hidroxa

Page 1

Chp

Ma

b

a

ARRAA

KHCHM

1

ibahb[bbtb(

TT

C

h2

Journal of Asian Ceramic Societies 3 (2015) 287–291

Contents lists available at ScienceDirect

Journal of Asian Ceramic Societies

HOSTED BY

j ourna l ho me page: www.elsev ier .com/ loca te / jascer

arbonate-containing hydroxyapatite synthesized by theydrothermal treatment of different calcium carbonates in ahosphate-containing solution

asanobu Kamitakaharaa,∗, Takuya Nagamoria, Taishi Yokoia, Koji Iokub

Graduate School of Environmental Studies, Tohoku University, 6-6-20 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, JapanFaculty of Economics, Keio University, Building #2-101B, 4-1-1, Hiyoshi, Kohoku-ku, Yokohama 223-8521, Japan

r t i c l e i n f o

rticle history:eceived 2 March 2015eceived in revised form 28 April 2015ccepted 5 May 2015vailable online 28 May 2015

eywords:ydroxyapatitearbonate

a b s t r a c t

Carbonate-containing hydroxyapatite (CHA) particles were synthesized by the hydrothermal treatmentof calcium carbonates in a phosphate-containing solution. Three types of calcium carbonates were syn-thesized: aragonite particles with rough surfaces, calcite particles with rough surfaces by heat treatmentof aragonite particles with rough surfaces, and aragonite particles with smooth surfaces. The effects ofthe calcium carbonate crystal phase and morphology on the synthesized CHA were investigated and mor-phological changes in the formed CHA particles were observed. The reaction rates varied depending onthe calcium carbonate crystal phase and morphology and this difference in reaction rates mainly affectedthe morphology of the synthesized CHA. On the other hand, the carbonate content and lattice constants

ydrothermal synthesisorphology

of formed CHA were almost independent of the calcium carbonate crystal phase and morphology. Thisis because the equilibrium mainly governed the composition of the synthesized CHA. The characteristicsof the starting material were highly important factors in controlling the morphology of the synthesizedCHA.

© 2015 The Ceramic Society of Japan and the Korean Ceramic Society. Production and hosting byElsevier B.V. All rights reserved.

. Introduction

Bone is an important organ that supports the body, protectsnternal organs, and stores ions. In order to repair damagedones, artificial materials that are free from pathogen and avail-ble in unlimited amounts are required. Sintered stoichiometricydroxyapatite (HA, Ca10(PO4)6(OH)2) ceramics are widely used asone-repairing materials because they can bond to natural bone1–4]. The sintered HA ceramics, however, show low bioresorba-ility and can hardly be replaced by new bone [5–7]. The idealone-repairing material should induce the regeneration of boneissue and be resorbed permitting the bone formation. As such aioresorbable bone-repairing material, carbonate-containing HACHA) has received attention because of its higher bioresorbability

∗ Corresponding author at: Graduate School of Environmental Studies,ohoku University, 6-6-20 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan.el.: +81 22 795 7375; fax: +81 22 795 7375.

E-mail address: [email protected] (M. Kamitakahara).Peer review under responsibility of The Ceramic Society of Japan and the Korean

eramic Society.

ttp://dx.doi.org/10.1016/j.jascer.2015.05.002187-0764 © 2015 The Ceramic Society of Japan and the Korean Ceramic Society. Produc

compared to stoichiometric HA [8,9]. There are many reports aboutthe synthesis of CHA from calcium carbonates by the reaction in thesolutions containing phosphate ions [10–18]. We focused on thehydrothermal process because the morphology and compositionof the formed HA can be controlled by the starting materials andsynthetic conditions [19–22]. It has been reported that the proper-ties of HA depend on the exposed crystal faces (i.e. the morphologyof the HA crystal) [23]. The composition and morphology of theHA ceramics affect their biological behavior in vivo [24,25]. Themechanical properties of HA ceramics are also affected by the mor-phology of their constituent particles [26]. Therefore, the control ofthe composition and morphology is important for the synthesis ofCHA. Moreover, in the case that calcium carbonate particles wereused not only as the calcium sources but also as the template of theresultant CHA [12,14,17,18], the control of the CHA formation reac-tion is important. In the present study, we intended to investigatethe effects of the types of calcium carbonates on the resultant CHAto obtain the fundamental knowledge for the CHA synthesis.

