東北大学機関リポジトリTOUR

IMR KINKEN Research Highlights 2007

著者

東北大学金属材料研究所

year

2007

Highlights of Collaborated Research

out in consideration of the joint research program. Accepted applications are performed in following each division and facility of IMR.

1) Research Laboratories

Joint research conducted by out side researchers and IMR staff members at each research laboratory. Two categories; “Research in priority areas” and “Exploratory Research” are prepared.

2) International Research Center for Nuclear Materials Science

This facility is open to university scientists all over Japan to support experiments using Japan Material Test Reactor, JMTR, the experimental breeder reactor, JOYO and the test reactor, JRR-3 operated by JAEA (Japan Atomic Energy Agency). The overseas reactor, BR2, located at the Belgian Nuclear Research Center, is also used for irradiation experiments. This facility acts as a hub for international collaborations; specimens irradiated in overseas reactors are accepted here for post-irradiation examination by participating university researchers. Research subjects covered here include fundamental studies and R&D on fusion structural materials, high heat-flux materials, and a variety of functional materials, as well as engineering-oriented studies for the safety of light water reactors and basic researches supporting them. Materials studies utilizing radio-isotopes are also being conducted.

3) Advanced Research Center of Metallic Glasses

This center was established in May 1987 as the research center for the development of new advanced materials for the 21st century, reorganized in May 1996 and April 2005. The main purpose of the research in this laboratory is to establish fundamentals and control techniques for synthesizing artificial substances, especially the metallic glasses, and to seek for the possibility of applying them as multi-functional materials for high technology, such as materials for energy saving, materials for environmental and ecology, structural materials, electronic information materials, biomaterials and materials for social welfare, etc.

4) High Field Laboratory for Superconducting Materials

This laboratory restarted in 2001, succeeding “the High Field Laboratory for Superconducting Materials” which was established in 1981. The main equipment is a hybrid magnet which generates steady high magnetic fields up to 31T. In addition, many cryocooled superconducting magnets which have been developed by our laboratory are installed. The laboratory also provides instruments for measuring various physical properties. These facilities are open to scientists and engineers on superconductors and other materials research.

(IFCAM)

The material sciences are fundamentally interdisciplinary and global in nature, involving a variety of academic and engineering fields, in order to create better, less-expensive and more eco-friendly functional matter. Clever development and wise use of advanced materials are the foundation of all improved human activity in this new century. Based on the extensive and successful research record of IMR, IFCAM was approved and founded in 2002 with a mission to serve as an international think-tank in materials research. To foster further international collaboration, liaison offices of IFCAM have been established at Cambridge University, Harvard University, Stanford University, the Swedish Royal Institute of Technology, and the Institute of Physics in China.

Fig. Low energy electron microscope in IFCAM.

6) Osaka Center for Industrial Materials research

The Osaka Center was established at Osaka as a special unit in IMR in April 2006 based on the agreement between Tohoku University and Osaka Prefecture Government. It is a laboratory dedicated to research and development of nano-structured metallic materials from the application viewpoints based on the basic research in the material science, chemistry and physics associated with research organizations in the Kansai region. Research in the Osaka Center is focused in particular on the understanding of the fundamental properties of nano-structured metallic materials and the rapid realization of their application to industry, in particular, the small and medium enterprises in Osaka area.

7) Center for Computational Materials Science

This center was developed from the Laboratory of Materials Information Science which was established in February 1989. Its main tasks are 1) administration and maintenance of the supercomputing system in this Institute, 2) maintenance of a supercomputing system network, 3) general support for the usage of the supercomputing system, 4) support of materials design by supercomputing simulation with vectorization and parallelization, 5) construction of factual database for materials in nonequilibrium phase, and 6) support of activities of Nanotechnology group on Sinet3 and Asian Consortium on Computational Materials Science Virtual Organization on NAREGI.

By using a hot deformation process found by Nakajima et al., Si single crystal wafers can be now deformed in spherical, cylindrical, or any kind of curved shape. Present evaluation using X-ray diffraction showed that the curvature of lattice plane can be controlled precise enough for X-ray optics.

Shaping semiconducting single crystals into spheres or cylinders to obtain ideal lenses for X-ray optics is a classical but having never been fully satisfied dream. As early as in 1933, Johansson proposed 'A new, exactly focusing X-ray spectrometer' for focusing X-ray, potentially without loosing angular resolution. Since then, many attempts have been made to realize efficient and precise monochromating crystals. If such ideally focusing monochromaters are provided, the efficiency of X-ray diffractometer or spectrometer should surge up. However, a precise deformation and the quality of crystal are generally not compatible. Therefore, such monochromating crystals have been prepared by soft bending of hard crystals, like elastic bending of Si, or hard bending of soft crystals, like large deformation of ionic crystals.

Recent results by Nakajima et al.[1-3] showed that hot deformation can be a key factor to obtain curved single crystals with both precise shape and good crystal quality. Figure 1 shows curved crystals deformed into cylindrical or spherical shape with as small radius of curvature as up to 30 mm. The photograph clearly shows that the crystals still maintain mirror surfaces. By choosing some appropriate deformation conditions, these crystals have proven to keep their (111) plane parallel to the surface without severe deterioration. Figure 2 shows relationship between the curvature of (111) lattice plane measured by 333 diffraction and the designed curvature in the spherically deformed wafers. The figure indicates that the curvature of the (111) lattice planes of the samples agree well with the designed curvature. For example, the curvature agrees for wide range of about 30 degrees for the wafer with the radius of 40 mm[4]. The full-width at half maximum (FWHM) of the diffraction peak for the sample was still about 0.2 degree. This corresponds to a coverage of 0.017 of solid angle. Although the deformation conditions of the present samples were not yet completely optimized, the results already suggests that

the crystals examined in the present measurements already fulfill basic requirements to be used for monochromating crystals with extraordinarily large acceptance.

From another viewpoint, we found an interesting phenomenon concerning surface morphology. When the deformation condition is not optimal, the surface morphology of the deformed samples exhibits a pattern specific to the crystallographic symmetry of the wafer. For example, a strongly deformed (111) wafer gave a well-developed 6-fold pattern of slip lines near the edge of the wafer, whereas (100) and (110) wafers gave a 4- and a 2-fold pattern respectively. Further examination on this point would help us understand how the deformation process goes on.

To summarize, evaluation of Si single crystals prepared by a hot pressing method revealed that the spherically or cylindrically processed crystals fulfill the performance required for X-ray optics with acceptance angle more than one order of magnitude larger than the conventional ones. Further examination and tuning of the process is now under way[5]. References

[1] K.Nakajima, K.Fujiwara, W. Pan and H.Okuda, Nature Materials 4, 47 (2005).

[2] K. Nakajima, K. Fujiwara, and W. Pan, Appl. Phys. Lett. 85, 5896 (2004)

[3] K. Nakajima, K. Fujiwara, and W. Pan, J. Electron. Mater. 34, 1047 (2005)

[4] H.Okuda, K.Nakajima, K.Fujiwara and S.Ochiai, J. Appl. Crystall. 39, 443 (2006)

[5] H.Okuda, K.Nakajima, K.Fujiwara and S.Ochiai, PCT/ JP2006/325491

Contact to

Hiroshi Okuda (International Innovation Center, Kyoto University) e-mail: [email protected]

Kazuo Nakajima (Crystal Physics Division) e-mail: [email protected]

Si Wafers Having Quadric Surfaces with Large Curvatures

for X-ray Optics

Fig. 1. Photograph of plastically bent Si (111) crystals deformed either in cylindrical or spherical form. The radius of curvature is (a)30 mm, (b)40 mm and (c)100 mm, respectively.

