Ir and Co catalysts are effective to obtain vertically aligned single-walled carbon nanotubes (VA-SWCNTs), but the growth condition of grown SWCNTs are much different; In alcohol catalytic chemical vapor deposition (ACCVD), VA-SWCNTs were grown from Co catalysts on Al₂O₃ support layers in the wide temperature range from 700 to 850°C [1], while VA-SWCNTs were obtained above 800°C with Ir catalysts on SiO₂ support layers [2]. In addition, SWCNTs grown from Ir catalysts have diameters less than ~1 nm, which were much smaller than those from Co catalysts. This suggests that the reaction process between ethanol and catalysts is different between the two catalysts. To clarify it, we investigated growth process of CNTs by ACCVD from Ir catalysts on SiO₂ and Co catalysts on Al₂O₃ by using in situ diffuse reflectance infrared Fourier Transform (DRIFT) measurements and in situ X-ray absorption fine structure (XAFS) spectroscopy. In situ DRIFT investigation showed that intermediate species formed on the catalysts during CNT growth were different between the two catalysts; formation of ethoxy was seen on Ir catalysts, while acetates were observed on Co catalysts. In addition, the amount of species adsorbed on Co catalysts were much larger than those on Ir catalysts. In situ XAFS result showed that Ir catalysts were kept metallic during SWCNT growth, while Co catalysts were partially carbonized (Fig. 1). From these results, we proposed a growth process model of SWCNTs from Ir and Co catalysts. We also performed ultra-violet photoelectron spectroscopy (UPS) to investigate the d band center of Ir 5d and Co 3d levels of catalyst particles. Our results showed that the d band center of Co 3d level is 0.8 eV lower than that of Ir 5d level. The lower binding energy of the d band center of Co catalysts would lead to high efficiency in feedstock dissociation, resulting in the wide temperature range in SWCNT growth.

[1] K. Cui et al. Nanoscale 8 (2016) 1608-1617.

[2] T. Maruyama et al. Appl. Surf. Sci. 509 (2020) 145340.

Graphene is a two-dimensional material with a sp²-carbon network. Its outstanding characteristics such as very-high electron mobility for field effect transistors [1] and gas-barrier properties have been remarked [2]. To improve these characteristics like impurity doping or find new one, the development of its modification methods is indispensable; however, the current ones might destroy or reduce properties: implantation, adsorption, and chemical modification. We have been using low-energy ion attack of photoemission-assisted Townsend discharge (PATD) plasma [3]. PATD is a discharge style of our photoemission-assisted plasma-enhanced chemical vapor deposition (PA-PECVD) system, which is a DC plasma system with the aid of UV-photoemission. In conventional radio-frequency discharge plasma, a sheath electric field may cause irreversible and severe damage to graphene. Because of the displacement current, both current and voltage are difficult to measure independently and precisely. Power in watt, which is a product of current and voltage, is used as a variable. However, the current is an extensive variable and is a factor of kinetics of chemical reactions. The voltage is an intensive variable and is a factor of thermodynamics. Thus, we can expect precisely-controlled graphene by PATD.

Figure 1 shows Raman spectra of graphene sheets treated by PATD plasma at 800 Pa and 300 V in a 100-sccm argon flow. The D band, which represents disorder of the graphene sheet and is observed around 1340 cm^-1, is controlled by quantity of electric charge passed. Treatment in a low electric charge (Fig. 1(b)) suppresses the D band, suggesting that minute and precise control is accomplished on graphene.

References: [1] S. Takabayashi et al., Diamond Relat. Mater. 22, 118 (2012); [2] S. Ogawa et al., J. Phys. Chem. Lett. 11, 9159 (2020); [3] T. Takami et al., e-J. Surf. Sci. Nanotechnol. 7, 882 (2009).

In recent years, two-dimensional transition metal dichalcogenides (TMDCs) have been actively studied because of their potential applications in thermoelectric materials and MISFETs utilizing their excellent mobility and Seebeck coefficient [1, 2]. Among them, WS₂ is known to have relatively high mobility and stability [3] and has attracted much attention. However, the structural disorder of WS₂ thin films, which is considered to have a significant impact on device performance, has not fully been understood yet from experiments, and thus computational calculation to deepen such understanding is desirable. In this study, we aim to clarify the structural disorder in WS₂ thin films due to S defects using the high-dimensional neural network potential (HDNNP) [4], which can achieve high prediction accuracy and feasible computational cost simultaneously.

Molecular dynamics (MD) calculations with the HDNNP and NVT ensemble at the temperatures of 500, 1000 and 1500K were performed for WS₂ slab models. The ratio of S/W was set to 1.7 or 1.9 based on experimental observation. The MD calculations show that the introduced S vacancies were spontaneously transformed into structures consisting of non-six-membered ring structures (in most cases, 5-membered and 9-membered rings), as shown in Figure 1. These structures moved around frequently during the MD calculations, and were also confirmed to be more stable than the S point defect (i.e. atomic vacancy) structures. In the presentation, correlation between neighboring defect structures will also be discussed.

This study was supported by KAKENHI Grant (Nos. 21H05552 and 23H04100), MEXT, Japan.

Reference

[1] G, Kogo, et al., Sci. Rep. 10, 1067 (2020)

[2] T. Hamada, et al., Jpn. J. Appl. Phys. 61 (2022) SC1007

[3] W. Zhang, et al., Nano Res. 7, 1731 (2014)

[4] J. Behler, and M. Parrinello, Phys. Rev. Lett. 98, 146401 (2007)

Introduction

Carbon nanotubes (CNTs) are potential candidates for flexible thermoelectric (TE) materials because of their flexibility and good thermoelectric property. The as-grown CNT sample is a mixture of metallic and semiconducting CNTs in a ratio of 1:2. This ratio can be controlled when a film is formed with the CNTs and significantly influence the TE performance of the film. According to the recent experimental work by Ichinose et al, the single-walled CNTs (SWCNTs) with semiconductor purity over 99 % exhibit a Seebeck coefficient of 200~300 µV/K when optimal carrier doping is achieved using a FET setup [1]. The Seebeck coefficient in this experiment was very high, but the electrical conductivity was less than 1 S/m. On the other hand, in the TE experiment on aligned SWCNTs with metal purity over 96 %, the electrical conductivity was above 100 S/m, which is about 10 times larger than that of non-aligned film, whereas their Seebeck coefficient was only about 30 µV/K.

These experimental results indicate that the TE performance of CNT films highly depends on both semiconductor purity and CNT alignment. In addition, the CNT films seem to have a trade-off relationship between the electrical conductivity and the Seebeck coefficient, resulting in the difficulty of enhancing TE power.

Modeling and Optimization Method

To address the tradeoff relationship mentioned above, we explored the optimal semiconductor purity and structure using a Thermoelectric Random Stick Network (TE-RSN) method [2] and Nondominated Sorting Genetic Algorithm Ⅱ (NSGA-Ⅱ) [3]. Our recently developed TE-RSN is a simulation method that combines a random stick network model and electrical and thermal circuit network equations [2]. The NSGA-Ⅱ is among the multiobjective genetic algorithm methods.

In the present simulation, the CNT film size has been fixed to 5 µm in length and 5 µm in width. The number of CNTs was 1250, the length was 500 nm and the diameter was 1.4 nm. Also, the semiconductor purity of the film, as well as the position and alignment angle of each CNT, were chosen randomly. We put in place 200 initial structures under these conditions and evaluated TE performance using the TE-RSN method. We then repeated the loops of the genetic algorithm (non-dominated sorting, crowded distance sorting, crossover and mutation) until the TE performance converged.

Results

As a result of optimization up to 2,000 generations, an averaged electrical conductivity increased from 100 S/m to 300 S/m and an averaged Seebeck coefficient increased from 60 µV/K to 120 µV/K compared to the initial generation. Thus, we succeeded in resolving the trade-off relationship between the electrical conductivity and the Seebeck coefficient. Furthermore, we identified the following 3 characteristic features by analyzing samples from the 2,000 generation.

・The semiconductor purity of CNT films existing in the 2,000 generation is about 90 %.

・The CNT alignment angle θ (|θ| ≦ 90°) in the CNT film is aligned in the direction of the electric field with a standard deviation σ = 50°.

・Some high-density regions appear in the CNT film parallel to the electric field and temperature gradient.

References

[1] Y. Ichinose et al., Nano Lett. 19, 7370 (2019).

[2] J. Kobayashi and T. Yamamoto, Jpn. J. Appl. Phys. 61, 095001 (2022).

[3] K. Deb et al., IEEE Trans. Evol. Comput. 6, 182 (2002).

Carbon nanotubes (CNTs) are potential candidates for use in nanoelectronics due to their outstanding electrical and mechanical properties. Especially in high-frequency device applications, not only the averaged current characteristics such as impedance but also the current fluctuation such as shot noise are essential. Until recently, the current fluctuation has been considered mere noise being an obstacle to signal detection, but recent developments in nonequilibrium statistical mechanics with respect to current fluctuations such as the fluctuation theorem (FT) and the thermodynamic uncertainty relation (TUR) suggest that they can be controllable signals and useful information. In addition to formal frameworks of the nonequilibrium statistical mechanics, it is desirable to development new theoretical and computational methods that allows quantitative evaluation of current fluctuations in a specific target material with high precisely.

In this study, we provide a new simulation method that can simulate the time-dependent current flowing through a material on an atomistic level. Furthermore, we apply the new method to CNTs and evaluate the averaged current, the current fluctuation (current variance) and higher orders of cumulant expansion. Fig. 1(a) shows the time-dependent current in metallic (5,5) carbon nanotubes with three typical lengths of L=51 nm, 281 nm and 2,757 nm at T=300 K, where L=281 nm is almost the mean free path L_m of electrons in the (5,5) CNTs due to the electron-phonon scattering. As we can easily expect, the averaged current <J>decreases with increasing L. On the other hand, the current fluctuation, i.e., the variance σ²=<(J-<J>)²>, increases monotonically with L and exhibits the maximum near L=L_m(=281 nm) as shown in Fig.1(b). The new finding is extremely instructive in controlling the current noise of CNT-based electric devices. Furthermore, to give a physical interpretation of the new finding, we perform theoretical analysis based on the quantum scattering theory combined with Buttiker’s fictitious probe method [1] that can incorporates the electron-phonon scattering effects phenomenologically. The details are presented in the presentation.

Reference

[1] M. Büttiker, IBM J. Res Dev., 32, 63 (1988).

Since the successful synthesis of graphene, single-element monolayer sheets have attracted much interest because of their distinctive electronic properties. One example is a monolayer sheet of boron (B) (a group 13 element) called “borophene”. Compared to graphene (a sheet of group 14 element), each boron atom is deficient of one electron, thus theoretically borophene layers are predicted to have various types of atomic structures, consisting of triangular lattices and hexagonal hollows. On the other hand, it is reported that a graphene-like honeycomb borophene has been synthesized on aluminum (Al) (111) substrate, based on scanning tunneling microscopy (STM) experiments [1], although in general the honeycomb structure of borophene is inherently unstable. Preobrajenski, et al. have proposed two different structural models for this honeycomb structure[2]; one is a “free-standing model”, in which the honeycomb borophene is simply on Al(111) substrate, and the other is an “AlB₂ model” in which the boron and aluminum atoms form into AlB₂ compositions at the topmost surface. However, the actual structure has not yet been determined and the origin of the honeycomb-like structure has not been clarified either. In this study, the atomic structure of honeycomb borophene on Al(111) was investigated by total-reflection high-energy positron diffraction (TRHEPD).

For sample preparation, B was deposited on a clean Al(111) surface at ~210°C. The quality of the samples was confirmed by using low-energy electron diffraction (LEED) and reflection high-energy electron diffraction. Figure 1(a) shows that the LEED pattern of the B/Al(111). Besides the Al-1×1 spots (red circles), one can see that many distinct fractional spots appeared. As shown in the inset (enlarged view of the area enclosed by the square) of Fig.1(a), we identified spots specific to the honeycomb borophene (yellow circles), which is consistent with the study by Preobrajenski, et al. [2]. The TRHEPD measurements were performed at the Slow Positron Facility, KEK. Figure 1(b) shows the TRHEPD rocking curves obtained before and after the B deposition under one-beam (OB) condition, where the diffraction intensities of the specular (00) spot are plotted as a function of θ. After the B deposition, the peak position does not significantly change while the intensity at 3.5° is suppressed. Under the OB condition, the diffraction intensities reflect the interlayer distance and the atomic density of each layer. Therefore this result indicates that the borophene layer has formed on the topmost surface and interferes with the beam diffracted from the Al(111) substrate. In this presentation, we will present details of the sample preparation conditions and the result of the structural analysis, and discuss the origin of the structure for honeycomb borophene on Al(111).

References

[1] W. Li, et al., Sci. Bull. 63, 282 (2018).

[2] A. B. Preobrajenski, et al., ACS Nano 15, 15153 (2021).

Proximity-induced superconductivity in graphene has been recently studied with great interest because the introduction of superconductivity to the Dirac fermions is very attractive in terms of both fundamental physics and technological applications. In this case, the group 14 elements Pb (T_c = 7.2 K) and Sn (T_c = 3.7 K) is preferred as candidate materials to induce superconductivity. Previously we have studied the superconducting properties for Pb-deposited bilayer graphene on SiC(0001) substrate by ultralow-temperature scanning tunneling microscope (STM) measurement, and we observed that Pb atoms form islands in the VW growth mode. We also observed a superconducting gap on graphene, which was induced by the proximity effect from the Pb islands. This is similar to a previous study by another group [1], using a quasi-free-standing monolayer graphene obtained by large scale hydrogen intercalation underneath graphene on 6H-SiC.

Here we focus on Sn-deposited graphene. Kessler et al. reported that Sn atoms form amorphous-like islands on exfoliated graphene [2]. They also demonstrated superconducting behavior of electrical conductivity. On the other hand, Kim et al. reported Sn atoms intercalate into the graphene/SiC(0001) and form an atomic-layer between graphene and SiC substrate [3]. To elucidate the origin of superconducting properties in Sn-deposited graphene on SiC, it is necessary to clarify the morphology of the surface during the Sn-deposition process.

In this study, we demonstrate the surface morphology of bilayer graphene/SiC(0001) in samples with different amounts of Sn deposition, observed by using STM. Figure 1(a) shows the STM image of about 10 ML Sn-deposited bilayer graphene. The Sn atoms were deposited on bilayer graphene at room temperature. One can see that crevasse-like groove structures are formed on graphene while no island-like structures were observed. These grooves are created across the terrace steps of the SiC substrate. As seen in Figure 1(b), its depth is estimated to be roughly 0.7 nm, which corresponds to an interlayer distance of graphene layers. These results suggest that Sn atoms penetrate the graphene by partially disrupting the graphene during the initial process of Sn deposition. In my presentation, we will present the results of STM experiments measured on samples with different amounts of Sn deposition and discuss the possibility of Proximity-induced superconductivity in bilayer graphene on SiC.

[1] F. Paschke, et al., Adv. Quantum Technol, 3, 2000082 (2020).

[2] B. M. Kessler, et al., Phys. Rev. Lett. 104, 047001 (2010).

[3] H. Kim, et al., J. Phys. D: Appl. Phys. 49. 135307 (2016).

Toward Society 5.0, next generation wireless communication (Beyond 5G) is indispensable as the infrastructure. 70 trillion Beyond 5G devices are needed and contain high environmental elements, such as In, As, and Cu. The realization of Society 5.0 will threaten global environment. Therefore, for realizing Beyond 5G, we should create Beyond 5G devices with low-environmental cost. We proposed a novel monolithically integrated devices on SiC consisting of graphene based antenna and GaN based high electron mobility transistor (GaN-HEMT) (Patent application number 2021 207486). As the first step, we are developing an antenna using graphene grown on SiC for the purpose of the monolithic integration.