We previously synthesized CHA from calcium carbonate byhydrothermal treatment and revealed that the character of thestarting material and the synthetic conditions were important fac-tors to control the morphology and composition of the resultant

tion and hosting by Elsevier B.V. All rights reserved.

Page 2

2 sian C

Cooccpttiamacw

2

2c

pbabhf

bs8nccccA

htow

Csmws1f9

2c

aavpTro

88 M. Kamitakahara et al. / Journal of A

HA [27]. Although we previously reported that the crystal phasef the starting material, calcium carbonate, affected the morphol-gy of the formed CHA, its formation mechanism and carbonateontent were unknown. In the present study, we prepared differentalcium carbonate particles with different crystal phases and mor-hologies. Calcite and aragonite were selected as crystal phases ofhe starting materials because the solubility of aragonite is higherhan that of calcite [28]. It is speculated that the reactivity of start-ng materials governed by their crystal phases and morphologiesffects the formed CHA. These calcium carbonates were hydrother-ally treated in an aqueous solution containing phosphate ions,

nd the effects of the crystal phase and morphology of the calciumarbonates on the morphology and compositions of the formed CHAere investigated.

. Materials and methods

.1. Synthesis of calcium carbonate particles with differentrystal phases and morphologies

Three types of calcium carbonates were synthesized: aragonitearticles with rough surfaces, calcite particles with rough surfacesy heat treatment of aragonite particles with rough surfaces, andragonite particles with smooth surfaces. The reagents, sodium car-onate, calcium nitrate tetrahydrate, calcium hydroxide, sodiumydroxide and diammonium hydrogen phosphate were obtained

rom Wako Pure Chemical Industries, Ltd., Japan.Aragonite particles with rough surfaces (A R) were synthesized

y a method detailed in a previous study [29]. A 0.10 mol dm−3

odium carbonate solution of 200 cm3 was kept in a water bath at0 ◦C and stirred for 30 min. Then, 20 cm3 of a 1.0 mol dm−3 calciumitrate solution was preheated to 80 ◦C and added to the sodiumarbonate solution at a rate of 5 cm3 min−1 by a pump. Stirringontinued for 1 h after adding the calcium nitrate solution. The pre-ipitates were collected by filtration, and washed with ethanol. Theollected product was dried at 90 ◦C for 1 d. This sample was named

R.It has been reported that aragonite transforms to calcite by a

eat treatment [30]. In order to obtain calcite particles (CAL) withhe same morphology as the obtained aragonite particles A R, thebtained A R sample was heated at 500 ◦C for 10 min. This sampleas named CAL.

In order to investigate the effect of the morphology on theHA formation by comparing A R sample, aragonite particles withmooth surfaces (A S) were also synthesized by the followingethod. A 1.5 mol dm−3 of calcium hydroxide suspension of 50 cm3

as mixed with 50 cm3 of a 2.5 mol dm−3 sodium hydroxideolution at 80 ◦C. A 0.50 mol dm−3 sodium carbonate solution of50 cm3 was added at a rate of 3 cm3 min−1 by a pump and stirredor 3 h. The precipitates were collected by filtration, and dried at0 ◦C for 1 d. This sample was named A S.

