Introduction

Surface roughness is an important parameter related to the properties of materials. Surface roughness affects the appearance of a material from a macroscopic perspective. For example, a material surface that is polished very flat will have a mirror surface. Furthermore, since surface roughness affects the adhesion of materials, surface roughness affects the mechanical properties and electrical properties. Microscopic surface roughness is also important such as in the semiconductor field, where atomic-level roughness has a strong influence on electrical properties. Because it is such an important property, ISO also stipulates a method for evaluating surface roughness. Conventionally, on the macro scale, surface roughness meters, laser microscopes, etc. have been used. On the other hand, the resolution is limited by the wavelength of light, making it difficult to evaluate surface roughness on a nanoscale. Traditionally, Scanning Probe Microscopes (SPM) and Atomic Force Microscopes (AFM) have been used to measure surface roughness at the nanoscale. These techniques visualize and measure the 3D shapes of tiny area on the surface. However, it is difficult to find a suitable field of view for measurement, and measuring large areas takes a lot of time. In contrast, Scanning Electron Microscope (SEM) offers versatility such as efficient field-of-view searching at low magnifications and elemental analysis capabilities. However, it has faced challenges in measuring 3D shapes. Currently, one of the methods for 3D reconstruction used in SEM is Photometric Stereo method (PS method) using a multi-segment backscattered electron detector. The PS method calculates the tilt angle of the sample surface from signal intensity of the detector and reconstructs the 3D shape of the sample surface, allowing for measurements to be taken in a short time. This method is not suitable for quantitative measurement because the error in height accuracy increases when the surface inclination angle is about 70° or more. On the other hand, relatively good 3D reconstruction is obtained when the surface inclination angle is between 0 and 70°. Therefore, it is considered that this method provides sufficient accuracy for measuring surface roughness even on samples with flat-like surfaces such as plating, where the tilt angles are small. Since the surface roughness of a plating surface varies depending on the thickness, it is possible to prepare samples with the same composition but different roughness. This allows for discussions to be made without errors caused by the sample composition. In this study, based on the above motivations, we used SEM to measure the surface roughness of plating.

Experiment

In the PS method, surface shapes are reconstructed using backscattered electron signals. Changes in SEM observation conditions may affect the height accuracy. Therefore, in this study, the observation conditions were kept constant. Furthermore, we measured the same field of view using a laser microscope and used those results for calibration. The samples consisted of electroless NiP plating on copper plates, and we measured the surface roughness at various film thicknesses.

Result

The results are shown in Fig. 1. Figure 1 (a) shows the height image measured with a laser microscope, and Figure 1 (b) shows the height map image measured with SEM. By comparing the two images, it was confirmed that the SEM height image could reproduce a shape relatively close to that obtained by the laser microscope. We also calculated the surface roughness from the height map image and obtained results that correlated highly with those from a laser microscope. Therefore, we believe that applying this method could provide surface roughness of plating measurement results equivalent to those obtained with a laser microscope.

1) Introduction: Hydrogen peroxide (H2O2) has attracted much attention as an environmentally friendly oxidant as well as a promising liquid fuel. Photocatalytic process is a potentially sustainable method for H2O2 production alternative to the current anthraquinone oxidation process. However, the activity is still limited, and new materials and approaches for photocatalytic H2O2 production are highly desired. Our group demonstrated that application of metal-organic framework (MOF) materials for photocatalytic H2O2 production via oxygen reduction for the first time^[1]. Thanks to the advantages of the unprecedented design flexibility of MOFs, we developed the MOF materials to improve the photocatalytic activity for H2O2 production by the linker functionalization, the missing-linker defects in MOF framework, and the utilization of a two-phase reaction system with hydrophobic MOFs^[1-9].

2) Ni/MIL-125-NH2: Co-catalyst deposition is an efficient approach to improve photoctalytic H2O2 production^[1,2]. Ti-based MIL-125-NH2 was synthesized using titanium isopropoxide and 2-aminoterephthalic acid and deposited with Ni nanoparticles (Ni/MiL-125-NH2). Catalysts were dispersed in an O2-saturated acetonitrile solution of benzylalcohol and irradiated with visible light (λ > 420 nm). The amount of H2O2 produced on Ni/MiL-125-NH2 was much higher than that of the pristine MIL-125-NH2. Ni nanoparticles improve the selectivity for the two-electron reduction of oxygen to H2O2 through the fast disproportionation of the superoxide radicals which are the intermediate of oxygen reduction.

3) UiO-66-NH2 with missing-linker: Defect engineering for MOFs is a promising process that can modulate their electronic structure, surface chemical properties and porosity. UiO-66-NH2 consisting Zr-oxo clusters and 2-amino-terephtalate linkers was prepared. The missing-linker terminated by acetate ligands in UiO-66-NH2 were introduced by adding acetic acid during the solvothermal synthesis. Addition of missing-linker in UiO-66-NH2 framework improve photocatalytic H2O2 production^[2,5]. The improvement of photocatalytic H2O2 production is attributed to not only promotion of linker-to-cluster charge transfer but also suppression of H2O2 decomposition.

4) Two-phase reaction system with hydrophobic MOF: One critical issue of the above single-phase reaction system needs an energy-consuming separation process following the H2O2 production to give separate solutions of H2O2 and the oxidation products. To solve these problems, our group developed a two-phase reaction system, which is composed of water and BA, for photocatalytic H2O2 production utilizing hydrophobic MOFs^[2,6-9]. As hydrophobic modified MOFs, the linker-alkylated hydrophobic MIL-125-R was obtained by modification of the linkers in MIL-125-NH2 with alkyl anhydride. And the cluster-alkylated hydrophobic OPA/MIL-125-NH2 was prepared by modification of the clusters in MIL-125-NH2 with octadecylphosphonic acid (OPA). These samples were dispersed in an O2-saturated two-phase system composed of water and BA, and irradiated with visible light (λ > 420 nm). This two-phase reaction system realized spontaneous separation of H2O2 formed to the aqueous phase and of the benzaldehyde formed to the BA phase. The spatial separation of H2O2 and photocatalytsts improve the yield of H2O2 by preventing the H2O2 decomposition. Furthermore, the limitation in MOF stability was overcome by utilizing a hydrophobic MOF in the two-phase reaction system because the MOF particles were separated from the low pH aqueous phase. The H2O2 production in the two-phase reaction system was dramatically enhanced by the hydrophobic modification with changing cluster alkylation OPA/MIL-125-NH2 from linker alkylation MIL-125-R^[6-9].

View PDF for the rest of the abstract.

Transparent conductive oxide (TCO) films are widely used as transparent electrodes in optoelectronic devices, such as flat panel displays, and solar cells. In recent years, with the development of devices that utilize light in the near-infrared (NIR) region, TCO films with transparency in the NIR wavelength range are in demand. High transparency in the NIR wavelengths can be achieved by reduction in free carrier absorption. Therefore, these applications require high- carrier mobility (µ) TCO films (because a high µ results in a high conductivity at a low carrier concentration) to thus provide high transparency with alleviated parasitic optical losses in the visible and NIR regions. Conventional Sn-doped In₂O₃ (ITO), which have been used as TCO materials optoelectronic devices, exhibit a µ of 40 cm²/Vs. By contrast, solid-phase crystallized H-doped In₂O₃ (spc-IO:H) exhibit an extremely high-µ (over 100 cm²/Vs). These properties result in a low optical loss over a wide optical range, which makes this material a promising candidate for many applications. spc-IO:H films are fabricated by a two-step process. In the first step, amorphous precursor films are deposited on a substrate without intentional heating via magnetron sputtering (MS). In the second step, the film is solid-phase crystallized via thermal or photo annealing, then exhibits high carrier mobility. Therefore, it is necessary to optimize the deposition conditions of the precursor, to achieve high quality spc-IO:H films. In this study, we investigated the effects of deposition conditions for precursor on the properties, such as texture evolution, and carrier transport of spc-IO:H films.

In₂O₃ precursor films were deposited on glass substrates via radio frequency (RF), and pulsed-direct current (Pulsed-DC)-MS without intentional heating. We employed In₂O₃ ceramic targets at deposition power levels of 20 to 100 W. The H₂O vapor pressure before deposition was set to 1×10^-4 Pa. The deposited precursor thin films were annealed at 200 ℃ for 30 min in vacuum.

Figure 1 shows plan-view scanning ion microscope (SIM) images of IO:H films under various power level of RF-MS and Pulsed-DC-MS with power of 20 W. The upper and lower images correspond to as-deposited, and post-annealed films, respectively. The inset value shows the Hall mobility (µ_H). Figure 1 clearly shows, as-deposited films contain small crystallites, and its density increased with increasing RF deposition power. Meanwhile, crystallites density of as-deposited IO:H film by Pulsed-DC-MS exhibit similar to those of RF-MS with power of 80 W. These difference in crystallite density is probably due to the difference in the energy magnitude of the deposition particles. An increase in the deposition power corresponds to deposition particles with greater kinetic energy, which favors the diffusion of ad-atoms, in which atoms rearrange into more stable structures, leading to increased generation of crystallites. Figure 1 also clearly shows that the IO:H films contained with a lower crystallite density were formed larger crystal grains after annealing.

For all the films, µ_H increased upon crystallization. However, the magnitude of µ_H was significantly affected by precursor deposition conditions. The highest µ_H of 125 cm²/Vs under consideration in this study was achieved at RF power of 20 W, 40 W and 60 W.

View PDF for the rest of the abstract.

Introduction

Oxide semiconductors are attracting significant interests for next-generation electronic devices, including flexible and thermochromic devices, owing to their advantageous properties, such as low-temperature processing, transparency in the visible light range, and biocompatibility. Flexible devices require low-temperature processes because they use polymers and other flexible materials as substrates.

The low-temperature polycrystalline silicon (LTPS) used in organic light-emitting diode driving devices has high mobility (>100 cm²/Vs), whereas amorphous silicon (a-Si) and amorphous oxide semiconductors, such as amorphous indium gallium zinc oxide (a-IGZO), exhibit lower mobilities of approximately 1 cm²/Vs and 10 cm²/Vs, respectively [1]. Recent advancements have led to the development of oxide thin-film transistors (TFTs) with high mobility of 60 cm²/Vs by employing a stacked structure comprising an ultra-flat amorphous metal gate and an IGZO channel [2], this represents a promising alternative to the LTPS with enhanced characteristics. Although numerous reports exist on oxide TFTs with multilayer structures [3], the underlying mechanisms by which these structures enhance the electrical properties remain poorly understood.

To understand these phenomena, we measured the resistance of two-terminal devices in real time during the deposition of oxide multilayer films and evaluated the resulting resistance changes. Additionally, we fabricated oxide multilayer structure TFTs to investigate the influence of multilayer films on their transfer characteristics.

Experimental method

ZnO was selected as the oxide channel film of TFTs, and Al₂O₃ was used as the upper multilayer of the channel film. These films were fabricated by pulsed laser ablation. The resistance changes caused by stacking these multilayers were investigated through the sequential growth of the upper multilayer film.The growth conditions involved the use of an Nd: YAG laser (fourth harmonic wavelength of 266 nm) to irradiate ZnO ceramic (99.99%) and Al₂O₃ ceramic (99.9%) at a frequency of 10 Hz and an output power of 200 mW. A 5 nm ZnO thin film was deposited onto a glass substrate at room temperature, followed by the formation of an Al electrode using a vacuum evaporation system to prepare the sample. The sample size was 5 mm × 15 mm with a 2 mm gap between the electrodes. The Al electrode of the prepared sample was connected to the vacuum flange terminal, and a digital multimeter was connected to the external terminal to acquire data at 0.5-second intervals using a personal computer. Resistance measurements of the two-terminal device were conducted before deposition and continued during the 16-minute deposition period until the film was exposed to the atmosphere post-deposition. A schematic diagram of the measurement in progress is shown in Fig. 1(a).

Results and discussion

The measurement results are shown in Fig. 1(b). The resistance of the ZnO thin film before Al₂O₃ deposition was measured at 4.65 kΩ. This value was set as the reference value R₀, and the reduction in resistance was normalized using R₀. A rapid decrease in resistance was observed immediately after the initiation of Al₂O₃ deposition. The resistance decreased by 50% within 5 min and by nearly 60% within 10 min of starting deposition. This indicated an increase in the carrier density owing to the formation of O₂ vacancies and other donor-type defects in the ZnO thin film, caused by the displacement of adsorbed molecules at the Al₂O₃/ZnO interface by high-energy Al₂O₃ particles.

View PDF for the rest of the abstract.