We will present the study on the graphene based antenna on SiC (Fig. 1). The antenna consists of a coplanar waveguide (CPW) and antenna elements. The characteristic impedance Z0 of the CPW was matched to the characteristic impedance of the measurement system (50 ohm) to minimize return loss. The structure-dependence of the characteristic impedance was investigated and succeeded in optimizing the structure to set 50 ohm. The return loss (RL) and transmission power efficiency of this CPW from 300 kHz to 9 GHz showed that RL is below -37.3 dB and transmission power efficiency 99.97%. This means that a very low-loss CPW was fabricated. In addition, this is the first successful fabrication of the graphene antenna on SiC resonating at 40.7 GHz by using the scattering (S) parameter measurement for high-frequnecy electrical measurement. The transmission power efficiency was measured to be 97.19%. The transmission power efficiency at the resonant frequency is 90% in the previous report. Our antenna is more efficient by 7%, compared to that in the previous report. Our work demonstrates the capability of the graphene antenna on SiC as the highly efficient antenna for Beyond 5G.

In 2019, we developed the hollow-cone Foucault imaging method as a new method for observing the magnetization of specimen materials using transmission electron microscopes [1-3]. This method can be regarded as third Lorentz microscopy, which makes it possible to simultaneously visualize both magnetic domains and domain walls as well as acquire contrast-inverted hollow-cone Foucault images under both bright-field and dark-field conditions. Furthermore, the schlieren images are obtained under the special boundary condition, schlieren condition, between the bright-field and dark-field conditions, we observed a contrast caused by the electromagnetic field in the space around the specimen [4]. In this study, we precisely controlled the schlieren condition on the basis of the inclination angle of the illumination electron beams and successfully observed the electromagnetic field around the specimen in the vacuum region.

Under the magnetic field-free condition for the specimen with the lens current of the objective lens turned off, the electron beam transmitted through the specimen was made to crossover at the selected-area (SA) aperture position. The SA aperture functioned as a reciprocal space aperture (diffraction aperture) to shielding out the deflected electron beams at and around the specimen, and this asymmetric imaging led to the contrast caused by the electromagnetic fields. To analyze the vector component of the electromagnetic fields, the image was recorded every 10 degrees of the azimuth angle by the illumination beams and used for image processing instead of the full-azimuthally integrated images as in the previous hollow-cone Foucault imaging method. This recording procedure is similar to that of laminography in X-ray experiments. The transmission electron microscope used was the JEM-2100 Plus, with an acceleration voltage of 200 kV.

We numerically reconstructed the series of azimuthally separated schlieren images. Figure 1(a) shows the distribution of the electric field from latex balls of 2 mm in diameter on a 50 nm thick Si₃N₄ membrane. The intensity of the electric field is indicated by brightness, and the direction is indicated by color. The electric charges are generated at the surface of the latex balls by the irradiation of the electron beams, and it can be seen that the electric field distribution is similar around the latex balls even with varying numbers of balls. Figure 1(b) shows the magnetic field leaking from the amorphous magnetic film into the vacuum. The orientation of the magnetic field is indicated by the azimuth of arrowheads and colors, and the intensity of the magnetic field is indicated by the size of the arrowheads. The colored area inside the thin film is the region where the electron beam was transmitted and the magnetization distribution inside the material was detected. Although the intensity of the recorded electron beams differed greatly between the vacuum and the inside of the specimen, simultaneous visualization in one micrograph should be feasible by further analysis for data processing.

We have demonstrated that our developed schlieren electron microscopy using thermal electron beams can be used to observe a wide area of electric and magnetic fields, which are difficult to observe even with electron holography using field emission electron beams.

Superlubricity due to the incommensurate contact between twisted crystal lattices of layered material reduces friction to the order of magnitude of less than several hundreds of pN [1, 2]. Twisted graphene is one of such typical superlubric interfaces, and can be formed by rotating a graphene sheet stacked on a graphite substrate as shown in Fig. 1(a). While the sheet is slid on the substrate, the nanoscale real contact region between the sheet and the substrate plays an important role in superlubricity, but its microscopic mechanism has yet to be discussed. Because of the microscopic roughness, the real contact area is smaller than the apparent contact area. Since the frictional force is represented by the sum of the force to shear the real contact region, in order to control the nanoscale friction, it is important to clarify the behavior of the real contact region during the sliding process of the sheet. Therefore, in this work, the mechanical behavior of the contact state of a twisted graphene interface is studied by a molecular simulation of the sliding process of the sheet. Furthermore, the mechanics of the contact state in a general twisted crystal lattices of layered material is mathematically discussed. By applying this discussion to the twisted graphene, the contact state behavior observed in the simulation is theoretically explained.

First, the molecular simulation of the sliding process of the sheet of the twisted graphene interface was performed. Fig. 1(b) shows the model of the twisted graphene. Twisted graphene interface is comprised of the pseudo AA and AB stacking regions which can be locally approximated as AA and AB stacking regions, respectively. The pseudo AA stacking regions are arranged on the lattice points of a triangular lattice as shown as red circles in Fig. 1(b). Similarly, the pseudo AB stacking regions are arranged on the lattice points of a hexagonal lattice as shown as blue circles in Fig. 1(b). Here, in AA stacking, every carbon atom in the sheet locates on the substrate carbon atom and the contact state is energetically unstable. By contrast, in AB stacking, every other carbon atom in the sheet locates on the substrate carbon atom and the contact state is stable. Thus, the contact state of the twisted graphene interface is not energetically flat although graphene is geometrically flat in nanoscale. In addition, from the results of the simulation, it was found that the stacking pattern moves while the sheet is slid. It indicates that the contact state of the twisted graphene interface varies during the sliding process of the sheet.

Next, for a general twisted crystal lattices of layered material, we consider a case when the sheet is rotated by the misfit angle θ and is translated by the sliding vector Δτ. When the deviation between the crystal lattices of the sheet and the substrate is expressed by σ₁ and σ₂, any lateral position τ can be written by the formula as a function of the rotation angle θ, the translational movement vector Δτ, and the crystal lattice deviations σ₁ and σ₂. According to the formula obtained above, when the sheet is slid, the stacking pattern also slides in the direction rotated by π/2 - θ/2 from the sliding direction of the sheet. This transelational movement of the stacking pattern can explain the smulated results. Thus the formula obtained above in this work can give a clear explanation to the contact mechanics of the twisted graphene interfaces.

[1] M. Hirano and K. Shinjo, Phys. Rev. B41, 11837 (1990).

[2] M. Dienwiebel et al., Phys. Rev. Lett. 92, 126101 (2004).

In recent years, there has been growing interest in transition-metal dichalcogenides (TMDs) due to their remarkable physical properties and potential applications of nanodevice materials. In 1T-TaS₂, a member of the TMD family, Ta atoms are aligned in the so-called “Star of David” pattern, caused from the periodic modulation of electronic charge density: charge density wave (CDW). At room temperature, 1T-TaS₂ is in the nearly commensurate CDW (NC-CDW) phase, whereas in the commensurate CDW (C-CDW) phase below 180 K [1].

We demonstrate direct observations of the CDW order in 1T-TaS₂ by non-contact atomic force microscopy (NC-AFM) as in Fig. 1. Although there are a lot of previous studies detecting the CDW orders by X-ray diffraction (XRD) and scanning tunnel microscopy (STM), observation by NC-AFM has not been reported. XRD can detect lattice distortion accompanied to the CDW phases; STM image indicates distribution of electronic states. In contrast, AFM can detect force between the tip and the sample including the electrostatic force, so it enables a possible direct visualization of the charge density distribution of the CDW.

Single crystals of 1T-TaS₂ used in this experiment were grown with the chemical vapor transport method and were cleaved under ultra-high vacuum conditions. AFM measurements were also performed in an ultra-high vacuum chamber at room temperature. The CDW triangle lattice structure was visualized through the attractive interaction between tip and sample. The period of the observed CDW order is approximately 1 nm, which is consistent with the previous results of STM measurements. In this presentation, we will discuss why the CDW can be observed in NC-AFM.

References:

[1] R. E. Thomson, B. Burk, A. Zettl, and John Clarke. Phys. Rev. B 49, 16899 (1994).

1. Introduction

Friction is a familiar phenomenon between two sliding surfaces and appears from macro to micro-scale. The effect of friction becomes more enhanced at the microscopic scale than at the macroscopic scale. To understand the microscopic-scale friction, experimental and theoretical studies on the peeling of atomic-scale graphene sheets adsorbed onto graphite substrate surface have been reported [1-4]. Although the molecular mechanics simulation studies on the atomic-scale peeling of the graphene sheet have been performed, the substrate surface discussed in these studies was focused on perfect crystal surfaces. Therefore, in this paper, the relationship between nanoscale peeling mechanics and surface defects is investigated using the model with surface defects.

2. Models and Methods of simulation

As shown in Fig. 1(a), we used a simulation model of the monolayer graphene sheet physically adsorbed onto the rigid graphite substrate surface. A molecular mechanics simulation of the peeling of graphene sheets adsorbed onto the graphite substrate with surface defects was performed. The peeling forces and energies acting on the graphene sheets were simulated to study effect of the size of the surface defect and the graphene sheet on the peeling process.

3. Results and Discussions

Figure 1(b) shows the critical peeling height, Δ z_s, where the transition from the surface to the line contact appears as a function of the ratio of the defect diameter (2r) to sheet width (W), 2r/W. The critical peeling height Δ z_s markedly changes depending on the number of sheet atoms N and the ratio 2r/W. As shown in Fig. 1(b), it is found that the mechanics of the free edge of the sheet can be classified into the following two cases: The transition to a line contact occurs (1) ‘after’ or (2) ‘before’ the free-edge passes through a defect. In the case (1), Δ z_s is nearly constant and the same as that for 2r/W=0.0. On the other hand, in the case (2), Δ z_s decreases compared to that in the case (1). Furthermore, during the peeling process, when the sheet slips just after the free-edge reaches the defect, the free edge jumps discontinuously to avoid the lattice defect. The energy dissipation is also observed simultaneously with discontinuous jumps. The energy dissipation increases as the defect size ratio 2r/W increases.

4. Conclusion and outlook

In this study, we investigated the effect of surface defects on the peeling of the graphene sheet. The remarkable effects of sheet size and surface defect size on the adhesion properties and the peeling process, i.e., the transitions from the surface contact to the line contact, were found. From the case of a sheet passing over a defect while maintaining the surface contact, the relationship between the energy dissipation and the defect size was found to affect frictional properties. Clarification of the relationship between surface defects and mechanical properties may lead to a new method of measurement and control of the nano-scale objects.

References

[1] N. Sasaki, H. Okamoto, N. Itamura, and K. Miura, e-J. Surf. Sci. Nanotech. 8, 105 (2010).

[2] M. Ishikawa, M. Ichikawa, H. Okamoto, N. Itamura, N. Sasaki, and K. Miura, Appl. Phys. Express 5, 065102 (2012).

[3] N. Sasaki, T. Ando, S. Masuda, H. Okamoto, N. Itamura, and K. Miura, e-J. Surf. Sci. Nanotech. 14, 204 (2016).

[4] R. Okamoto, K. Yamasaki, and N. Sasaki, Mater. Chem. Front. 2, 2098 (2018).

The realization of a novel lubric system, making it possible for nano- and micromachines to move easily, has been strongly desired. Such a lubricant can also contribute to moving a variety of macroscopic objects. We have recently developed graphite/C₆₀/graphite sandwiched systems [1], C₆₀ intercalated graphite films [2], and those consisting of alternating close-packed fullerene monolayers and graphite layers [3], which are expected to provide an exciting breakthrough in industrial development. In this paper, the mechanism of superlubricity of graphite/C₆₀/graphite interface is numerically studied using molecular mechanics [4,5].

As a model of the graphite/C₆₀/graphite interface, the close-packed C₆₀ monolayer inserted between two rigid graphite sheets is used. The periodic boundary condition within the (0001) plane is applied to the 1×1 unit cell. For each graphite interlayer distance, the metastable structure of the graphene/C₆₀/graphene interface is calculated by minimizing the total energy using the structural optimization, Polak-Rebiere-type conjugate gradient (CG) method.

First, the simulated interlayer distances of about 1.3 nm are in good agreement with previous experimental results. Next, the frictional feature along the commensurate direction of the graphene/C₆₀ /graphene interface is investigated. It is clarified that the small rotation and elastic contact of C₆₀ molecules are the origins of the superlubricity of the graphene/C₆₀/graphene interface along the commensurate scan direction. Anisotropy of the superlubricity of the C₆₀ bearing system is obtained [4,5]. Furthermore, the effect of the molecular orientation of C₆₀ on the Amonton-Coulomb's law is also studied, which tells us the close relationship between molecular friction and the molecular-scale local structures.

References

[1] K. Miura, S. Kamiya and N. Sasaki, Phys.Rev. Lett. 90, 055509 (2003).

[2] K. Miura, D. Tsuda, and N. Sasaki, e-J. Surf. Sci. Nanotech. 3, 21 (2005).

[3] M. Ishikawa, S. Kamiya, S. Yoshimoto, M. Suzuki, D. Kuwahara, N. Sasaki, and K. Miura, J. Nanomat. 2010, 891514 (2010).

[4] N. Itamura, K. Miura, and N. Sasaki, Jpn. J. Appl. Phys. 48, 060207(R) (2009).

[5] N. Sasaki, N. Itamura, H. Asawa, D. Tsuda, and K. Miura, Tribology Online 7, 96 (2012).

Introduction

Carbon nanotubes (CNTs) have excellent mechanical properties such as lightweight, strength, and elasticity. These mechanical properties make CNTs promising materials as nanodevices. However, on the nanoscale, nanodevices are subject to fracture due to frictional forces because of the enhanced surface effect. To solve these problems, it is necessary to study mechanics of CNTs on the nanoscale. In this study, double-wall CNT (DWCNT) is discussed in order to elucidate the mechanical properties of CNTs. Molecular mechanics simulation for rotation of the inner tubes of DWCNT is performed. The relationship between the stacking structure of the inner and outer tubes of DWCNT and the interaction energy was analyzed to reveal the nanomechanics of the inner-layer tube.

Model and method of simulation

In this work, the rigid DWCNT comprised of the inner tube ( (3,3) armchair-type single-walled CNT (SWCNT)) and the outer tube ( (7,7) armchair-type SWCNT), was adopted as the simulation model. The models of SWCNTs were created by structural optimization using the conjugate gradient method.

Simulated results

As shown in Figure 1(a), the interaction energy of the DWCNTs showed a periodicity of about 17° with respect to the rotation angle of the inner-layer CNT. The interaction energy took the minimum values at θ=4°, 21°, 38°, 55°, 72°, ... and the maximum values at θ=12°, 30°, 47°, 64°, 81°, ... . The above simulated results can be explained by focusing on the coaxial cylindrical structure of DWCNT.

Discussions

As illustrated in Figure 1(b), the inner- and outer-layer CNTs are hexagonal and 14-square prisms, respectively. Therefore, when the inner-layer hexagonal prism rotates, the set of the planes facing parallel to each other changes between the inner-layer hexagonal and the outer-layer 14-square prisms. Therefore, the planes of the inner- and outer-layer CNTs become parallel to each other every 8.6°of rotation angle of the inner-layer CNT. Furthermore, when the inner- and outer-planes become parallel, the lattice stacking between the inner and outer planes becomes the same every about 17°. Thus the simulated results can be explained by the periodicity of the geometry and lattice stacking of DWCNT for the rotation of the inner-layer CNT.

Conclusions

In this study, the relation between the interaction energy of DWCNT and its structure was clarified as a function of the rotation angle of the inner-layer CNT. The interaction energy was found to reflect the structural information such as geometry and lattice stacking between the inner- and outer-layer CNTs.

