Hydrothermal synthesis of magnetite/hydroxyapatite

©2008 The Ceramic Society of Japan950

Journal of the Ceramic Society of Japan 116 [9] 950-954 2008

Hydrothermal synthesis of magnetite/hydroxyapatite composite material for hyperthermia therapy for bone cancer

Setsuaki MURAKAMI, Toshiaki HOSONO, Balachandran JEYADEVAN, Masanobu KAMITAKAHARA and Koji IOKU†

Graduate School of Environmental Studies, Tohoku University, 6-6-20, Aoba, Aramaki, Aoba-ku, Sendai 980-8579

The magnetite/hydroxyapatite (HA) composite that facilitates direct bonding to bones through HA and generation of heat from magnetite exposed to AC magnetic field is considered suitable for hyperthermia therapies of cancer in bones. Porous magne-tite/hydroxyapatite (HA) composite was prepared by a hydrothermal treatment of α -tricalcium phosphate (α -TCP) and nano-sized magnetite powder mixture having magnetite concentrations varying between 10 and 50 mass% at 120°C under saturated vapor pressure for 24 h. This composite had micro-sized pores of about 400 μm and submicron-sized pores of about 0.2 μm in size and magnetite particle aggregates were strongly trapped in the cages of rod-shaped HA particles only when the mag-netite contents were 30 mass% or less. At magnetite concentrations higher than 30 mass%, the composite cannot hold the par-ticles completely and considered unsuitable for biomedical applications. When magnetite/hydroxyapatite (HA) composites with various magnetite concentrations were exposed to high frequency magnetic field for 10 min, the temperature rise in the com-posite was a function of magnetite concentration and the temperature rise in 30 mass% was 55°C, which is well above the tem-perature required for treatment. However generation of higher temperatures, which could be achieved by controlling the magnetic field strength, frequency, and exposure time, may be required because the heat loss due to blood flow, etc., is consid-ered during hyperthermia therapies of cancer in bones.©2008 The Ceramic Society of Japan. All rights reserved.

Key-words : Hydrothermal, Hydroxyapatite, Magnetite, Composite, Magnetic hyperthermia therapy

[Received March 5, 2008; Accepted July 17, 2008]

1. IntroductionThe current therapies available to treat cancer patients mainly

rely on surgical operation and chemical therapy. However, the quality of patient’s life is affected greatly by the trauma and side effect caused by these therapies. In recent years, hyperthermia, a treatment technique based on deactivation of cancer cells by raising the temperature in the range of 43°C and 48°C has gained much attention as an alternative therapy to treat cancer.1) Here, cancer is preferentially heated, as nerve and blood systems in cancerous cells are not fully developed. Therefore, hyperthermia therapy is considered an effective non-invasive technique to treat cancer. At present, supersonic-therapy, hot-water-therapy and infrared radiation-therapy are used to elevate the malignant cell temperature.1) However the serious disadvantage of these methods is that the heat generated during such treatments not only kills cancer cells but also normal cells, because it is difficult to heat deep-seated cancerous cells effectively and locally. Therefore, therapeutic procedures that can heat carcinomatous area effectively and selectively are urgently needed. As ferro-magnetic nanoparticles have the ability to generate heat in high frequency magnetic field, magnetite is one of the most promising candidates for hyperthermia. If magnetite can be made to accu-mulate only in the cancer tumor, they can specifically heat the cancer cells. Many attempts are being made to use fine magnetic particles for hyperthermia,1)–8) and the successful use of magne-tite (Fe3O4) fine particles to treat liver cancer has been reported. There are the cancers of many body parts. Though the frequency

of bone cancer is less compared to cancer in other parts of the body, the consequences are grave and the patients undergo resec-tion that limits their motility functions. Thus, materials that can destroy the cancer cells and at the same time induce bone regen-eration are desired. The bone has a bigger space and the blood streams in comparison with other organs. Moreover, the magne-tite particles do not have the directivity to the cancer cells, the magnetite particle can not be fixed to the cancerous area in the bone tissue. Therefore, porous hydroxyapatite (HA) composites with magnetite dispersion are expected to be useful, where HA possesses bone-bonding ability owing to its similarity with min-eral constituents found in teeth and bones9),10) and also can serve as a medium to hold the magnetite that can act as thermal seeds whenever necessary. Conventional porous HA ceramics are pre-pared by sintering at high temperature.11),12) Incorporation of magnetite to form magnetite/hydroxyapatite (HA) composite using the above process is not viable as magnetite would convert to non-magnetite hematite by oxidation. The present authors13)–16)