.2. Hydrothermal synthesis of carbonate-containing HA fromalcium carbonate particles

An amount of 3.3 mmol of each starting material (A R, CAL

nd A S) was put into a 50 cm3 Teflon® vessel with 20 cm3 of

0.10 mol dm−3 diammonium hydrogen phosphate solution. Theessel was tightly sealed in a stainless steel autoclave and thenut in an oven at 160 ◦C for 24 h for the hydrothermal treatment.he final products were named CHA-A R, CHA-CAL, and CHA-A S,espectively. In order to examine the temporal changes, a numberf autoclaves were used and taken out at the determined periods.

eramic Societies 3 (2015) 287–291

2.3. Characterization of materials

The crystal phases of the samples were examined by X-raydiffraction (XRD, RINT-2200VL, Rigaku Co., Japan). The internalstandard method using silicon (640d, NIST, USA) was applied fordetermination of lattice constants. The diffraction lines between25◦ and 53◦ were used for the determination of lattice parametersa and c using the software Jade 6 (Rigaku Co., Japan). The specificsurface areas of the samples were examined by the N2-BET methodusing an Autosorb-iQ ASIQM0000-3 (Quantachrome Instruments,Florida, USA). The structure of the products was examined byFourier transform infrared spectroscopy (FT-IR, FT/IR-6200, JASCO,Japan). For the FT-IR measurements, the samples were mixed withKBr and pressed into pellets before the transmittance method wasapplied. The morphology was observed by scanning electron micro-scope (SEM, SU8000, Hitachi, Japan) and transmission electronmicroscope (TEM, HF-2000, Hitachi, Japan). The carbonate contentwas calculated from the carbon content examined by CHNS/O ele-mental analysis (2400II, Perkin Elmer, USA).

3. Results and discussion

Figs. 1 and 2 show XRD patterns and SEM photographs of thestarting materials, respectively. All the particles showed rod-likemorphology. The crystal phase of samples A R and A S was com-posed mainly of aragonite with a small amount of calcite. Althoughthe relatively large peaks of calcite were observed in the A S sam-ple, the main crystal phase the A S sample was aragonite becausethe calcite provides higher intensity in XRD than aragonite. Thecrystal phase of the CAL sample was pure calcite. The CAL samplehas almost same morphology as the A R sample. The A R samplehas a slightly rougher surface whereas the A S sample has a rela-tively smooth surface. The specific surface areas measured by BETmethod of the CAL, A R and A S samples were almost same andabout 2 m2 g−1. Comparing the CAL and A R samples as well as A Rand A S samples, the effects of crystal phase and morphology canbe discussed, respectively.

The temporal changes in crystal phase and morphology wereexamined in order to reveal the mechanism of CHA formation fromdifferent calcium carbonates. The temporal changes in crystal phaseand morphology are shown in Figs. 3 and 4, respectively. The crys-tal phases of the products after the reaction were only CHA and/or�-tricalcium phosphate (�-TCP, Ca3(PO4)2), and the patterns wereshown in the range 2� = 20–40◦ in Fig. 3. Fig. 5 shows TEM pho-tographs of the samples after the hydrothermal treatment at 160 ◦Cfor 24 h.

Comparing the CHA-CAL and CHA-A R products, this differencemight have appeared due to the differences in their crystal phases.From the SEM photographs of the products treated for 15 min, it canbe seen that the reaction started from the surfaces of the particlesfor both the CAL and A R samples. Due to the treatment, how-ever, the shape of the formed aggregates became different betweenthe CAL and A R samples. For the product from the CAL sample,the flake-like nanoparticles were formed on the surfaces of thestarting materials maintaining the original shape (rod-shape) toform agglomerates, as shown in SEM and TEM photographs. Onthe other hand, the original shape of the starting material wasnot retained for the product from the A R sample which formedglobular agglomerates composed of flake-like nanoparticles. Themorphologies depend on the crystal growth of CHA. The crystalgrowth conditions were changed by the differences in the crys-

tal phase. The XRD patterns of the samples after the 6 h treatmentshowed that the transformation to CHA was almost completed forA R, unlike CAL in which calcium carbonate remained. These resultsindicate that the reaction rate of CAL was slower than that of A R.

Page 3

M. Kamitakahara et al. / Journal of Asian Ceramic Societies 3 (2015) 287–291 289

20 4030 20 4030 20 4030

Inte

nsity / a

.u.

Inte

nsity / a

.u.

Inte

nsity /

a.u

.

2θ / ° (CuKα 2) θ / ° (CuKα 2) θ / ° (CuKα)

Calcite A_SA_RCAL Aragonite

Calcite

Aragonite

Calcite

Fig. 1. XRD patterns of starting materials before hydrothermal treatment.