Magnetic skyrmions, shown in Fig. 1(a), are noncollinear magnetic structures. They are stable despite their nanometer scale size because they are topologically protected. Besides, they can be driven easily with small currents [1]. With these properties, they are promising for applications for memory technologies such as racetrack memories [2]. It is necessary to write and delete individual magnetic skyrmions for this application. This was achieved with spin-polarized current, electric field and magnetic field gradient [3, 4, 5]. As the information density should increase in the next generation memories, more localized methods for manipulation will be preferred. We focused on external force as a local means of manipulation. The aim of the present study is to control magnetic skyrmions with external force and to quantify it.

Individual magnetic skyrmions appear on Fe/Ir(111) when external magnetic fields are applied [4]. In the present study, local external force was applied to individual magnetic skyrmions on Fe/Ir(111) by bringing atomic force microscopy (AFM) tips closer to them. When an external magnetic field was applied to Fe/Ir(111), an individual magnetic skyrmion was observed with scanning tunneling microscopy (STM), as shown in Fig. 1(b). It had a bean-like shape, similar to those seen in the previous study [3]. The AFM tip was brought closer to the sample surface, and this resulted in annihilation of the magnetic skyrmion. Forces acting between the tip and the sample during the tip approach and the tip retraction were measured. The measurement clarified that they are different before and after the annihilation. In this presentation, we discuss what kind of force is acting between the tip and the sample during the annihilation to reveal the origin of the skyrmion annihilation. In addition, the creation and annihilation mechanism of individual magnetic skyrmions proposed in the previous research will be compared.

[1] N. Nagaosa and Y. Tokura, Nat. Nanotechnol. 8, 899 (2013). [2] A. Fert, V. Cros and J. Sampaio, Nat. Nanotechnol. 8, 152 (2013). [3] N. Romming et al., Science 341, 636 (2013). [4] P.-J. Hsu et al., Nat. Nanotechnol. 12, 123 (2017). [5] A. Casiraghi et al., Commun. Phys. 2, 145 (2019).

The frictional force is the sum of the forces required to break the real contact points produced by the adhesion of the asperities of two facing objects. It is known that the real contact area increases as the load increases. [1] However, the motion of atoms during a growth process of the real contact areas has not been fully discussed yet. Therefore, in this study, we discuss a structural change during indentation process and stick-slip motion on the 6H-SiC (0001) substrate surface, which has recently attracted much attentions as a power-semiconducting material that can significantly reduce heat loss and electrical energy.

The simulation model comprised of 6H-SiC(0001) substrate and a virtual rigid tip with a radius of curvature of 1.5 nm composed of helium atoms was adopted. The periodic boundary condition was applied to the unit cell within a lateral plane. As model potentials, Lennard-Jones and Tersoff potentials were used. The temperature, the time step, and the tip sliding velocity were set as T=100 K, t=0.1 fs, and v=10 m/s, respectively. The velocity-verlet method was adopted. First, the static optimized structure was prepared using the conjugate gradient method. Next, the thermal equilibrium state of the substrate surface was achieved using molecular dynamics simulation without tip for 10 ns. Then simulations of vertical and horizontal tip scans were performed. For the vertical scan, the tip was brought close to the substrate surface at a speed of 10 m/s to simulate the relationship between the loading force and the indentation depth of the tip apex atom. For the horizontal scan, the initial surface structure was prepared simulating the thermal equilibrium state a under a constant loading condition of 100 nN for 10 ns before the lateral scan of the tip was performed. Then, the lateral scan of the tip for the scan length of 3 nm was performed.

The simulated results are described below. First, under the initial thermal equilibrium state without a tip, the mean thermal oscillation amplitude of the surface atoms was reduced to 1 pm under T=100 K. Figure 1 shows the relationship between the loading force and the z-height of the tip atom during the vertical indentation process. An observed trend is that the loading force increases with decreasing tip height. However, once the loading force reached a peak value at the tip height of 0.173 nm, it rapidly decreases with a decrease of the tip height. The reason for this can be explained as follows: As the surface-atom height directly below the tip atom becomes lower, the number of substrate surface atoms except for those directly below the tip atom in contact with the tip increases, and whole the number of atoms supporting the tip increases. During the horizontal scan process, a friction force curve with a periodicity of approximately 0.3 nm was obtained, as shown in Figure 2. It is clarified that this period of stick-slip motion corresponds to the lattice constant of 6H-SiC(0001) of 0.308 nm.

[1] J. H. Dieterich and B. D. Kilgore, Pure and Appl. Geophys. 143, 283 (1994).

Introduction

Sum frequency generation (SFG) vibrational spectroscopy, which detects the light of the sum frequency of near-infrared (NIR) and mid-infrared (MIR) which resonates with molecular vibrations, is a powerful technique to elucidate structures and orientation of molecules on surfaces [1]. However, far-field SFG spectroscopy does not have enough spatial resolution, down to sub-μm, due to the diffraction limit of the light. Therefore, it has been difficult to observe the molecular behaviors that govern the reactivity of heterogeneous catalysts with nm-resolution. To tackle this problem, we have been developing the technique of tip-enhanced (TE)-SFG vibrational spectroscopy that induces the second-order nonlinear optical processes in the nanogap between a scanning tunneling microscope (STM) tip and a substrate [2]. In our previous developments, we have dealt with systems in which gold (Au), a coinage metal, is used for both the tip and the substrate, in order to obtain a large electric field enhancement effect of plasmons. On the other hand, in many cases, non-coinage materials, such as platinum (Pt) or nickel (Ni), were used in the field of heterogeneous catalysis [3]. In this study, we have attempted TE-SFG vibrational spectroscopy of adsorbed molecules on the surface of Pt, a typical non-coinage metal.

Methods

Pt-deposited mica substrates were immersed in 4-methylbenzenethiol (MBT) ethanolic solution, and an MBT-adsorbed Pt substrate (MBT/Pt) was prepared as a sample. An Au-deposited mica substrate was also used as a reference sample. The STM tip was fabricated by electrochemical etching of Au wire. The light source was a Yb-fiber laser whose output was split into a narrow-band filter and an optical parametric oscillator to generate narrow-band NIR light and tunable MIR light, respectively. The coaxially aligned NIR and MIR beams were focused on the tip-substrate gap, and the SFG signals were detected by a spectrometer. The contrast between the spectra acquired during the tip's approach to the sample in the constant-current mode of the STM and its retraction from the sample by 30 nm was extracted as the tip-enhanced signal. In addition, we estimated the intensity of the electric field enhancement on Au and Pt substrates by an electromagnetic simulation based on the finite difference time domain (FDTD) method [2].

Result & Discussion

In the case that vibrationally non-resonant MIR light (3000 cm^-1) was employed, we observed non-resonant TE-SFG signals from the MBT/Pt with the same shape as from the Au. These signals were observed only when the tip was approached to the substrate in the tunnel current region of STM. In a previous study of tip-enhanced Raman spectroscopy (TERS) [4], no signals were detected from the Pt substrate in the tunnel current region. The result suggests the existence of an enhancement mechanism unique to the SFG process, which was not observed in TERS. The result validated by the FDTD simulation can be explained by the comparable enhancement for both the Au and Pt substrates in the region of the NIR to MIR.

In the MIR region around 2900 cm^-1, where the CH resonance is situated, a dip structure was observed in the spectrum from the MBT/Pt, which is not found in Au. This dip is caused by the interference of the resonant SFG signal of methyl groups with the non-resonant signal of the Pt. Thus, we succeeded in obtaining the vibrational resonance TE-SFG signals originating from molecules adsorbed on Pt substrates. The applicability of TE-SFG vibrational spectroscopy to molecular systems on non-coinage surfaces has been demonstrated. This opens the way to the observation of heterogeneous catalytic reaction systems.

View PDF for the rest of the abstract.

Group-V elemental two-dimensional materials are attracting increasing attention because of their intriguing properties related to a topologically non-trivial phase. Recently, a flat honeycomb structure of bismuthene was distinguished on Ag(111), which was grown and kept at a low temperature [1]. The flat bismuthene on Ag(111) exhibit a (2×2) superstructure with Bi atoms located at the hollow sites of Ag(111), indicating that the distance between two Bi atoms is 108.7% of Bi(111) layer. The band calculation showed that a large SOC-induced gap was opened at the K point. Although the STS measurement revealed the metallic density of states (DOS) at both zig-zag and armchair edges near the Fermi level, knowledge of the electronic band structure in a wide energy and momentum range is still necessary to understand the intriguing electronic states of the ultraflat-Bi structure. In this work, two-dimensional band dispersion of (2×2) superstructure with Bi grown on Ag(111), which has been urged as an ultraflat hexagonal bismuthene, is investigated using angle-resolved photoemission spectroscopy (ARPES).

All measurements were performed at the plane grating monochromator (PGM) station of the Saga University Beamline (beamline 13) at the Saga Light Source [2]. Ag films with thicknesses of 110, 11, and 4 ML were prepared by depositing Ag onto Si(111) (√3×√3)-B surface at 100 K. The sample was then annealed at 300 K to yield a well-ordered Ag film. The deposition of Bi was performed at 100 K with a deposition rate of 0.05 ML/min, resulting in a (2×2) superstructure (Fig. 1(a)).

ARPES along the Γ-K line of the 110-ML-thick Ag(111) film on the Si(111) (√3×√3)-B surface is shown in Fig. 1(b). In addition to the Shockley surface state and the broad background intensity due to the indirect transition from the sp-band, a structure of the sp-band direct transition is observed; it crosses the Fermi level at the wave number of 1.1 Å^-1 when a photon energy of 19 eV is used. As shown in Fig. 1(c), the Shockley surface state disappears, and the intensity near the Fermi level at the Γ point also decreases after the growth of the (2×2)-Bi surface. The structure attributed to the sp-band direct transition is observed at the same position, whereas its intensity is slightly decreased. Notably, a band dispersing to the high binding energy side with tops at the first and second K points appears on the (2×2)-Bi surface. The observed dispersion is consistent with the calculated band with p_xy character in ultraflat-Bi of the (2×2) structure on 3 ML Ag(111) layers [1]. The Bi p_xy band of ultraflat-Bi grown on 11 and 4 ML thick Ag(111) is located at the same binding energy as that of the 110 ML thick Ag(111).

The Bi 5d peaks of (2×2)-Bi on Ag(111) films show apparent asymmetry, displaying a high binding energy tail. The asymmetric tails toward high binding energies reflect gapless excitations for metallic systems. The asymmetry index a at the (2×2)-Bi surface is 0.057, reflecting the interaction between the Ag sp and Bi sp states; however, it is smaller than the asymmetry index of 0.090 at the (√3×√3) surface, where one-third of the Ag atoms in the Ag(111) surface are substituted by Bi atoms to form a Ag₂Bi surface alloy. The smaller electronic coupling on the (2×2) surface is also consistent with the reported structure, where Bi atoms are arranged at the hollow sites of Ag(111) with (2×2) periodicity without substituting Ag atoms.

[1] S. Sun et al., ACS Nano 16, 1436 (2022). [2] K. Takahashi et al., AIP Conf. Proc. 2054, 040011 (2019).

When a topological insulator becomes a superconductor, Majorana particles, which are themselves antiparticles, appear on the surface or edges of the material. Since they are robust to environmental disturbances and can memorize the history of particle exchanges (non-commutative anyons), they have attracted much attention as a group of materials that can contribute to next-generation quantum computer devices. The candidate materials for topological superconductors (TSCs) include Cu_xBi₂Se₃ (chiral p-wave superconductor) [1] and Pb/TlBiSe₂ (s-wave superconductor/topological insulator junction) [2]. However, the superconducting transition temperature T_C of these materials is still low, below 3.8 K and 10 K, respectively. Meanwhile, Fe(Se,Te), which was demonstrated as a TSC by ARPES and STS in 2018, has the highest T_c among TSCs, reaching up to 14 K when the Se/Te ratio is adjusted [3].

Fe(Se,Te) is a van der Waals material, and it is thought that intercalation (IC) of foreign atoms and molecules between the layers is possible in addition to adsorption on the surface. Therefore, it is expected that both the carrier density and Debye frequency can be increased by IC with light elements such as hydrogen and alkali metals. In fact, in the sister system FeSe, the T_C increased from 8 K to 44 K by IC of hydrogen in a bulk sample using the solution IC method [4]. We have attempted to realize TSCs having even higher T_C by modifying Fe(Se,Te) with hydrogen and alkali metals.

First, the cleaved surface of bulk Fe(Se,Te) was irradiated with cracked hydrogen atoms under ultra-high vacuum (UHV) environment, and the temperature dependence of electrical resistance was measured using an in-situ independently driven four-probe electrical transport measurement system. As shown in Figure 1, the normal resistance of the sample exposed to 10⁵ Langmuir hydrogen decreases from 0.52 Ω/Sq. to 0.31 Ω/Sq. and the T_C increases from 10.2 K to 12.3 K. This is thought to be due to the increase in carrier density and/or in Debye frequency caused by hydrogen modification. In the presentation, we will discuss results based on more systematic data.

Reference: [1] Y. S. Hor et al., Phys. Rev. Lett. 104, 057001 (2010). [2] C. X. Trang et al., Nat. Commun. 11, 159 (2020). [3] M.H. Fang et al., Phys. Rev. B 78, 224503 (2008). [4] Y. Meng et al., Phys. Rev. B 105, 134506 (2022).

Majorana fermions, particles that are their own antiparticles, are important building blocks for topological quantum computation due to the non-Abelian statistics. The existence of Majorana zero mode (MZM) in the interface between a s-wave superconductor (SC) and a topological insulator (TI) has been predicted [1] and was observed in Bi₂Te₃/NbSe₂ heterostructures experimentally [2]. On the other hand, in contrast to conventional TIs with a single Dirac cone, a topological crystalline insulator (TCI) hosts multiple Dirac cones in the Brillouin zone [3-5]. Recently, a superconducting TCI was predicted to possess multiple MZMs [6, 7], which can encode more qubits in topological quantum computation. To create and manipulate multiple MZMs effectively in a SC/TCI heterostructure, a high-quality homogeneous interface and surface are essential. Unlike SC/TI systems, very few experiments have been performed for SC/TCI systems due to their novelty and difficulties in sample fabrication.

SnTe is known as a typical TCI and owns a rock-salt crystal structure. Unlike van der Waals materials, this type of structure requires lattice matching at the interface of SC/TCI heterostructures, which greatly increases the difficulties of sample fabrication due to the limited material choices. In 2018, a-few-layer-α-Sn(111) (namely stanene) was reported to be a superconductor when grown on PbTe(111) although neither bulk α-Sn nor PbTe shows superconductivity [8, 9]. This enlightened us because SnTe and PbTe have the same rock-salt structure and similar lattice constants, and even better SnTe is a TCI while PbTe is a trivial semiconductor. Therefore, if α-Sn(111) can be grown on SnTe(111) and exhibit superconductivity, it is expected to become a candidate for a SC/TCI heterostructure with multiple MZMs.

In this study, we have successfully fabricated a few-layer-thick single crystal α-Sn(111) on SnTe(111) using the underlayer of Bi₂Te₃(111)/Si(111) by molecular beam epitaxy. The schematic cross-section image of deposited layers is shown in Figure 1(a). Figure 1(b) shows the temperature dependence of resistance with varying thickness of α-Sn. Superconductivity emerges around 1 K in the samples with more than a 3-layer of α-Sn even though SnTe is not a superconductor at this temperature range. The observed maximum transition temperature is ~ 1.5 K in the sample with 4-layer-α-Sn. Furthermore, we found that our samples show two-dimensional superconductivity checked by measuring critical magnetic fields with rotating the out-of-plane angle of magnetic fields and by analyzing I-V curves with the Berezinskii–Kosterlitz–Thouless (BKT) theory. In the presentation, we will report more details about the superconducting properties including a one- to two-band superconducting transition and a huge in-plane critical magnetic field beyond the Pauli limit.