have reported the synthesis of porous HA ceramics prepared by hydrothermal methods at low temperature for bone regeneration. The synthesis conditions are mild enough to prevent the conversion of magnetite into any non-magnetic iron oxide forms. In our previous research,17) we reported the preparation of micron-sized magnetite particles dispersed porous HA ceramics treating a mixture of α -tricalcium phosphate and magnetite par-ticles under hydrothermal conditions. However, the composite did not generate enough heat desired for use in cancer treatment. Heat generation of magnetite under a high frequency magnetic field depends on its crystallite size of the magnetite.6) In the present study, we have attempted to prepare the magnetite/hydroxyapatite (HA) composite by using the nano-sized magne-

† Corresponding author: K. Ioku; E-mail: [email protected]. ac.jp

Paper


951

JCS-Japan

tite particles that could generate much heat in alternating magnetic field and discuss their properties from the synthesis and heating capacity perspectives.

2. Experimental2.1 Preparation of magnetite/hydroxyapatite com-

positePowder of α-tricalcium phosphate (α -TCP Ca3(PO4)2: Taihei

Chem. Ind.), and magnetite (Fe3O4) nano-sized particles (about 15 nm) prepared by the method reported by Atsumi et al.6) (Fig. 1(a)) were used as the starting materials. After the addition of 0–50 mass% magnetite particles to α -TCP, the mixed powder (0.2 g) was molded into cylindrical shape of about 8 mmφ × 3.5 mmh with PLLA fibers with 400 μm in diameter under the com-pressive pressure about 100 kPa by adding about 50 mg of water. After molding, PLLA fibers were pulled out to make intercon-necting micro-sized pores. The specimens were set in 105 cm3

autoclave with 20 cm3 of water and then they were exposed to water vapor at 120°C under saturated vapor pressure for 24 h. As a reference, the magnetite/hydroxyapatite composite with addi-tion of 30 mass% magnetite of about 1 μm in size (Wako Chem-ical Co.) (Fig. 1(b)) was also prepared by the same process.

2.2 Characterization of materialsThe phases of the specimens were identified by powder X-ray

diffractometry with Cu Kα radiation, operating at 40 kV and 40 mA (XRD; RINT–2200VL, Rigaku Corporation). The micro-structure of specimens was observed using a scanning electron microscope (SEM; S–4100, Hitachi).

2.3 Evaluation of heat generationThe temperature changes in the specimens were measured

under the magnetic field (THAMWAY). The frequency and amplitude of alternating magnetic field are 600 kHz and 3.2 kA/m, respectively. The temperature of the composite was measured by an alcohol thermometer.

2.4 Soaking in simulated body fluidThis composite was soaked in 20 cm3 of a simulated body

fluid proposed by Kokubo et al.18),19) for 3 weeks to estimate its bone-bonding ability. The bone-bonding of a material is often evaluated by examining the ability to form apatite on its surface in SBF with inorganic ion concentrations nearly equal to those of human body plasma.19) Then, the sample was taken out from the fluid and gently washed with water. Finally, the microstruc-ture of the surface of specimens was observed by SEM.

3. ResultsThe XRD patterns of magnetite/hydroxyapatite (HA) compos-

ite samples after hydrothermal treatment are shown in Fig. 2. The peaks ascribed to HA were observed for the hydrothermally treated sample of magnetite free α -TCP. For the samples con-taining magnetite, in addition to peaks corresponding to HA, the peaks of magnetite were also observed. Furthermore, the peak intensities of HA and magnetite depended on their concentration in the composite. When the samples were precisely measured by XRD and the crystallite size of the magnetite was calculated using the Scherrer’s equation, before and after hydrothermal treat-ment, the XRD patterns of magnetite were not changed by the hydrothermal treatment and the crystallite size of the magnetite was about 12 nm before and after the hydrothermal treatment.