Introduction

Recently, oxide semiconductors have attracted considerable attention as next-generation semiconductor materials. In₂O₃, one of the oxide semiconductors, is used as a material for low temperature polycrystalline oxide (LTPO) because it is a single crystal with high mobility of 160 cm²/Vs [1] and can crystallize at low temperatures below 200 ℃ and has conductivity. LTPO is generally used as a switching thin-film transistor (TFT) in organic light-emitting diode (OLED) displays because of its low off-leakage current and is expected to be applied to TFTs to drive OLEDs by improving their mobility. However, because of the trade-off between mobility and operational stability, achieving both has been challenging. Therefore, we focused on InGaO (IGO), which is a mixture of In₂O₃ with high mobility, and Ga₂O₃ with excellent insulating properties. We fabricated TFTs using a solution method combining a carbon-free aqueous precursor solution and deep ultraviolet excimer light and confirmed the superiority of the excimer light-assisted process by evaluating various characteristics [2, 3]. In this report, we described the details of the excimer light-assisted process, the surface observation results obtained by atomic force microscopy (AFM) with and without excimer light irradiation after In₂O₃, Ga₂O₃, and IGO thin films were fabricated using aqueous precursor solutions, and the characterization of TFT characteristics fabricated using the excimer light-assisted process.

Experimental method

In₂O₃, Ga₂O₃, and In_0.5Ga_0.5O precursor solutions were prepared to form the thin films, and the In₂O₃ precursor solution was prepared by mixing indium nitrate trihydrate (In(NO₃)₃・3H₂O) and ultrapure water at 0.3 mol/L. The Ga₂O₃ precursor solution was prepared by mixing gallium nitrate octahydrate (Ga(NO₃)₃・8H₂O) and ultrapure water at 0.3 mol/L. The In_0.5Ga_0.5O precursor solution was prepared by mixing In(NO₃)₃・3H₂O, Ga(NO₃)₃・8H₂O, and ultrapure water at a composition ratio of 1:1. Next, the solutions were applied to the hydrophilized EAGLE XG glass substrates to form In₂O₃, Ga₂O₃, and In_0.5Ga_0.5O thin films using the spin-coating method, and the glass substrates were rotated at 2000 rpm for 30 s to form the thin films. Subsequently, the samples were prepared with and without excimer light irradiation for 90 min.

Results and discussion

Fig. 1 illustrates the principle of the excimer light-assisted process and the results of surface observations of the In₂O₃, Ga₂O₃, and In_0.5Ga_0.5O thin films obtained by AFM. An aqueous solution of indium nitrate is decomposed by excimer light irradiation. Free radicals (・OH) generated during decomposition modify indium ions, which are converted to In₂O₃ by heat treatment in air. This effect is expected to promote the formation of In₂O₃ from the aqueous indium nitrate solution (Fig. 1(a)). Comparing the root mean square (RMS) roughness of the samples without and with excimer light irradiation, the RMS roughness of the samples without excimer light irradiation ranged from 2 to 6 nm, while the RMS roughness of the samples with excimer light irradiation was less than 1 nm (Fig. 1(b)).

View PDF for the rest of the abstract.

We recently proposed a unified Si oxidation reaction model mediated by point defect generation [1-3]. Vacancies produced at the SiO₂/Si interface (V⁰) become chemically active sites (V⁺ and V⁻) after carrier trapping, and O₂ dissociation at V^+/− subsequently occurs, which induces the strain and the following V⁰ generation at the SiO₂/Si interface, thus forming a reaction loop. This reaction model demonstrated two reaction loops: a rapid loop that proceeds with a single step (loop A) and a slow loop with a double step (loop B). However, it is still unclear what causes the branching of the two loops. In this study, we investigated the mechanism behind the branching of these loops, focusing on chemisorbed O₂ at the SiO₂/Si interface.

XPS observations were performed with the surface reaction analysis apparatus (SUREAC2000) at BL23SU, SPring-8. Sb-doped n-type Si(001) with 2 Ωcm resistivity wafer was prepared. O 1s and Si 2p spectra were alternately obtained by a real-time XPS measurement for the Si sample during the sample was irradiated with a 0.06 eV O₂ supersonic molecular beam at room temperature (RT).

Fig. 1 (a) shows the time evolution of the intensity of O 1s spectra (I_O-1s). This uptake curve can be fitted with a Langmuir-type adsorption model, I_O-1s = I_sat[1-exp(-kt)], where I_sat is a saturation level. From the fitting, the surface (before point G) and interface (after point G) oxidation region, as shown in Fig.1(a), were determined. For each point of A−M in Fig. 1(a), we performed curve fitting as shown in Fig. 1(b, c). The O 1s spectra can be separated into five components of a~e. The components a, b, and c correspond to ins (Si-O-Si), tri (Interstitial O), and ad (Si-O), respectively. d and e are assigned to paul (chemisorbed O₂) state. Because the paul with ins O on the backbond has a comparatively long lifetime, the d and e components can be observed in the XPS spectra. The paul is still observed even at point M in the interface oxidation region. Since the surface is entirely covered with SiO₂ at the interface oxidation region, the paul can be assigned to the chemisorbed O₂ at the P_b1 center (P_b1-paul); a dangling bond with backbond O at the SiO₂/Si interface. Based on the P_b1-paul and our unified oxidation model, we constructed a branching model of the loops A and B (Fig. 1(d)). In our model, the oxidation process requires minority carrier (hole) trapping. Loop A proceeds via hole trapping, while spontaneous dissociation of paul without hole causes loop B. Here, the branching ratio of loop A can be nearly 0 at RT because of small k_mCT (reaction coefficient of minority carrier trapping). This can be confirmed by fitting the time evolution of SiO₂ thickness (X_O). Therefore, the oxidation at RT dominantly proceeds via loop B. Because loop B involves P_b1-paul in the process and the dissociation of P_b1-paul can be a rate-limiting step due to small k_mCT at RT, the growth rate of SiO₂ thickness (dX_O/dt) can be expressed as the equation in Fig.1 (e). a and N_Pb1-paul are a constant and the amount of P_b1-paul, respectively.

View PDF for the rest of the abstract.

The cryopump is a gas accumulation type vacuum pump that condenses or adsorbs gas molecules on a cryogenic panel placed inside a vacuum vessel to perform pumping. It is widely used in semiconductor manufacturing equipment, FPD and optical thin film deposition equipment, and research chamber applications, as it has the advantages of high pumping speed for water and hydrogen, and the ability to achieve oil-free and clean ultra-high vacuum. The cryopump has unique specification defined by its pumping principle and structure, such as gas storage capacity and tolerance values for heat load (throughput, crossover), which are not found in other vacuum pumps. However, although there are overseas standards (AVS, Pneurop) regarding performance measurement methods for cryopumps, there was no JIS standard, and there had been no international standardization.In May 2019, a proposal for ISO standard development on performance measurement methods for cryopumps was made by China at the ISO TC112 Vacuum Technology Committee Kyoto Conference. It was approved in October 2020 and officially launched as a project. In Japan, ULVAC CRYOGENIC INCORPORATED, CANON ANELVA CORPORATION, and Sumitomo Heavy Industries, Ltd. took the lead in developing the standard, and after discussions at ISO international conferences, it was officially issued as ISO 21360-6 in December 2023. The adopted measurement methods as ISO standards generally follow the existing overseas standards, but new term definitions (base pressure cryogenic vacuum pump) and measurement methods (setup using gate valves) have been added, and it is considered to be a content that satisfies the Japanese side as well. In this presentation, we will first explain the overview of cryopumps and introduce the flow of ISO standard development and the contents of the issued standard.

1. Introduction

A Ti-Zr-V non-evaporable getter (NEG) coating, which was developed at CERN, is a breakthrough vacuum technology because it makes the vacuum chamber wall a getter surface by a low activation temperature around 180-300 degree C [1, 2]. The chamber itself can function as a vacuum pump by applying a NEG coating to the inner wall of the chamber. However, sticking probability for limited gas species, mainly such as H₂ and CO, has been reported in the previous research. In this research the pumping speed measurement for the H₂, CO, O₂, CO₂, and N₂ was performed by the orifice method using the NEG-coated chamber of 100 mm diameter and 200 mm length. Furthermore, the pumping speed deterioration by repetitive air exposure and re-activation was evaluated for such gas species.

2. Experiment

The pumping speed was measured by repeating 15 times in one set "After opening to atmosphere → Vacuum evacuation → Baking (Activation) with 20 degree C for 8.5 hours → pumping speed measurement". Figure 1 shows the configuration of the measurement setup. P1 chamber (NEG coated) and P2 chamber were connected by an orifice of 1.75x10^-2m³/s(N₂), and gas was induced from the P1 side. Figure 2 shows the measurement results for various gases. It was found that while the pumping speed for H₂, CO, CO₂, N₂ other gases was degrading, O₂ was degrading slowly and was still above 1000 L/s at the 15 times cycle. In the conference, we report the details of the experiment and the measurement results.

Fig. 1 Schematic diagram of the apparatus used to measure pumping speed. TMP: turbomolecular pump; DSP: dry scroll pump; EXG: extractor gauge, RGA: residual gas analyzer; and CDG: capacitance diaphragm gauge.

Fig.2 Pumping speed measurement results for various gases

References

[1] C. Benvenuti, et al., VACUUM 50, 57 (2001).

[2] O. B. Malyshev, et al., Journal of Vacuum Science & Technology A 27, 321 (2009).

In QST Rokkasho Fusion Energy Institute, the accelerator driven fusion neutron source A-FNS project is ongoing for the fusion reactor material study. This accelerator system will irradiate high current deuteron beam to the free surface liquid lithium (Li) target (Fig. 1) and produce intense neutron flux by the nuclear reaction between deuteron and Li. Moreover, it will have to realize Ultra High Vacuum (<10^-5 Pa) to keep the availability of superconducting accelerator and High Vacuum (>10^-3 Pa) at the target system to prevent the Li boiling. However, contamination and vacuum degradation by liquid Li are concerned because the A-FNS accelerator consists of wide cross-sectional beam ducts for transporting a high current beam with a strong space charge effect. Therefore, the experimental validation of outgas rate, residual gas type and Li vapor behavior are required for individual condition of operational temperature and flow speed of the liquid Li.

For the vacuum characteristics study, we designed the dedicated experimental setup to demonstrate the interface between the beam transport system and the free surface liquid Li target system. Although this setup was 1/10 scale length of the A-FNS accelerator design, the cross-sectional size and the configuration of test ducts were designed same ratio as the actual system. In addition, the vacuum pumps layout and the pressure distribution were confirmed by the 3D Montecarlo vacuum simulation code Molflow+. In order to reduce the Li vapor contamination, Ar gas shield method was introduced by referring the gas jet curtain method in RIKEN [1]. Also, vacuum gauges, thickness monitors (QCM: Quartz Crystal Microbalances) and a residual gas analyzer (QMS: Quadrupole Mass Spectrometers) were installed to observe the vacuum characteristics. Finally, the assembled setup was attached to a liquid Li loop system (Fig. 1).

As a result of this experiment, although the reaching pressure is higher than the design value due to huge outgassing from Li target, a stable Li flow was demonstrated under large differential pumping from 10^-2 Pa to 10^-4 Pa. In addition, the base pressure and the partial pressure of the accelerator vacuum was kept regardless of the operational temperature (220-250 ℃) and the flow speed (13-18 m/s) of the liquid Li. However, the inflow of Li vapor was detected by the QMS and it was enhanced by rising the Li temperature of 220 ℃ to 250 ℃. On the other hand, the Li vapor flux was decreased by increasing the Li flow speed of 13 m/s to18 m/s. The relation between the vapor flux and the flow speed was verified by the analysis with a 2D thermal conductive equation for the Li flow surface. Regarding the Ar gas shield, the significant Li vapor suppression was not effective. After the experiment, the detector of the thickness monitor in the accelerator vacuum was observed by a Scanning Electron Microscope (SEM) to visualize the vapor condition inflowing from the liquid Li target system. The Li particle seemed droplets which composed of micron clusters. The large droplets did not affect the vacuum pressure but behave as a contaminate particle.

In summary, the designed differential pumping can be realized for the interface between the beam transport system and the free surface liquid Li target system. However, the contamination by the Li vapor will be depended on the operational condition of the Li target. In order to minimize the inflow of the Li vapor, we can conclude that the Li temperature and flow speed should be kept under 250 ℃ and over 18 m/s, respectively. The detail of this experiment and result will be presented.

References

[1] H. Imao et al., “Charge stripping of ²³⁸ U ion beam by helium gas stripper” Phys. Rev. ST Accel. Beams, 15, 123501 (2012).