Reference

[1] L. Fu and C. L. Kane, Phys. Rev. Lett 100, 096407 (2008).

[2] H.-H. Sun et al., Phys. Rev. Lett 116, 257003 (2016).

[3] T. H. Hsieh et al., Nature commun. 3, 982 (2012).

[4] Y. Tanaka et al., Nature Phys. 8, 800 (2012).

[5] R. Akiyama et al., Nano Research 9, 490 (2016), J. Phys.: Conf. Ser. 568, 052001 (2014).

[6] X.-J. Liu et al., Phys. Rev. B 90, 235141 (2014).

[7] T. Mizushima et al., Phys. Rev. B 90, 184516 (2015)

[8] M. Liao et al., Nature Phys. 14, 344 (2018).

[9] J. Falson et al., Science 367, 1454 (2020).

Rare-earth mono-pnictides (REPn: RE = rare-earth element, Pn = N, P, As, Sb, and Bi) with the NaCl-type crystal structure have attracted much attention for a long time because of the characteristic magnetic properties originating from the interaction between RE’s localized 4f orbitals and Pn’s p orbitals, namely p-f mixing [1], such as an extreme magnetoresistance (XMR) [2], and complex magnetic phase transitions known as a "devil’s staircase" [3]. Very recently, some of these compounds have also been predicted to be topologically non-trivial due to the band inversion of RE’s d-orbitals and Pn’s p-orbitals at the X point of the bulk Brillouin zone, which has been experimentally demonstrated [4]. This fact makes REPns fertile enough to investigate the interplay between strong electron correlation and topology.

As a member of REPn, in addition to XMR [5], ytterbium mono-antimonide (YbSb) is also expected to be a heavy fermion system with a magnetically ordered ground state [6]. So far, an antiferromagnetic (AFM) ordering at 0.32 K and a mixing-type antiferro-quadrupolar ordering at 30 K in YbSb have been reported with M&oumlssbauer spectroscopy, specific heat measurement, and nuclear magnetic resonance (NMR) using stoichiometric polycrystalline samples [7-9]. However, there is no report on single-crystalline YbSb bulk samples because of the difficulty of the synthesis.

In this work, we fabricated YbSb(001) thin films by co-evaporating Yb and Sb on GaSb(001) substrates using a molecular beam epitaxial method and detected the electronic structure using an angle-resolved photoelectron spectroscopy (ARPES). We monitored the lattice constant during the sample fabrication using a reflection high energy electron diffraction. We confirmed that the fundamental diffraction streaks of the fabricated thin films appeared at the same positions as those of the substrate. This result suggests that the fabricated films have a similar lattice constant to that of GaSb (6.095 Å [10]), which is consistent with the predicted rock salt structure of YbSb (6.081 Å [11]). In the ARPES data of a 14-ML-thickness sample at the temperature of 7.5 K using 21-eV synchrotron light with p-polarization, we observed that the Sb 5p hole band is split into several bands like quantum-well states around the Γ point (Figure 1), as reported in other RESb thin films [12]. These results suggest the successful fabrication of single-crystalline YbSb thin films. In the poster session, we report the sample fabrication details and the observed ARPES data of YbSb.

[1]H Takahashi and T Kasuya, J. Phys. C: Solid State Phys. 18, 2697, 2709, 2721, 2731, 2745, 2755 (1985).

[2] L. Ye et al., PRB 97, 081108 (2018).

[3] T. Chattopadhyay et al., PRB 49, 15096 (1994).

[4] L-L. Wang et al., Commun. Phys. 6, 78 (2023).

[5] W. Xie et al., PRB 101, 085132 (2020).

[6] H. R. Ott et al. Solid State Commun. 55, 113 (1985).

[7] K. Hashi et al., JPSJ. 70, 259 (2001).

[8] A. Oyamada et al., JPSJ. 71, 297 (2002).

[9] A. Yamamoto et al., PRB 70, 220402(R) (2004).

[10] M. E. Straumanis et al., APL 36, 3822 (1965).

[11] J. Hayashi et al., Philo. Mag. 84, 3663 (2004).

[12] S. Chatterjee et al., Sci. Adv. 7, eabe8971 (2021).

In van der Waals (vdWs) materials, the engineering of stacking structures is a powerful method to control the electronic states for novel quantum phenomena or future electronic devices. To change the stacking structure, atomic intercalation is a promising way. This is because the atomic intercalation can modulate the stacking structure without a highly technical protocol nor a limitation for the sample size [1-4]. However, there is little knowledge of the atomic dynamics during the intercalation process related to the stacking structures.

Here, we report that Li-intercalation into epitaxial graphene (EG) on SiC (0001) drives topological domain wall motions associated with stacking change in graphene/buffer layers [5]. In-situ ultrahigh-vacuum (UHV) aberration-corrected low-energy electron microscopy (LEEM) has been applied to real-time imaging of the Li-intercalation process. The buffer layer and the top graphene layer forming a carbon bilayer configuration (named the bilayer system thereafter) are used for the observation of Li-intercalation dynamics (Fig. 1(a)). AB and BA stacking domains are alternatively distributed and sandwich topological domain walls (TDWs) in between (Fig. 1(b)). The TDWs are topologically protected due to the difference in the number of unit cells in the buffer and graphene layers. The crossing points of multiple TDWs have a topologically protected AA stacking structure. The top of Fig. 1(c) is a dark-field LEEM image of the pristine bilayer system. The dark and bright domains correspond to AB and BA stacking domains, respectively. They are consistent with the top illustration of Fig. 1(b).

We deposited Li on the bilayer system with a low flux rate to capture the Li-intercalation dynamics. At 7 min of the Li-intercalation, Li-intercalation firstly occurred at AA stacking points, which changes the contrast of the bright-field LEEM image to bright dots as shown in stage 1 in Fig 1(c). At 9 min (stage 2 in Fig 1(c)), Li-intercalated domains with the bright contrast grow into AB stacking domains (surrounded by or between blue dashed lines in Fig 1(c)). Further Li-intercalation makes the Li-intercalated domains extend into BA stacking domains (surrounded by or between red dashed lines in Fig 1(c)) as stage 3 in Fig 1(c). Finally, the Li-intercalated domains cover most areas, whereas they do not combine with each other and are divided by dark lines (i.e. TDW regions).

To elucidate the mechanism of the stacking-dependent Li-intercalation, we performed density functional theory calculations. Li adsorption energy into AA, AB, and BA stackings is in the order of AA < AB < BA. Thus, it is energetically preferable for Li to intercalate the AA stacking points first and then selectively intercalate AB domains instead of BA domains as observed in the LEEM snapshots. In addition, we calculated the stable stacking structure of the Li-intercalated bilayer system and figured out that it is AA stacking.

Next, we look into the evolution of the stacking distribution during the Li-intercalation in the stripe microstructure as seen in the pink rectangle regions in LEEM snapshots of Fig. 1(c), with molecular dynamics simulation. When Li-intercalated domains start to grow into AB regions, the intercalated regions change their stacking structure to AA (Stage 2 of Fig. 1(d)). Further Li-intercalation changes the BA stacking to AA stacking structure. Finally, Li-intercalated domains with AA stacking fill the whole region, while there is a TDW region in between them as shown in Stage 3 of Fig. 1(d). The TDW region can not have AA stacking structure by the topological constraint, and never disappear.

View PDF for the rest of the abstract.

When ferromagnetism is induced into a topological insulator (TI), the Zeeman gap is created in the surface Dirac Cone. By tuning the Fermi level to be in this gap, the dissipation-less conduction state, called the chiral edge state appears at the sample edge by quantum anomalous Hall effect (QAHE). QAHE has attracted much attention because of its potential for energy-saving spintronics devices that operate even in zero magnetic fields and has already been reported in (Bi_1-xSb_x)₂Te₃(Z₂-type TI), doped with magnetic atoms such as Cr and V [1]. On the other hand, it has been theoretically suggested that doping ferromagnetism on the surface of topological crystalline insulator (TCI) also results in QAHE [2]. TCI is a new type of TI, and whereas nontriviality is guaranteed by the time-reversal symmetry in a conventional Z₂ type TI that in TCI is protected by the mirror symmetry of a crystal possessing multiple Dirac Cones (DCs). It is expected that this TCI’s unique characteristic will lead to the realization of high-Chern-number QAHE which can be enabled by multiple DCs applicable to novel next-generation energy-saving devices with multiple switchable channels [3,4].

We fabricated sandwich structures Cr_x(Bi_1-ySb_y)_2-xTe₃ (CBST)/ Pb_xSn_1-xTe (PST)/CBST as shown in Fig. 1 (a) for introducing ferromagnetism into the TCI system. PST is a typical TCI material, and its Fermi level and the size of the non-trivial gap are tunable by changing Pb/Sn ratio because SnTe and PbTe are p- and n-type, respectively. According to previous studies on Z₂ type TI, the observation temperature of QAHE in a sandwich structure using magnetic modulation doped layers is much higher compared to a random doped system. Hence, for aiming for QAHE in a TCI system, we also adopt the modulated doped sandwich structure instead of a random doped system. The ferromagnetic layer Cr_x(Bi_1-ySb_y)_2-xTe₃ (CBST) is the modulation doping layer of Cr, and it should be an insulator. For that, we investigated the effects of the amount of Cr doping, resistivity, carrier type, and carrier density. So as to realize QAHE in TCI, it is necessary to fine-tune the Fermi level to be the gap and to show ferromagnetism enough with keeping crystallinity also.

Regarding magnetic properties of sandwich structures, we observed clear hysteresis ~ in magnetization-magnetic field (M-H) curves as shown in Fig. 1(b), and coercivity (~10 kOe at 2 K) is more than five times larger than Sn_0.9Cr_0.1Te with randomly doped magnetic elements [4]. About electrical transport properties, on the other hand, no clear hysteresis due to anomalous Hall effect was observed in this CBST/PST/CBST structure, although QAHE was reported in a similar structure based on Z₂ TI, CBST/BST/CBST[5]. In this presentation, we would like to discuss this cause.

[1]C. -Z. Chang, et al. Science. 340, 167 (2013). [2]C. Fang, et al. PRL. 112, 048801 (2014). [3]R. Adhikari, et al. PRB. 100, 124422 (2019). [4]F. Wang, et al. PRB. 97, 115414 (2018).[5]M. Mogi, et al. APL. 107, 182401 (2015).

Introduction

Organic doping has recently been attracting renewed attention as a carrier enhancement method to improve the performance of devices using organic semiconductors. The efficiency of carrier injection into organic semiconductors is considered to be determined by the difference between the molecular orbital energy levels in addition to the relative position of each molecule. At metal electrode interfaces, the molecular orientation at the surface and the distance of the energy levels of the frontier molecular orbitals from the Fermi level is important. In this study, the structural changes of the perylene monolayer on the Au(111) surface induced by the introduction of bromine molecules as a hole-dopant were observed by scanning tunneling microscopy (STM) in ultrahigh vacuum (UHV).

Experiment

Perylene was deposited on a cleaned Au(111) surface under UHV at room temperature (RT). Bromine was introduced by electrolysis of heated AgBr. STM observations were performed under UHV at RT with PtIr tips.

Results and Discussion

Figure 1a shows the STM image of a monolayer of perylene formed on an Au(111) surface. The molecules are adsorbed with the carbon skeleton plane parallel to the substrate, forming a superstructure with a period of (4x4) [1]. Figures 1b-1d show STM images after the introduction of bromine. In the initial stage of bromine introduction, bromine is considered to adsorb on perylene in molecular form and induce partial charge transfer. In Figure 1b, the perylene molecule retains the (4x4) structure, but the contrast is changed due to the adsorbed bromine molecules. A model of an upright adsorption of a bromine molecule on top of each molecule is shown in Figure 1b.

In another region of the perylene monolayer with the same amount of bromine dosage, the molecules changing in-plane orientation and alignment are observed as shown in Figure 1c. Some protrusions in the rows indicated by the arrows are brighter than those in other rows, suggesting that the molecular planes are tilted with respect to the surface. On the other hand, the molecules in the other rows rotate in the in-plane direction. These changes are caused by bromine molecules penetrating beneath the perylene monolayer. In fact, carbon K-edge X-ray absorption fine structure spectroscopy (C K NEXAFS) and X-ray photoelectron spectroscopy of the bromine 3d orbital (Br 3d XPS) show that the perylene molecules tilt as the bromine amount increases and the bromine is dissociatively adsorbed on the gold surface [2].

The adsorbed bromine pushes up the work function of the gold surface, causing the highest occupied molecular orbital (HOMO) level of the perylene to exceed the Fermi energy and transfer electrons to the substrate, resulting in cations [2]. Figure 1d shows the STM image when an efficient amount of bromine is introduced. Since the unit cell is about twice the size of (4x4), two molecules should be included. Cofacial pi-pi stacking structure model is proposed as shown in Figure 1d. Providing the bromine and perylene are separated and form a one-dimensional stacking structure, it will be an interesting model system for charge-transfer complexes.

References

[1] C. Seidel, et al., Phys. Rev. B, 64, 195418(2001).[2] O. Endo, et al., J. Phys. Chem. C 126, 15971 (2022).

Background

Diarylethene (DAE) molecules are known for its photochromism and are expected to be used as switching devices due to their exceptional thermal stability and reaction rate compared to other photochromic molecules [1]. Two isomers are called closed-form (C-DAE, Fig. 1(a)) and open-form (O-DAE, Fig. 1(b)). To date, the reversible and controlled isomerization reactions of single molecules have not been realized on metallic substrates. This is attributed to the interactions between the adsorbed molecule and substrate metal, such as charge transfer and orbital hybridization, which induce modifications to the intrinsic characteristics of the system. In order to examine the effects of molecule-substrate interactions and intermolecular interactions on the adsorption behaviors, the present study undertakes a comparative analysis of the adsorption structures of the DAE molecules on three noble metal surfaces, specifically Au(111), Ag(111), and Cu(111). Previously we reported the adsorption structures of DAE molecules on Cu(111) and Au(111) [2]. In this study, we aim to gain insight into how the intermolecular and molecule-substrate interactions affect their adsorption on Ag(111).

Method

Experiments were performed using a scanning tunneling microscope (STM) operated in ultrahigh-vacuum (UHV) at cryogenic temperatures (4 K). A single crystal of Ag(111) was cleaned by repeated cycles of Ar ion sputtering for 20 minutes and annealing at 500 ^oC for 30 minutes. Subsequently, O-DAE molecules were deposited using a K-cell, heating at 103-105 ^oC for 10 seconds. The temperature of the Ag surface during the deposition was kept either below 0 ^oC or at about 50 ^oC (Hereafter we call these samples A and B, respectively). We also observed samples A and B after annealing at 50 ^oC for 30 minutes. This is to investigate the effect of isomerization by mild-annealing [2] on the formation of clusters or superstructures.

Results and Discussion

An overview STM image of sample A is shown in Fig. 1(c). There exist some characteristic cluster structures on this surface. Enlarged images are also shown in Figs. 1(d) and 1(e). The manipulation of the clusters using an STM probe tip initiated their decomposition. This process unveiled that the cluster, as shown in Fig, 1(d), comprised seven C-DAE and six O-DAE molecules. Another type of cluster, as shown in Fig. 1(e), has a zig-zag structure wherein linearly aligned O-DAE molecules bent due to the attachment of two C-DAE molecules. After the mild-annealing, another type of three-fold symmetric structure was observed as shown in Fig. 1(f). This cluster consists of nine C-DAEs.