SEM photographs of the surfaces of hydrothermally prepared magnetite/hydroxyapatite (HA) composites with and without magnetite are shown in Fig. 3. All porous materials were com-

(a)

(b)

Fig. 1. (a) TEM photograph of the nano-sized magnetite particles. (b) SEM photograph of the magnetite particles about 1 μm in size.

Fig. 2. XRD patterns of the samples after hydrothermal treatment. (a) 0 mass% magnetite addition, (b) 10 mass% magnetite addition, (c) 30 mass% magnetite addition and (d) 50 mass% magnetite addition.

Murakami et al.: Hydrothermal synthesis of magnetite/hydroxyapatite composite material for hyperthermia therapy for bone cancer

952

JCS-Japan

posed of rod-shaped HA particles elongated along the c-axis. Length of rod-shaped HA particle decreased as their concentra-tion in the composite decreased. Though the presence of agglom-erated magnetite particles on the surface of the composite was not detected for magnetite concentration less than 30 mass%, this became prominent for the sample with 50 mass% magnetite. Figure 4 shows the changes in the aspect ratios of HA rods as a function of magnetite concentration. The aspect ratios of HA crystals in the composites decreased for any increase in the mag-netite concentration.

SEM photographs of the polished surface of the composite

with 30 mass% magnetite under different magnifications are shown in Figs. 5(a) and (b). This composite had pores of about 400 μm in size, which are interconnected and had a porosity of about 65%. Rod-shaped HA particles were observed even on the polished surface. This means that almost parts of the composite are composed of rod-shaped HA particles.

Figure 6 shows the temperature rise against exposure time for magnetite/hydroxyapatite (HA) composite with different magne-tite concentration in the magnetic field whose intensity and frequency were 3.2 kA/m and 600 kHz, respectively. For any exposure time, the temperature rise in the magnetite/hydrox-yapatite (HA) composite was a function of the magnetite con-centration. The temperature rose above the target temperature of 43°C within 10 min for magnetite/hydroxyapatite (HA) compos-ites whose magnetite concentration was 30 and 50 mass%. On the other hand, magnetite/hydroxyapatite (HA) composite with large sized magnetite (1 μm) did not generate heat.

The magnetite/hydroxyapatite (HA) composite with a mag-netite concentration of 0 and 30 mass% was soaked in SBF for 3 weeks to study the effect of magnetite in the bone-bonding ability of HA. Figure 7 shows SEM photographs of the surfaces of SBF soaked composites with (a) 0 and (b) 30 mass% mag-netite. As seen in the photographs, the formation of leaf-like par-ticles was observed on the surface of rod shaped HA present in magnetite/hydroxyapatite (HA) composites with 0 and 30 mass% magnetite. These leaf-like particles are believed to be the bone-like apatite. The weight of the composite increased on one hand, and the concentrations of Ca and P decreased in spent SBF.

(a) 0 mass% magnetite addition (b) 30 mass% magnetite addition

Fig. 3. SEM photographs of the surfaces of magnetite/HA composites prepared by hydrothermal method. (a) 0 mass% magnetite addition and (b) 30 mass% magnetite addition.

Fig. 4. Changes in aspect ratios of HA rod-shaped particles depending on the magnetite addition.

(a) (b)

Fig. 5. SEM photographs of (a) the polished surface of magnetite/HA composite with 30 mass% magnetite addition (b) the magnified photograph of the polished surface.


953

JCS-Japan

4. DiscussionAfter the hydrothermal treatment, α -TCP in all samples

changed into HA and as a consequence the specimens hard-ened.20) Magnetite addition did not inhibit the formation of HA from α -TCP by the hydrothermal treatment. When the crystallite size of the magnetite was calculated using the Scherrer’s equation,the crystallite size of the magnetite was not changed by the hydrothermal treatment. Moreover, hematite was not detected in

the composite. The HA in the composite was calcium deficient, because HA in the composites decomposed into β -TCP with sto-ichiometric value of Ca/P molar ratio of 1.50 and stoichiometric HA with Ca/P of 1.67 after heating at 900°C for 3 h in air. Ioku et. al.21) reported that Ca/P ratio could be determined from the intensity ratio of XRD patterns of the stoichiometic HA and β -TCP. From this insight, Ca/P ratio of HA in the composite with 30 mass% magnetite was determined to be 1.53, which is similar to natural bone in composition. Therefore, this composite is expected to bind to natural bone rapidly.