Vacuum firing, which is a heat treatment at high temperature in a high vacuum furnace, is known as the method for the outgassing reduction of the vacuum materials, such as stainless steel, titanium, etc. The outgassing rate of the vacuum-fired stainless steel is known to be low after ordinal baking at 150-200℃. In this research, the effect of the vacuum firing (850℃ for 10 h) on the stainless steel and titanium is investigated from the vacuum and surface point of view. The build-up test of the vacuum chambers clearly showed the outgassing suppression by the vacuum firing. Especially, the hydrogen outgassing, which was the main component after baking, was much reduced. Thermal desorption spectroscopy showed that the vacuum firing reduced the desorption of H₂, H₂O, CO, and CO₂ with high desorption energy even after air exposure. Especially the effect on the H2 was very large. X-ray photoelectron spectroscopy (XPS) showed the increase of ferric oxide and the decrease of chrome oxide on the near surface of the vacuum-fired stainless steel. On the other hand, the XPS also showed that the chrome oxide was systematically increased by heating from 200℃ to 400℃. These results support the outgassing reduction mechanism by the vacuum firing that the hydrogen is reduced from the bulk due to the diffusion to the vacuum phase during the vacuum firing and the surface metal oxides are reformed as a diffusion barrier from the gas phase to the bulk. For the titanium, the surface titanium oxide film was once removed by the vacuum firing and the reformed oxide film is thinner. Thus the it is possible that the vacuum fired titanium chamber easily have the getter funcution by the baking.

As 2D metal-free materials, we have experimentally synthesized rhombohedral boron monosulfide (r-BS) and its exfoliated nanosheets of BS [1, 2]. r-BS is found to be as a p-type semiconductor [3, 4], and have a great electrocatalytic property for oxygen evolution reaction (OER) in alkaline solution by mixing with graphene [5, 6]. Furthremore recent studies show photo-catalytic activity of r-BS [7], and long-term stability of r-BS mixed with graphene for OER by using Ni-foam [8, 9]. In this talk, details of synthesis, characterization and properties of boron monosulfide nanosheets and r-BS will be introduced.

Acknowledgement: This work was done with Prof. M. Miyauchi, Prof. M. Otani, Dr. S. Hagiwara, Dr. L. Li, Mr. C. Jiang, Dr. F. Kuroda, Mr. N. Watanabe, Prof. E. Nishibori, Mr. H. Kusaka, Dr. M. Toyoda, Prof. T. Tokunaga, Prof. A. Yamamoto, Prof. T. Fujita, Dr. M. Miyakawa, Prof. S. Saito, Prof. K. Sugawara, Dr. T. Taniguchi, Dr. K. Watanabe, Dr. S. Shinde, Prof. T. Sakurai, Ms. M. Lima, Mr. K. Miyazaki, Mr. K. Matsushita, Dr. M. Masuda, Dr. S. Ito, and Prof. H. Hosono. and other collaborators.

Reference:

[1] H. Kusaka, et al., J. Mater. Chem. A 9, 24631 (2021).

[2] T. Kondo, Vacuum and Surface Science 65 (2022) 302 (in Japanese).

[3] N. Watanabe, et al., Molecules 28, 1896 (2023).

[4] K. Sugawara, et al., Nano Lett. 23, 1673 (2023).

[5] L. Li, et al., Chem. Eng. J. 471, 144489 (2023).

[6] S. Hagiwara, et al., ACS Appl. Mater. Interfaces 15 (2023) 50174.

[7] K. Miyazaki, et al., Scientific Reports 13 (2023) 19540.

[8] L. Li, et al., Sci. Technol. Adv. Mater. 24 (2023) 2277681.

[9] L. Li and T. Kondo, Vacuum and Surface Science 67 (2024) 333 (in Japanese).

Vanadium dioxide (VO₂) thin films exhibit a metal–insulator transition (MIT) with a change in resistance of three orders of magnitude nearby room temperature (~340 K). The resistive transition in VO₂ is accompanied by a structural change from a low-temperature semiconducting monoclinic phase to a high-temperature metallic tetragonal rutile phase. Still, there is ongoing debate about the MIT's initial process and mechanism, with in-depth research employing surface scientific methods like C-AFM and Raman spectroscopy.

In this study, we grew VO₂ thin films on TiO₂(110) substrates with step and terrace structures to investigate their crystallinity and MIT dynamics in real space, by using X-Ray Diffraction (XRD) and Infrared Scanning Near-field Optical Microscopy (IR-SNOM) with Atomic Force Microscopy(AFM) simultaneously. VO₂ thin films were grown by pulsed laser deposition under the substrate temperature, the laser frequency, and the partial oxygen pressure of 723 K, 2 Hz, and 0.95 Pa, respectively. The crystallinity was characterized using temperature-dependent XRD while heating and cooling the sample between 60 ℃ ~ 120 ℃.

As a result, a noticeable shift in the VO₂(220) diffraction peak was observed (Fig.1(a)), which is attributed to the structural transition from the insulating M1 phase at lower temperatures to the metallic R phase at higher temperatures. Regarding the AFM/IR-SNOM measurements, we succeeded to visualize the dynamics of the VO₂ metal-insulator transition in real space (Fig.1(b)). During the presentation, we will delve into the process of MIT on the surface, by focusing on series of temperature-dependent IR-SNOM images, coupled with pertinent morphological information obtained through tapping-mode AFM.

1. Introduction

Oiliness agents with polar groups form the adsorption film, reducing friction and wear on sliding surfaces [1]. Previous studies suggested that the friction reduction effect of oiliness agents involves not only the adsorption film but also solvation layer at the friction interface. Watanabe et al. incorporated a frictional system into an SFG spectroscopy system and observed changes in the adsorbed structure due to friction. They demonstrated that in a solution of n-dodecane with stearic acid, n-dodecane molecules were constrained on the adsorption film of stearic acid and aligned in the shear direction [2]. Therefore, to understand the friction reduction of oiliness agents, it is necessary to consider the role of the solvation layer. In this study, we developed a new atomic force microscopy (AFM) system combining FM-AFM and LFM, which measures repulsive force and lateral force simultaneously, to investigate the friction reduction mechanism of n-hexadecane with stearic acid. By combining FM-AFM and LFM, it is also possible to calculate the friction coefficient distribution along the direction of the depths. This report presents the results of simultaneous measurements of solid-liquid interfacial structure and frictional properties using the developed FM-AFM/LFM system.

2. Experimental detail

2.1. FM-AFM/LFM simultaneous measurement method

Fig. 1 shows a schematic of the simultaneous measurement system for FM-AFM and LFM. The equipment consisted of an SPM-8100FM (SHIMADZU, JP) and lock-in amplifier (LI5660, NF, JP). LFM measurements were performed simultaneously with regular FM-AFM measurements. The sample was oscillated in Y direction by XYZ scanner using AC signal from a function generator. The torsion signal of the cantilever, caused by the oscillating sliding of the sample, was input to the lock-in amplifier. Using the reference signal from the function generator, the lock-in amplifier extracted only the lateral force signal from the torsion signal, which contained noise. These measurements were performed with the cantilever constantly excited in Z direction for FM-AFM measurements.

2.2. Experimental conditions

In the FM-AFM/LFM simultaneous measurement method, it is possible to obtain repulsive force and lateral force against Z axis at 25°C using a silicon cantilever (PPP-NCLAuD, spring constant C: 20 N/m, resonance frequency f: 78-80 kHz). The repulsive force was calculated from resonance frequency changes of the cantilever using Sader's formula [3].

3. Result & Discussion

As a result, we were able to measure repulsive force (Fig. 2(a)(b)) and lateral force (Fig. 3(a)(b)) simultaneously. From these results, the friction coefficient against the Z axis (Fig. 4) was obtained using a single test method. The friction coefficient and phase difference between the sample and cantilever (Fig. 5(a)(b)), provide new insights into the low-friction area induced by the solvation structure in n-hexadecane with stearic acid. It was confirmed that the repulsive force, friction force, and phase difference decreased around Z = 4.6. This suggests that in a solution of n-hexadecane with stearic acid, n-hexadecane molecules were constrained on the adsorption film of stearic acid and behaved like a solid, thereby protecting the mica surface and reducing friction.

4. Conclusion

[1] The FM-AFM/LFM simultaneous measurement method enables the acquisition of both repulsive force and lateral force at the same time.

[2] It is possible to investigate the relationship between friction coefficient and solid-liquid interface structure, estimated from repulsive force measurements.

[3] n-hexadecane molecules were constrained on the adsorption film of stearic acid and behaved like a solid, thereby protecting the mica surface and reducing friction.

View PDF for the rest of the abstract.

Methane is a promising source molecule for the production of value-added chemicals. However, their inertness, which originates from the most stable C-H bond among alkanes, hinders their utilization because high temperatures are generally required for the oxidation/dehydrogenation of methane [1]. Recently, it was reported that methane is readily activated on the (110) surface of IrO₂, a platinum group metal oxide (PGMO), even at temperatures below room temperature [2]. This high activity is attributed to the strong bonding interactions between the IrO₂ surface and CH₃ fragments formed after C-H bond cleavage [3]. Such PGMO surface-CH₃ interactions correlate with the activation barrier, and PtO₂ has been identified as more favorable for the activation of methane [3]. Our group has focused on the inherently high surface sensitivity of nanosheet structures and has successfully observed methane oxidation on IrO₂ nanosheets at room temperature [4]. In this study, a series of PGMO nanosheets (IrO₂, RuO₂, and PtO₂) was systematically studied in terms of their methane activation ability.

The nanosheet films were fabricated on SiO₂/Si substrates by the layer-by-layer deposition or the transfer of vacuum-filtrated films. The change in the electrical resistance of the fabricated films upon methane exposure was measured in situ, and two factors affecting the electrical resistance were spectroscopically investigated: the bonding state of the nanosheets and deposition of reaction products. The change ratio of the electrical resistance was found to be in the order RuO₂ < IrO₂ < PtO₂ (Figure 1). In the case of PtO₂, monolayered films were found to be unnecessary for observing the resistance decrease caused by the reduction of PtO₂ to Pt during methane oxidation because PtO₂ is an insulator, unlike metallic IrO₂ and RuO₂. The consumption of lattice oxygen for methane oxidation was confirmed by near ambient pressure X-ray photoelectron spectroscopy (AP-XPS). The reduction of PGMO did not occur in RuO₂ without a temperature increase, whereas partial and most reductions were observed even at room temperature in IrO₂ and PtO₂, respectively. The presence of amorphous carbon deposited on nanosheet surfaces as a product of the methane oxidation reaction was confirmed by Raman scattering spectroscopy. The peak enhancement ratios before and after methane exposure were in the order of RuO₂ < IrO₂ < PtO₂, which is consistent with the electrical and AP-XPS results shown above. In conclusion, the high oxidation activity of PtO₂ for methane was confirmed even in the nanosheet form. The material dependence of the methane activation ability of PGMO nanosheets found in this study supports the theoretically determined mechanism based on the PGMO-CH₃ interaction, which may provide a guideline for the development of catalytic materials with high methane activation abilities.

[1] J. L. G. Fierro, Catal. Lett. 22, 67 (1993).

[2] Z. Liang et al., Science. 356, 299 (2017).

[3] Y. Tsuji et al., J. Phys. Chem. C. 122, 15359 (2018).

[4] Y. Ishihara et al., Adv. Mater. Interfaces. 10, 2300258 (2023).

Carbon neutrality, aiming for zero greenhouse gas emissions and reducing fossil fuel dependence, is a global trend. Renewable energy is advancing rapidly, but stable supply is challenging. Hydrogen energy offers a solution by conserving, storing, and transporting energy. MgH₂ is promising hydrogen storage materials due to Mg's abundance, high hydrogen density (up to 7.6 wt%), and thermodynamic stability (ΔH = -74.5 kJ/mol), allowing safe transport. Challenges include slow hydrogen desorption and absorption due to strong Mg-H bonds and reduced hydrogen capacity with repeated use. Additionally, during the utilization of hydrogen, high temperatures are needed for hydrogen desorption. Supporting finely processed MgH₂ on two-dimensional materials with catalysts, for example Ni, improves repeatability and sorption kinetics of H₂ [1,2].

Hydrogen boride (HB) nanosheet is two-dimensional material comprising a negatively charged hexagonal boron network and positively charged hydrogen atoms with a stoichiometric ratio of 1:1 [3]. In a previous study, it was reported that HB nanosheets reduce metals with redox potentials higher than Ni to form metal nanocomposites (Ni/HB) [4]. We then reported the spontaneous formation of highly dispersed Ni nanoclusters on HB nanosheets [5]. Based on the results of transmission electron microscopy, X-ray photoelectron spectroscopy, and X-ray diffraction, size of Ni nanoclusters on the HB nanosheet was found as small as 1-3 nm diameter [5].