Fig. 2. Relationship between the bending angle of (111) by X-ray diffraction measured from the center of the samples and that calculated from designed radius of curvature for spherically deformed Si (111) wafers.

-20 -15 -10 -5

0

5

10 15 20

-20

-15

-10

-5

0

5

10

15

20

Si (111) wafer

Δω

/ degree

Δω

/ degree

Fabrication of Fullerene Field-Effect Transistor Devices

on Thin Films and Nanometer Scale

Field-effect transistor (FET) devices with thin films of fullerenes have been fabricated and their FET characteristics have been investigated. The maximum field-effect mobility, μ, value of the FET with C60 thin films reaches 0.5 cm2 V-1 s-1

in which the Eu electrodes exhibiting small work function are used for effective carrier injection from electrodes.

Field-effect transistor (FET) devices with thin films of organic molecules are very attractive because of possible application for plastic electronics such as electronic papers and flexible displays, and their properties have been rapidly improved during the past 10 years [1, 2]. The typical device structure is shown in Figure 1. In this study, various types of FET devices are fabricated with thin films of fullerene molecules (C60, C70, C76, C78, C82, C84, C88 and fullerene

derivatives) and several types of dielectric gate insulators (SiO2, BaxSr1-xTiO3, polyimide, polyvinyl alcohol (PVA),

polyvinyl phenol (PVP), and parylene). The maximum field-effect mobility, μ, of the fullerene device in our study is 0.5 cm2 _V-1 _s-1 _{for the C60 thin film FET with Eu electrodes and}

SiO2 gate insulator. In this device the carrier injection was

very effective because of the Eu electrode with small work function, φ, of 2.5 eV, where Ohmic contact is formed owing to the Fermi level, εF = –2.5 eV, for Eu and LUMO level, εc =

-3.6 eV, of C60 thin films. Furthermore, the low-voltage

operation was realized in the C60 FET with high dielectric

gate insulator, BaxSr1-xTiO3. The FET characteristics are

clearly observed below source/drain voltage, VDS, of 10 V

because of an accumulation of high concentration of carriers. The effective carrier injection from electrodes and effective accumulation of carriers at channel region are very important for the realization of high-performance FET devices.

We have performed a quantitative study on carrier injection from electrodes to active layer with the C60 FET with

Au electrodes modified by 1-alkanethiols (CnH2n+1SH, n= 4,

6, 8, 10, 12, 16). In these devices the large carrier injection barriers were formed at the interfaces between the electrodes and C60 thin films. The carrier injection barriers are mainly

produced by additional tunneling barriers through the insulating CnH2n+1SH. The effective Schottky barrier heights, φBeffs, for the C60 FET devices with Au electrodes modified

by C6H13SH, C10H21SH and C16H33SH were determined to

be 0.338 ± 0.003, 0.373 ± 0.001 and 0.51 ± 0.0e eV, respectively, within the framework of thermionic emission model for double Schottky barriers. The tunneling efficiency, β, of electrons through insulating alkyl chain from electrodes to C60 was ~1 Å-1 for CnH2n+1SH. This value is the same order

as that for vacuum barrier. The Schottky barrier height, φB,

for the pure Au – C60 junction was 0.09 – 0.22 eV whose

value can be estimated from the temperature dependence of φBeff. This analysis is very effective to know electronic

structures at the interface between electrodes and active layer. The inverter circuit composed of n-channel C60 and

p-channel pentacene FET devices has also been fabricated for an application toward logic gate circuits based on organic devices. The inverter circuit showed an effective operation. Now the low-voltage operation in this inverter is tried with high dielectric gate insulator. In this study, nanometer scale device fabrication with C60 films has also been investigated.

The nanometer scale structures of C60 and other fullerenes

were fabricated on well-defined Si substrates by use of scanning tunneling microscope (STM), and the application of their structures toward nanometer-scale electronic devices has been tried. The electron/hole injections into C60

close-packed surfaces from STM tip produced C60 polymer rings,

(Fig. 2) whose outer/inner diameters expanded with an increase in bias voltage. This reaction is closely associated with the propagation of electrons/holes in the surface [3]. This result suggests an existence of field-effect chemical reaction on nanometer scale

Fig. 2. Polymer ring of C60 formed by electron injection from STM tip.

References

[1] C. D. Dimitrakopoulos and D. J. Mascaro, IBM J. Res. Dev. 45, 11 (2001).

[2] C. D. Dimitrakopoulos, P. R. L. Malenfant, Adv. Mater. (Weinheim Ger.) 14, 99 (2002).

[3] R. Nouchi, K. Masunari, T. Ohta, Y. Kubozono and Y. Iwasa, Phys. Rev. Lett. 97, 196101 (2006).

Contact to

Yoshihiro Kubozono (Research Laboratory for Surface Science, Okayama University)

e-mail: [email protected]

Yoshihiro Iwasa (Low Temperature Condensed State Physics Division)

e-mail: [email protected]

Interfacial Electronic Structure of Spin Tunneling Junctions based on

Perovskite Oxides Studied by in-situ Photoemission Spectroscopy

Heterointerfaces based on perovskite oxides have opened a wider horizon of possibility for creation of new multifunctional properties in ways that would not be possible in single-phase bulk materials. For controlling the novel functionalities, the detailed information on interfacial electronic structure is necessary.

The half metallic nature of optimally hole-doped manganese oxide La0.6Sr0.4MnO3 (LSMO) makes the

manganite intriguing for potential applications in spintronic devices in the form of heterostructures, such as tunneling magnetoresistance (TMR) devices and ferromagnetic field-effect transistors. However, the performance of TMR devices based on LSMO and SrTiO3 (STO) barriers is far worse than

what would be expected from the high spin polarization of LSMO, suggesting the formation of an interface layer with reduced spin polarization [1]. Since the performance of spintronic devices is very sensitive to interfacial electronic states, a detailed investigation of the interfacial electronic structure is crucial for designing spintronic devices. Here, we report on the interfacial electronic structure of STO/LSMO spin tunneling junctions studied by the Mn 2p-3d resonant photoemission (RPES), core-level photoemission, and x-ray absorption spectroscopic (XAS) [2,3].

The fabrication of LSMO/STO multilayers and their synchrotron-radiation analysis were performed at the in-situ photoemission spectroscopy system combined with a combinatorial laser molecular-beam epitaxy (laser MBE), which is installed at the soft-X-ray undulator beamline BL-2C at the Photon Factory as an end-station (Fig. 1). The distinctive feature of this system is the direct connection from the spectrometer to the combinatorial laser MBE chamber: Thin film samples can be transferred quickly into the photoemission chamber without breaking ultra high vacuum. Figure 2 shows the Mn 2p-3d RPES spectra of STO/LSMO multilayers. The elemental selectivity of these techniques enables us to extract the 3d partial density of states of LSMO layer in the vicinity of the interface with the STO overlayers [2]. We have found that the spectral intensity of Mn 3d eg

states near the Fermi level is drastically reduced when the LSMO film is capped with the STO overlayer, indicating the occurrence of hole doping into Mn sites in LSMO layers close to STO layers. By contrast, Ti 2p core level and XAS spectra are indicative of typical Ti4+ _{states with octahedral crystal}

symmetry (not shown). The preservation of Ti4+ _{states in the}

STO capping layer may be due to the high chemical stability of TiO2 plane in the STO. These results suggest that the hole

doped into the LSMO layer close to the heterointeface originates from the chemical carrier-concentration modulation at the valence-mismatched interface composed of the stacking sequence –TiO2–SrO–MnO2–La0.6Sr0.4O–, which is

proper to multilayers based on perovskite oxides.