Sample B, with a substrate temperature during the deposition higher than that of sample A, exhibited a greater abundance of 3-fold symmetric structures in comparison to sample A. Subsequent to annealing, nearly all clusters on sample B bore a remarkable resemblance to those present on sample A. This can be attributed to the probable isomerization of all O-DAE molecules to C-DAE upon their adsorption on the substrate at 50 °C. The post-annealing process thus did not have any effect on sample B. The numbers of molecules contained within the clusters exceed those observed in the case on Au(111) [2], which implies that the intermolecular interactions on Ag(111) play a dominant role in the formation of clusters on Ag(111).

Conclusion

The deposition of two isomers of DAE on Ag(111) resulted in the formation of several distinct types of highly symmetric clusters, depending on the deposition and post-annealing conditions.

View PDF for the rest of the abstract.

Fe₃O₄ nanoparticles are good catalysts for, e.g., ammonia synthesis and the water-gas shift reaction. Catalytic efficiency varies depending on the surface orientation and molecular activity at adsorption sites, and thus clarifying surface structures of Fe₃O₄ nanoparticles at the atomic scale is crucial for a better understanding of catalytic reaction mechanisms. Although numerous studies reported the observation of nanoparticles using transmission electron microscopy [1], they only revealed shapes and sizes of nanoparticles, but not the surface structure. Because Fe₃O₄ nanoparticles are often functionalized by organic ligands such as oleic acid, direct observation of surface structures is challenging. In this study, we investigate, using scanning tunneling microscopy (STM), the effect of ultraviolet (UV) ozone treatment and annealing in ultra-high vacuum (UHV) on the removal of oleic acid from the surface of Fe₃O₄ nanoparticles. Because flat samples are essential for STM observations, we employed a Langmuir-trough to fabricate monolayers of Fe₃O₄ nanoparticles. The solution of Fe₃O₄ nanoparticles (Ocean NanoTech, LLC) was 100 times diluted with chloroform. The diluted solution was gently placed on the water subphase with a microsyringe and transferred to the Au (111) thin film on mica. The prepared sample was first treated by the UV ozone cleaner for 20 min. We then checked the overall sample morphology using amplitude-modulation atomic force microscopy (AFM) in ambient. As shown in Fig. 1(a), we confirmed that the nanoparticles existed as monolayers. The monolayer was composed of tiny islands (approximately 50-60 nm wide), which did not appear before the UV ozone treatment. It may indicate the removal of oleic acid. Next, we introduced the sample to an UHV-STM and conducted STM observations. We found Fe₃O₄ nanoparticles, but the probe tip was easily contaminated during the scan. This implies that oleic acid still remains, or air-gas molecules are adsorbed on the nanoparticle surfaces. We thus annealed the sample up to 510 ℃ in UHV until the base pressure was recovered. After the annealing procedure, we were able to scan more stably, and Fig. 1(b) shows an example of STM images. Since the diameters of nanoparticles are approximately 15 nm according to the company specification, the island shown in Fig. 1(b) is likely an aggregate of three nanoparticles. Although the Fe₃O₄ nanoparticles are supposed to exist as polyhedral structures [2], the shapes of nanoparticles varied. On the surface of nanoparticles, we found many protrusions separated by1-2 nm. The distance between protrusions is larger than the expected atomic arrangement of Fe₃O₄, but the density is smaller than that of oleic acid on the Fe₃O₄ (001) surface [1]. These differences may imply the removal of oleic acid, but there is a possibility that residual organic compounds, decomposed oleic acids, cause surface protrusions. Further verification is needed.

References

[1] Dreyer, A., et al. Organically linked iron oxide nanoparticle supercrystals with exceptional isotropic mechanical properties. Nat. Mater. 15, 522-528 (2016).

[2] Meng, Y. et al. Prediction on morphologies and phase equilibrium diagram of iron oxides nanoparticles. Appl. Surf. Sci. 480, 478-486 (2019).

Detailed knowledge of metal-carbon interactions is indispensable for generating graphene-based nanostructures for application in catalysis, electrodes, and magnetic storage[1,2]. Graphene has been grown on many transition metal surfaces by chemical vapor deposition, where the layer forms a moiré superstructure due to the mismatch between the lattice constants of graphene and those of the underlying metal surface[3]. Consequently, the structure and electronic states of graphene are modulated depending on the region in the moiré unit cell. Such moiré patterns have been used as templates to grow uniformly sized metal clusters. During cluster formation, metal atoms impinge to the surface, followed by thermal diffusion and nucleation. In this process, the metal atoms prefer to interact with specific sites in the unit cell, which was predicted to be responsible for the formation of uniform cluster arrays. In this work, we studied the interaction of Ag atoms with graphene supported on Rh(111) by using a scanning tunneling microscope (STM)[4]. Because of their unique localized plasmon response, Ag nanoclusters have been studied as optical materials. Therefore, we also investigated the correlation between the structure of Ag clusters and their optical properties.

Upon adsorption at 15 K, the Ag atoms existed mainly as monomers at specific sites in the moiré unit cell because thermal diffusion was suppressed. The Ag atoms exhibited a dI/dV peak at ∼0.5 V above the Fermi level, which was assigned to Ag(5s)-C(2p_z) anti-bonding state. At elevated temperature to 78 K, the Ag adatoms were thermally activated to diffuse and form clusters, whose sizes were smaller than ~5 nm. At room temperature, the clusters did not exist on graphene terrace and they grew preferentially at the steps. In the corresponding extinction spectra, surface plasmon responses due to Ag clusters were observed, and their peak energies changed depending on the growth temperature.

[1] X. Liu et al., Progress in Surface Science 90, 397-443 (2015).

[2] A.T. N’Diaye et al., Phys. Rev. Lett. 97, 215501 (2006).

[3] J. Wintterlin and M.-L. Bocquet, Surface Science 603, 1841-1852 (2009).

[4] H. Okuyama, D. Yamamoto, S. Hatta, T. Aruga, Carbon 210, 118032 (2023).

Among more than 200 polytypes of SiC, the cubic zinc blende (3C) polytype and he hexagonal 4H and 6H polytypes have been most intensely studied. The polar surfaces of SiC are technologically important because they are commonly used for SiC epitaxial growth. Although there is a variation of the fundamental energy gap by about 1 eV between the 3C and 4H polytypes, the 3C-SiC(111) and hexagonal nH-SiC(0001) with n Si-C bilayers in a unit cell exhibit many similarities. The (√3&times√3)R30^&ordm reconstruction is one of the intrinsic surface phases of 3C-SiC(111) and nH-SiC(0001) surfaces and is observed experimentally [1]. The model with Si adatom on the T4 site which is the on-top site of an underlying second layer is most favorable for the Si-terminated (√3&times√3)R30^&ordm reconstruction [2]. Our previous calculations have shown energetics for surface steps with the Si-terminated (1&times1) reconstructed SiC(111) surface and the microscopic atomic structures have been identified [3,4]. We considered the five types of steps appearing on the (111) vicinal Si-terminated surfaces inclined toward either the <-12-1> or <-101> direction. We found that the <-12-1> straight step is energetically more favorable than the straight <-101> step, and that either the straight or the meandering step edge, depending on the inclined direction of the vicinal SiC surfaces, emerges. We report here first-principles calculations that clarify the dependence for the surface steps on Si-terminated 3C-SiC(111) surfaces on the surface reconstructions.

The calculations are performed using RSDFT code [5] which is based on the real-space finite-difference pseudopotential method and is the most suitable for the current massively parallel architecture. Figure 1 is a schematic illustration of five distinct atomic steps (Si2, Si3, C1, C2, and SC) on the Si-terminated 3C-SiC(111) surface with an (1&times1) reconstruction. In the case of the (√3&times√3) reconstruction, we use the periodic-array slabs containing 400–600 atoms to model Si2+Si3 steps, Si2+C1 steps, and 2 SC steps as the step-edge configurations. There are several distinct structural arrangements for each pair of steps, depending on the distances between the step edge and the Si adatoms on the terrace. We have found that the geometries in which the Si adatoms are located at off-edge sites are generally lower in energy. We observe the rebonding of the Si bonds between the upper- and lower-terrace Si edge atoms, which is also observed in the case of the (1&times1) reconstruction. The calculated formation energies for the Si-rich and Si-poor conditions are considered. We have found that the order of the formation energy for each step pair is identical to that obtained for the case without the (√3&times√3)R30^&ordm terrace reconstruction. The formation-energy differences between the most and the next stable step pairs are the same in the range of less than 70 meV/&Aring. We conclude that the structural feature and the energetics of steps on 3C-SiC(111) surfaces are essentially the same for (1&times1) and (√3&times√3)R30^&ordm reconstructions on the terrace.

[1] P. M&aringrtensson, F. Owman, and L. I. Johansson, Phys. Status Solidi B 202, 501 (1997).

[2] M. Sabisch, P. Kr&uumlger, and J. Pollmann, Phys. Rev. B 55, 10561 (1997).

[3] K. Seino and A. Oshiyama, Appl. Phys. Exp. 13, 015506 (2020).

[4] K. Seino and A. Oshiyama, Phys. Rev. B 101, 195307 (2020).

[5] J.-I. Iwata et al., J. Comput. Phys. 229, 2339 (2010).

The authors have been working on integrating atom probe (AP) mass analysis and structural observation based on field ion microscopy (FIM) [1]. FIM has an advantage in image quality against ion mapping by 3D-AP data. To combine the FIM and the AP, the image gas ion must be blocked to go into the ion detector of the mass spectrometer. In the conventional FIM-AP system, the toroidal lens tuned for the energy of the field evaporated ion served as the energy filter. Though FIM observation and AP analysis can be done in an identical system in this way, there is one disadvantage the probed region itself is not visible. The authors are trying to eliminate the difficulty by introducing an image deflector driven by an AP analysis system.

The deflector installed between the specimen and the phosphor screen with a probe hole shifts the FIM image projected on the screen to show the probed region image not visible during AP analysis. During the AP triggering sequence, the deflectors are not biased to make no influence on the trajectory of image gas ions. When a field-evaporated ion is detected, the acquisition system stops the trigger. Then the deflectors are biased at +/-1 kV and the CCD camera captures the shifted FIM image. After the image acquisition, the system reverts to AP analysis. Thus, the image data of probed region and the mass analysis data are synchronized in the analysis.

For the improvement of the preliminary system reported elsewhere [1], the authors have been working on the following points; (1) the boost of deflector bias voltage to obtain the efficient shift of the FIM image, (2) the reduction of back pressure during FIM-AP analysis. The bias voltages of +/-1 kV were achieved. As the image shift is almost proportional to the amplitude of applied voltage on the deflector plates, the shift stroke of the image is 1.5 times of probe hole diameter as shown in Figure 1. There is no overlapping of the probed region and the probe hole.

The reduction of back pressure in the time of flight mass spectrometer is required for the improvement of the detection efficiency of the AP system. The image gas (helium, typically) of the 10^-3 Pa region is introduced for the FIM image observation. The ionized gas emitted by the specimen is eliminated by the energy filter (toroidal lens) in front of the ion detector but the neutral species also disturb the flight of the field evaporated sample ions. To minimize the dose gas amount, the dose tube of 0.8 mm I.D. with pre-cooling was mounted at a distance of 5 mm from the sample tip. The evacuation pump layout was also arranged to reduce the back pressure in the spectrometer. By these modifications, the back pressure measured in the toroidal lens was reduced to a 10^-4 Pa order. Other results will be also shown in the presentation.

[1] M. Taniguchi, Y. Yamauchi, K. Yoshikawa, J. of Vac. Sci. Technol. B, 41, 042806 (2023). https://doi.org/10.1116/6.0002607

Alloys between different kind of metals on surfaces often exhibit novel structures comparing to those in the bulk. For example, Ni and Au do not form alloys in the bulk form, but Au adsorbed on the topmost surface layer of Ni is known to induce surface two-dimensional alloying by replacing some of the Ni atoms on the surface with Au [1]. Since ultrathin films on metal surfaces will a unique structure, it will have high expectations for the discovery of a novel growth modes from the conventional thin film growth modes (FM, VW, SK) and for the development of new functionalities.

It is known that Ni-Ag, like Ni-Au, does not mix atoms in bulk form, and that Ag thin films on a Ni(111) surface grow in layer-by-layer and form a moire pattern due to the lattice mismatch [2], and that Ag(111) is formed on a Ni(100) surface has been confirmed in our laboratory. Although there have been many studies on metal heteroepitaxial growth by STM since the 1990s, there have been few observations of thin films of fcc metals at the atomic level, and the growth pattern of thin films, including interdiffusion and alloy formation, has not been clearly established at present. In this study, we studied the Ag-induced surface structures on the Ni(110) surface (Ni(110)-Ag) that has not been observed by ultra-high vacuum scanning tunneling microscopy (UHV-STM) and low-energy electron diffraction (LEED), depending on the growth and annealing conditions.

UHV-STM observation of the Ag-deposited surface on Ni(110) substrates at room temperature showed that Ag did not alloy with the substrate Ni atoms, but formed a reticulate structure. When deposited at 200°C, the reticular structure was aligned to form a stripe structure along the [001] direction. A 10×10 nm² STM image (Fig. 1) of the RT deposition shows that three regions coexist. Region A is a bright region with a periodic protrusion in the [001] direction, which corresponds to the substrate atomic rows. Therefore, Ag atoms are considered to grow epitaxially in the [11-0] direction to form bright bands in the [001] direction. They occasionally form wavy structures with a 3× periodicity in the [11-0] direction. Region B is noisy, where Ag atoms do not match the periodicity of the substrate in the [11-0] row and they are considered to fluctuate in the row. Region C is a trough structure with a depression of about 120 pm and is considered to be a region where no Ag atoms exist.

LEED observation showed one-tenth of the satellites spots near integer-order spots of the Ni(110) 1×1 structure in the [11-0] direction, indicating that epitaxially grown Ag atoms have a super structure along the [11-0] row.

References

[1] L. Pleth Nielsen et al., Phys. Rev. Lett. 74, 1159 (1995).

[2] S. Nakanishi et al., Phys. Rev. B 62, 13136 (2000).

One of the indispensable fundamental technologies supporting the current information society is semiconductor devices with minute three-dimensional (3D) shapes, and the exploration of the physical properties derived from the 3D surfaces and 3D shapes that are the stage for electron transport etc. This is extremely important for improving performance and developing new functions, which requires a 3D surface with neatly ordered atoms and sharp edges. We have created µm-scale 3D Si structures with both edge geometry and surface ordering by combining processing and surface treatment technologies, and explored the physical properties derived from the edges and 3D surfaces, such as unique electrical conduction properties [1-3].

The quantum size that is expected to be localized in the specific region such as the boundary edge region of the 3D surface is about 10 nm, and more integrated sub-µm-scale 3D Si structures are required for the coupling and manifestation of quantum phenomena through the faceted surface. In general, creating atomically ordered Si {111} faceted surfaces requires flash annealing at over 1000℃. On the other hand, the sharp edge shape cannot be kept by heating above 1000℃ because the melting point drop becomes significant in the sub-µm-scale 3D Si structure due to the size effect. Therefore, we aimed to create atomically ordered faceted surfaces and sharp-edged structures in sub-µm-scale 3D Si line structures by introducing pretreatment of RCA cleaning[4], which is expected to form surface order at heating below 1000℃.

Three-dimensional facet line samples were fabricated on a Si (001) substrate with equivalent {111} planes by dry etching and wet etching, and the edge-to-edge spacing (pitch size) was 0.4 µm. The specimens were then cleaned with a piranha solution and immersed in hydrofluoric acid (25%) for 5 to 10 minutes. The sample surface was hydrogen-terminated by this RCA cleaning. The sample was degassed (550℃) for 3 hours in an ultra-high vacuum (base < 10^-8 Pa), followed by flash annealing (50℃ increments above 800℃, one flash only). The diffraction spot intensity was evaluated from LEED observations (30-120 eV), and the amount of change in edge width was evaluated from SEM observations.