In this study, MgH₂-Ni/HB composites were synthesized by ball milling of the mixture of MgH₂ with Ni/HB. The ratio of Ni to MgH₂ was controlled by varying the mixing ratio of Ni_0.05/HB (synthesized at a ratio of Ni:HB = 0.05:1) and MgH₂. DSC (differential scanning calorimetry) was performed under several hydrogen pressures with different heating rates. The DSC results provide information on the reaction heat during hydrogen desorption and absorption. Hydrogen desorption was observed by increasing the temperature up to 400 °C, followed by hydrogen absorption by decreasing the temperature down to 100 °C. Fig. 1 shows the DSC results for the first and the fourth cycles, which show that the two profiles were equivalent. A detailed discussion will be given in the presentation.

[1] Dan, L. et al. ACS Appl. Energy Mater. 5, 4976 (2022).

[2] Xu, N. et al. Int J Miner Metall Mater 30, 54 (2023).

[3] Nishino, H. et al. J. Am. Chem. Soc. 139, 13761 (2017).

[4] Ito, S. et al. Chem. Lett. 49, 789 (2020).

[5] Noguchi, N. et al. Molecules 27, 8261 (2022).

Majorana fermion is an exotic particle that is identical to its own antiparticle. Due to its peculiar nature based on the “non-Aberian” statistics, the Majorana fermion is expected to be used for fault-tolerant topological quantum computing. The vortex core of the topological superconductors is one of the ideal platforms of the Majorana quasiparticle that is predicted to appear as a zero-energy vortex bound state (ZVBS). Although several experimental efforts have been conducted to detect the ZVBS in several topological superconductor candidate materials [1-3], the existence of the Majorana quasiparticle is still controversial. In this talk, we will present the challenge of detecting the Majorana quasiparticle in the vortex core of topological superconductor Fe(Se,Te), using the ultra-low temperature scanning tunneling microscope [4]. We found that a certain number of vortices possess the ZVBS within 20 &mu eV energy resolution, which suggests its Majorana origin [Fig. 1(a)]. We also found that some vortices do not host the ZVBS [Fig. 1(b)] and that the fraction of vortices with the ZVBS decreases with an increasing applied magnetic field [Fig. 1(c)]. Based on this observation, we argued that the Majorana-Majorana interaction in a disordered vortex lattice might split the zero-energy Majorana-bound state into finite energies [6]. Even though these results are consistent with the picture that the observed ZVBS is attributed to the Majorana zero mode, the detection of another signature peculiar to the Majorana quasiparticle is indispensable for more consolidated experimental evidence.

References

[1] J.-P. Xu et al, Phys. Rev. Lett. 114, 017001 (2015)

[2] Q. Liu et al., Phys. Rev. X 8, 041056 (2018)

[3] D. Wang et al., Science 362, 333 (2018)

[4] T. Machida et al., Rev. Sci. Instrum. 89, 093707 (2018)

[5] T. Machida et al., Nat. Mater. 18, 811 (2019)

[6] C.-K. Chiu et al., Sci. Adv. 6, eaay0443 (2020)

Graphene-based superconductors are promising for various applications due to their optical transparency, mechanical strength, and flexibility, with twisted bilayer graphene being a prominent example [1]. This system exhibits unconventional superconductivity characterized by strong correlation effects and a van Hove singularity (VHS) in the electronic structure.

An alternative approach to achieving graphene-based superconductors is the synthesis of interlayer compounds, similar to graphite intercalation compounds. These compounds are typically understood to exhibit conventional superconductivity via electron-phonon interactions, but the presence of VHS in the electronic structure remains a topic of interest. VHS has often been observed using angle-resolved photoemission spectroscopy (ARPES) following the intercalation of various metals into the interface of epitaxial graphene and its substrate. Despite the potential for VHS-driven unconventional superconductivity in intercalated systems, experimental confirmation has been limited by the absence of integrated vacuum systems combining sample growth, ARPES, and cryogenic measurements.

In this study, we investigated the correlation between VHS and superconductivity in calcium-intercalated bilayer graphene [2-4] using an all-in-one multi-probe system that includes molecular beam epitaxy, photoelectron spectroscopy, and electrical transport measurements at cryogenic temperatures. Our findings indicate that in the dense calcium phase, an epitaxial layer of calcium forms at the graphene-SiC interface, enhancing the superconducting transition temperature (T_c). Both theoretical and experimental analyses revealed a metallic state at the interface and its hybridization with one of the Dirac cones. This leads to a distinctive VHS state that increases the density of states near the Fermi level. Two types of attractive interactions are proposed to contribute to the enhanced T_c through VHS: increased electron-phonon coupling via low-energy phonon modes and direct Coulomb interactions between electrons and holes.

Furthermore, in lithium-intercalated bilayer graphene, we demonstrated that the VHS of the Dirac band is also situated at the Fermi level. Interestingly, the Fermi level slightly shifts from the VHS as the number of graphene layers increases. This phenomenon suggests potential for superconductivity and exploring various electronic ground states associated with the VHS.

References:[1] Y. Cao et al., Nature 556, 43 (2018).[2] S. Ichinokura et al., ACS Nano 10, 2761 (2016).[3] H. Toyama et al., ACS Nano 16, 3582 (2022).[4] S. Ichinokura et al., ACS Nano, 18, 13738 (2024).[5] S. Ichinokura, et al., Physical Review B 105, 235307 (2022).

Open-shell nature of molecules are origins of their unique characters, and the characters result in the molecular functions. In the materials, the molecules are interacted with so many solids, and the functions will be varied by the solid surfaces. The varied functions are applied to catalysts, fuel cells, magnets, batteries, and so on (Fig. 1). However, the theoretical investigations on the interfaces between solids and open-shell molecules are difficult, and we have not fully understood the interactions. The deepening our understanding of the solid-molecule interactions will therefore unveil a roadmap for reaching brand-new materials.

The reason why the theoretical investigation of solid/molecule interfaces is difficult is simple: solid/molecule interfaces are also physics/chemistry interfaces. For molecules, we investigate them using chemistry such as molecular orbital calculations, and this approach show high accuracy for isolated systems. However, the computational cost is too high to apply it to periodic systems. On the other hand, using physics, we can investigate the solid systems. Band calculation by density functional theory with plane-wave basis (DFT/plane-wave) is a high-throughput approach for periodic systems, but it is difficult to apply it to open-shell molecules due to the lack of correction and analysis scheme. Therefore, theoretical development was necessary to handle the systems with solid/open-shell molecule interfaces.

My challenge is that the band calculation results are analysed and corrected using techniques of quantum chemistry. Namely, when investigating the solid/molecule interfaces, the results by physics are discussed within knowledge of chemistry. To achieve it, I developed spin projection schemes, which are methods for correction and analysis of open-shell molecules in quantum chemistry, for DFT/plane-wave method [1-20]. The developed scheme has opened diradical chemistry and physics on solid-molecule interfaces. I will present the essence of theoretical development, some of applications, and remained issues to achieve ab initio data-driven science for complexes of solid and open-shell molecules.

[1] K. Tada, H. Koga, M. Okumura, S. Tanaka, Chem. Phys. Lett. 701, 103 (2018)

[2] K. Tada, H. Koga, M. Okumura et al., Mol. Phys. 17, 2251 (2019)

[3] K. Tada, T. Maruyama, H. Koga, M. Okumura, S. Tanaka, Molecules 24, 505 (2019)

[4] K. Tada, M. Okumura et al., Appl. Surf. Sci. 465, 1003 (2019)

[5] K. Tada, S. Tanaka, T. Kawakami, Y. Kitagawa, M. Okumura, K. Yamaguchi, Appl. Phys. Express 12, 115506 (2019)

[6] K. Tada, S. Tanaka et al., Chem. Lett. 49, 137 (2020)

[7] K. Tada, T. Kawakami, S. Tanaka, M. Okumura, K. Yamaguchi, Adv. Theory Simul. 3, 2000050 (2020)

[8] K. Tada, S. Tanaka et al., Mol. Phys. 119, e1791989 (2021)

[9] K. Tada, Y. Kitagawa, T. Kawakami, M. Okumura et al., Phys. Chem. Chem. Phys. 23, 25024 (2021)

[10] K. Tada, M. Okumura, S. Tanaka et al., Chem. Phys. Lett. 765, 138291 (2021).

[11] K. Tada, Y. Kitagawa, T. Kawakami, M. Okumura, S. Tanaka, Chem. Lett. 50, 392 (2021)

[12] K. Tada, M. Mori, S. Tanaka, Chem. Lett. 50, 1057 (2021)

[13] K. Tada, H. Sakurai, M. Kitta, K. Yazawa, S. Tanaka, J. Solid State Chem. 304, 122593 (2021)

[14] K. Tada, M. Kitta, S. Tanaka, Chem. Lett. 51, 157 (2022)

[15] K. Tada et al., e-J. Surf. Sci. Nanotech. 20, 59 (2022)

[16] K. Tada, T. Masese, G.M. Kanyolo, Comput. Mater. Sci. 207, 111322 (2022)

View PDF for the rest of the abstract.

[Introduction]

Ionic liquids (IL) have excellent characteristics such as ionic conductivity, thermal and chemical stability, and nonvolatility. It is known ionic liquids as chemical reaction field can be used. The metal nanoparticles can be synthesized by sputtering metal atoms into ionic liquids [1]. Moreover, the transition metal nanoparticles as catalyst progress to form nano carbon materials by chemical vapor deposition. The shape of this nanoparticles is possible to influence the properties of nanocarbon materials [2]. The understanding of the process of nanoparticle formation by transition metal atoms and the control technique of its shape are important. On previous work, we have observed that the behavior of the ionic liquid molecules and transition metal atoms may be different with each surface structure (Si-rich or C-rich) on transition metal deposited IL/SiC surface. So, we focused on Si-rich surfaces, the surface structure was observed to clarify the behavior of transition metal atoms on the IL covered Si surface after annealing and the formation process of nanoparticles. In this study, the nanoparticle formation was carried out in a vacuum environment, but in order to simplify the process, formation of nanoparticles was attempted in the air.

[Experimetal]

The sample was Si(100) substrates. The ionic liquid was 1-ethyl-3-methylimidazolium bis(triflumethylsulfonyl)imide ([EMIm]Tf₂N). At first, ionic liquid dropped on the substrate in the air. Next, the platinum (Pt) as transition metal material was deposited on the surface by a magnetron sputtering system. The samples were annealed in the air at below 350 ^oC for 10 minutes. The surface structures were observed by scanning electron microscope (SEM), Energy Dispersive X-ray Spectroscopy (EDS) and X-ray diffraction (XRD). The annealing temperature was measured by pyrometer.

[Result]

The samples with varying Pt film thickness from 10 to 40 nm were prepared. Figure 1 shows an SEM image of the Pt (20 nm) deposited [EMIm]Tf₂N/Si(100) surface with annealing at 200 ^oC for 10 minutes in the air. The many nanoparticles were observed on the surface. SEM and EDS results of this surface suggested that Pt nanoparticles covered with ionic liquid was dispersed on the surface. XRD observation of the sample further indicated the possibility of Pt silicide formation. In this presentation, we will report the observation results of the surface structure when the amount of Pt deposited is varied up to 40 nm and the substrate annealing temperature is also varied.

[References]

[1] T. Torimoto, et. al., Appl. Phys. Lett., 89, 243117 (2006).

[2] T. Ikari et al., Surf. Rev. Lett., 16, 761 (2009).

Surface chemistry of carbon dioxide (CO₂) is one of the topics that have received much attention because of its impact on catalytic CO₂ reduction, including methanol synthesis, reverse water gas shift, Sabatier reaction, and so on [1]. Copper is an essential element for the methanol synthesis catalyst, and thereby, the reaction mechanism of methanol synthesis from CO₂ on the Cu-based catalyst surface has been intensively investigated [2]. One of the plausible mechanisms is the pathway where formate species play a critical role. While formate is recognized to be formed via the reaction of CO₂ with an adsorbed hydrogen atom [2], some studies have proposed the presence of carbonate (CO₃) species as a precursor for formate [3]. The recent in-situ spectroscopic studies on the well-defined Cu model catalyst have supported the CO₃-mediate pathway [4]. Despite such importance of CO₃, the works focusing on the detailed properties of CO₃ on the well-defined Cu surface have been little reported. In this study, we investigated the adsorption structure of CO₃ on the Cu(111) surface using scanning tunneling microscopy (STM) and in-situ infrared reflection absorption spectroscopy (IRAS).

The clean Cu(111) surface was prepared by several cycles of Ar⁺ sputtering and annealing at 670 K. In the present STM experiment, Cu(111) was exposed to 0.1 Pa of CO₂ at 300 K in a load-lock chamber. The CO₂-exposed sample was transferred to the STM chamber, and the STM measurements were carried out at 80 K. In the in-situ IRAS measurement, the vacuum chamber was separated from any vacuum pump, and CO₂ was introduced through a pulse valve. The IRAS spectra were continuously measured under CO₂ at 300 K. The pressure of CO₂ was kept at 0.01 Pa during the FT-IR measurement.