The surfaces of the composites with 0, 10 and 30 mass% magnetite was composed of rod-shaped HA particles, and had submicron-sized pores formed through tangling of rod-shaped HA particles. So nano-sized magnetite particle agglomerates were held inside the submicron-sized pores. However, the surface of the composite with 50 mass% magnetite was not fully covered by rod-shaped HA particles. This suggested that there is a limit to the concentration of magnetite that could be held within the composite.

The composite had micro-sized pores of about 400 μm in size formed using the PLLA fibers which were removed after mold-ing. The size of micro-sized pores depended on the diameter of the fiber, and took interconnecting tunnel structure. These micro-sized pores facilitate the entrance of the cells for bone formation and rapid integration of the composite with natural bone.

Cancer cells are destroyed when they are heated up to about 43°C.1) The temperature of the composites with 30 and 50 mass% magnetite addition reached over 43°C within 10 min under the magnetic field at 600 kHz and 3.2 kA/m. On the other hand, the composite with 10 mass% magnetite and the composite with the large sized magnetite did not generate enough heat. On the basis of crystallite size and TEM photograph (Fig. 1(a)) of magnetite, the used magnetite nano-sized particles were supreparamagnetic. On the contrary, the crystallite size of the magnetite of about 1 μm in size was several hundreds nanometer. This indicates that the magnetite of about 1 μm in size was fer-romagnetic.6) The mechanisms of heat generation in ferro-magnetic and superparamagnetic particles are different. Heat generation of ferromagnetic is caused by hysteresis loss. On the contrary, heat generation of superparamagnetic is caused by the delay in Neel relaxation of the magnetic moment.6) The heat gen-erated cannot be fully utilized for heating the cells as part of it is lost due to the bloodstream in the vicinity. Therefore, the com-posite must generate enough heat to raise the temperature of the surrounding tissue above 43°C. Furthermore, the heat generated in the composite can be controlled by the amount of the magne-tite, the size of magnetite particles, magnetic field frequency and the magnetic field strength. In practice, it would be wise to implant the magnetite/hydroxyapatite (HA) composite that pos-sessed the potential to generate sufficient heat that could be easily controlled by magnetic field parameters. Magnetic field generators that can operate under an AC frequency of 50 kHz ~1.2 MHz and maximum magnetic field strength of 15 kA/m are technically feasible. In our study, we have limited our experi-ments to AC frequency of 600 kHz and magnetic field strength of 3.2 kA/m, which is considered safe for human body.6)

The composite with 0 and 30 mass% magnetite addition was soaked in SBF for considering the effect of magnetite addition. Although the magnetite was added, leaf-like apatite was formed on both the samples with 0 and 30 mass% magnetite after soak-ing in SBF for 3 w. The results of the study undertaken to inves-tigate the binding property of the magnetite/hydroxyapatite (HA) composite with natural bone suggested that the 30 mass% mag-

Fig. 6. Temperature changes of the composite in the magnetic field.

(a) 0 mass% magnetite addition

(b) 30 mass% magnetite addition

Fig. 7. SEM photographs of the surface of the composite with (a) 0 and (b) 30 mass% magnetite addition after soaking in SBF.

Murakami et al.: Hydrothermal synthesis of magnetite/hydroxyapatite composite material for hyperthermia therapy for bone cancer

954

JCS-Japan

netite has high binding potential.19) It is known that magnetite shows passivity22) and hardly dissolve in the neutral pH range.23)

Therefore, magnetite particles in the composite hardly dissolve in SBF and body environment. Indeed, particles or bulks of mag-netite have been used in magnetic hyperthermia and reported to be safe.7),8)

Considering the entrapment of magnetite particles and heat generation, the composite containing 30 mass% of magnetite is expected to be suitable for the hyperthermia therapies of bone cancer.