Fig. 2. SEM photographs of starting materials before hydrothermal treatment.

Inte

nsity / a

.u.

Inte

nsity / a

.u.

Inte

nsity /

a.u

.

° (C

CAL A_R A_S

403020

24 h

6 h

1 h

Before

403020 403020

24 h

6 h

1 h

Before

24 h

6 h

1 h

Before

CHAβ-TCPAragoniteCalcite

CHAAragonite

CHACalcite

fter d

TpdtirsAi

otorut

2θ / ° (CuKα 2) θ /

Fig. 3. XRD patterns of samples before and a

he CHA formation process is speculated to be the dissolution-recipitation reaction. The calcium and carbonate ions produced byissolution of calcium carbonates react with phosphate ions nearhe surfaces of the calcium carbonate particles. The lower solubil-ty of calcite might induce a slower reaction from the surface andetain the original shape in the CHA-CAL product. However, thehapes of primary particles which construct the CHA-CAL and CHA-

R products after the treatment for 24 h were flake-like and almostdentical, as shown in the TEM photographs.

Comparing the CHA-A R and CHA-A S products, the morphologyf the particles is the primary difference. From the SEM pho-ographs of the products treated for 15 min, precipitates were

bserved on A R, while they were not observed on A S. The loweactivity of A S might be due to its smooth surface. For the prod-ct from the A S sample, precipitation of nano-sized particles onhe surface was observed after the 1 h treatment. Comparing the

uKα 2) θ / ° (CuKα)

ifferent periods of hydrothermal treatment.

XRD patterns of the products from the A R and A S samples afterthe treatment for 1 h, the rate of CHA formation from the A S sam-ple was slower than that from the A R sample, and �-TCP wasformed only from the A S sample. For the product from the A Ssample, the original shape remained and the shape of the particles,which construct the aggregates, was different from the A R sample.Aggregates of uniformly-sized flake-like particles were observedfor the product from the A R sample, while aggregates of rod-likeand whisker-like particles with different sizes were observed forCHA-A S, as shown in SEM and TEM photographs. It can be seenfrom the XRD result of the product from the A S sample after thetreatment for 1 h, multiple crystal phases, i.e. aragonite, calcite and

�-TCP were observed. In this situation, the rate of CHA formationshould have been different. When the reaction rates are different,uniform nucleation and crystal growth of the CHA do not occur. Thecrystal phase with low reactivity is used as a supplier of the CHA

Page 4

290 M. Kamitakahara et al. / Journal of Asian Ceramic Societies 3 (2015) 287–291

Fig. 4. SEM photographs of products after different periods of hydrothermal treatment.

er hydrothermal treatment at 160 ◦C for 24 h.

clfap

it

scowsotstTacts

0

1

2

3

4

5

6

7

8

9

10

CHA-CAL CHA-A_R CHA-A_S

Ca

rbo

na

te c

on

ten

t/

ma

ss%

Fig. 5. TEM photographs of products aft

rystal growth at the later stage, and this situation provides the rod-ike and whisker-like particles with different sizes. The slower CHAormation rate might have induced the preferential crystal growthlong the c-axis, along which HA can easily grow to form rod-likearticles.

From these results, we can say that the difference in the reactiv-ty due to the crystal phase and surface structure should producehe observed differences in the morphology of the formed CHA.

The carbonate content determined by elemental analysis ishown in Fig. 6. All the CHA products contained about 7 mass% ofarbonate ions and their carbonate contents were almost same. Inrder to examine the state of carbonate ions in CHA, FT-IR analysisas conducted. It has been reported that the carbonate ions can

ubstitute both of the OH− and PO43− sites of HA, and the position

f the carbonate ions can be determined by the wave number ofhe absorption bands [31,32]. The FT-IR spectra of the products arehown in Fig. 7. In order to show the state of the carbonate ions inhe products, the FT-IR spectra from 400 to 1600 cm−1 were shown.he CHA samples showed the absorption bands due to the carbon-

te ions. The existence of the absorption band at 875 cm−1 due toarbonate ions substituted at PO4