In the STM images after the CO₂ exposure for 1 min, the depletions having 3-fold rotational symmetry were observed. When the exposed time was increased to 4 min, the size of the depletion did not change, and the number of them increased. This indicates that the feature originates from the species formed by the surface reaction of CO₂. The most probable candidate is CO₃ having the molecular plane parallel to the surface. In in-situ IRAS spectra, the vibrational peak at 1288-1292 cm⁻¹ was developed with the elapsed time. The calculational results indicate that this peak can be assigned to the asymmetric mode of CO₃ with the molecular plane parallel to the surface. The absence of the peak at the 1500-1800 cm⁻¹ region, which is the feature of carbonyl group, also supported the parallel orientation of CO₃. The observation of asymmetric mode indicates that CO₃ adsorbed on Cu(111) belongs to the C_s point group rather than C_3v. The parallel orientation and the C_s symmetry strongly suggest that CO₃ is adsorbed at the bridge site of Cu(111) judging from the surface normal dipole selection rule.

The present STM and in-situ IRAS results are consistent with each other. Based on these results, we concluded that CO₂ is chemisorbed as carbonate Cu(111) at the near-ambient condition at 300 K. It is adsorbed at the bridge site with the molecular plane parallel to the surface. The contribution of carbonate species to methanol synthesis should be further investigated.

Reference

[1] U. Burghaus, Prog. Surf. Sci. 89, 161 (2014).

[2] G. Cui et al., Catalysts 14, 232 (2024).

View PDF for the rest of the abstract.

Graphene produced by chemical vapor deposition (CVD) is now widely used. It has a polycrystalline structure, in which carbon atoms form 5- and 7-membered rings at the domain boundaries [1]. The relationship between such domain boundaries and transport properties is still poorly understood. Doping effects and conduction properties induced by thin films of organic molecules are of great importance in the device application.

In the present work, we focused on oligothiophene (6T) and phtyalocyanine (Pc) molecules because of their excellent electronic properties and other functionalities. It has been reported that thin films of 6T on CVD-graphene in field effect transistors (FETs) slightly improve their performances [2]. However, detailed knowledge of how the molecular adsorption structure and electronic states near the interface affect the performance has not been clarified. Here, we investigated the interaction between graphene and 6T or Pc thin films by atomic force microscopy (AFM), micro-Raman spectroscopy, photoluminescence, X-ray photoelectron spectroscopy (XPS), and synchrotron radiation photoemission spectroscopy (KEK BL13B, SPring-8 BL23SU).We also fabricated a FET structure and measured transport properties.

Fig. 1. XPS spectra of 6T molecularly adsorbed CVD graphene. Fig. 1 shows the XPS spectra, which clearly show the S2s core level of the 6T molecule. Thickness dependence of the peak intensity is apparent. On the other hand, spectral decomposition of the C1s core level between the graphene substrate and the molecule is rather difficult. Hence, we applied photoelectron spectroscopy with synchrotron radiation to obtain the high-resolution core levels as well as the valence states. Annealing the 6T/Pc films has led significant structural changes, corresponding to the slight changes in the electronic states. We further analyzed the transport properties of a monolayer graphene layer using the FET structure. For example, we found the change in the On/Off ratio by the adsorption of the 6T thin films. Here, we also investigated Pc thin films grown on graphene, and compare the interfacial structures, electronic states and transport properties with those of 6T thin films.

[1] S. M. Fus et al., Prog. Surf. Sci. 92 (2017) 176.

[2] T.-J. Ha et al., Appl. Phys. Lett. 101 (2012) 033309.

Bismuthene is a two-dimensional material with a honeycomb structure of Bi. Because Bi is heavy element(Z=83) with a strong spin-orbit coupling (SOC), bismuthene is considered to be a two-dimensional topological insulator possessing a huge nontrivial gap [1], [2], [3].

Recently, a honeycomb structure of bismuthene grown on Ag(111) has been reported [1], where Bi was deposited at low temperatures. The atomic structure and topological edge states of the synthesized ultraflat bismuthene are observed at the atomic scale using STM/STS. This experiment reported that the honeycomb structure is completely flat with a lattice constant of around 5.7 Å [1]. The electronic structure of this flat bismuthene with honeycomb structure on Ag(111) was elucidated by angle-resolved photoemission spectroscopy using synchrotron radiation [4]. However, the structure determination of the honeycomb bismuthene has not been performed yet.

We examined the atomic structure of the honeycomb bismuthene using dynamical low-energy electron diffraction (LEED) analysis. Intensity-voltage I(V) curves have been measured on (2×2)-Bi grown on Ag(111) at about 180 K. We have found that the flat honeycomb structure of Bi shown in Fig.1 reproduces the experimental IV curves well, and that two Bi in the unit cell lay on the same plane within an experimental error. The layer distance between Bi and Ag is 2.43 Å, shorter than that for Bi/SiC (2.75 Å) [3]. The short layer distance in Bi/Ag would account for the strong interference between Bi p-states and Ag substrate.

[1] S. Sun et al., ACS Nano, 16, 1436, (2022).

[2] M. Zhou et al., Proc. Natl. Acad. Sci. U.S.A., 111, 14378, (2014).

[3] F. Reis et al., Science, 357, 287, (2017).

[4] K. Takahashi et al., Phys Rev B, 108, L161407, (2023).

Recent research has focused on ferroelectric (FE) thin films due to their scalability and compatibility with modern field-effect transistors, as well as their potential to enhance energy-efficient electronic devices [1]. Hafnium zirconium oxide thin film (Hf_xZr_1-xO₂, HZO) is a promising FE material for silicon-based applications, particularly because it exhibits good ferroelectricity on Si [2]. However, the performance of HZO in devices is influenced by the quality of the interface between the thin film and silicon, which can be affected by impurities, defects, or vacancies introduced during fabrication.

In this study, we use terahertz emission spectroscopy (TES) to non-destructively assess the properties of HZO/Si interface [3]. TES measures the electric potential change at the interface. Our setup involves exciting photocarriers in a p-type silicon sample covered with a 30~50 nm thick HZO layer using a femtosecond laser, as shown Fig. 1(a). The photocarriers are accelerated by the electric field due to band bending at the interface, generating a transient current that produces a THz wave. This THz wave is then detected in the time domain.

Our results, shown in Figure 1(b), indicate that the THz emission amplitude is highest for the sample with an Hf content of x = 0.05, followed by the sample with x = 0.25. Samples with higher Hf contents of x = 0.5 and 0.75 exhibit lower THz amplitudes, similar to the emission from p-Si without HZO. These findings suggest that an Hf content of x < 0.5 leads to stronger silicon conduction band bending and higher THz emission. This stronger band bending may be due to interface states caused by vacancies or impurities. We will also present the waveforms of THz emissions from ITO/HZO/Si heterojunctions under the various bias voltages.

This work was supported by JST, CREST Grant Number JPMJCR22O2, Japan.

References

[1] S. S. Cheema et al., Nature 604, 65–71 (2022).

[2] J. Muller et al., Nano Lett., 12, 4318-4323 (2012)

[3] D. Yang, A. Mannan, F. Murakami, and M. Tonouchi, Light. Sci. Appl. 11, 334(2022).

WS2 (Tungsten disulfide) is one of the transition metals dichalcogenides (TMDs) and possesses numerous unique mechanical and optoelectronic properties [1–3], such as a direct bandgap (in monolayer form), high mobility, 2 eV bandgap emission, and has been studied for a wide range of applications. The fabrication process of two-dimensional materials mainly utilizes the Chemical Vapor Deposition (CVD) method [4,5]. This method is widely used in the industry to produce various semiconductor and metal thin films. The CVD process of TMDs involves the evaporation of metal oxides, typically using argon as a carrier gas to transport the precursors for deposition onto the substrate. The transition metal reacts with chalcogen elements on the substrate, forming TMDs[6]. In the CVD process, since precursors are crucial for the reaction, it is essential to strictly control parameters such as the gas concentration and distribution of the precursors during the deposition process. In this regard, computational fluid dynamics (CFD) can be introduced to simulate the reaction conditions inside the chamber during the process.

In this study, substrates at three different positions typically show that the first and third substrates have better film deposition conditions. Still, the film deposition on the substrate surface is not uniform. WS2 usually deposits in a triangular shape, growing in a unidirectional or bidirectional manner. Within the same process, different positions yield different results. It is learned through the simulation results when WO3 powder sublimates, more WO3 diffuses onto the first substrate. This diffusion trend matches the actual film deposition direction. Observations from the CFD model indicate that the third substrate is located at the end of the heating tube, where the temperature drops sharply. However, the overall gas flow density is higher at this position, resulting in varying growth conditions.

Reference

Sun, Z., Martinez, A., and Wang, F., 2016, “Optical Modulators with 2D Layered Materials,” Nat. Photonics, 10(4), pp. 227–238.[2] Balandin, A. A., 2011, “Thermal Properties of Graphene and Nanostructured Carbon Materials,” Nat. Mater., 10(8), pp. 569–581.[3] Lv, R., Chen, G., Li, Q., McCreary, A., Botello-Méndez, A., Morozov, S. V., Liang, L., Declerck, X., Perea-López, N., Cullen, D. A., Feng, S., Elías, A. L., Cruz-Silva, R., Fujisawa, K., Endo, M., Kang, F., Charlier, J.-C., Meunier, V., Pan, M., Harutyunyan, A. R., Novoselov, K. S., and Terrones, M., 2015, “Ultrasensitive Gas Detection of Large-Area Boron-Doped Graphene,” Proc. Natl. Acad. Sci., 112(47), pp. 14527–14532.[4] Xu, Z., Lv, Y., Li, J., Huang, F., Nie, P., Zhang, S., Zhao, S., Zhao, S., and Wei, G., 2019, “CVD Controlled Growth of Large-Scale WS ₂ Monolayers,” RSC Adv., 9(51), pp. 29628–29635.[5] Yan, J., Lian, S., Cao, Z., Du, Y., Wu, P., Sun, H., and An, Y., 2023, “CVD Controlled Preparation and Growth Mechanism of 2H-WS2 Nanosheets,” Vacuum, 207, p. 111564.[6] Stand, N., Mendoza, C. D., and Freire, F. L., 2022, “Synthesis of WS2 by Chemical Vapor Deposition: Role of the Alumina Crucible,” Crystals, 12(6), p. 835.

In the SuperKEKB accelerator, sputter ion pumps (SIPs) and cold cathode gauges (CCGs) are employed as vacuum pumps and gauges, respectively. These components are extensively used in large-scale vacuum systems due to their high reliability. Typically, ion pumps receive 5-7 kV to the Penning cells inside the vessel, while CCGs receive 1-3 kV to an inverted magnetron-type cell, through high-voltage feedthroughs. Under suboptimal conditions, such as the high humidity (low dew point) sometimes found in environments like underground tunnels, these high-voltage feedthroughs are prone to abnormal discharge due to creeping discharge and corrosion at the insulator brazing sections, potentially leading to insulation breakdown. To address these issues, we have conducted accelerated tests in a laboratory setting, wherein commercially available and commonly used high-voltage feedthroughs (connectors) in SIPs and CCGs are subjected to prolonged high-voltage application in high-humidity environments to investigate their high-voltage characteristics and durability under such conditions [1].

The experiment commenced in November 2023. Symbols, applied voltages, brief descriptions of the tested high-voltage feedthroughs, and their appearances are tabulated in Fig. 1 (a) and (b), respectively. The connection cables are RG-58A/U (Type-SHV-10kV), RG-59/U (Type-SHV), and RG-59B/U (others). The feedthroughs have been exposed to environments with a humidity higher than 80% and a temperature of approximately 20 ºC. The applied voltages are 5 kV or 3 kV (Type-MHV, Type-SHV). The test results up to July 2024 are presented in Fig. 1 (c), where the period during which the voltage was applied and the creepage distance of insulators on the atmosphere side for each feedthrough are summarized. As can be seen in Fig. 1 (c), the shorter the creepage distance of the insulator, the sooner the breakdown occurs. The breakdown of Type-A occurs at the same position as the breakdown observed in the actual accelerator tunnel. The Type-A_C and Type-A_CD, which have a sealed cover on the outside of Type-A, exhibit much better durability against humidity. The Type-B and Type-SHV-10kV show the highest durability to date. Among commonly used high-voltage connectors, Type-SHV appears to be weak against humidity at 5 kV. However, Type-SHV and Type-MHV are viable at 3 kV. Suitable high-voltage feedthroughs will be utilized in the appropriate places, considering these results along with their costs and availability.

References

[1] Y. Suetsugu, K. Shibata, T. Ishibashi, M. Shirai, S. Terui, Y. Mu Lee, presented in the 21th Annual Meeting of Particle Acc. Soc. of Japan, Jul. 30 – Aug. 2, 2024, Yamagata, Japan, WEP078 (2024).