Fig. 1. A schematic bird’s eye view of the in situ synchrotron-radiation photoemission spectroscopy system combined with a combinatorial laser MBE thin film growth system.

*OUFOTJUZ
BSC
VOJUT

#JOEJOH
&OFSHZ
F7

Mn 2p-3d Resonance

STO / LSMO

     #JOEJOH
&OFSHZ
F7

3 ML

Fig. 2. Mn 3d spectra of LSMO layers in the vicinity of an interface with STO. Mn 3d spectra of LSMO (x = 0.4 and 0.55) films are also presented for comparison. The Mn 3d spectra were obtained by subtracting the off-resonance spectra from the on resonance spectra as shown in the inset.

References

[1] H. Yamada, Y. Ogawa, Y. Ishii, H. Sato, M. Kawasaki, H. Akoh, and Y. Yokura, Science 305, 646 (2004).

[2] H. Kumigashira, A. Chikamatsu, R. Hashimoto, M. Oshima, T. Ohnishi, M. Lippmaa, H. Wadati, A. Fujimori, K. Ono, M. Kawasaki, and H. Koinuma, Appl. Phys. Lett. 88, 192504 (2006).

[3] M. Minohara, I. Ohkubo, H. Kumigashira, and M. Oshima,

Appl. Phys. Lett. 90, 132123 (2007).

Contact to

Hiroshi Kumigashira (Department of Applied Chemistry, The University of Tokyo)

e-mail: [email protected]

Masashi Kawasaki (Superstructured Thin Film Chemistry Division)

Perpendicular Magnetization of L1

0

-Ordered FePt Films in the

Thinnest Limit Probed by Circularly Polarized Soft X-rays

range were fabricated by alternate monatomic layer deposition of Fe and Pt on Pt (001) substrate and were capped by Pt. Perpendicular Fe magnetic moment under magnetic field, together with that under remanence, was detected by soft x-ray magnetic circular dichroism.

Thin films of L10-ordered FePt have attracted much interest

because the perpendicular magnetization is expected to open up such applications as ultrahigh-density magnetic storage media and spintronic device components. In the course of such applications, it is crucial how thin the magnetic layer can be made keeping the

room-temperature perpendicular magnetization. Since

ferromagnetism is a collective phenomenon, it could be diminished when the thickness is reduced, leading to lower Curie temperatures and possibly to the disappearance of perpendicular magnetization even in the ferromagnetic phase.

In order to determine the thinnest limit for L10-ordered FePt films

to have perpendicular magnetization, ultrathin films of FePt sandwiched by Pt, as shown in Fig. 1, were fabricated by alternate monatomic layer (ML) deposition of Fe and Pt[1]. When the thickness of the magnetic layer is as thin as a few ML, it is extremely difficult to measure magnetization precisely by conventional techniques such as SQUID because of a large background due to substrate. Therefore, we have utilized soft x-ray magnetic circular dichroism[2] of the Fe 2p→3d photoabsorption (Fe 2p→3d XMCD), which can yeild accurate Fe 3d magnetic moment even for Fe coverage of below 1 ML.

In Fig. 2 is shown the temperature and thickness dependence of the perpendicular Fe 3d magnetic moment [3] in Pt/(Fe 1ML/Pt

1ML)n/Pt(100) under 1.4 T (a) and under remanent magnetization

(b). In the latter, samples were first subjected to a magnetic field of 1.4 T and were measured without magnetic field. The drastic decrease from n=2 to 1 of the moment at room temperature (RT) under 1.4 T (see Fig. 2 (a)) might suggest that the Curie temperature

TC is above RT for n≥2 and below RT for n=1. On the other hand,

the perpendicular remanent magnetization at RT reaches zero already at n=2. Let us define the temperature under which

perpendicular remanent magnetization drastically increases as Trem.

Trem is between 250 K and 300 K for n=2 and Trem ~ 160 K for n=1.

These results, characterized by the lowering of both TC and Trem

with the reduction of thickness n, agrees well to the expectation that ferromagnetism should be weakened as the thickness is reduced.

A possible mechanism for the absence of remanent

magnetization between Trem and TC is the formation of stripe

domains characteristic of perpendicular magnetization. One cannot exclude other possibilities such as the spin reorientation transition to in-plane magnetization. In order to clarify the actual mechanism, it will be necessary to investigate magnetic domain structures by such methods as XMCD microscopy.

The observed Fe 3d magnetic moment of 2.6-2.7 μB in low

temperatures (see Fig. 2 (a)) is very near to the reported Fe magnetic

moment in the bulk L10-ordered FePt of 2.8 μB obtained by neutron

diffraction. Band structure calculation of L10-ordered FePt predicts

Fe magnetic moment of 3.0 μB. The contribution of the orbital

magnetic moment to the total moment, μorbital/(μspin+μorbital) is found

to be 5±1% for all studied n’s (n=1, 2, 3, 10). This suggests that the Fe 3d electronic state does not depend strongly on n.

In summary, Pt/(Fe 1ML/Pt 1ML)3/Pt (001) was found to show

perpendcular remanent magnetization at room temperature, which is expected to open up application in the field of magnetic storage and spintronics. Samples with thinner FePt layer showed perpendicular remanent magnetization in lower temperatures. Thickness dependence of perpendicular magnetization can be attributed to the

lowering of TC because ferromagnetism would be weakened due to

the crossover from three dimensions to two dimensions.

Fig. 2. Temperature and thickness dependence of the perpendicular

Fe 3d magnetic moment per atom in Pt/(Fe 1ML/Pt 1ML)n/Pt(100)

under 1.4 T (a) and under remanent magnetization (b). Markers represent the measured values and the solid lines are guides for eyes.

References

[1] T. Shima, T. Moriguchi, S. Mitani and K. Takanashi, Appl. Phys. Lett. 80, 288 (2002).

[2] S. Imada, T. Muro, T. Shishidou, S. Suga, H. Maruyama, K. Kobayashi, H. Yamazaki and T. Kanomata Phys. Rev. B, 59, 8752 (1999). [3] S. Imada, A. Yamasaki, S. Suga, T. Shima and K. Takanashi, Appl. Phys. Lett. 90, 13250/1-3 (2007).

Contact to

Shin Imada (Graduate School of Engineering Science, Osaka University)

e-mail: [email protected] Koki Takanashi (Magnetic Materials Division) e-mail: [email protected]

Fig. 1. Schematic view of ultrathin FePt films sandwiched by Pt, where gray and blue balls represent Pt and Fe atoms, respectively, for Pt/(Fe 1ML/Pt 1ML)n/Pt(100), with n=1 (a), n=2 (b), and n=3 (c).

Evaluation of Amorphous Calcium Phosphate Coating Films Fabricated on

Titanium by Using RF Magnetron Sputtering for Medical Applications

on titanium substrates by using radiofrequency magnetron sputtering was evaluated in vivo for investigating its applications in dental and medical implants. Titanium implants coated with the amorphous calcium phosphate film were implanted into the mandibles of beagle dogs. The percentage of bone-implant contact in coated titanium was greater than that in uncoated titanium.

