher magnifications reveal sharp-edged crystals with clearly visible growth steps (Fig. 4(b)). Each crystal was covered with a thin layer of irregular shaped particles, most probably this phase is CDHA, as seen in the diffractogram at around 2θ ¼32°. The nitrogen adsorption measurements showed a specific surface area for the amorphous sample of 209 m2/g while the β-TCP sample had a surface area of 65 m2/g. For the monetite samples, the specific surface area was 3.6 m2/g.
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
Fig. 2. a and b. SEM images of samples consisting of amorphous calcium phosphate.
The liquid ester from the β-TCP sample with 0.27 g water added in the precursors was analyzed with Karl Fischer titration, in order to measure the total amount of water remaining in the liquid part. This analysis showed a water content of approximately 0.1 g. Accordingly, of the 0.27 g added water approximately 0.17 g was absorbed in the TCP powder and 0.1 g in the ester. In order to assess the stability of the β-TCP at higher temperatures, a sample was heated to 1000 °C for 12 h. The XRD diffractograms for the sample before and after heat treatment is shown in Fig. 5. Vertical bars indicate reflections for β-TCP (JCPDS 09-0169). As seen from the diffractograms, both samples consist of β-TCP with nearly identical peak positions. The heat treated sample has slightly more narrow peaks which indicates sintering and crystal growth of the particles. Also, small peaks at 2θ ¼31.8° and 49.5° seen for the heat treated sample coincide with the XRD pattern for hydroxyapatite (JCPDS 09-0432), with reflections (2 1 1) and (2 1 3), respectively. Due to the low intensity of these peaks, the type of apatite which was formed is hard to determine. The liquid part from the synthesis was investigated with gas chromatography. A mixture of equal weights of triethyl phosphate and ethyl propionate was prepared, this chromatogram is seen in Fig. 6, lower image (A). Triethyl phosphate had a retention time of 15.82 min, whereas ethyl propionate had a retention time of 7.04 min. Samples from the synthesis were also analyzed, shown in the upper image (B) of Fig. 6. As seen, this sample consists mainly of ethyl propionate, with a minor content of propanoic acid (approx. 1%, seen directly to the right of the ethyl propionate
14063
Fig. 3. a and b. SEM images of samples consisting of crystalline TCP.
peak). This shows that the triethyl phosphate is completely decomposed during the synthesis. However, there was the possibility that heavier phosphorous containing compounds were dissolved in the mixture without showing up on GC, so ester samples were further investigated with ICP-AES. These measurements revealed a total phosphorous content of 0.25 wt%, which further corroborates that a very high amount of the phosphate in the triethyl phosphate is converted to TCP. In contrast to the calcium phosphate, the ester formation was not affected by the water content of the precursors, ethyl propionate was the main product, regardless of the water amount. The amount of water in the precursors had a direct effect on the formed calcium phosphate polymorph. As seen from Table 1, ACP was formed at anhydrous conditions and up to a Ca/H2O molar ratio of 3.33, above this water content β-TCP was the synthesis product. It can also be seen from this Table that the formation of β-TCP occurred in an interval between Ca/H2O¼3.33–1.67. Multiple experiments in this range revealed that the border between ACP and βTCP formation was quite sharp, and no diffractogram showed a coexistence between ACP and β-TCP. In contrast, the crystalline phases were mixed in certain intervals, at Ca/H2O41.67, CDHA together with β-TCP was formed. The effect of water on phase transformations of ACP is described in the literature, but this phase transformation is normally direct from ACP to CDHA [23]. The reason to the phase transformation at certain Ca/H2O ratios is unclear, and will be subject to further investigation in the future. The ester formation on the other hand was not affected by the amount of water, ethyl propionate was the main product
14064
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065
Fig. 6. GC diagrams of a 50/50 mixture of ethyl propionate and triethyl phosphate (diagram A) and sample from the synthesis (diagram B).
decrease in specific surface area (and an increase in crystal size). An interesting possibility of this synthesis method would be to use a dicarboxylate such as calcium succinate as a calcium source and use phosphorylated glycerol as the phosphorous source, to produce an in situ polymerized, biodegradable polymer/TCP scaffold.
4. Conclusions
Fig. 4. a and b. SEM images of samples consisting of calcium phosphate (Monetite).
The calcium phosphate formation from the reaction between calcium propionate and triethyl phosphate was investigated. The inorganic part of the synthesis products showed to be highly sensitive to the water content of the precursors. Amorphous calcium phosphate was formed at anhydrous conditions, whereas addition of a small amounts of water created β-TCP. Higher amounts of water created increasingly higher amounts of CDHA and monetite. The formed ACP and β-TCP were nanosized with high specific surface areas, which makes this method potentially useful to create in situ polymerized ACP/polymer or β-TCP/polymer scaffolds.
References Fig. 5. XRD diffractograms of a β-TCP sample, as synthesized (lower diffractogram) and the same sample heat treated at 1000 °C (upper diffractogram). Vertical bars show reflections for β-TCP from literature data. (2 1 1) and (2 1 3) are hydroxyapatite reflections.