A pressure-anomaly detection system utilizing machine learning for the vacuum system of the SuperKEKB accelerator has been developed [1-3]. The system identified abnormal pressure behaviors among approximately 600 vacuum gauges before triggering the conventional alarm system, facilitating the early implementation of countermeasures. By comparing the recent pressure behaviors of each vacuum gauge with previous behaviors, the program detected anomalies using the decision boundary of a feed-forward neural network (FNN), as shown in Fig. 1 (a). As reported last year, a preliminary test using 2022 data yielded promising results [1]. Recently, another FNN was implemented in the system, capable of predicting possible causes of the anomalous behaviors [2, 3]. Figure 1 (b) illustrates an example of an anomaly in pressure behavior, where the cause is predicted to be a discharge. One feature of the system is its use of realistic regression models for pressure behaviors, enabling high-accuracy anomaly detection and cause prediction with a relatively small amount of data. An example of anomaly detection and the corresponding pressure trend is shown in Fig. 1 (c). The program, implemented in Python, has been operational since April 2024 for testing. The recent status of the anomaly detection system will be reported.

References

[1] Y. Suetsugu, presented in Annual Meeting of the Japan Soc. Vac. and Surf. Sci. 2023, Oct. 31 – Nov. 2, 2023, Nagoya, 2Ip01-07 (2023).

[2] Y. Suetsugu, Phys. Rev. Acc. Beams, 063201 (2024).

[3] Y. Suetsugu, presented in the 21th Annual Meeting of Particle Acc. Soc. of Japan, Jul. 30 – Aug. 2, 2024, Yamagata, Japan, WEP079 (2024).

Resistive Random Access Memory (ReRAM) has various advantages such as low power consumption, high integration, and fast read out time. PCMO is one of the promising candidates of materials used in ReRAM. Although ReRAM has already been partially in practical use, its operating mechanism is not yet fully understood.

In this study, we focused on the oxygen deficiency model, which is one of the models of ReRAM operation, and researched how the physical properties of PCMO thin films are affected by the oxygen flow rate during film deposition.

PCMO(Pr_0.7Ca_0.3MnO₃) thin films were deposited by RF magnetron sputtering with the oxygen flow rates of 0, 0.8, and 1.6 sccm during deposition. During this process, the flow rate of argon gas was kept constant, and the total pressure was maintained at 2.0 Pa. STO(100) and LAO(100) substrates were used and films were deposited at a substrate temperature of 650°C.STO(100) and LAO(100) have lattice constants of 0.3975 nm and 0.3821 nm, respectively. These lattice constants are close to that of PCMO, but the tensile and compressive stresses are introduced on STO and LAO substrates, respectively. To evaluate the crystallinity, chemical bonding states, and film thickness of these thin films, X-ray diffraction (XRD), Electron Spectroscopy for Chemical Analysis (ESCA), and stylus profilometry were used.

The thickness of the PCMO film was approximately from 10 nm to 45 nm.

Figure 1 shows the XRD patterns of PCMO thin films fabricated on the LAO(100) substrate. The diffraction peak from PCMO deposited on the LAO(100) substrate appeared at slightly lower angles than that of the substrate peak. This indicates that the PCMO grows epitaxially. Furthermore, with increasing the oxygen flow rate during deposition, the intensity of PCMO(200) peak decreases. Oxygen was introduced to compensate for oxygen deficiencies during deposition. The present result, however, shows a deterioration of the crystallinity of the PCMO thin films by the supply of oxygen. Therefore, it is required to investigate the chemical states of Mn and O atoms in PCMO film to discuss the reason of the deterioration of the crystallinity.

The details of results of the evaluation of chemical bonding states by Electron Spectroscopy for Chemical Analysis (ESCA) will be shown in the conference.

Vanadium dioxide (VO₂) has attracted attention due to its reversible insulator-to-metal transition (IMT) characteristics. VO₂ exhibits an abrupt resistance change of more than 3–4 orders of magnitude around 68 °C, accompanied by a change in crystal structure from a low-temperature monoclinic phase to a high-temperature tetragonal phase [1]. However, vanadium is a multivalent metal that forms various oxide phases [2]. Therefore, the selective growth of VO₂ film by reactive sputtering is difficult. When the amount of oxygen is excessive even slightly, the deposited films easily become V₂O₅. In our previous study, VO₂ thin films were prepared by reactive high-power pulsed magnetron sputtering (r-HPPMS), controlling the oxygen flow rate alternately. As a result, we confirmed a resistance change of 1 order of magnitude during heating and cooling cycles with one of the samples [3]. In this study, we attempted to improve the crystallinity of VO₂ and the resistance vs temperature (R-T) characteristics by changing the deposition parameter of HPPMS.

VO₂ was deposited on c-Al₂O₃ substrates using the r-HPPMS system. A vanadium metal (φ50 mm, 99.9%) was used for the target. The Ar gas flow rate and pressure were 5.0 sccm and 1.0 Pa, respectively. The O₂ flow rate was initially set to 1.5 sccm, and a fixed voltage was applied to the HPPMS power source, to achieve the time-averaged HPPMS discharge power of 10 W in oxide mode. During deposition, the voltage was adjusted to keep the 10 W power. Then the O₂ gas flow rate was controlled by software in the following manner: The higher O₂ flow rate of 1.5 sccm was set to achieve oxide mode at r HPPMS discharge power of 10 W. After the oxide mode was stabilized, VO₂ deposition was started. After holding the rate for 28 seconds, it was decreased at a rate of 0.50 sccm/sec, until it became zero. After waiting for 4 seconds, the O₂ flow rate was increased again at the rate of 0.50 sccm/sec until 1.50 sccm. This O₂ flow rate cycle was repeated for 20 min of the deposition run. The substrate temperature was 400 °C. The distance between the target and the substrate was 60 mm. The repetition frequency of the HPPMS discharge was XXX Hz, and the duty ratio was changed between 8.5% and 6.5%. The crystallinity of the deposited films was evaluated by the X-ray diffraction (XRD) in θ-2θ scan mode. Two- or four-probe methods were applied to investigate the R-T characteristics. The sample film temperature was changed by the Peltier device. The temperature range was between RT (almost 25 °C) and 100 °C in this study.

The R-T characteristics of the films are shown in Figure 1 (Left). The resistance change in three orders of magnitude against temperature was confirmed with the first 7% sample. The 8 % sample exhibited the resistance change in two orders of magnitude. The 8.5% and the second 7% samples revealed the resistance change in one order of magnitude. Figure 1(Right) shows the XRD patterns of the samples. The diffraction peak from the VO₂ (020) plane was confirmed in all samples except the one with the 6.5% duty cycle. The peak was located at 2θ=39.74° (8.5%), 39.86° (8.0%), 39.88° (7.5%), 39.94° (first 7.0%) and 39.87° (second 7.0%). The VO₂ (020) diffraction peak of the first 7.0% duty ratio sample showed the largest diffraction intensity and the largest diffraction angle. The peak shift suggested that the film with the 7.0% duty ratio sample was under compressive stress. These results indicated that the crystallinity and the electrical properties were improved by adjusting the duty ratio.

References:

[1] J. Morin, Appl. Phys. Lett., 3 (1959) 34.

View PDF for the rest of the abstract.

The oxidation of alcohols to aldehydes or ketones is an essential chemical process. For this reaction, noble metals such as Au [1] and Pd [2] have been reported as highly efficient catalysts. However, noble metals are expensive and not abundant. To solve these problems, development of noble metal substitute catalyst is strongly desired. In this work, Ni, which has advantages of costs and availability over noble metals, is focused on as an active component. To improve the catalytic activity of Ni, two techniques were introduced, the one is nanoclusterization and the other is addition of second element. Zn modified NiOx nanocluster was prepared by colloidal method using polyvinylpyrrolidone (PVP) and NaBH₄ as a stabilizer and reducing agent, respectively. Obtained colloid was poured into Al₂O₃ support under N₂ atmosphere to support it. After immersing it, washed by distilled water to form NiOx nanoclusters on Al₂O₃. Prepared NiOx nanocluster catalyst showed extremely high catalytic activity compared to bulk Ni(OH)₂ as shown in Table 1. Addition of Zn into NiOx nanoclusters improved the catalytic activity. Only addition of 5.3% Zn relative to Ni improved the acetophenone yield from 40% to 53%. X-ray absorption fine structure (XAFS) analysis revealed that the Ni species with high oxidation state was generated in the catalyst matrix by the addition of Zn as shown in Fig. 1. Catalysis will be explained in terms of Zn and Ni local structure.

References

[1] E. Pakrieva et al., Nanomaterials 10, 151 (2020).[

2] Y. Lu, H. Zhu, J. Liu and S. Yu, ChemCatChem 7, 4131 (2015).

Two-dimensional (2D) materials have attracted attention as future nanodevice materials because they are structurally stable despite their atomic-scale thickness and exhibit unique electrical and optical properties [1]. GaSe is known to be a highly photo and electrically responsive 2D material. Considering that a transistor is controlled by the gate voltage, which turns the current between source and drain on and off, the response of GaSe nanoribbons to electron irradiation is an interesting phenomenon because it can work as a switch (on and off). The response of WS₂ nanoribbon suspended between electrodes against electron irradiation has been reported [2], but the results are different from those expected due probably to contaminations, and the intrinsic response has not been confirmed, yet.

In this study, we aim to investigate the intrinsic electrically response of suspended GaSe nanoribbon to irradiating electrons by our uniquely developed in-situ transmission electron microscopy (TEM) sample holder [3,4].

A special device on a silicon chip was fabricated for TEM: it has a window for TEM observation, a gold electrode of about 10 μm width crosses over a silicon nitride film covering the window, and a nanogap of about 100-200 nm is formed by cutting the electrode using focused ion beam. Thin GaSe nanoribbons were suspended over the nanogap according to the following dry transfer process. Thin and small GaSe flake, obtained by mechanical exfoliation of bulk crystal, was picked up with a viscoelastic resin, PDMS, and transferred to the nanogap by stamping (Fig. 1(a)). Figure 1(b) shows a TEM image of the GaSe nanoribbon.

The current passing through the GaSe nanoribbon was measured by applying a bias voltage of 2 V between the nanogap. The cycle of irradiating electrons for 10 seconds and stopping for 60 seconds was repeated. As shown in Figure 1(c), the current varied sharply with this cycle, showing the electrically response of suspended GaSe nanoribbon clearly. The current density was 7600 A/m² during electron irradiation, while it was only 30 A/m² during non-irradiation.

References:[1] K. S. Novoselov, et al., Nature 438, 197–200 (2005). [2] Y. Fan, et al., ACS Appl. Mater. Interfaces 8, 32963 (2016). [3] C. Liu, et al., Carbon 165, 476-483 (2020). [4] C. Liu, et al., Nanotechnology 32, 025710 (2020).

Carbon nanowalls (CNWs) have a wall structure consisting of multilayered graphene that stands perpendicular to the substrate. CNWs have properties of a large surface area, high electrical conductivity, and chemical stability. Therefore, CNWs are expected to be used as an electrode material for electric double layer capacitor (EDLC). However, there are two problems in using CNW as an electrode material in EDLCs, because the top of the wall is sharp. The one is nonuniformity of the electric field. Electric field at the top of the wall is larger than that at the surface of the wall. The other is large hydrophobicity. CNWs have a large hydrophobicity due to the sharp top of the wall. Therefore, effective surface area of CNW electrode, because an electrolyte solution can not penetrate between the walls. The nonuniformity of the electric field and the large hydrophobicity would reduce the capacitance of EDLC using CNW electrode. It is expected to solve these two problems by coating the top of the walls with a hydrophobic insulator film such as SiO₂. In order to achieve that, a technique to coat only the top of the walls with insulator film. Therefore, in this work, we have studied the conditions to coat the top of the wall with oxide film in CNWs.

CNWs were fabricated on crystalline silicon substrates by hot-wire CVD using CH₄ and H₂. NiO films were deposited on the CNW samples by rf-magnetron sputtering with rf-power of 100 W and 150 W for 30 min. The line-scan profile of Ni along the cross-section of the NiO deposited CNW samples were measured using SEM-EDX.

Figure 1 shows the line profiles of Ni-Kα X-ray intensities for NiO-deposited CNW samples with rf-power of (a) 100 W and (b) 150 W. Here, The deposition rate of NiO films was 0.065 nm/sec and 0.072 nm/sec at rf-power of 100W and 150 W, respectively. In both samples, the signal was detected in whole CNW region. In the NiO-deposited CNW sample at rf-power of 100 W, the signal intensity was almost the constant. On the other hand, the signal intensity at the bottom of the wall (near the substrate) was larger than that at the top of the wall in the NiO-deposited CNW sample at rf-power of 150 W. These results suggest that the high deposition rate would be required in the coating only the top of wall.