Since titanium and its alloys can be directly connected to living remodeling bones at the light microscopic level, i.e., osseointegration, they have been widely used in dental and medical implants. In fact, almost all dental implants are made of commercially pure titanium or α+β type titanium alloys such as Ti-6Al-4V. The fixation between the dental implants and bones might be influenced by the state of the bones and the possible length of the implants. The coating of titanium implants with calcium phosphate is one of the methods used to improve the osseointegration of the implants.

Many experimental coating processes for calcium phosphate have been investigated, including plasma spraying, sputtering, pulsed laser deposition, dip coating, sol-gel coating and electrophoretic deposition. Among these processes, radiofrequency (RF) magnetron sputtering has several advantages such as a low processing temperature and excellent adhesion to metallic substrates, as shown in Fig.1 [1]. In the fabrication of calcium phosphate films on titanium, their phase, crystallinity and preferential crystallographic orientation can be controlled by changing the sputtering parameters [2]. Amorphous calcium phosphate (ACP) fabricated by RF magnetron sputtering is a candidate coating material for titanium. The ACP coating film on titanium exhibits bioactivity with the rapid formation of apatite crystals in simulated body fluids because of the coating film’s bioresorbability.

An ACP film with a thickness of 0.5 μm was coated on titanium implants. The coated titanium implants were inserted into bone cavities created in the mandibles of beagle dogs. The animals were sacrificed 2, 4, 8 or 12 weeks after the

implantation, and the percentage of bone-implant contact was measured. Figure 2 shows the percentage of bone-implant contact [3], which reflects the biocompatibility of implants. The percentage of bone-implant contact for the ACP-coated specimen was greater than that for the uncoated specimen 8 - 12 weeks after implantation. The application of sputtered ACP films as coatings on titanium implants was suggested to be effective for implants’ rapid and strong fixation with bones. As compared with other surface modification processes of titanium that involve high-temperature processes or post heat treatments such as sol-gel coating and NaOH treatment method, the low processing temperature required in the sputtering process of ACP films could prove to be an advantage for its use as coating on implants made of β-type titanium alloys; this is because the microstructure of β-type titanium alloys changes due to the high processing temperature during coating. In addition, thin calcium phosphate films fabricated with RF magnetron sputtering might be able to maintain the roughness of titanium implants such as dental implants.

References

[1] T. Narushima, J. Jpn. Soc. Biomater. 25, 252 (2007). [2] T. Narushima, K. Ueda, T. Goto, H. Masumoto, T. Katsube, H. Kawamura, C. Ouchi and Y. Iguchi, Mater. Trans. 46, 2246 (2005).

[3] K. Ueda, T. Narushima, T. Goto, H. Nakagawa, H. Kawamura and T. Katsube, Mater. Trans. 47, 307 (2006).

Contact to

Takayuki Narushima (Department of Materials Processing, Tohoku University)

e-mail: [email protected]

Takashi Goto (Multi-Functional Materials Science Division) e-mail: [email protected]

Fig. 1 Comparison between calcium phosphate coating processes on titanium. Adherance Poor Good Coating Area, S / cm2 0.1 1 10 100 1000 Dip coating Electrophoretic Thermal spraying PLD Sputtering IBDM Biomimetic Sol-gel R.T. 300 600 900 1200 Thickness of Coating, t / μm Process Temperature, T / C° 0.1 1 10 100 1000 Dip coating Electro-phoretic Thermal spraying Biomimetic Sol-gel Sputtering PLD IBDM 0 _{2 - 4 w 8 - 12 w} 10 20 Bone-Implant contact, C (%) 30 40 50

*

Fig. 2 Percentage of bone-implant contact for ACP-coated and unACP-coated (control) titanium implants. The differences between the ACP-coated and control titanium implants were statistically analyzed by Student's t-tests (*p<0.05).

Microwave Activated Reactions of Hydrides

for Advanced Hydrogen Storage Systems

borohydrides have been systematically investigated. The former TiH2 shows rapid heating and the later LiBH4 rapidly

desorbs more than 13.8mass% hydrogen by microwave irradiations. The energy saving microwave process might be applied to advanced hydrogen storage systems.

The required properties on hydrogen storage materials are low reaction temperatures and fast kinetics. In order to decrease the reaction temperatures, the stabilities of the hydrides are controlled by alloying for metal hydrides or fabricating appropriate composites for borohydrides. For improving the kinetics of the reactions, catalysts such as transition metals and oxides are added. In the conventional way, the dehydriding and hydriding reactions of hydrogen storage materials were carried out mainly by external heating in electric furnaces. We have proposed the application of the energy saving microwave process to the hydrogen storage system and the effects of microwave irradiation were investigated on the dehydriding reactions of metal hydrides and alkali borhydrides.

Figure 1 (a) shows the temperature changes indicating the dehydriding reactions for several metal hydrides and alkali borohydrides [1]. A rapid increase of temperature is observed in TiH2 and LiBH4 by microwave irradiation. TiH2 is

heated up to 600 K for only 3.6 min and LiBH4 shows a rapid

heating above 380 K. Dehydriding reaction of LiBH4 upon

heating is shown in Fig. 2 (b).To investigate these microwave enhanced reaction, the high frequency complex permittivity spectra εr* = εr’ - iεr” were measured for these hydrides using

a coaxial line which is sealed in the airtight sample holder.

Fig. 1 (a) Temperature changes of metal hydrides and alkali borohydrides as a function of microwave irradiation time, and (b) overall dehydriding reaction of LiBH4 upon heating.

Fig. 2 Dielectric constants εr’ and loss values εr” as a function of frequency at room temperature and 380 K. Inset shows the temperature dependence of εr’ and εr” of LiBH4 at 2.16 GHz.

The complex permeability spectra of TiH2 and LiBH4 are

shown in Fig. 2. Based on this result, it can be concluded that the microwave heating is originated by the conductive loss in TiH2 and by the dielectric loss in LiBH4 at room temperature.

In spite of the rapid heating, the skin depth effect disturbs the penetration of microwaves and a small amount 0.16 mass% of hydrogen is desorbed in TiH2. On the other hand, LiBH4

was found to be heated by the conductive loss above 380 K. This can be attributed to the ionic electrical conduction due to the crystal structural transformation from orthorhombic to hexagonal, in which the (BH4)- tetrahedron is rearranged

along the c-axis. Approximately 13 mass% of hydrogen can be desorbed in this process.

To improve the dehydriding properties, the composite materials of these two hydrides were investigated [2]. The temperature change of the TiH2 + 2LiBH4 composite by

microwave irradiation is also shown in Fig. 1(a). The temperature of the composite gradually increases up to 9 min and rapidly increases above 350 K by the microwave irradiation. This behavior is similar to a superposition of the individual behaviors of TiH2 and LiBH4. Thus the suitable

composite fabrication is useful for controlling the rapid heating and for maintaining the amount of desorbed hydrogen. References

[1] Y. Nakamori, S. Orimo and T. Tsutaoka, Appl. Phys. Lett.

88, 112104 (2006).

[2] M. Matsuo, Y. Nakamori, K. Yamada and S. Orimo, Appl.

Phys. Lett. 90, 232907 (2007).

Contact to

Takanori Tsutaoka (Hiroshima University) e-mail: [email protected]

Yuko Nakamori (High-Temperature Materials Science (Environmental Materials Science) Division)

e-mail: [email protected]

(a) (a)

(b) (b)

Anomalous Josephson Effect due to Odd-frequency Pairs

superconductor / normal metal / superconductor junctions, where the p-wave pairing symmetry is assumed in superconductors. The amplitudes of the Josephson current become much larger than those in the s-wave junctions because of a cooperative effect between the midgap Andreev resonant states and the proximity effect. We show that the Josephson current is carried by the unusual odd-frequency Cooper pairs in normal metals.