The results of heating the RCA-cleaned facet line sample with a pitch size of 0.4 µm to 850℃ are shown from the SEM images and LEED spots in the figure. The SEM images show that the edge shape is maintained, and the LEED spots confirm the surface order of a 7×7 structure. The flash annealing temperature dependence of the edge width change in the facet line samples shows that the edge shape changes significantly with increasing heating temperature with or without RCA cleaning. On the other hand, the flash annealing temperature dependence of the surface ordering was examined, and diffraction spots showing strong surface ordering were observed in the RCA-cleaned sample from heating at lower temperatures than those in the unwashed sample. This result indicates that RCA cleaning is effective for the formation of surface order from low temperature heating.

In conclusion, we determined that the optimal heating temperature for the RCA-cleaned facet line sample with a pitch size of 0.4 µm is 850℃, and succeeded in fabricating a sub-µm-scale Si line structure with atomically ordered faceted surfaces and sharpen edges.

[1] S. Takemoto et al., Jpn. J. Appl. Phys., 57 (2018) 085503.

[2] A. Irmikimov, K. Hattori, et al., ACrystal Growth Des. 21 (2021) 946.

[3] K. Hattori, et al., e-JSSNT 30 (2022) 214.

[4] A. Ishizuka and Y. Shiraki, J. Electrochem. Soc. 133, 666 (1986).

The morphology and structure of metal films on oxide underlayers were studied fruitfully in the field of electric field-controlled devices such as MOSFETs. 3D metal islands were consistently observed in the metal/oxide system due to the large surface energy difference between them. How to eliminate the bumps of the ultrathin metal films was an important issue in ultrathin and nanostructure devices. For metal film deposition on the Al₂O₃ underlayer, the molybdenum seeding layer was a good choice for epitaxial growth. The aim of this study is to investigate the early stage of the growth of Mo film on the Al₂O₃(0001) substrate.

Two interesting phenomena were observed in the study. First, a flattened molybdenum circular disk was obtained on the Al₂O₃ (0001) substrates as the thickness(t) and growth temperature (Tg) of Mo are lower than 10nm and around 450^oC, respectively. The disk-like shape Mo islands with a body-centered cubic (bcc) crystal structure as shown in Fig. 1(A) for the morphology and Figs. 1(B)-1(D) for different thicknesses of 5, 20, and 50nm, respectively. The surface energy of Mo and Al₂O₃ are 2907 erg/cm² [2] and 690 erg/cm² [3], respectively. Disk-like morphology conflicts with the 3D islands that are usually observed in a higher surface energy (Es) material growth on a lower one.

Secondly, a Kurdjurmov-Sadis (KS) or a Nishiyama-Wassermann (NW) relationship between Mo(110) and Al₂O₃(0001) is expected as shown in Figs. 1(H) and 3(I), respectively. Little strain energy should be contributed from the crystal asymmetry between the rectangular and hexagonal lattices. In this system, a Mo fcc(111) was observed for thicknesses larger than 100nm as depicted in Figs. 1(E)-1(G) for thicknesses= 100, 200, and 500nm, respectively. A strain-release process seems to take place as films get thicker.

REFERENCES

[1] K. N. Tu, J. W. mayer,& L. C., Electronic thin film science for electrical engineers and materials scientists(1992): p.113-124

[2] Vitos, L., et al., The surface energy of metals. Surface science, 1998. 411(1-2): p.186-202.

[3] Magkoev, T. and G. Vladimirov, Aluminium oxide ultrathin-film growth on the Mo (110) surface: a work-function study. Journal of Physics: Condensed Matter, 2001. 13(28): p. L655

[4] Rashid, H., et al., Physical and electrical properties of molybdenum thin films grown by DC magnetron sputtering for photovoltaic application. Results in Physics, 2019. 14: p. 102515.

An amorphous carbon film has a lot of useful properties such as chemical/mechanical stability, bio-compatibility, and so on. Then, it is one of promising coating materials for medical apparatus such as stent and artificial organs. For medical applications, amorphous oxidized carbo films are required to add the hydrophilicity to the films [1]. Although plasma oxidation of amorphous carbon films is a useful method, this process is two-step process. One step process is desirable; a deposition technique of amorphous oxidized carbon films is required. Then, the oxidized carbon films are deposited with plasma enhanced chemical vapor deposition (PECVD) method using ether-molecules as sources. Ether molecules have a molecular backbone of -O-C-O-. The backbone is terminated with alkyl groups. Alkyl groups includes ethyl, propyl, butyl. If these alkyls are chosen, the carbon contents of deposited film can be changed. In this study, we compare the deposited films using diethylether, dipropylether, dibutylether, as source molecules. To understand the deposition process, in-situ and real-time multiple internal reflection infrared spectroscopy (MIR-IRAS) was adopted. Plasma was generated by feeding 30W rf power in the molecular pressure of 50 mTorr. Fig. 1 shows the IR spectra acquired of the deposited films. Each deposited film has O-H stretching mode, C-H stretching mode and C=O stretching mode. The formations of these species are due to generation in the plasma; source molecules are decomposed and polymerized in plasma. Compared with C-H peak, the relative intensity of O-H peak is strong in the films with diethyl-ether. It indicated the source molecules are affected to the density of C=O in the deposited film.

Reference [1] T. Nakatani, et al., J. Photopolym. Sci. Technol. 21, 225 (2008).

Research on materials with high charge mobility is indispensable in developing organic semiconductors. In typical organic semiconductor materials such as polyacene, the highest occupied molecular orbital (HOMO) plays an essential role in improving mobility. Among these materials, BTBT derivatives are especially effective as a semiconductor material because of their high flexibility and mobility due to spatial overlap between HOMO of adjacent molecules. However, previous studies have shown that in BTBT derivatives, it is not always possible to represent the electronic state accurately by HOMO alone, and it is necessary to consider the contribution of the second highest occupied molecular orbital (HOMO-1), which is one orbital below HOMO [1].

In addition to the contribution of molecular orbitals, we focus on the influence of the relationship between the substrate and adsorbed molecules on the electron conduction mechanism. To elucidate this, we adopted the structure that DPh-BTBT is adsorbed on Ag (Fig. 1(a)) and performed structural optimization by VASP calculation based on density functional theory. Subsequently, we calculated the change in charge distribution and the orbitals of isolated DPh-BTBT molecule and investigated the electron transfer behavior in the adsorbed state. In addition, we performed structural optimization for several structures with different adsorption distances. Separately from these analyses using VASP, we are currently constructing several structures with various angles between two isolated DPh-BTBT molecules and are trying to identify the intermolecular angle which is the boundary between the isolated system and the integral system for the bimolecular system.

First, we optimized the structure that a single DPh-BTBT molecule is adsorbed on Ag and converged with an adsorption distance of 3.03 Å and an adsorption energy of -2.81 eV. In subsequent charge distribution calculations, charge changes were observed in the intermediate region between the Ag substrate and the DPh-BTBT molecule (Fig. 1(d)). This suggests orbital hybridization between Ag and DPh-BTBT. In addition, the shape of the region where the charge distribution was changed (Fig. 1(e)) was similar to that of HOMO-1 calculated by Gaussian (Fig. 1(f)), suggesting the contribution of HOMO-1 to the electronic state. On another front, we also optimized the non-parallel adsorption structure of DPh-BTBT on Ag, which reached convergence at the adsorption energy of -1.60 eV, which is much less stable than that of -2.81 eV for the parallel structure. To investigate the cause of this unnatural instability and elucidate the formation process of the molecular thin film, we again optimized the structures for several adsorption distances, which reached convergence for the structure with adsorption distance of 3.00 Å with the DPh-BTBT molecule tilted in the parallel direction. In addition, regarding the Gaussian analysis for the structure of two DPh-BTBT molecules, which is currently in progress, according to the currently available data, it is presumed that the electronic state is an isolated system when the intermolecular angle is 0°, that is, when the two molecules are parallel, and by contrast, it is regarded as an integral system at 20°. Based on this, we will further investigate the angular region between 0° and 20° and discuss the results.

References

[1] Y. Kuroda, H. Ishii, S. Yoshino, and N. Kobayashi, Jpn. J. Appl. Phys. 58, SIIB27 (2019).

Atomically thin indium layers grown on the Si(111) substrate have attracted much interest and have been extensively studied, since the two-dimensional (2D) superconductivity was reported to appear below about 3 K on the so-called ‘rect’ structure of the Si(111)-√7×√3-In surface (referred to as √7×√3-rect). In our previous studies, the adsorption of organic molecules on the √7×√3-rect reconstruction was investigated [1-2].

In the present study, the formation process of the √7×√3-rect is studied in detail by varying the post-heating temperature after indium deposition on the Si(111) substrate, using the STM and the LEED. We find that at a low post-heating temperature of about 400 ºC, 2D indium islands of tens of nm in size are formed, consisting of a triple indium layer with a √7×√3 reconstruction different from the √7×√3-rect reconstruction (referred to as √7×√3-TL), as shown in Fig. 1(a) and 1(b). Although the atomic arrangements in the topmost indium layers of the √7×√3-TL and the √7×√3-rect reconstructions are considered to be nearly the same, their dIt/dVs(Vs) curves look different from each other. Surface defects in the disordered region around the √7×√3-TL islands seen in Fig. 1(a) originate from the partially remaining 7×7 reconstruction at the interface between the Si substrate and the indium layers. Additional indium atoms evaporated on the √7×√3-rect surface with the √7×√3-TL islands form flat wide single indium layers incorporating the √7×√3-TL islands. However, the layers do not adopt the √7×√3-TL reconstruction, but the incommensurate ~5.4×~5.4 reconstruction [3].

References

[1] T. Suzuki, et.al., Nanoscale 11 21790 (2019).

[2] T. Suzuki, et.al., Phys. Chem. Chem. Phys. 22 14748 (2020).

[3] T. Suzuki, et. al., Surf. Sci. 726 122174 (2022).

TiO₂ is a material that has various applications, such as photocatalyst, electron transport layer for perovskite solar cell, and so on. Representative crystal phases of TiO₂ are rutile, anatase, and brookite among its crystal phases. Anatase and brookite are metastable states, thus it is not simple to obtain the epitaxial thin film of these. Gas phase methods, like pulse laser deposition [1], molecular beam epitaxy [2], and so on, are typically used for fabrication of epitaxial thin films of anatase TiO₂. It has been reported that the epitaxial orientation is (001)[100]TiO₂||(001)[100]substrate for using LaAlO₃ (001) and SrTiO₃ (001) as a substrate. On the other hand, for the fabrication of (001)-faceted anatase TiO₂ crystal by hydrothermal method, the effectiveness of fluoride ions has been reported [3]. Recently, we reported the epitaxial growth of anatase TiO₂ by solvothermal method on LaAlO₃ (001) with a precursor containing fluoride ions and the local structure analysis around dopant atoms of the fabricated thin film [4].

To further the local structure analysis for understanding of the relation between photocatalytic activity and dopant, it was found that the change of the substrate is necessary. In this presentation, the successful of epitaxial growth of anatase TiO₂ thin film on SrTiO₃ by the invented solvothermal method and the growth dependence on precursor compositions will be presented.

Schematic drawings of the experimental procedures are shown in Figure 1 (a). The results of crystallinity and orientation evaluation by X-ray diffraction are shown in Figure 1 (b) and (c). It shows that the fabricated thin film was anatase TiO₂ and was grown epitaxially. In addition, the dependence of the crystallinity on the precursor compositions was found. Fluoride ions added as a surfactant remained on the surface of the fabricated thin film, as shown by X-ray photoelectron spectroscopy.

Continuing the growth investigation, structural analysis will be performed to reveal the local structure around the dopant and the adsorption structure of fluorine atoms on the surface in the future.

References

[1] H. Sakama, G. Osada, M. Tsukamoto, A. Tanokura, and N. Ichikawa, Thin Solid Films 515, 535 (2006).

[2] M. Murakami, Y. Matsumoto, K. Nakajima, T. Makino, Y. Segawa, T. Chikyow, P. Ahmet, M. Kawasaki, and H. Koinuma, Appl. Phys. Lett. 78, 2664 (2001).

[3] H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng, and G. Q. Lu, Nature 453, 638 (2008).

[4] K. Ono, K. Kimura, T. Kato, K. Hayashi, R. M. G. Rajapakse, and M. Shimomura, Chem. Eng. J. 451, 138893 (2023).

Vanadium dioxide (VO₂) transits to a tetragonal crystal structure from a monoclinic crystal structure at a temperature of about 68ºC concomitant with a resistance change over four orders of magnitude, so-called insulator-metal transition (IMT) ¹⁾. It is also known that VO₂ changes not only its resistance but also transparency abruptly due to the IMT. Thus, VO₂ is expected to have a wide range of applications such as sensors, actuators, and smart windows. If VO₂ can be grown on a flexible substrate, applicability of VO₂ will be highly enhanced. In this study, polyimide films, which have high heat resistance and chemical resistance, were employed as flexible substrates. In addition, zinc oxide (ZnO), which has a lattice matching with VO₂, was employed as a buffer layer to promote VO₂ crystal growth²⁾. Further, ZnO nanorod (NR) was employed in order to enhance crystallinity of VO₂ films. Because NR’s crystalline grain size is larger than that of zinc oxide seed film prepared by the sputtering method, it allows large in-plane crystalline growth in VO₂ film.

In sample preparations, polyimide films with a thickness of about 10 µm were prepared by spin coating with polyimide varnish followed by heat treatment. Then, ZnO_seed with a thickness of 460 nm was deposited on the polyimide by sputtering method. Next, ZnO_seed substrates were placed in a solution of HMTA (C₆H₁₂N₄) and zinc nitrate 6-hydrate (Zn(NO₃)₂・6H₂O) at 90°C for 120 min to fabricate ZnO_NR layer with a height of about 400 nm. The VO₂ films were deposited on the ZnO_seed or ZnO_NR buffered polyimide films by reactive sputtering with substrate biasing at low temperature of 250 °C. Low temperature growth of 250 °C was effective for preventing the diffusion of ZnO during VO₂ deposition^3,4). VO₂ deposition conditions were Ar and O₂ total pressure of 0.5 Pa, O₂ flow rate of 1.0 sccm, RF power of 200 W, bias power of 20 W, and deposition time of 40 min.

Fig. 1 shows the SEM image of ZnO_NR from the top, in which ZnO_NR growth with a hexagonal crystal system can be seen. The grain sizes of the nanorods were 150 - 200 nm. Fig. 2 shows the XRD patterns of VO2/ZnO_NR/polyimide (a) and VO2/ZnO_seed/polyimide (b). The (020) and (040) peaks of VO2 were confirmed in both samples. In addition, values of FWHM of the rocking curve were 5.95° for VO2/ZnO_NR/polyimide and 6.53° for VO2/ZnO_seed/polyimide, suggesting that the sample using ZnO_NR as the buffer layer had higher crystallinity of VO2. Fig. 3 is a comparison of the resistivity-temperature characteristics of VO2/ZnO_NR/Polyimide/quartz (a) and VO2/ZnO_seed/Polyimide/quartz (b). Both the ZnO_NR buffer sample and the ZnO_seed buffer sample showed a resistivity change of about three orders of magnitude with temperature, but the ZnO_NR buffer sample showed a steeper transition than the ZnO_seed buffer sample. In the presentation, we will also report more detailed surface morphology images and transmittance-temperature measurement results.

1) F. J. Morin, Phys. Rev.Lett 3, 34 (1959). 2) K. Kato et al., Jpn. J. Appl. Phys. 42, 6523 (2003). 3) N. H. Azhan et al., J. Appl. Phys. 117, 185307 (2015). 4) Y. Miyatake et al., J. Vac. Sci. Technol A, 40, 043406 (2022).