Details will be reported at the presentation including other results.

This work was partially supported by JSPS KAKENHI Grant Number 23K03924.

The nanomaterials often have a single crystalline structure; their mechanical properties exhibit both size and orientation dependence. Recently, we developed a Microscopic Nanomechanical Measurement Method (MNMM) that allows us to precisely obtain the equivalent spring constants (force gradients) of nanomaterials while observing their atomic structures [1]. However, Young's modulus can vary locally depending on the cross-sectional area, meaning that the overall shape and size can only provide an average Young's modulus for the entire system, not a size-dependent one. In a previous study, Zhang et al. found that a {111} layer was introduced at the narrowest constriction every ca. 0.24 nm elongation; from the stiffness difference before and after the introduction, Young’s modulus was estimated for the introduced {111} layer [2]. In this study, we propose a method to determine the local Young's modulus of nanomaterials by measuring the initial interval and displacement of a specific region in TEM images during stretching (Fig. 1): the local Young's modulus can be estimated when the ratio of the displacement of the local region with respect to the overall elongation is calculated. In contrast to the previous study, the present method can measure the local Young's modulus at various positions within the same nanocontact. This enables us to demonstrate the size dependence of local Young's modulus within the same nanocontact. Our estimated local Young’s modulus for gold [111] nanocontacts exhibited a similar size dependence to the previous findings for each introduced {111} layer [2].

[1] J Zhang, K Ishizuka, M Tomitori, T Arai and Y Oshima, Nanotechnology 31 205706 (2020)

[2] J Zhang, K Ishizuka, M Tomitori, T Arai and Y Oshima, PHYSICAL REVIEW LETTERS 7 (2022)

Spindt-type field emitter arrays (FEAs) are FEAs with an integrated gate electrode fabricated using semiconductor manufacturing processes and have been applied to a variety of devices.[1] Spindt-type FEAs are fabricated by the deposition of emitter material inside a cavity through a hole formed in its ceiling, which works as the gate electrode. To fabricate sharp emitter shapes, it is important that deposited particles of emitter material must be injected along the normal to the substrate, and a vacuum evaporation technique with a long source-substrate distance has been used because it enables this. However, even in vacuum evaporation, there are particles incident on the substrate at an angle tilted from the substrate normal. Therefore, large equipment with a sufficient distance between the source and the substrate is required to increase the proportion of particles in the normal direction. This reduces the efficiency of material use and makes it hard to fabricate large-area FEAs. In contrast to vacuum evaporation, sputter deposition can fabricate a uniform film over a large area. In addition, while it is hard to deposit compounds with vacuum evaporation, sputter deposition can easily prepare compounds such as nitrides by using a metal target and reactive gases. However, most sputtered particles are electrically neutral, and the angle of incidence cannot be controlled. Therefore, we used a triode high-power pulsed magnetron sputtering (t-HPPMS), which can improve the directionality of sputtered particles. [2] In HPPMS, short pulse power with a low duty ratio is applied to the target to generate high-density plasma, which efficiently ionizes the sputtered particles for deposition. Furthermore, in t-HPPMS, the plasma potential can be controlled by the positively biased cap electrode. By increasing the plasma potential, the increased potential difference between the plasma and grounded substrate improves the directionality, enabling the fabrication of Spindt-type emitters. In this study, Hf and HfN Spindt-type emitters were fabricated by t-HPPMS, and their shapes were compared. The electron emission of Hf emitters was also measured; since the sputtering yields of HfN and Hf are different, the ionization rates, which are important for t-HPPMS, are also different. Therefore, the directionality of the sputtered particles changes and affects the emitter shape. First, HfN and Hf Spindt-type emitters were fabricated using t-HPPMS with an Hf target at an Ar gas pressure of 0.6 Pa. HfN emitters were fabricated using an Ar and N₂ gas mixture. When the plasma potential is set too high, compressive stress in the deposited film becomes stronger and delamination occurs. The cap voltage was set at 10 V to form sharp emitters without exfoliation. Comparing the shape of each emitter, the Hf emitter had a higher aspect ratio. This is because the sputtered Hf particles were ionized more. Unfortunately, the HfN emitters did not work as an emission device because many particles adhered to the surface during HfN deposition. Therefore, emission measurements were performed only for Hf emitters. Emission characteristics were evaluated in an ultra-high vacuum chamber at a pressure of about 10^-8 Pa. The gate electrode was positively biased and the emitter was grounded. Emission currents were collected at an anode placed above the emitter. Hf emitters started emission when 60 V was applied to the gate electrode. A maximum emission current of 6 nA was confirmed when 100 V was applied. The Fowler-Nordheim plot obtained from the measured current shows a straight line, which represents that the current is due to the field emission.

References:

[1] C. A. Spindt, J. Appl. Phys, 39, 3504-3505 (1968).

[2] Nakano, et al., JVSTB, 40, 063102-1-9 (2022).

Perovskite solar cells (PSCs), which utilize perovskite-structured material as the active layer, have demonstrate remarkable power conversion efficiency. A solar cell must have a built-in electric field suitable for separating and extracting photo-generated electrons and holes in the active layer for their highly efficient operation. However, the band structures at the interface and carrier dynamics in PSCs remain insufficiently understood. Recently, we evaluated the surface state of perovskite solar cells using terahertz (THz) emission spectroscopy (TES) that measured the waveforms of THz emissions from the samples excited by femtosecond laser pulses [1]. In this study, the samples were prepared by sequentially spin coated MeO-2PACz (hole transport layer) and Cs_0.05(FA_0.76MA_0.24)_0.95Pb(I_0.76/Br_0.24)₃ (perovskite layer) on ITO (transparent electrode), and introduced the IPA solution as a passivating agent at the interface of the perovskite layer for one sample, while the other was prepared without IPA solution. The effect of the IPA solution on the photocarriers was investigated using TES with these two kinds of samples. The photocarriers in the samples were photoexcited using a femtosecond laser as illustrated in Fig. 1(a). The photocarriers were accelerated by the built-in electric field at the interface and by internal diffusion, generating a transient current that produced THz waves. The THz waves were measured using an optical pump-probe technique to investigate the photoexcitation process in the picosecond time domain. As shown in Fig. 1(b), THz waves were observed from both samples, and were in phase with the signals from InAs. This suggests that the THz emission results from the photo-Dember effect, which is caused by the difference in diffusion velocities of photo-generated electrons and holes in the vicinity of the surface, leading to an ultrafast diffusion current. Additionally, as observed around 22 ps, the passivated sample exhibited faster relaxation of THz waves compared to the non-passivated sample, indicating that passivation may accelerate the relaxation of the current. Our results suggest that TES provides valuable insights into the role of interface passivation in enhancing the carrier dynamics and overall efficiency of PSCs.

References

[1] T. Mochizuki et al., Photonics 9, 316 (2022).

Introduction Recently, we have developed a novel functional analysis system for ion channel proteins, in which we can apply a lateral voltage to the hydrophobic core of artificial cell membranes. Using this system, we reported that the activity of the human voltage-gated Na ion channels can be recovered from their non-conductive state by the application of the lateral voltages [1]. This finding suggests that lateral voltage may be a novel parameter to enhance the activity of channels that have been difficult to measure due to their susceptibility to inactivation. However, there have been two challenges with the lateral voltage devices that have prevented them from becoming established as a functional analysis system for ion channels. The short electrode lifetime (about 30 minutes) due to oxidation of the electrode material (Ti) and the low fabrication yield (less than 10%) due to cracks in an insulating SiO₂ layer. In this study, we aimed to develop a highly efficient fabrication process for longer-lived and more stable devices by using Au as the electrode metal and CYTOP resin as the flexible insulating layer.

Fabrication Method

Micropores were formed in the Teflon film by electric sparking, and two Ti/Au thin films for the electrodes were sputtered around the micropores. The electrode pattern was controlled by a Ni mask. The film was then washed in chloroform using an ultrasonic cleaner. To form an insulating layer over the electrode, the Teflon film was dipped in CYTOP (CTL-109AE) except for the electrode contact area, pulled up with a dip coater at a rate of 1 mm/s, and baked. This process (dip, pull, and bake) was repeated 2-3 times. The CYTOP resin clogged in the micropores was removed by RIE (Reactive Ion Etching) using a Ni mask with a 120 µm aperture. The electric properties of the fabricated devices were evaluated by measuring the contact resistance of the electrode area connected to the lateral voltage source and the resistance between the two electrodes. In addition, the effect of lateral voltage on ion channels was evaluated. A lipid bilayer was formed in the micropore of the device by the folding method, and human Na channels were incorporated by the vesicle fusion method. Ion channel currents were measured and compared with and without the application of lateral voltage.

Results and Discussion

To investigate the lifetime of the fabricated electrode devices, the contact resistance was measured after applying a lateral voltage to the devices for a given time. The contact resistance of the Au electrode remained <10Ω even after the voltage (DC 4 V) was applied for more than 6 hours, which was found to be sufficient for the functional analysis of ion channels. Next, the resistance between the two electrodes was measured in buffer solution to examine the insulating properties of the CYTOP layer. When the device was coated twice with the CYTOP layer, the electrode-to-electrode resistance of the device exceeded 250 GΩ. After optimizing the CYTOP concentration, the device fabrication yield was improved up to 60%. Using the fabricated device, we applied the lateral voltage to lipid bilayers containing Na channels whose activities disappeared during the repetitive applications of transmembrane voltage pulses. Upon the application of the lateral voltage, the channel activities appeared again, showing frequent opening and closing events. The device fabricated by this process proved to be a useful tool for the novel analysis of inactivated ion channels. The establishment of such a system for analyzing ion channel function is expected to contribute to the discovery of new knowledge about ion channels and drug discovery.

Introduction

Praja1 (PJA1) is an intrinsically disordered protein (IDP) consisting of 643 amino acid residues. Different from structured proteins, IDPs are characterized by the lack of a fixed or ordered three-dimensional structure in the absence of thier binding partners. PJA1 has been reported to promote ubiquitination and degradation of aggregate-prone proteins in neurons although the details of protein-protein interactions remain unclear [1,2,3]. Whereas most structured proteins on metal surfaces would lose their functionality by disruption of their higher-order structure, PJA1 is anticipated to retain its functionality in terms of protein interactions due to its inherent properties as an IDP. If PJA1 maintains its function on a solid surface, then studying its behavior leads to understanding and exploiting its interactions with aggregated proteins. In this work, we investigated the behaviors of PJA1 adsorbed on gold substrate by using electrochemical methods: changes in the state of PJA1 by the interaction with alpha-synuclein (α-syn), which is also an IDP and known to be one of interaction partners of PJA1, were examined from the electrode reaction of ferricyanide added as a marker.

Method

Human PJA1 protein was put on a gold substrate (a vacuum-deposited gold thin film on silicon wafer) by drop-casting its aqueous solution, corresponding to 3.7 μg/mm² on the surface, for 1 h at room temperature. To remove non-adsorbed components, the substrate was rinsed with deionized water and dried with N₂ blowing. The interaction with α-syn (140 amino acid residues) was investigated by putting its solution on the substrate followed by incubation at room temperature. After non-reacted component was removed by rinsing, the substrate was dried with N₂ blow. Cyclic voltammetry and electrochemical impedance spectroscopy were carried out with these gold substrates as a working electrode, a Pt wire counter electrode, and an Ag/AgCl (3 M, NaCl) reference electrode, in the electrolyte of 5 mM [Fe(CN)₆]^3- in 0.1 M KCl.

Results

In the cyclic voltammograms for the gold electrodes with PJA1 at different concentration of α-syn (0, 0.01, and 0.1 μg/mL), a pair of peaks attributed to the oxidation and reduction of [Fe(CN)₆]^3-/4- was observed at ~0.24 and ~0.16 V, respectively, for all three conditions. The magnitude of peak current for 0 μg/mL α-syn was much smaller than that for the case without PJA1, while peak currents increased under the presence of α-syn, suggesting that the suppression of [Fe(CN)₆]^3-/4- electrode reaction by the presence of PJA1 is reduced by the interaction with α-syn. From the electrochemical impedance measurement, the magnitude of charge transfer resistance was suggested to be largest for 0 μg/mL α-syn among the three conditions and decreased under the presence of α-syn. These results may be caused by the exposure of the gold surface as a consequence of the conformational change in PJA1, and thus demonstrated that PJA1 can interact with α-syn even in its adsorbed state on gold substrate. Details will be presented in the poster session.

References

[1] K. Watabe, M. Niida-Kawaguchi, M. Tada, Y. Kato, M. Murata, K. Tanji, K. Wakabayashi, M. Yamada, A. Kakita, and N. Shibata, Neuropathology, 42, 6, 488-504 (2022).