While the Josephson effect was originally discovered in a tunnel junction, any conductor can support a supercurrent in equilibrium. When a diffusive normal metal is inserted between two superconductors, the proximity effect is responsible for the Josephson current. Recent theoretical studies showed the sensitivity of the proximity effect to pairing symmetries of superconductors [1-3]. In this article, we show anomalous behaviors of the Josephson current in superconductor / normal metal / superconductor (SNS) junctions as shown in Fig. 1, where superconductors are characterized by the p-wave symmetry. The midgap Andreev resonant state (MARS) penetrating into a normal metal causes the anomalous Josephson effect. We also show that odd-frequency Cooper pairs carry the anomalous Josephson current.

In Fig. 2(a), we show calculated results of the Josephson current as a function of phase differences across the SNS junctions. The insulating barriers are inserted at two interfaces between superconductors and a normal metal. The transmission probability of the insulator is about 0.1. The current-phase relations (CPR) are surprisingly close to J ∝ φ at low temperatures. We note that such CPR is expected only in ballistic SNS junctions and indicates large contributions of the multiple Andreev reflections to the Josephson current. For comparison, the CPR of the s-wave SNS junctions is shown with a solid line, where the Josephson current is amplified by 10. In the s-wave junctions, the CPR is almost described by a sinusoidal function because the multiple Andreev reflections are suppressed due to the impurity scatterings in a normal metal and the insulating barriers at the interfaces. Thus a quasiparticle seems to be free from any scatterings by impurities.

Fig. 1 Schematic illustration of a SNS junction. Green circles represent impurities.

Since an electron obeys the Fermi-Dirac statistics, the pairing function of Cooper pairs must be anti-symmetric under the interchange of two electrons, i.e.,

Fω(r,r')=-F-ω(r',r).

The spin part of the pairing function is suppressed because they are always odd for the s-wave and even for the p-wave symmetry by interchanging two electrons. The interchange of the two electrons results in the negative sign on the right hand side of the equation. In the p-wave superconductors, the equation is satisfied because of the odd-parity pairing symmetry. The odd-parity in the orbital part, however, cannot persist in a normal metal and only the s-wave component survives. This is because the pairing function must be isotropic in both real and momentum spaces due to impurity scatterings. In Fig. 2(b), we show the pairing function in a normal metal as a function of Matsubara frequency (ω) in the p- and s-wave SNS junctions. The real and imaginary parts of the pairing function remain finite values in the s- and p-wave junctions, respectively. The results show that the pairing function is an even function of ω in the s-wave junctions, whereas it is an odd function of ω in the p-wave junctions. Thus we conclude that the odd-frequency pairs are responsible for the anomalous Josephson current. The p-wave pair potential has the nodes in the orbital part of its pairing function. The proximity effect transfers the nodes in momentum space to the nodes in time.

Fig. 2 (a) Current-phase relation of p-wave SNS junction. (b) The pairing function versus frequency.

In summary we studied the Josephson effect in SNS junctions of the p-wave pairing symmetry. The results show that the anomalous Josephson current is carried by the odd-frequency pairs in a normal metal

References

[1] Y. Asano, J. Phys. Soc. Jpn. 71, 905 (2002).

[2] Y. Tanaka and S. Kashiwaya, Phys. Rev. B 70, 012507 (2004).

[3] Y. Asano, Y. Tanaka and S. Kashiwaya, Phys. Rev. Lett. 96, 097007 (2006).

Contact to

Yasuhiro Asano (Department of Applied Physics, Hokkaido University)

e-mail: [email protected]

Sadamichi Maekawa (Theory of Solid State Physics Division) e-mail: [email protected]

Development of Novel Solid Electrolyte Capacitors

-Dielectric Properties of Anodically Oxidized Film on Nb-Ti

density are improved in the anodically oxide film on Nb-Ti alloys, comparing with the conventional one on Nb. Therefore, the novel electrolyte capacitors can be developed by using Nb alloy-type valve metals with a suitable additional element and composition.

Niobium (Nb) has become of considerable practical interest as a valve metal, which is readily anodized and form stable oxide films for solid electrolyte capacitors [1,2]. To apply the Nb capacitor for a wider field, it is required to further improve the capacitance and leak current density of the anodic oxide film. Here, titanium (Ti) as well as

Nb is known as a valve metal, on which the anodic oxide films, TiO2,

shows an excellent relative permittivity of about 55 – 80 [3]. Therefore, we propose to use Nb-Ti alloys as valve metals in order to develop a novel capacitor, and the purposes of the current study are to systematically evaluate the structure and dielectric properties of the anodic oxide film on Nb-Ti alloys as a function of Ti content, and thereby to clarify the effects of the alloying element on the properties of anodic oxide film.

Nb-Ti solid solution alloys with nominal compositions of Nb - 1 to 15 at.% Ti, together with pure Nb, were prepared by argon arc melting These ingots obtained were annealed, followed by cold-rolling and cutting into plates. The specimens were anodized in an

aqueous electrolyte of 0.6 mass % H3PO4 at 333 K at the voltage of

12, 16 and 20 V for approximately 25.2 ks.

Fig. 1 shows the CV values per unit surface area of the anodized

Nb and Nb-Ti alloys. Here, the “CV value” exhibits the capacitance from the practical viewpoint of solid electrolyte capacitors. The CV

value of the anodized Nb was calculated to be about 11.5 μF V/cm2_,

and that of the anodized Nb-Ti alloy was enhanced by the addition

of titanium, showing a maximum of about 12.3 μF V/cm2 _{at the Ti}

content of 5 at.%, and then, it decreased gradually to 11.8 μF V/cm2

at the Ti content of 15 at.%. The change of the CV values can be explained in the terms of the relative permittivity, ε, and the growth rate of the anodic oxide film, d, which depend on the composition of the Nb-Ti alloys substrate: the higher CV value of the anodized Nb -5at.% Ti alloy arises from an appropriable combination of the ε and

d. On the contrary, the decrease of the CV value for the anodized

Nb-Ti alloys with Ti content of more than 5 at.% is due to the fact of

the formation of a thicker film, in spite of higher ε. Therefore, it might

be stated that the Nb -5 at.% Ti alloy is one of the most suitable choice for the solid electrolyte capacitor in the Nb-Ti alloys with the Ti content between 0 and 15 at.%.

Fig. 2 shows the leak current density of anodized Nb and Nb-Ti

alloys, indicating that the leak current density tends to decrease with the Ti content of the alloy, although the values measured were significantly deviated as indicated by error bars. It should be noted that the deviation was also reduced by the increase of Ti content of the alloy. The leak current is generally explained in terms of the ionic conductivity of the anodic oxide and the immersion of the electrolyte into the defects like micro-cracks of the anodic oxide film, which arise from the stress generated by formation of the anodic oxide. The large deviation of the leak current value observed in this study can mainly be interpreted by the latter term, because the distribution and the size of cracks induced by anodizing varies from specimen to specimen. We found that the anodic oxide film on Nb is subject to cracks during anodizing and that the defect is suppressed as the Ti content of the Nb-Ti alloy increases (not shown). Therefore, it is proposed that the decrease of the leak current is caused by the suppression of the defect. One of the possible reasons why the Ti addition to the substrate Nb leads to the formation of non-defective

film is that the stress applied in the Nb2O5 / substrate interface is

accommodated by TiO2, because the density of TiO2 is between

that of amorphous Nb2O5 and substrate Nb-Ti alloy.