regardless of water content. As Karl Fischer titration showed, parts of the water was present in the formed ester, and some parts were present in the TCP powder. The fact that the amorphous composition did not form CDHA directly with water may be due to the presence of the ester compounds. These evidently bind water and may slow down the ripening process from ACP-TCP-CDHA. Formation of monetite at higher water concentrations is unexpected, excess water should produce mainly CDHA. Monetite formation is mainly favored at lower pH [24], so the reason may be that a low pH is created by the formation of small amounts of propanoic acid. The transformation from ACP to monetite was also accompanied by a significant
[1] H.A. Lowenstam, S. Weiner, On Biomineralization, Oxford University Press, New York, 1989. [2] F.H. Albee, H.F. Morrison, Studies in bone growth, triple calcium phosphate as a stimulus to osteogenesis, Ann. Surg. 71 (1920) 32–39. [3] M. Mathew, L.W. Schroeder, B. Dickens, W.E. Brown, The Crystal Structure of α-Ca3(PO4)2, Acta Crystallogr. B 33 (1977) 1325–1333. [4] E.W.H. Bodde, J.G.C. Wolke, R.S.Z. Kowalski, J.A. Jansen, Bone regeneration of porous β-tricalcium phosphate (ConduitTM TCP) and of biphasic calcium phosphate ceramic (Biosel1) in trabecular defects in sheep, J. Biomed. Mater. Res. A 82A (2007) 521–776. [5] G. Russmueller, D. Moser, E. Spassova, R. Plasenzotti, P.W. Poeschl, R. Seemann, S. Becker, K. Pirklbauer, C. Eder-Czembirek, C. Czembirek, C. Perisanidis, R. Ewers, C. Schopper, Tricalcium phosphate-based biocomposites for mandibular bone regeneration – a histological study in sheep, J. Cranio Maxillofac. Surg. 43 (2015) 696–704. [6] K.S. TenHuisen, P.W. Brown, Formation of calcium-deficient hydroxyapatite from α-tricalcium phosphate, Biomaterials 19 (1998) 2209–2217. [7] H. Monma, S. Ueno, T. Kanazawa, Properties of hydroxyapatite prepared by the hydrolysis of tricalcium phosphate, J. Chem. Tech. Biotechnol. 31 (1981) 15–24. [8] X. Yang, Z. Wang, Synthesis of biphasic ceramics of hydroxyapatite and β-tricalcium phosphate with controlled phase content and porosity, J. Mater. Chem. 8 (1998) 2233–2237.
P. Kjellin et al. / Ceramics International 42 (2016) 14061–14065 [9] M. Ermrich, F. Peters, X-ray powder diffraction data of synthetic β-tricalcium phosphate, Mat.-Wiss.Werkstofftech 37 (2006) 526–529. [10] S. Somrani, C. Rey, M. Jemal, Thermal evolution of amorphous tricalcium phosphate, J. Mater. Chem. 13 (2003) 888–892. [11] E.D. Eanes, Thermochemical studies on amorphous calcium phosphate, Calc. Tiss. Res. 5 (1970) 133–145. [12] S. Raynaud, E. Champion, D. Bernache-Assollant, P. Thomas, Calcium phosphate apatites with variable Ca/P atomic ratio I. Synthesis, characterisation and thermal stability of powders, Biomaterials 23 (2002) 1065–1072. [13] W.J. Stark, S.-E. Pratsinis, M. Maciejewski, S.F. Loher, A. Baiker, Flame synthesis of metal salt nanoparticles, in particular calcium and phosphate comprising nanoparticles, US Patent 7,879,303. [14] S. Loher, W.J. Stark, M. Maciejewski, A. Baiker, S.E. Pratsinis, D. Reichardt, F. Maspero, F. Krumeich, D. Günther, Fluoro-apatite and calcium phosphate nanoparticles by flame synthesis, Chem. Mater. 17 (2005) 36–42. [15] J. Bae, Y. Ida, K. Sekine, F. Kawano, K. Hamada, Effects of high-energy ballmilling on injectability and strength of β-tricalcium-phosphate cement, J. Mech. Behav. Biomed. Mater. 47 (2015) 77–86. [16] B. Nasiri-Tabrizi, A. Fahami, Mechanochemical synthesis and structural characterization of nano-sized amorphous tricalcium phosphate, Ceram. Int. 39 (2013) 8657–8666.
14065
[17] J.-S. Bowa, S.-C. Lioub, S.-Y. Chen, Structural characterization of room-temperature synthesized nano-sized β-tricalcium phosphate, Biomaterials 25 (2004) 3155–3161. [18] T. Toyama, K. Nakashima, T. Yasue, Hydrothermal synthesis of β-Tricalcium phosphate from amorphous calcium phosphate, J. Ceram. Soc. Jpn. 110 (2002) 716–721. [19] R. Vani, E.K. Girija, K. Elayaraja, S.P. Parthiban, R. Kesavamoorthy, S.N. Kalkura, Hydrothermal synthesis of porous triphasic hydroxyapatite/(α and β) tricalcium phosphate, J. Mater. Sci.: Mater. Med. 20 (2009) S43–S48. [20] J. Tao, W. Jiang, H. Zhai, H. Pan, X. Xu, R. Tang, Structural components and anisotropic dissolution behaviors in one hexagonal single crystal of β-Tricalcium phosphate, Cryst. Growth Des. 8 (2008) 2227–2234. [21] L. Galea, M. Bohner, J. Thuering, N. Doebelin, C.G. Aneziris, T. Graule, Control of the size, shape and composition of highly uniform, non-agglomerated, submicrometer β-tricalcium phosphate and dicalcium phosphate platelets, Biomaterials 34 (2013) 6388–6401. [22] K. Pielichowska, S. Blazewicz, Bioactive polymer/hydroxyapatite (nano)composites for bone tissue regeneration, Adv. Polym. Sci. 232 (2010) 97–207. [23] S.V. Dorozhkin, Amorphous calcium orthophosphates: nature, chemistry and biomedical applications, Int. J. Mater. Chem. 2 (2012) 19–46. [24] B. Jokic, M. Mitric, V. Radmilovic, S. Drmanic, R. Petrovic, D. Janac’kovic´, Synthesis and characterization of monetite and hydroxyapatite whiskers obtained by a hydrothermal method, Ceram. Int. 37 (2011) 167–173.