[2] W. Onodera, K. Kawasaki, M. Oishi, S. Aoki, and T. Asahi, J. Mol. Evol., 92, 1, 21-29 (2024).

[3] S. Aoki, K. Kawasaki, K. Imadegawa, M. Oishi, T. Asahi, and W. Onodera, BioRxiv, 2024.06.10.598176 (2024).

Nanotube stamping is a technique for delivering target molecules directly into adherent cells through nanotubes by inserting an array of nanotubes into the cell membrane [1]. The ability for direct intracellular delivery has the potential to evaluate the effectivity or toxicity of drugs on cells, and even to add new functions to cells by introducing functional proteins (Fig. 1). While conventional methods such as viral vectoring and electroporation have drawbacks such as affects by viruses and voltage pulses, nanotube stamping has the advantage of achieving both a higher delivery rate (85%) and cell viability after stamping (90%). However the nanotube stamping requires subcellular control of the relative positioning (height and inclination of the nanotube membrane) because it involves inserting many nanotubes from above into cells adhering to the scaffold. Specifically, deep insertion can crush the cells, while shallow insertion introduces few molecules. When stamping on cells with weak adhesion such as neurons, more precise control of the nanotube membrane position is required.

In this study, we evaluated the feasibility of functionalization (measurement of neural activity by fluorescent calcium imaging), especially for weakly adherent neurons, by precise control of the relative positioning of the nanotube membrane to the cell/scaffold interface using a control system integrating microscope optics, image processing, and stepper motor control. The nanotube stamping system was constructed to control the relative distance between the nanotube membrane and the cell membrane in combination with real-time image processing, which enabled molecular introduction with an accuracy of approximately 1 µm without relying on the operator’s skill. In addition, the membrane tilt was corrected to make the membrane parallel to the scaffold surface, and the focus of the phase contrast microscopy observation for viewing the cells was aligned with the coaxial epi-illumination for viewing the nanotube membrane.

Fluorescent molecule Calcein (concentration, 1.6 mM; molecular weight, 622.55 Da) was introduced into fetal 18-day-old rat cerebral cortex cells. In this experiment, the membrane was held for 10 minutes after nanotube insertion to deliver Calcein to neurons by diffusion. After releasing the nanotubes from the cells, the neurons were rinsed twice to remove fluorescent molecules in the medium. Fig. 1 shows clearly that the fluorescent molecules were successfully introduced into the cytoplasm of the neurons. In the presentation, we discuss the dynamics on the nanotube insertion and the cell/scaffold on interface.

Reference

[1] Miyake, Analytical Chemistry(2024)

Organisms such as microorganisms and plants have cell walls that encircle their cells and execute diverse functions by generating turger pressure inside the cell. For instance, in Colletotrichum orbiculare, which is one of a type of plant-pathogenic fungus, spores differentiate into appressoriums on the plant surface and make a turger pressure to punch a hole in the plant cell wall for mycelium to invade. The cell wall structure of the appressorium that supports such strong turger pressure attracted extensive interest from numerous research fields. Despite this, the nanoscale structure of the cell wall has yet to be elucidated through microscopic techniques, such as optical microscopy and electron microscopy. Thus, a method for visualizing this is strongly required.In this study, we developed a method for the nanoscale analysis of appresorium cell walls using atomic force microscopy (AFM). We used appressoria of Colletotrichum orbicular with both wild-type and melanin mutant (where the melanin gene was knocked out) as our model sample. Melanin mutant cells are known to have lower turger pressure in the appressorium than wild-type cells, and we expected differences in the cell wall structure as well. The appressoria were grown on a plastic dish and immersed in ultrapure water. We performed AFM measurements using a NanoWizard4 (JPK, Bruker) and 240AC (OPUS) cantilever. The cantilever was brought close to the appressorium surface, and the surface topography image was obtained using the QI mode in JPK. The obtained high resolution AFM images of the cell wall surface of both wild-type and melanin mutant appressoria, revealing the fine fibrous structures that make up the cell wall with a resolution of less than 10 nm. Our analysis found that the diameter of the fibrous structures was larger in the wild-type cell wall than in the melanin mutant and that particle-like structures with a size of several tens of nanometers were present in the wild-type cell wall, forming a robust membrane structure that creates strong turger pressure inside the appressorium. This method is expected to provide new insights into the mechanism by which appressoria can form such high turger pressures and contribute to the field of biology and materials science.

The cell membrane is a reaction field for material transport and signal transduction via membrane proteins. Among membrane proteins, ion channels play an important role in the regulation of ion concentrations and signal transduction in a living organism. Ion channels can maintain their correct conformation when encapsulated in lipid bilayers and most of them achieve their functions by assembling molecule in lipid bilayers. Therefore, experimental methods to evaluate their association states in membranes are useful for the development of expression methods of ion channels and elucidation of their drug effects. In this study, we aim to observe ion channel molecules by atomic force microscopy (AFM) using the supported lipid bilayer (SLB) system, which is an artificial cell membrane. Natural lipids extracts, such as asolectin, are used in cell-free membrane protein synthesis systems. However, they are a mixture of multi-component lipids, and thus it is difficult to fabricate uniform SLBs with good reproducibility. Because the outer membrane region of ion channels has a size of approximately 1 - 3 nm, it is necessary to fabricate a smooth SLB with a surface roughness of less than a nanometer order, to observe the molecular images of the ion channels. Therefore, we aimed to optimize the sample preparation and AFM observation suitable for the observation of the ion channels.

Vesicle suspensions were prepared by adding buffer (KCl 120 mM, HEPES 10 mM, pH 7.2/ KOH) to a vacuum-dry film of asolectin, which is a crude lipid from obtained from soybean, and a dye-labeled lipid (Rb-DPPE (Ex/Em: 560/583 nm)) at a ratio of 0.995:0.005 (w/w). SLBs were fabricated on mica substrates by the vesicle fusion method, and fluorescence microscopy, fluorescence recovery after photobleaching (FRAP) and AFM observations were performed in the buffer solution.

Fluorescence images showed that asolectin-SLB was formed, as uniform fluorescence intensity was observed on the mica substrate after the SLB formation operation. However, white bright spots (Fig. 1a, white arrows) and dark regions (Fig. 1a, black arrows) were also observed. These bright spots are unruptured adsorbed vesicles. Because the particle size of the vesicles is ~50 nm, the presence of adsorbed vesicles on the SLB may interfere with the observation of smaller size ion channels. The dark regions observed in the fluorescence images tended to appear more easily when the asolectin reagent was stored at -20 °C for more than 2 months, suggesting that they are intramembrane domains containing oxidized lipids. The presence of oxidized lipid domains is undesirable in ion channel observations because the composition and state of the lipid bilayer around membrane proteins can affect their activity. Reducing these adsorbed vesicles and domain regions is required for SLBs for molecular imaging by AFM. We investigated SLB preparation conditions by changing the temperature, time, and lipid concentration during the SLB formation process, as well as the storage method of asolectin reagent. As a result, SLBs with almost no adsorbed vesicles or dark domains were successfully prepared on a mica substrate (Fig. 1b). The roughness of the SLB surface observed by AFM was 0.69 nm (rms). This value is sufficiently smaller than the height of the outer membrane region of the ion channels (1 - 3 nm), and thus the asolectin SLB was suitable for their molecular imaging.

Introduction

Hydrogen separation membranes using vanadium can supply high-purity hydrogen compactly and at low cost. There is currently high demand for improved hydrogen permeation performance in vanadium membranes. In this study, two types of hydrogen visualization methods are used to observe hydrogen permeating through the vanadium membranes in situ. The first is the observation of hydrogen ions using an operando hydrogen microscope. The second is the observation of silver deposited by the silver decoration method. As a result, we confirmed that the hydrogen permeation characteristics depend on the crystal orientation of the vanadium membrane surface.

Experiment

We prepared a rolled vanadium (99.936%) plate material for the hydrogen permeation measurement. The grain size was controlled to 50-150 μm by heat treatment at 1173 K for 2 h. The structure and crystal orientation were confirmed by electron backscatter diffraction (EBSD). The specimen shape is the same for both experiments, a circular plate with a diameter of 16 mm and a thickness of 0.5 mm, with a mirror-polished surface. In the operando hydrogen microscope, we used the electron stimulated desorption (ESD) method to observe hydrogen ions desorbing from the metal surface in real time during the hydrogen gas permeation experiment. In the silver decoration experiment, the sample cell was placed on the stage of an optical microscope and observed in situ [1].

Result

First, we observed hydrogen permeation at a sample temperature of 573 K and a hydrogen supply pressure of 100 Pa using an operando hydrogen microscope. We found that grains that appear bright in SEM image (secondary electron image) have a relatively high concentration of hydrogen distribution [2]. Next, we observed the deposition of silver in a silver decoration experiment at a sample temperature of 296 K and a current density of -5.3 mA/cm². As time passed, bright and dark areas appeared due to the deposition of black spots (Fig. 1 (a)). Elemental analysis using energy dispersive X-ray spectroscopy (EDX) confirmed that the deposited black spots corresponded to the distribution of silver (Fig. 1 (c)). Finally, EBSD analysis identified the crystal orientation of each grain. A comparison of the silver distribution in Fig. 1 (a) with the IPF map in (d) shows the anisotropy of the silver distribution on the surface. The amount of silver distribution is high in the crystal grains with (101) and (111) orientations, and low in the crystal grains with (001) orientation. The areas with high silver distribution correspond to areas with high hydrogen permeation flux. Furthermore, the SEM image of the same area in Fig. 1(b) shows that the crystal grains that appear dark in the SEM image correspond to the (001) orientation. These results show similar trends in both the operando hydrogen microscope experiment and the silver decoration experiment.

Conclusion

In this study, we succeeded in visualizing the behavior of hydrogen that permeated through a vanadium membranes by using two different methods. EBSD analysis established a clear correlation between the hydrogen permeation flux through the vanadium membranes and the crystal orientation of the surface. It is presumed that the difference in hydrogen permeation flux on the observed surface is due to the difference in diffusion coefficient due to crystal orientation affecting hydrogen diffusion. This study provides a new understanding of the mechanism of hydrogen permeation and clarified the effect of crystal orientation on hydrogen permeation efficiency.

References

[1] A.N. Itakura, T. Kusawake, T. Fujimaru, S. Miyai, Y. Matsumoto, Y. Murase, e-Journal of Surface Science and Nanotechnology 22, 174–178 (2024).

View PDF for the rest of the abstract.

Higher-capacity lithium-ion batteries are required for applications of electronic vehicles. Lithium cobalt oxides (LiCoO₂, LCO) with a layered rock salt-type structure is a popular cathode material. LCOs are typically operated below 4.2 V (vs. Li/Li+) when combined with organic electrolyte. In this voltage range, there is no significant decrease in capacity after charge-discharge cycles. Operating at higher voltages can further increase the battery capacity because the amount of deintercalated Li-ion increases. However, high voltage operation leads to the capacity fade after cycling. This is suggested to be due to irreversible changes in the LCO structure.

The capacity fades due to high voltage cycling can be reduced by coating the cathode surface with metal oxides [1]. Since most metal oxides have low electron and ion conductivity, excessive coating seems to lead to suppress the Li (de)intercalation. The understanding of the coating effect has been desired toward optimizing the coating. Typical cathode is a mixture of polycrystalline LCO, conductive additives, and binders. This results in various interfaces, such as LCO−LCO with different orientations, LCO−additives, and LCO−binders. The complicated interfaces make difficult it to interpret the coating effect of metal oxides to LCO.

In this study, we investigated the effect of metal oxides coating on the structural change of LCO after high-voltage operation. Epitaxial growth was used to construct a model battery with almost a single crystalline LCO. Zirconium (Zr) oxides were chosen as a typical coating material. Scanning transmission electron microscopy (STEM) was used to examine the local structural changes in the LCO films with and without Zr oxides coating by high voltage cycling.

LCO cathode was epitaxially grown on a substrate by pulsed laser deposition. The cathode surface was coated with Zr oxide by further deposition of Li₂ZrO₃ (LZO/LCO). STEM observation of the LZO/LCO showed island growth of Zr oxides on the cathode surface (Fig.1). Two type of batteries were constructed with the bare-LCO or LZO/LCO cathode, organic electrolyte, and lithium metal anode. The batteries were operated for 25 charge-discharge cycles at 4.5 V. Although the discharge capacity of the bare LCO decreases continuously, that of LZO/LCO was maintained almost constantly.

After a further 100 cycles, the cathodes were processed for STEM observations. Bare-LCO showed the irreversible structural change into spinel-type Co₃O₄ at the surface. In contrast, irreversible structural changes were partially suppressed in the vicinity of the Zr oxides islands on LZO/LCO. The interface of LZO and LCO showed a disarranged structure. This suggests the interphase of LZO-LCO, which may realize both the ionic conductivity and protection of LCO.