Thus, we can state that the both the CV value and the leak current density can be improved by using the Nb-Ti alloys containing between 1 and 10 at.% Ti as a valve metal in comparison with the capacitor composed of pure Nb. To summarize, we demonstrated that the novel solid electrolyte capacitor can be developed by using an alloy-type valve metal with a suitable choice of the sort and the amount of the additional element.

References

[1] K. Kukli, M. Ritala and M Leskela, J. Electrochem. Soc. 148, F35 (2001).

[2] S. Semboshi, N. Masahashi, T.J. Konno and S. Hanada, Metal. Mater. Trans. A, 37 A, 1301 (2006).

[3] T. Hurlen and S. Hornkjol, Electrochem. Acta. 36, 189 (1991).

Contact to

Satoshi Semboshi (Department of Materials Science, Osaka Prefecture University)

e-mail: [email protected]

Naoya Masahashi (Osaka Center for Industrial Materials Research) e-mail: [email protected]

Fig. 1 CV values plotted as a function of the Ti content of the Nb-Ti

alloy specimens anodized in the 0.6mass% H3PO4 electrolyte at

333 K 25.2 ks. The CV value of the anodized Nb-Ti alloys shows a maximum at the Ti content of around 5 at.% of the alloys.

Fig. 2 Leak current density plotted as a function of the Ti content of the anodized Nb-Ti alloy, showing that the leak current density decrease as the Ti content of Nb-Ti alloys.

Development of Fabrication Technology for Low Activation

Vanadium Alloys as Fusion Blanket Structural Materials

joints and pressurized creep tubes were fabricated from the reference high-purity V-4Cr-4Ti ingots designated as NIFS-HEAT. Control of impurities such as C, N, O was the key to maintaining the good mechanical properties of the products.

Vanadium alloys are promising candidates for fusion blanket structural materials, because of their low activation property, high temperature strength and high resistance to neutron irradiation. In the development of the alloy for fusion blanket application, joining and tubing are the key necessary technologies. Joining and tubing technologies are also crucial for fabricating Pressurized Creep Tubes (PCTs), which are the only available means to evaluate creep deformation during irradiation. Coating with W is thought to be necessary for application to fusion first wall for the purpose of protecting the wall of vanadium alloy from heat and particle loadings originated from the core plasma. Japanese Universities and National Institute for Fusion Science (NIFS) have a collaboration program for producing high purity reference heat of V-4Cr-4Ti alloy (NIFS-HEAT) [1]. After production of the ingot, collaboration activities were promoted for fabricating various products and characterizing the alloys and the products including irradiation tests. In this program, plates, sheets, wires and tubes were fabricated. Also carried out was the production of W-coating, laser weld joints and pressurized creep tubes. The products were exhibited in Fig. 1. Part of the achievement of this program was reported in Ref. [2].

The common key issue in the fabrications is the control of impurities such as C, N and O. The increase in the level of C, N and O is known to degrade the ductility of the alloy. The mechanical properties of the products were improved significantly not only by reducing the impurity levels of C, N and O, but also by controlling density, size and distribution of the Ti-C, N, O precipitates. In the laser weld joining, atmospheric control of the welding area was essential. Also important was

the post-weld heat treatment, which can cause redistribution of the impurities introduced during the welding and the Ti-C, N, O precipitates. For the tubing, vacuum level and prior surface cleaning at the intermediate heat treatments were crucial.

Because the size, density and distribution of Ti-C, N, O precipitates influence strongly the mechanical properties, efforts were made to increase the high temperature strength by distributing high density of fine Ti-C, N, O precipitates in the matrix by series of controlled heat treatments. In addition to the increase in the Yield and Ultimate Tensile Strength, increase in thermal creep resistance was derived by the fine precipitation.

Through the production of PCTs, technology for small scale welding with controlling the grain size was enhanced [3]. The PCTs fabricated are being used for characterizing thermal and irradiation creep properties in JOYO and HFIR reactors. Recent results showed that the consistency of the data was much better than those obtained using the PCTs fabricated previously.

The present achievements enhanced coordination for promoting international collaboration on developing vanadium alloy-based test blanket modules to be installed into International Thermonuclear Experimental Reactor (ITER). References

[1] T. Muroga, Mat. Trans. 46, 405 (2005).

[2] T. Nagasaka, T. Muroga, K. Fukumoto, H. Watanabe, M. L. Grossbeck and J. Chen, Nuclear Fusion 46, 618 (2006). [3] K. Fukumoto, H. Matsui, M. Narui, T. Nagasaka and T. Muroga, J. Nucl. Mater. 335 103 (2004).

Contact to

Takeo Muroga (National Institute for Fusion Science) e-mail : [email protected]

Yuhki Satoh (Nuclear Materials Engineering Division) e-mail: [email protected]

Novel Magnetic Ordering in Neptunium Dioxide : A Mystery for Half a

Century is Addressed using the Nuclear Magnetic Resonance Technique

The phase transition in Neptunium dioxide (NpO2) was originally

discovered in 1953. Since this discovery, however, it has remained a most-lasting mystery in the physics of actinide compounds. In recent works, the possibility that this phase transition can be described as one of a new class of phase transition associated with the orbital degrees of freedom has been discussed. With this in mind, we have tackled this problem anew using the nuclear magnetic resonance (NMR) technique.

Actinide compound are very exciting, because they have most of the interesting effects that are being studied in condensed matter physics nowadays. They are very complex systems to understand, and they are the least understood. One of the reasons that these systems have not been studied much is because of their radioactivity. Therefore, they require special facilities and special places. There are only a few places in the world where these compounds are studied, and International Research Center for Nuclear Materials Science, Institute for Materials Research Science is one of them.

A complexity of actinide compound arises from the fact that the orbital moment of f-electron is never quenched by the crystal field, which differentiates the f-electron systems from the d-electron systems. Furthermore, because of the strong spin-orbit interaction in a f-electron shell, their spin and orbital degrees of freedoms are tightly coupled with each other. In such the case, the relevant degrees of freedom for one ion are its magnetic and electric multipole moments, that is, charge 0), dipole 1), quadrupole (rank-2), and octupole (rank-3) etc. The interactions between these multipole moments bring complex phase diagram to actinide compounds, even if the f-electrons stay in the localized limit.

The phase transition in NpO2 was originally discovered through

specific heat and magnetic susceptibility measurements, which appeared to suggest a transition to a magnetic dipolar ordering state

at T0 =26 K. However, in the 1980s, such a dipole order scenario has

been unambiguously ruled out by both neutron elastic scattering and Mösbauer spectroscopy measurements. In 2000, the possibility of ordering of higher rank of magnetic multipole, i.e. octupole was proposed theoretically. Soon after that, resonant x-ray scattering measurements suggested the occurrence of longitudinal triple-q type

antiferro-octupolar (AFO) order below T0 . This AFO ordered state has

also been suggested by recent microscopic calculations based on the j-j coupling scheme by K.Kubo and T. Hotta (Fig.1) [1].

In order to elucidate the microscopic properties of this exotic

ordered state, we have initiated the first NMR studies in NpO2 [2,3].