5. ConclusionMicrostructure designed nano-sized magnetite/hydroxyapatite

porous composites were prepared by hydrothermal method. The composite was found to strongly hold 30 mass% magnetite in the cages of rod-shaped HA particles. This composite had micro-sized pores of about 400 μm and submicron-sized pores of about 0.2 μm in size. Furthermore, the temperature of the composite with 30 mass% magnetite rose to 55°C within 10 min, which could be considered sufficient for any further control through magnetic field parameters depending on the therapeutical envi-ronment.

Acknowledgement The authors thank Prof. K. Tohji of Tohoku University for his honest support.

References1) M. Shinkai, H. Honda and T. Kobayashi, J. Chem. Eng. Japan,

64, 14–16 (2000)[in Japanese].2) R. K. Gilchrist, R. Medal, W. D. Shorey, R. G. Hanselman, J.

G. Parrott and C. B. Taylor, Annals of Surgery, 146, 596–606 (1957).

3) M. Kawashita, M. Tanaka, T. Kokubo, Y. Inoue, T. Yao, S. Honda and T. Shinjo, Biomaterials, 26, 2231–2238 (2005).

4) M. Ma, Y. Wu, J. Zhou, Y. Sun, Y. Zhang and N. Gu, J. Mag-netism and Magnetic Materials, 268, 33–39 (2004).

5) T. Yao and T. Kokubo, Ceramics Japan, 28, 679–683 (1993)

[in Japanese].6) T. Atsumi, B. Jeyadevan, Y. Sato and K. Tohji, J. Magnetics

Society of Japan, 30, 555–560 (2006)[in Japanese].7) K. Konishi, T. Maehara, T. Kamimori, H. Aono, T. Naohara,

H. Kikkawa, Y. Watanabe and K. Kawachi, J. Magnetism and Magnetic Materials, 272–276, 2428–2429 (2004).

8) M. Johannsen, U. Gneveckow, L. Eckelt, A. Feussner, N. Waldofner, R. Scholz, S. Deger, P. Wust, S. A. Loening and A. Jordan, Int. J. Hyperthermia, 21, 637–647 (2005).

9) M. Jarcho, Clin. Orthop., 157, 259–272 (1981).10) K. Ioku, J. Soc. Inorg. Mater. Japan, 3, 412–418 (1996)[in

Japanese].11) T. Kanazawa, T. Umegaki, H. Monma and K. Yamashita,

Gypsum and Lime, 210, 261–273 (1987)[in Japanese].12) M. Akao and H. Aoki, Gypsum and Lime, 209, 225–228

(1987)[in Japanese].13) K. Ioku, M. Fukuhara, H. Fujimori and S. Goto, Korean J.

Ceram., 5, 115–118 (1999).14) K. Ioku, S. Yamaguchi, H. Fujimori, S. Goto and M.

Yoshimura, Solid State Ionics, 151, 147–150 (2002).15) K. Ioku, G. Kawachi, S. Sasaki, H. Fujimori and S. Goto, J.

Mater. Sci., 41, 1341–1344 (2006).16) K. Ioku, G. Kawachi, N. Yamasaki, M. Toda, H. Fujimori and

S. Goto, Key Eng. Mater., 288–289, 521–524 (2005).17) K. Ioku, S. Murakami, T. Atsumi, B. Jeyadevan and E. H.

Ishida, Key Eng. Mater., 309–311, 1039–1042 (2006).18) S. B. Cho, K. Nakanishi, T. Kokubo, N. Soga, C. Ohtsuki, T.

Nakamura, T. Kitsugi and T. Ymamuro, J. Am. Ceram. Soc., 78, 1769–1774 (1995).

19) T. Kokubo and H. Takadama, Biomaterials, 27, 2907–2915 (2006).

20) H. Monma and T. Kanazawa, J. Ceram. Soc. Japan, 84, 209–213 (1976).

21) K. Ioku, T. Murakami, Y. Ikuma and M. Yoshimura, J. Ceram. Soc. Japan, 100, 1015–1019 (1992)[in Japanese].

22) M. Kuroda, S. Tyan and T. Yamaue, Environmental Conser-vation Engineering, 36, 64–71 (2007)[in Japanese].

23) H. Tamura, S. Takasaki and R. Furuichi, The Japan Society for Analytical Chemistry, 47, 397–403 (1998)[in Japanese].

Documents

Hydrothermal synthesis of magnetite/hydroxyapatite