1. Introduction

Hematite iron oxide (Fe₂O₃), which is an n-type semiconductor with a bandgap of 2.0~2.2 eV, is harmless for human and extremely abundant on the earth. In previous woks, Cu-doped p-type iron oxide thin films and homojunction solar cells with n-type iron oxide have been successfully fabricated by electrochemical deposition(ECD) at constant-potencial and two-step pulse potencial.¹⁾²⁾ Therefore in this work, Cu-doped p-type iron oxide and homojunction solar cells were fabricated by constant-currrent ECD. The advantage of the constant-current method is that the effects of the first layer on the second layer deposition can be diminished. Moreover, continuous deposition of p-type iron oxide was performed by adding copper sulfate (to dope Cu) to the solution in which n-type iron oxide was deposited.

2. Experimental methods

Tin doped indium oxide (ITO)-coated glass sheet was used as the deposition substrate, platinum as the counter electrode, and Ag/AgCl as the reference electrode. The n-type iron oxide deposition solution contained 50 mM iron(II) sulfate heptahydrate and 100 mM sodium sulfate dissolved in water. The deposition current density was -0.12 mA/cm² and the deposition time was 30 minutes. The p-type iron oxide deposition solution was prepared by adding 5 mM copper sulfate to the n-type iron oxide solution. The deposition current density was -0.64 mA/cm² and the deposition time was 10 minutes. When fabricating pn-junction, the substrate was not removed from the deposition solution after n-type iron oxide deposition, and copper sulfate was added to the solution under the same deposition current and deposition time as above. Then, annealing was performed for 1 hour at 400 °C in an atmospheric atmosphere. Photoelectrochemical (PEC) measurements were performed on each of the p- and n-type iron oxides. Current-voltage (I-V) measurements were performed on the fabricated pn-junction.

3. Results and discussion

The results of the PEC measurements are shown in Fig. 1. If the sample is p-type, a negative photocurrent is observed during a negative potential sweep. For n-type, a positive photocurrent is observed during a positive potential sweep. The figure shows that the iron oxide exhibits n-type photoresponse, while the Cu-doped iron oxide exhibits p-type photoresponse. The results of I-V measurements for the p-n junction are shown in Fig. 2. Although the rectification properties are not good, weak photovoltaic properties were observed during AM1.5 irradiation, resulting in an open circuit voltage of 30.2 mV, a short circuit current of 2.81 μA, and a power generation efficiency of 2.67 × 10^-5 %. In summary, iron oxide homojunction solar cells were successfully fabricated by continuous deposition using the constant current ECD.

References

1) S. Kobayashi and Ichimura, Semicond. Sci. Tecnol. ,33,(2018),105006.

2) R. Takayanagi and Ichimura, Jpn. J. Appl. Phys. ,59,(2020),111002.

Vanadium dioxide (VO₂) shows a sharp insulator-metal transition (IMT) with resistance change over 4 orders at around 68°C concomitant with a structural phase transition (SPT). The VO₂ film also shows rapid decrease in resistance with current jump when certain voltage is applied between facing electrodes on the film. This current jump possesses a negative resistance region, resulting in the phenomenon of self-sustained oscillation. Self-sustained oscillation is expected to be applied to neuromorphic devices¹⁾. Further, it is known that parallel connection of self-sustained oscillators realizes coupled oscillation, in which their phases are automatically synchronized ^2,3). The coupled oscillations are expected to be applied to various issues in mathematical engineering field, such as graph coloring problem. ⁴⁾ In this study, the coupled oscillation was successfully demonstrated in a planar device with two self-oscillation circuits. VO₂ film was reactively sputtered on Al₂O₃ (001) substrate with size of 20 × 20 mm² at 400℃. The VO₂ film with thickness of 100 nm showed IMT with resistance change over three orders. As electrodes, 10 nm-thick Ti was coated on the VO₂ film followed by 200 nm-thick Au deposition. In the planar electrodes, the width and the gap were 5000 μm and 10 μm, respectively, as shown in Fig. 1. In order to investigate the coupled oscillation, self-oscillation circuits were composed between 2 and 3, and between 3 and 4, respectively. The two self-oscillation circuits were connected with a 22 nF-capacitor as shown in Fig. 2. Through this capacitance, two circuits interact each other and adjust their phases automatically. An AC voltage was supplied through an internal resistance of 1 kΩ by using curve tracer (Kokusai Denki Co. Ltd.). In the experimental results, anti-phase synchronization of two oscillating waves, in which voltage across VO₂ (2 - 3) rose during off period of voltage (3-4), was found as shown in Fig. 3. Thus, cooperative oscillation was realized in the coupling of two oscillators. Coupled oscillation frequency was around 10 kHz at a capacitance of 22 nF. The anti-phase synchronization appeared at the supply voltage from 100 to 92 V, and the amplitude between minimum and maximum values was 12.8V. In the present study, we directly observed changes of coupling mode by using curve tracer. In the presentation, we will show minute results of synchronization of two oscillators. References 1)Liu et al., J. Appl. Phys., 120, 124102 (2016). 2) M. S. Mian et al., J. Appl. Phys., 117, 215305 (2015). 3) Tobe et al., J. Appl. Phys., 127, 195103 (2020). 4) A.Parihar et al. Sci.Rep. 7, 911 (2017).

Introduction

Recently, we found that, in addition to the conventional transmembrane voltage, the voltage applied laterally to the membrane can reactivate once inactivated human voltage-gated Na ion channel membrane proteins using an artificial cell membrane system [1]. This finding suggests that lateral voltage may be an important novel parameter for promoting the activity of ion channels prone to inactivation, which are difficult to measure by conventional methods. However, the short electrode lifetime and low yield of lateral voltage-applying devices are major bottlenecks in the establishment of a new lateral voltage-based ion channel function analysis system. The usable lifetime of the lateral voltage application devices is as short as 20 minutes due to oxidation of the Ti electrodes, and the fabrication yield is less than 10% due to cracks in the SiO₂ layer used as the insulating layer. In this study, to address these issues, we aimed to develop a highly efficient fabrication process for more stable and longer-lived devices by using Au, a non-oxidizing metal, as the electrodes and CYTOP, a flexible insulating coating, as the insulating layer.

Method

A micropore was formed in a Teflon film by an electric spark. The Teflon film was covered with a Ni mask with an electrode pattern, and a Ti/Au thin film used for electrodes was formed on the film by sputtering. To form an insulating layer on the electrodes, the Teflon film was then dipped into CYTOP (CTL-109AE) except for the electrode contact region, pulled up by a dip coater at a speed of 1 mm/s, and baked. We evaluated the electrical properties of the fabricated devices by measuring the resistance between the electrodes and the contact resistance of the electrode region that was connected to the lateral voltage source. In addition, the effect of lateral voltage on ion channels was evaluated using the fabricated devices. Artificial cell membranes were formed in the micropores of the devices by the monolayer-folding method, and human Na-ion channels were embedded in the membranes via proteoliposome fusion. Ion channel currents were measured and compared with and without the application of the lateral voltage.

Results and discussion

To investigate the electrode lifetime of the fabricated devices, the contact resistance was measured after that lateral voltage was applied to the devices for a given time. The result showed that the contact resistance of the Au-based electrodes remained low (< 10 Ω) even after the lateral voltage was applied for more than 2 hours, which was sufficient for the functional analysis of ion channels. We next investigated the insulation property of the CYTOP layer through the measurement of resistance between the fabricated electrodes in buffer solutions. When the devices were coated with a single CYTOP layer, we observed current leakage through the buffer solution. When the devices were coated twice with the CYTOP layers, the resistance between the electrodes increased to more than 250 GΩ even in the buffer. Thus, it was found that the double CYTOP coating was suitable for obtaining a high resistance of the insulating layer. The concentration of the CYTOP was also important to improve the fabrication yield. By adjusting these conditions, we were able to improve the device fabrication yield to 60%. Using the fabricated devices based on the Au electrodes and CYTOP layers, we applied lateral voltage to inactivated Na channels and observed that the channel functions were re-activated by the lateral voltage. It was demonstrated that the devices fabricated by the new process could be useful as a novel tool for the analysis of inactivated ion channels.

View PDF for the rest of the abstract.

Thin films of metal-organic frameworks (MOFs) have potential applications as gas sensors, filters, and electrocatalysts due to its efficient and specific gas capturing properties. Various types of monolayer and thicker MOF thin films have been fabricated, but the structural details remain unclear. In this study, we fabricated one of the MOF thin film called NAFS-1 using a Langmuir-trough, following the recipe in previous study[1], and carried out the structural characterization of monolayer and thicker films using infrared scattering-type scanning near-field optical microscopy (IR s-SNOM).

For sample preparation, 5,10,15,20-tetrakis(4-carboxyphenyl)porphyrinato-cobalt(II) (CoTCPP) and pyridine were first dissolved in mixed chloroform/methanol solvent. This CoTCPP/pyridine solution was spread onto CuCl₂ subphase, which spontaneously formed the NAFS-1 at the air-liquid interface. The NAFS-1 film was then transferred onto a Si substrate by the horizontal dipping method.

We observed these samples using the amplitude-modulation atomic force microscopy (AFM) and confirmed the fabrication of the gathered islands of NAFS-1 films as shown in Figure 1(a). The thickness of the film is approximately 1.5 nm, which is consistent with the expected film thickness based on the structural model[1].

The same sample was further characterized by IR s-SNOM with a broad-band infrared laser ranging from 1000 to 1700 cm^-1 and a PtIr5-coated tip. Figure 1(b) shows the IR s-SNOM spectra obtained on the NAFS-1 film. Each spectrum was recorded on the spot indicated in the AFM image in Figure 1(a). As indicated by dashed black line in Figure 1(b), these spectra show almost the same characteristics. Three IR s-SNOM spectra show characteristic IR peaks of the NAFS-1 around 1400 cm^-1 (corresponding to C=O symmetric stretching mode of carboxylate ion in paddlewheel structure), 1620 cm^-1 (corresponding to C=O asymmetric stretching mode of carboxylate ion in paddlewheel structure), and 1680 cm^-1 (corresponding to C=O stretching mode of carboxy group). It indicates the nanoscale structural homogeneity of the fabricated NAFS-1 films. These results demonstrate the capability of the IR s-SNOM in characterizing MOF thin films. This technique is expected to be readily applicable to various types of other organic porous thin films. Attribution of the other peaks shown in Figure 1(b) and further experiment to reveal the dependence on the number of layers will also be discussed.

References

[1] R. Makiura et al., Nature Mater. 9, 565 (2010).

Recently, various objects have become targets for sensing in order to realize an IoT society, and sensors are required higher sensitivity and miniaturization. In application of many automotive, aerospace, and industrial fields, there is a high need for compact sensors to measure mechanical quantities such as strain, stress. Strain gauge are intended to measure the magnitude and direction of the strain and the magnetostrictive effect in ferromagnetic films can be applied to strain detection. In this research, a GIG (Granular in Gap) structure [1] is used for a new strain gauge. It has a structure in which a granular film is sandwiched between soft magnetic yokes. When the strain is applied to the device, the change of strain is detected as a resistance change in the granular film. Co-AlO and amorphous(a)-FeSiBNb were employed as the granular film and the soft magnetic yoke of the strain gauge. The thickness of Co-AlO film and a-FeSiBNb yoke was 300 nm. The structure of the strain gauge is indicated in the inset of Figure 1. The gap length was estimated to be approximately 4 µm. An AC voltage at 80Hz supplied from a lock-in-amplifier (LIA) is applied to the series circuit of the gauge and a variable resistor. When the strain is applied to the gauge, the strain is detected as a voltage of the gauge by LIA. The MR ratio of the Co-AlO granular film before processing into a GIG element was about 6%. The magnetostriction constant of a-FeSiBNb yoke was 30.4ppm. A magnetic field of 5 Oe must be applied in direction of H⊥gap in order to align the magnetic moments of yokes. The granular film becomes a low resistance state due to a magnetic field appeared in the gap yielded by magnetic poles at the edge of yokes. When the strain is applied in direction of H//gap, the magnetic moments of yokes change its direction, increasing the resistance of granular film owing to the decrease of the magnetic field in the gap. The dependence of output voltage on the applied strain ε is shown in Fig.1. The blue circles are the result of increasing strain, and the red circles are the result of decreasing strain. The output voltage became large with increasing strain, and gradually almost constant because the magnetic moments of yokes saturated in direction of H//gap. This indicates that the strain up to about 6.0 × 10^-5 can be detected. The gauge factor estimated was approximately 50, larger than that of a typical metal strain gauge. Reference [1] N. Kobayashi et al., J. Magn. Magn. Mater. 188. 30 (1998).

Since amorphous carbon thin films have interesting properties such as high hardness, a low friction coefficient, and chemical inertness, they have attracted a great deal of attention for a wide range of applications. Owing to their properties, amorphous carbon films are mainly used in industrial products such as coatings for the magnetic media of hard disk drives, machine parts for molds, cutting tools and so on. As amorphous carbon films formed by various vapor phase methods include hydrogen atoms, it was pointed out that the thermal stability was not good due to the hydrogen desorption [1]. In this study, the electronic structure in amorphous carbon thin films as a function of annealing temperature was investigated by photoelectron spectroscopy.

Amorphous carbon films were produced by the RF plasma method [2]. Photoelectron measurements were carried out at the BL 7B of NewSUBARU synchrotron radiation facility, University of Hyogo.

Before annealing, a broader peak at 285 eV was observed in the photoelectron spectroscopy spectrum of the C 1s core level as shown in Fig. 1. The photoelectron spectra of the C 1s core level became narrower with increasing annealing temperature. This indicates that the coordination of C atoms in amorphous carbon thin films was changed. By using the curve fitting analysis of the C 1s photoelectron spectra, the coordination of C atoms in amorphous carbon thin films was evaluated as a function of annealing temperature. In addition, the photoelectron spectra of the valence band as a function of annealing temperature were also measured. In the presentation, the electronic structure of amorphous carbon thin films as a function of annealing temperature is discussed.

References

[1] D. R. Tallant, J. E. Parmeter, M. P. Siegal, and R. L. Simpson, Diam. Relat. Mater. 4, 191 (1995).

[2] Y. Haruyama et. al., Jpn. J. Appl. Phys. 60, 125504 (2021).

INTRODUCTION

Physical property exploration using light has revealed many previously unknown physical properties. Electronic and nuclear spin states of surface-adsorbed molecules [1,2], characterization of electronic states using X-ray absorption spectroscopy (XAS) has been proposed [3,4]. Wavenumber-resolved photoemission spectroscopy is a powerful method for studying the electronic structure of crystals [5]. Advances in detectors have made it possible to obtain highly accurate 3D maps compared to conventional 2D maps obtained by Angle-resolved photoemission spectroscopy (ARPES). Since it can quickly measure from metals to organic thin films, all kinds of measurements are possible. A material design for spin orbitronics devices with a dramatic energy-saving function is desired. Atomic-layer crystals stacked via van der Waals forces are candidates for this. Bi2Se3 is a strong topological insulator, possessing a Dirac-conical surface dispersion connecting the conducting and valence states. It has been thought that the spin direction is determined by the electron momentum. However, recently, it was found that this spin state is maintained only under a specific optical geometry, and it was reported that the spin state in solids is an entanglement state [6].

RESULTS AND DISCUSSION

In this presentation, we calculated wavenumber-resolved photoemission spectra of layered materials such as Bi2Se2 and TiSe2. The initial and final states were calculated based on density functional theory (DFT), and the photoelectron intensity was calculated. At this time, boundary conditions on the surface were considered. Compared to the methods we have reported so far [7-10], it has the advantage that the surface electronic state can be considered in detail, and the scattering term in the final state can also be included.

REFERENCES

[1] K. Niki et al., Phys. Rev. B 77, 201404 (R) (2008).

[2] K. Niki et al., Phys. Rev. B 79, 085408 (2009).