3− sites and that at 880 cm−1 dueo carbonate ions substituted at OH− sites implied that all the CHAsynthesized from different calcium carbonates contained carbonate

Fig. 6. Carbonate content of products after hydrothermal treatment at 160 ◦C for24 h.

ions at both of the OH− and PO43− sites. Significant differences were

not observed among the spectra of the three CHA products.These results indicate that the state and content of carbonate

ions in CHA were independent of the starting materials. This is alsosupported by the lattice constants of the CHA products determined

Page 5

M. Kamitakahara et al. / Journal of Asian C

4006008001000120014001600

CHA-CAL

CHA-A_R

CHA-A_ST

ransm

itta

nce / a

.u.

PO43–

CO32–

(B type)

CO32–

(A type)

PO43–

CO32–

(A an d B type)

F

bCaattdsCia

gccspcpgu

4

oom

[[

[[

[

[

[[

[

[[

[

[

[[

[

[

[

[[[

[

[32] I.R. Gibson and W. Bonfield, J. Biomed. Mater. Res., 59, 697–708 (2002).

Wavenumber / cm –1

ig. 7. FT-IR spectra of products after hydrothermal treatment at 160 ◦C for 24 h.

y XRD. The lattice constants a and c of all the products CHA-CAL,HA-A R and CHA-A S were 0.943 nm and 0.690 nm, respectively,nd were almost identical. This similarity suggests that the statend content of carbonate ions in CHA are almost the same amonghe three products because it has been reported that the trend ofhe changes in the lattice constants by substitution with carbonatesepends on the substitution sites [33]. These lattice constants wereimilar to those reported by Lee et al. [34] who prepared porousHA ceramics by hydrothermal treatment of calcium carbonate

n an (NH4)2HPO4 solution and reported the lattice parameters = 0.9427 nm and c = 0.6887 nm.

Although the reaction rates to form CHAs and the morpholo-ies of the formed CHAs were different, the state and content ofarbonate ions in the CHAs were almost identical. The dissolvedalcium carbonate should have reacted with phosphates near theurface of the calcium carbonate and carbonates were easily incor-orated by the same manner regardless of the type of calciumarbonate. Moreover, as the total amount of carbonate and phos-hate ions was the same in the system, the equilibrium mainlyoverned the state and content of carbonate ions in the CHA prod-cts.

. Conclusions

The influence of calcium carbonate properties on the morphol-gy and composition of CHA has been investigated. The characterf the starting material was a highly important factor to control theorphology of the synthesized CHA.

[

[

eramic Societies 3 (2015) 287–291 291

Acknowledgements

This research was partially supported by JSPS KAKENHI(21300175, 23760630). We are grateful for the experimental helpof Prof. M. Kawashita and Ms. Y. Nakano of Tohoku University.

References

[1] M. Jarcho, J.L. Kay, R.H. Gumaer and H.P. Drobeck, J. Bioeng., 1, 79–92 (1977).[2] H. Aoki, Medical Application of Hydroxyapatite, Ishiyaku EuroAmerica, Tokyo,

St. Louis (1994).[3] L.L. Hench, J. Am. Ceram. Soc., 74, 1487–1510 (1991).[4] R.Z. LeGeros, Clin. Orthop. Relat. Res., 395, 81–98 (2002).[5] H.A. Hoogendoorn, W. Renooij, L.M. Akkermans, W. Visser and P. Wittebol, Clin.

Orthop. Relat. Res., 187, 281–288 (1984).[6] A. Uchida, N. Araki, Y. Shinto, H. Yoshikawa, E. Kurisaki and K. Ono, J. Bone Joint

Surg. Br., 72-B, 298–302 (1990).[7] S.V. Dorozhkin, J. Mater. Sci., 42, 1061–1095 (2007).[8] M. Hasegawa, Y. Doi and A. Uchida, J. Bone Joint Surg. Br., 85-B, 142–147 (2003).[9] E. Landi, G. Celotti, G. Logroscino and A. Tampieri, J. Eur. Ceram. Soc., 23,

2931–2937 (2003).10] D.M. Roy and S.K. Linnehan, Nature, 247, 220–222 (1974).11] M. Yoshimura, P. Sujaridworakun, F. Koh, T. Fujiwara, D. Pongkao and A.