Figure 2 shows the field angular (θ) dependence of the 17_{O NMR}

line splitting, which illustrates the angular variation of hyperfine (HF)

fields at 17_{O nuclei [3]. In the presence of longitudinal triple-q type}

order, the 17_{O NMR line splits into two inequivalent sites, i.e., the}

O(1) _{and O}(3) _{sites [2]. The nearly flat curve for the O}(1) _site

corresponds to the isotropic nature of its HF field. On the other hand,

the two other curves for the O(3) _{sites show a strong θ-dependence}

due to the anisotropic nature of their HF fields. Using the invariant form of the HF field contributions derived by O.Sakai et al. (TMU) by

assuming the longitudinal triple-q AFO ordered state for NpO2, we

can deduce the angular dependence of HF field at O sites. As shown in Fig. 2, we have found that the angular dependences of HF field are well reproduced by the model calculation for all three O sites. This excellent agreement obtained with just three scaling parameters provides microscopic evidence for the proposed longitudinal triple-q

AFO model for NpO2.

Compared with dipolar ordering, multipolar ordering is hard to investigate by conventional techniques. In the present work, we demonstrate that microscopic investigation of multipole ordering is possible by means of NMR through the HF interactions.

References

[1] K. Kubo and T. Hotta, Phys. Rev. B 71 140404(R) (2005); ibid. 72, 132411 (2005).

[2] Y. Tokunaga, Y. Homma, S. Kambe, D. Aoki, H. Sakai, E. Yamamoto, A. Nakamura, Y. Shiokawa, R.E.Walstedt, and H. Yasuoka, Phys. Rev. Lett. 94, 137209 (2005).

[3] Y. Tokunaga, D. Aoki, Y. Homma, S. Kambe, H .Sakai, S. Ikeda, T. Fujimoto, R.E. Walstedt, H. Yasuoka, E. Yamamoto, A. Nakamura, and Y. Shiokawa, Phys. Rev. Lett. 97, 257601 (2006).

Contact to

Yo Tokunaga (Advanced Science Research Center, Japan Atomic Energy Agency (JAEA))

e-mail: [email protected]

Yoshiya Homma (Radiochemistry of Metals Division) e-mail: [email protected]

Fig.1. Magnetic-octupolar ordering state of NpO2 calculated by K.

Kubo and T. Hotta [1]. Red and Blue colors on the surface indicate the weight of up- and down-spin states.

Fig.2. Field-angle dependence of the internal field at O sites in

NpO2. The solid and dotted lines are the results of model

calculations with and without the contribution from the field-induced magnetic octupolar ordering, respectively [3].

Ultrahigh Fatigue Strength, High Fracture Toughness and

Excellent Cutting Properties in Bulk Metallic Glasses

which are important properties for practical applications of bulk metallic glasses (BMGs), were investigated in the BMGs with different chemical compositions. The Ti-based BMG showed super high fatigue strength in comparison to the high strength crystalline alloys with high fatigue strength. The fracture toughness showed equivalent high values to the high tensile steel. The roughness of the machined surfaces exhibited the precise finishing level not being achieved in crystalline alloys.

Fatigue tests were carried out on nanocrystal dispersed Ti41.5Zr2.5Hf5Cu42.5Ni7.5Si1 (Ti-based), Cu60Zr30Ti10 (CuZrTi,

Cu-based) and Cu60Hf25Ti15 (CuHfTi, Cu-based) bulk metallic

glasses (BMGs) under axial loading at a stress ratio of 0.1 and a frequency of 10 Hz. The fatigue limit (σw = σmax. -σmin.)

and fatigue ratio (σw/σB, σB ; tensile strength) in the Ti-based,

CuZrTi and CuHfTi BMGs were 1610 MPa and 0.79, 980 MPa and 0.49, and 860 MPa and 0.40, respectively. In particular, the Ti-based BMG showed super high fatigue strength in comparison to the high strength crystalline alloys with high fatigue strength [e.g. Cr-Mo steel (JIS SCM435), σw

= about 1000 MPa] as shown in Fig.1 [1].

The plane-strain fracture toughness (KIC) test was carried

out using CT specimens of Zr50Cu40Al10, Zr50Cu30Al10Ni10,

Cu60Zr30Ti10 and Ti41.5Zr2.5Hf5Cu42.5Ni7.5Si1 BMGs (Zr-, Cu-

and Ti-based BMG, respectively). Both the Zr- and Cu-based BMGs showed considerably large KICs, 44 and at least about

40 MPa·m1/2_{, respectively, and the Ti-based BMG showed 23}

MPa·m1/2 _{as shown in Fig.2 [2]. On the specimen surfaces,}

small crack tip bluntings as the result of many plane-stress type shear slips near the fatigue crack tip were observed, and the maximum length of the shear bands was almost coincident with the plane-stress plastic zone size calculated by the fracture mechanics. On the fracture surfaces, in the case of the large KIC, stretched zones(SZs) (maximum SZ

width = several tens μm) with crack tip opening by the accumulation of plane-strain type long and short shear slips were observed.

The cutting characteristics of BMGs were examined by turning with different tool materials, nose radii (Rn) and cutting

speeds (V). Round bars of Zr65Cu15Ni10Al10 and

Pd40Cu30Ni10P20 BMGs were used as the workpieces. The

value of surface roughness (Ra) exhibited the precise finishing

level, which can not be achieved in crystalline alloys, and the value became smaller with increasing values of Rn (i.e.,

0.08μm). The chip of the BMGs showed an ideal flow type with very short and regular intervals formed by planar slip as shown in Fig.3, and revealed very homogeneous, flat and featureless back surfaces. From these observations of chips, it is presumed that the reason for the excellent cuttability of BMGs is due to a slipping-off mechanism at planes of very short intervals decided only by the maximum shear stress and non-built -up edges caused by a low glass transition temperature [3].

References

[1] K. Fujita, T. Hashimoto, W. Zhang, N. Nishiyama, C. Ma, H. Kimura and A. Inoue, J. Japan Inst. Metals, 70, 816 (2006). [2] K. Fujita, A. Okamoto, N. Nishiyama, Y. Yokoyama, H. Kimura and A. Inoue, J. Alloys and Compounds, 434–435, 22 (2007).

[3] K. Fujita, Y. Morishita, N. Nishiyama, H. Kimura and A. Inoue, Mater. Trans. 46, 2856 (2005).

Contact to

Kazutaka Fujita (Department of Mechanical Engineering, Ube National College of Technology)

e-mail: [email protected]

Hisamichi Kimura (Advanced Research Center for Metallic Glasses)

e-mail: [email protected]

Fig.1 S-N curves in BMGs and representative crystalline alloys.

D.Broek, Elementary Engineering Fracture Mechanics, Sijtoff &Noordhoff, 1983, p.301.

0 20 40 60 80 100 120 140 500 1000 1500 2000 Fracture toughness, 　K IC or KQ /MPa ･m 1/2

Tensile strength, σB/MPa 0 20 40 60 80 100 120 140 500 1000 1500 2000 Al-alloy Ti-alloy High tensile steel Maraging steel -Cu-based Ti-based Zr-Cu-Al Zr-CuAl-Ni Pd-based KIC KIC KIC KQ KIC 12 49 23 44 62

Fig.3 Chip formation mechanism in the BMGs.

Fig.2 Comparison between KICs or KQ in BMGs and Crystalline

Structure and Strength at the Bonding Interface of a Titanium-

biomedical Polymer Composite for Artificial Organs

To develop metal-polymer composite with high mechanical strength and flexibility for artificial implants, commercially pure titanium combined with segmented polyurethane (SPU) through (3-trimethoxysilyl) propylmethacrylate (γ-MPS), and the interface was characterized. Shear bonding stress of Ti-SPU composite was dramatically increased with the increase of the γ-MPS layer thickness.