[3] J. Kogo et al., J. Phys. Soc. Jpn. 91, 034702 (2022).

[4] T. Fujikawa et al., J. Electron Spectros. Relat. Phenom. 233, 57 (2019).

[5] S. Suga and A. Sekiyama, Springer Series in Optical sciences 176, (2013).

[6] K. Kuroda a et al., Phys. Rev. B 94, 165162 (2016).

[7] M. Kazama et al., Phys. Rev. B (R), 89, 045110, (2014).

[8] T. Fujikawa and K. Niki, J. Electron Spectros. Relat. Phenom. 206, 74 (2016).

[9] K. Niki et al., Vacuum and Surf. Sci. 63, 336 (2020).

[10] K. Baumgärtner et al., arXiv:2305.07773.

Among the family of rock-salt nitrides, niobium nitride (NbN) undergoes the superconducting transition at T_c as high as 16 K [1]. In contrast, regarding niobium monoxide (NbO), only its quasi-rock-salt structured bulk with ordered vacancies has been reported, with its superconductivity below 1.5 K [2]. However, recent technical progress has facilitated the epitaxial synthesis of rock-salt NbO through pulsed laser deposition [3]. The electrical transport property of these NbO thin films is highly sensitive to the film thickness. In the thinner film region, they exhibit insulating behavior, while in the thicker film region, they display metallic behavior with a superconducting transition at the maximum T_c = 6.0 K [3]. On the other hand, a comprehensive investigation into the electrical transport property of Nb(O, N), a rock-salt oxynitride possessing a composition intermediary between NbN and NbO, has not yet to be conducted systematically. In this study, we prepared epitaxial thin films of rock-salt Nb(O, N) and conducted experimental and theoretical analyses to investigate their electrical transport property.

Epitaxial thin films of NbO, NbO_0.5N_0.5 (nominal composition), and NbN were fabricated using a pulsed laser deposition system with rapid beam deflection (RBD-PLD)[4]. Polycrystalline pellets of NbO and NbN were used as targets. The thin films were grown onto MgO (100) substrates (with a lattice parameter of a = 4.212 Å) at a temperature of 500^oC without introducing any gases. A KrF excimer laser was employed for the deposition process with a fluence of 0.9 J cm^-2 and a frequency of 40 Hz. The crystal structure was characterized with X-ray diffraction (XRD), while the temperature-dependent resistivity (ρ-T) measurement was conducted using a physical properties measurement system (PPMS). First-principles calculations were also performed on the supercomputer MASAMUNE-IMR in Tohoku University using the Quantum ESPRESSO software package.

Figure 1 shows the results of reciprocal space mapping (RSM) measurements on 15-nm-thick films. These results confirmed their epitaxial growth with the relationship [110]_film // [110]_MgO and [001]_film // [001]_MgO, and indicated an in-plane compressive strain from the substrate, with each a-axis length all very close to that of the substrate. Accordingly, the variation of the c-axis length in these films became more remarkable: NbO_0.5N_0.5 corresponds to c = 4.313 Å, the value of which positioned between those observed for NbN (c = 4.295 Å) and NbO (c = 4.347 Å). The ρ-T curves, illustrated in Figure 2, demonstrate that NbO exhibited an insulating behavior which is characteristic for the thinner film region, whereas the resistivity of NbN in all the temperature region underwent a pronounced reduction, e.g., by about two orders of magnitude at RT, though it was still insulating or semiconducting behavior in the present result. On the other hand, NbO_0.5N_0.5 exhibited a sort of intermediate resistivity behavior positioned between those of NbO and NbN. The results of the density of states (DOS) estimated from DFT+U+V calculations (employing U = 7 eV) for the rock-salt type structures of NbO, NbO_0.5N_0.5, and NbN are displayed in Figure 3. Upon closer inspection of the DOS near the Fermi level, a band gap of approximately 1.5 eV becomes evident for NbO, while the DOS gradually increases from NbO through Nb(O, N) to NbN.

View PDF for the rest of the abstract.

FeGa alloys have large saturation magnetostriction constants (λ_s) and saturation magnetization, and their good mechanical properties are expected to be applied to sensors and power generation devices. Recently, there is a need to improve soft magnetic and high-frequency properties of FeGa films for application to high-frequency devices [1,2]. Addition of light elements such as B [3] and C [4] has been reported to decrease the coercive force (H_c) and increase the λ_s. In this study, we investigated structure and magnetic properties of FeGaN films.

(Fe₈₆Ga₁₄)_1-xN_x films were prepared on a micro cover glass by dc magnetron sputtering method, using a composite target consisting of Fe₈₀Ga₂₀ chips and an Fe disk. The sputtering power and time were fixed at 27 W and 1 hour, respectively. During deposition, a dc magnetic field of about 200 Oe was applied to the substrate. The sputtering gas was a mixture of Ar and N₂ gas, and a partial pressure ratio P_N2 / P_Total, where P_N2 is the partial pressure of N₂ and P_Total is the total gas pressure, was controlled. A 30 nm thick SiN film was deposited on the surface to prevent oxidation. The magnetic properties of the films were evaluated by vibrating sample magnetometer and torque magnetometer, and the crystal structure was confirmed by X-ray diffraction (XRD).

XRD patterns revealed that the films for P_N2 / P_Total < 0.100 were crystallized, and the films for P_N2 / P_Total > 0.120 were in amorphous state.

Figure 1 shows the H_c and theλ_s as a function of P_N2 / P_Total. The H_c decreased significantly for P_N2 / P_Total > 0.100. A minimum Hc of about 1.3 Oe was obtained at P_N2 / P_Total = 0.135. The decrease in Hc is due to the structural change from crystal to amorphous. For P_N2 / P_Total > 0.135, the Hc increased due to the perpendicular magnetic anisotropy caused by non-uniform compressive stress introduced during deposition or the formation of columnar structure. The λ_s increased with increasing P_N2 / P_Total, and showed the maximum value of about 30 ppm. At P_N2 / P_Total = 0.135, the minimum Hc and the maximum λ_s were obtained, simultaneously.

References

[1] D.Cao et al. : AIP Advances 7, 115009(2017).

[2] S.Muramatsu et al : The Papers of Technical Meeting on Magnetism, IEE Jpn, MAG-21-085 (2021).

[3] J.Lou et al. : Appl. Phys. Lett. 91, 182504 (2007).

[4] S.Muramatsu et al. : The Papers of Technical Meeting on Magnetism, IEE Jpn, MAG-22-085 (2022).

LaNiO₃ (LNO) is the most technologically relevant conductive metallic oxide because it exhibits a simple pseudocubic perovskite structure with a small degree of rhombohedral distortion as well as a relatively simple composition and stoichiometry. In addition, the pseudocubic perovskite structure of LNO has a lattice parameter very close to many monocrystalline substrates, and useful materials. Hence, LNO attracted significant attention, especially as a bottom electrode. In this study, we report the synthesis of the epitaxial LNO thin films on SrTiO₃ (STO).

Figure 1 shows the out-of-plane XRD patterns of annealed LNO thin films. The films were deposited on an unheated STO(111) substrate in a mixed Ar/O₂ atmosphere by using RF sputtering, and the total pressure during deposition was kept at 0.3 Pa. The RF power applied to the target was maintained at 50 W. After the deposition, the films were annealed at 500^oC in air with heating rate of 1.4-83^oC /min by using electric a muffle furnace. In Fig. 1, LNO 111 and 222 peaks were present, indicating the growth of (111)-oriented LNO epitaxial films. In addition, we found that low heating rate resulted in high crystallinity of epitaxial LNO. Furthermore, an XRD reciprocal space map of the LNO/STO(111) revealed that LNO (111)-oriented epitaxial thin-film grew on STO(111) and its orientation relationships were [1-10](111)LNO//[1-10](111)STO and [11-2](111)LNO//[11-2](111)STO (not shown here). Here, an as-deposited LNO film only showed peaks corresponding to the STO(111) substrate (not shown here), indicating that the LNO film was amorphous. Hence, these results indicate that the as-deposited LNO film was an amorphous precursor, and thereby the annealing at 500^oC in air with low heating rate allowed solid-state growth for LNO(111) epitaxial films.

Organic semiconductor devices have been receiving increasing attention because of their attractive properties, which include low weight, low-cost production, low-temperature processing, and mechanical flexibility, being distinguished from their conventional inorganic counterparts. Organic semiconductor thin film heterostructures are especially important in realizing multilayer devices such as solar cells. Epitaxial growth, in which a crystalline thin film is aligned in the in-plane direction with respect to the crystal orientation of an underlying layer, is one of the key strategies to control the organic heterostructures. Our group has reported that large grains of N,N'-Di-n-octyl-3,4,9,10-perylenetetracarboxylic Diimide (PTCDI-C8) with their sizes greater than 100 µm were formed as a result of crystallization via its liquid crystal phase during vacuum deposition [1]. In this study, we investigated the growth morphology and epitaxial relationship in the organic heterostructure of copper phthalocyanine (CuPc) grown over the large grains of a PTCDI-C8 underlayer. Firstly, a (001)-oriented PTCDI-C8 thin film with large grains was prepared at a substrate temperature of 180 °C on a Si(100) substrate covered with a thermally oxidized SiO₂ layer, by using an continuous-wave infrared (CW-IR) laser deposition technique [2] in a vacuum. After that, 100 nm-thick CuPc was deposited at 180 °C on the PTCDI-C8 thin film. The fabricated organic heterostructures were characterized using techniques such as out-of-plane X-ray diffraction (XRD), atomic force microscopy (AFM), and grazing incidence X-ray diffraction (GIXD) at SPring-8. Figure 1(a) shows the out-of-plane XRD pattern of a CuPc thin film with a thickness of 100 nm grown on a PTCDI-C8 large grain thin film. From this pattern, it was found that the grown CuPc thin film took the α-phase [3] with its (100)-orientation on the (001)-oriented PTCDI-C8 thin film. In AFM observation, the surface morphology of the CuPc thin film showed needle-like grains aligned in a unique direction on the PTCDI-C8 large grains, in contrast to the growth of CuPc alone where its random grains with an average size of ~100 nm were formed. (Fig.1(b)). To investigate the epitaxial relationship between the CuPc and PTCDI-C8 layers, we trimmed the surrounding thin film leaving a single PTCDI-C8 grain and successfully obtained an in-plane reciprocal space map for the single grain from GIXD measurement. (Fig.1(c)). As a result, an epitaxial growth of CuPc on the PTCDI-C8 large grain was confirmed with a relationship in which the c* direction of CuPc is parallel to the a* direction of PTCDI-C8. In the presentation, we will also discuss the initial growth behavior of CuPc on PTCDI-C8 large grains.

Introduction

Vanadium dioxide (VO₂) attracts attention due to its reversible insulator-metal transition (IMT) characteristics.¹⁾ VO₂ shows abrupt resistance changes of 3-4 orders of magnitude accompanying a structural phase transition from a low-temperature monoclinic structure to a high-temperature tetragonal structure at around 68°C. The IMT of VO₂ can be achieved by electrical and optical excitation, making it a promising material for application in various fields.²⁾ However, selective growth of VO₂ is a critical issue because of the presence of various oxide phases in the V-O phase diagram. In addition, it is also known that the VO₂ crystal phase showing IMT on an amorphous glass substrate is hard to prepare. In this study, we attempted to fabricate a single-phase VO₂(M1) film on the fused silica substrate with IMT by monitoring the discharge voltage and controlling the process over time.

Experiment

VO₂ thin film was deposited on fused silica glass substrates using a reactive direct current magnetron sputtering (r-DCMS) system. A DC power of 50 W was applied to the vanadium metal (φ50 mm, 99.9%) target. The Ar gas flow rate and pressure were set to 5.0 sccm and 1.0 Pa. Before the VO₂ deposition, the hysteresis behavior of the r-DCMS system was measured by increasing/decreasing the O₂ flow rate while monitoring the discharge voltage. Fig. 1 (a) shows the measured hysteresis curve. The deposition of VO₂ films was performed in the transition region of the O₂ flow rate decreasing profile. The deposition was conducted in two types of operations. One is a fixed discharge voltage operation at 380 V (O₂ 0.28 sccm) and 370 V (O₂ 0.32 sccm). Second was a transient voltage change operation: the discharge voltage was initially set at 390 V and waited for 20 seconds. Then, the O₂ flow rate was reduced to change the voltage to 360 V, 350 V, 340 V, and 320 V and waited for another 20 seconds. These operations were repeated during the deposition. The substrate temperature and the deposition time were 400 ± 5℃ and 10 minutes. The crystallinity of the prepared films was evaluated by X-ray diffraction (XRD). The resistance changes against the temperature (R-T characteristics) were investigated by the two-probe method.

Result and discussion

Fig. 1 (b) shows the XRD patterns for the deposited samples. The diffraction peak from VO₂(M) (011) plane at 2θ = 27.99° was observed only in the 390/350 V sample, indicating the crystal growth of VO₂ thin films on a fused silica glass substrate.

Fig. 1 (c) shows the R-T characteristics of the prepared samples. In all samples except 390/350 V, the IMT was not observed. In the 390/350 V sample, one order of magnitude change in resistance was achieved. These results correspond with the XRD results. Single-phase VO₂ (M1) is successfully obtained by controlling the discharge voltage though VO₂ was deposited on the fused silica glass substrate. The present results will contribute to the deposition of selective single-phase VO₂ (M1) films by reactive sputtering.

Refference

1) C. H. Griffiths, and H. K. Eastwood, J. Appl. Phys 45, 2201 (1974).

2) F. J. Morin, Appl. Phys. Lett., 3, 34 (1959)

INTRODUCTION

Tungsten trioxide (WO₃), a typical electrochromic (EC) material, has been actively studied for functional window applications because of its reversible color change [1]. We have studied the properties of relatively thicker WO₃ films (500~1500 nm) prepared by reactive magnetron sputtering to achieve superior light shielding performance [2]. In the previous study, we optimized the cyclic voltammetry (CV) conditions, such as voltage range and scan rate for films of 1000 nm thickness [3]. In this study, under the optimized CV conditions, we investigated the cyclic characteristics of WO₃ on coloration and bleaching up to 50 cycles. It was found that the coloration efficiency (CE) strongly depended on the WO₃ film thickness at the same CV condition.

EXPERIMENT

WO₃ thin film was deposited by reactive DC magnetron sputtering. ITO or quartz glass substrates were used. The deposition conditions were as follows: DC power was 50 W, Ar flow rate and pressure were 10.0 sccm and 3.0 Pa, and oxygen flow rate was 2.0 sccm. Three different thicknesses of WO₃ films (500, 1000, and 1500 nm) were prepared by adjusting the deposition time. CV measurements were performed in a tripolar cell with Pt as the reference and counter electrodes and WO₃/ITO glass as the working electrode. The electrolyte was a solution of 1 M LiClO₄ in propylene carbonate (50 mL). The voltage scan range was set to -1.2~+1.6 V, at a rate of 10 mV/s. 50 cycles were performed, with coloration and bleaching as one cycle. Transmission spectra were measured for the as-deposited state and at colored and bleached states at the initial and every 10 cycles from 10 to 50.

RESULT AND DISCUSSION

Figure 1 shows the coloration efficiency dependence on the number of cycles for each film thickness. The CEs of the samples of 1000 nm and 1500 nm thicknesses were about the same, while the 500 nm sample showed higher CE than the others. This indicates that lithium cations may be incorporated differently in the 500 nm sample than in the 1000 and 1500 nm samples. Additional measurements for 700 nm will be made to investigate the cation behavior in more detail.

REFERENCES

[1] Granqvist, Thin Solid Films 564, 1 (2014).

[2] Mian, et al., Physica Status Solidi (a) 219, 2100646 (2021).

[3] Hosaka et al., IVC22, Mon-PO1A-18.