Ahniyaz, Mater. Sci. Eng. C, 24, 521–525 (2004).12] M.Y. Ma, Y.J. Zhu, L. Li and S.W. Cao, J. Mater. Chem., 18, 2722–2727 (2008).13] K. Ishikawa, S. Matsuya, X. Lin, Z. Lei, T. Yuasa and Y. Miyamoto, J. Ceram. Soc.

Jpn., 118, 341–344 (2010).14] Y.J. Guo, Y.Y. Wang, T. Chen, Y.T. Wei, L.F. Chu and Y.P. Guo, Mater. Sci. Eng. C,

33, 3166–3172 (2013).15] D.P. Minh, S. Rio, P. Sharrock, H. Sebei, N. Lyczko, N.D. Tran, M. Raii and A.

Nzihou, J. Mater. Sci., 49, 4261–4269 (2014).16] D.P. Minh, A. Nzihou and P. Sharrock, Mater. Res. Bull., 60, 292–299 (2014).17] Y.P. Guo, T. Long, S. Tang, Y.J. Guo and Z.A. Zhu, J. Mater. Chem. B, 2, 2899–2909

(2014).18] Q. Sun, J.T. Lou, F. Kang, J.F. Chen and J.X. Wang, Powder Technol., 261, 49–54

(2014).19] K. Ioku, J. Ceram. Soc. Jpn., 118, 775–783 (2010).20] M. Kamitakahara, Y. Enari, N. Watanabe and K. Ioku, Trans. Mater. Res. Soc. Jpn.,

36, 405–408 (2011).21] T. Goto, I.Y. Kim, K. Kikuta and C. Ohtsuki, J. Ceram. Soc. Jpn., 120, 131–137

(2012).22] M. Kamitakahara, Y. Uno and K. Ioku, J. Mater. Sci. Mater. Med., 25, 239–245

(2014).23] T. Kawasaki, S. Takahashi and K. Ikeda, Eur. J. Biochem., 152, 361–371 (1985).24] T. Okuda, K. Ioku, I. Yonezawa, H. Minagi, Y. Gonda, G. Kawachi, M. Kamitaka-

hara, Y. Shibata, H. Murayama, H. Kurosawa and T. Ikeda, Biomaterials, 29,2719–2728 (2008).

25] Y. Gonda, K. Ioku, Y. Shibata, T. Okuda, G. Kawachi, M. Kamitakahara, H.Murayama, K. Hideshima, S. Kamihira, I. Yonezawa, H. Kurosawa and T. Ikeda,Biomaterials, 30, 4390–4400 (2009).

26] S. Murakami, K. Kato, Y. Enari, M. Kamitakahara, N. Watanabe and K. Ioku,Ceram. Int., 38, 1649–1654 (2012).

27] W. Park, M. Kamitakahara, T. Nagamori and K. Ioku, Phosphorus Res. Bull., 25,72–77 (2011).

28] J.Y. Gal, J.C. Bollinger, H. Tolosa and N. Gache, Talanta, 43, 1497–1509 (1996).29] J.L. Wary and F. Daniels, J. Am. Chem. Soc., 79, 2031–2034 (1957).30] J. Peric, R. Krstulovic, T. Feric and M. Vucak, Thermochim. Acta, 207, 245–254

(1992).31] C. Rey, V. Renugopalakrishnan, B. Collins and M.J. Glimcher, Calcif. Tissue Int.,

49, 251–258 (1991).

33] R.Z. LeGeros and J.P. LeGeros, in An Introduction to Bioceramics, Ed. by L.L.Hench and J. Wilson, World Scientific, Singapore (1993) pp. 139–180.

34] Y. Lee, Y.M. Hahm, S. Matsuya, M. Nakagawa and K. Ishikawa, J. Mater. Sci., 42,7843–7849 (2007).