Polymers are widely used as biomaterials because of their high degree of flexibility, biocompatibility, and technologic properties, while the polymers show insufficient strength and long-term durability for some purposes because of their structures. On the other hand, metals have good mechanical properties, especially toughness, and long-term durability. However, the biocompatibility of metals is generally inferior to that of polymers and ceramics because no biofunction is added to the metals during the manufacturing process. If a polymer and a metal could be bonded and used as a composite material, a new material having good biocompatibility and high mechanical strength could be created [1]. The objective of this study was to investigate the unequivocal relationship between the shear bonding strength and the chemical structure at the bonding interface of a metal-polymer composite through a silane coupling agent (3-(trimethoxysilyl) propyl methacrylate (γ-MPS)). As the base materials for the composite, Ti and a segmentated polyurethane (SPU) were employed. The chemical structure of the Ti/γ-MPS/SPU interface is illustrated in Fig. 1.

According to glow discharge optical emission spectroscopy (GD-OES), the intensity of S in the γ-MPS layer increased with the increase of the concentrations of the γ-MPS solution and immersion times. In other words, the number of the γ-MPS molecular unit and the thickness of the γ-MPS layer increased with the concentration of the γ-MPS solution and the immersion time.

Shear bonding stress of Ti/γ-MPS/SPU interface increased with the increase of the concentration of the γ-MPS solution only in the case of 1-min immersion. On the other hand, the shear bonding stress of the Ti/γ-MPS/SPU interface formed from 1.0 and 2.0% γ-MPS solutions significantly increased with immersion times.

If the number of molecular units per Ti surface area is small, each molecular unit keeps a distance from its neighbors and consequently falls down to the Ti surface. Also, -Si-O-Si- bonding network among molecular units is not sufficiently formed. On the other hand, if the number of molecular units per Ti surface area is large, the molecular units will crowd at the interface and stand perpendicular to the Ti surface. As a result, the thickness of the γ-MPS layer will increase. In this case, the -Si-O-Si- bonding network is sufficiently formed. A thick γ-MPS layer would be attributable to the increase in the shear bonding strength because a thick γ-MPS has more molecular units containing S-H groups bonded to SPU.

The Ti-SPU composite was fractured leaving the SPU component elements on the fractured surface, determined by XPS. However, more residual SPU on the fractured surface of the Ti-SPU composite with the γ-MPS layer existed than that without a γ-MPS layer. The SPU elements remained on the fractured surface as a result of the presence of the MPS layer. The thicker the γ-MPS layer was, the larger the SPU area fraction on the fractured surface was (Fig. 2).

In summary, the thickness of the γ-MPS layer is controlled by the concentration of the γ-MPS solution and the immersion time. The shear bonding stress of the Ti/γ-MPS/SPU interface increased with the increase in the thickness of the γ-MPS layer. The Ti-SPU composite was fractured inside the SPU, and the shear bonding stress of the Ti/γ-MPS/SPU interface increased with the increase in the SPU area fraction. The γ-MPS is very useful to improve the bonding strength of the Ti-SPU composite. The factor governing the shear bonding strength between Ti and SPU is the thickness of the γ-MPS layer. This study should lead to enhancements in the creation of metal-polymer composites for artificial organs [2].

References

[1] Y. Tanaka, H. Doi, Y. Iwasaki, S. Hiromoto, T. Yoneyama, K. Asami, H. Imai, and T. Hanawa, Mater. Sci. Eng. C27, 206 (2007). [2] H. Sakamoto, H. Doi, E. Kobayashi, T. Yoneyama, Y. Suzuki, and T. Hanawa, J. Biomed. Mater. Res. 82A, 304 (2007).

Contact to

Takao Hanawa (Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University)

e-mail: [email protected]

Hisamichi Kimura (Advanced Research Center for Metallic Glasses) e-mail: [email protected]

Fig. 1 Metal-polymer composite through a silane coupling agent.

Fig. 2 Schematic model of the fractured region before and after the shear bonding test in the case of a thin γ-MPS layer (top) and a thick γ-MPS layer (bottom).

Process Innovation for Producing Metallic Glass Components

-Rapid-Rotation Centrifugal

Casting-An innovative “Rapid-Rotation Centrifugal Casting Process” for production of the amorphous bulk metallic glassy alloys (BMGs) was developed. The molten alloys were poured into copper mold with rotation speed of 3000rpm which can cause 80 times as much centrifugal pressurization force as gravity, and the samples of small gear, screw and biomedical artificial tooth-root model of Zr or Fe alloy systems were fabricated within only a few seconds at one process.

Bulk metallic glassy alloys(BMGs) have the strong interest in material science and technology because they have the unique characteristics such as very high strength, high corrosion resistance, high wear resistance as well as the flexible workability based on viscous super-elasticity phenomenon of appearing near the glassy transition temperature Tg of BMG. However, too much deformation process during this secondarily working for BMG will cause the nano-microscopic change of amorphous structure as well as the degradation of BMG’s properties, therefore, new innovative material processing technique for mass-producing BMG components is becoming a important subject for commercial production of BMGs. To this problem, recently, Professor Y.Furuya(Hirosaki University), President (Professor) A.Inoue and Associate Professor H.S.Kimura of IMR of Tohoku University succeeded in development of an innovative “Rapid-Rotation Centrifugal Casting Process” which enables the reduction in cost to the components and commercial production of the amorphous bulk metallic glassy alloys. The schematic figure of the developed method is shown in Fig.1 The molten raw BMG alloys by high frequency induction heating method were poured into copper mold with rotation speed of 3000rpm which can cause 80 times as much centrifugal pressurization force as gravity, and the samples of small gear, screw and biomedical artificial tooth-root model of Zr or Fe alloy systems could be produced within only a few seconds at one process. The fabricated BMG component sample is shown in Fig.2. The produced two samples of artificial tooth and ring of Zr55Al10Ni5Cu30b are almost

amorphous state as shown in XRD pattern in Fig.2.

According to synergistic effect of rapid solidification and high pressurization casting in this innovative metal glass component manufacturing process, there are technical advantages of expansion of element ingredient composition, uniformity of the quality, reduction of internal defects and prevention of degradation at the time of secondary fabrication process of BMGs, which seem easy to automation, a large amount of cost cut and mass production in the very near future.

Fig.2 Product Samples (Artificial Tooth (A) Model, Ring(B)) and the XRD pattern of Zr55Al10Ni5Cu30

References

[1] A. Inoue, Bulk Amorphous Alloys, Trans. Tech. Publisher, Zurich (1998).

[2] Y. Furuya. Mat. Res. Soc. Symp. Proc. Vol. 604, Materials Research Society, 109 (2000).

[3] K. Hashimoto, T. Kubota, T. Mashiko, T. Okazaki, Y. Furuya and A. Inoue, J. Japan Inst. Metals. 71-7, 553 (2007)

Contact to

Yasufumi Furuya (Department of Intelligent Machines and System Engineering, Hirosaki University)

e-mail: [email protected]

Hisamichi Kimura (Advanced Research Center for Metallic Glasses)

e-mail: [email protected]

Fig.1 Schematic Figure of the developed “Rapid-Rotation Centrifugal Casting for BMG Components” (Furuya,Kimura,Inoue)