Table of contents for issue 2, volume 63, Japanese Journal of Applied Physics

Polarized terahertz (THz) sources are important components in THz technologies. This paper highlights and discusses recent progress and measurement methods in the monolithic generation of polarized THz radiation using intrinsic Josephson junction oscillators made of high-T_c superconductors. The polarized radiation is generated from three mesa designs: truncated-edge square, notched cylindrical, and rectangular mesa structures. The polarization control depends on the excitation of two orthogonal TM modes in these mesas, comprising stacked intrinsic Josephson junctions in single crystalline Bi₂Sr₂CaCu₂O_8+δ. This method maintains a high output intensity and low axial ratios while avoiding the signal loss associated with external polarimetric modulators prevalent in the THz frequency range. Moreover, it demonstrates the manipulation of terahertz wave helicity by adjusting the current injection position, with experiments substantiating the device’s capability to switch between left-handed and right-handed elliptical polarization at designated frequencies.

The expansion rate of single Shockley-type stacking faults (1SSFs) was examined in 4H-SiC under UV illumination in various basal plane dislocation (BPD) structures with 90° or 30° Si-core partial dislocations (PDs) at the expansion front. In the case of 30° Si-core PDs at the front, we found some BPDs with extremely slow expansion rates. Photoluminescence imaging revealed that the BPDs were accompanied by characteristic dim lines in the shallower parts of the epitaxial layers. We confirmed that the lines were threading edge dislocations by transmission electron microscopy. Additional high-resolution scanning transmission electron microscope analysis revealed that the leading partial was a 30° C-core instead of a 30° Si-core. This implies the large amount of C-core segments on the expanding PD might be the reason for the 1SSFs having very slow expansion rates. Moreover, the expansion rate of 90° Si-core PDs was obtained experimentally and compared with that of 30° PDs.

Heterogeneous chiplet integration has become a crucial performance enabler in the microelectronics industry by providing the flexibility of die disaggregation, and the ability to mix/match different IP blocks optimized on different Si nodes in a single package. It shows great potential in supercomputing, autonomous driving, artificial intelligence, and machine learning applications. With rising demand in high performance computing, the key focus in heterogeneous integration (HI) scaling has been to push interconnect density with increased bandwidth and improved power efficiency. In this paper, we provide an overview of embedded multi-interconnect bridge packaging technology scaling and discuss key considerations for advanced substrate packaging technologies to enable further HI applications.

Silicon carbide (SiC) is widely used in power semiconductor devices; however, basal plane dislocations (BPDs) degrade device performance, through a mechanism called bipolar degradation. Recently, we proposed that proton implantation could suppress BPD expansion by reducing BPD mobility. We considered three potential mechanisms: the hydrogen presence around BPDs, point defects induced by implantation, and carrier lifetime reduction. In this study, we discuss the mechanisms of proton implantation and its applicability to SiC power device production.

X-ray crystal truncation rod (CTR) measurements are commonly used to analyze surface or interfacial atomic structures quantitatively. However, it is difficult to expand the measurement range to determine three-dimensional atomic structures in real time. We developed a method for CTR measurements that involves the simultaneous intensity measurements of X-rays and electron diffraction under molecular beam epitaxial conditions. Using the proposed method, we observed the formation of liquid gallium ordering on the GaN(0001) surface and determined the ordered structure within each layer.

We investigate the effects of gamma-ray (γ-ray) irradiation on an In_0.53Ga_0.47As high electron mobility transistor (HEMT). After γ-ray radiation, the irradiated HEMT shows degradation of the maximum transconductance (g_m,max), the unity current gain cutoff frequency (f_T), and the maximum oscillation frequency (f_max)—about 12.8%, 18.0%, and 16.9%, respectively—because of an increase in on-resistance (R_on) in In_0.53Ga_0.47As HEMTs exposed to high-dose γ-ray radiation. Moreover, we obtain a minimum noise figure (NF_min) of about 1 dB from 8 to 40 GHz for the irradiated HEMT, which is a lower value compared to a non-irradiated HEMT; this is because the gate leakage current is reduced after γ-ray irradiation.

In situ laser-based photoemission electron microscopy observations with time-dependent dielectric breakdown measurements of TiN/Hf_0.5Zr_0.5O₂(HZO)/TiN ferroelectric capacitors were performed to reveal the dielectric breakdown (DB) mechanism. We succeeded in visualizing the hard DB spots through the top electrode. We found that capacitors with short- and long-lifetime distributions were broken down near and far from the edge of the capacitors, respectively. This indicates that the DB is either topography-dependent or film-quality-dependent. This work demonstrates an effective method of detecting DB in a non-destructive manner to provide an insight for achieving higher endurance HfO₂-based ferroelectric capacitors.

Among the various printing technologies available, reverse offset printing is the most promising for transferring fine patterns. This study reports the experimental results of interfacial adhesion and elastic deformation of a blanket during the printing process to demonstrate the fabrication of a high-resolution silver thin-film pattern. Additionally, a silicon wafer was used as the printing template, which allowed for the production of a high-resolution template more easily and at a lower cost than conventional glass templates. We successfully patterned line/space = 4 μm/4 μm using the proposed printing technique, and this resolution can be applied to X-ray diffraction gratings according to simulations.

We have demonstrated the growth of CsBH₄ (001) epitaxial thin films utilizing an IR laser deposition technique. The thin film exhibited coherent growth with a tensile strain from the substrate, resulting in its crystallinity superior to other kinds of metal borohydride thin films reported in earlier studies. Moreover, the notable chemical stability of CsBH₄ enabled surface morphological and optical characterizations even under atmospheric conditions. The optical transition energy values estimated from a Tauc plot were found smaller than the theoretical ones previously reported, and DFT calculations suggested a partial deficiency of Cs⁺ or [BH₄]⁻ created in the thin films.

Photoresponsivity is an important measure for applications as light absorbing layers because it is proportional to carrier lifetime. Previous studies have shown that the photoresponsivity of undoped BaSi₂ films increases by more than 10 times by post-annealing (PA) at 1000 °C, but decreases by moderate-temperature PA. Such BaSi₂ films are compressively strained in the normal direction, regardless of whether they are undoped or As-doped BaSi₂ films and show a distinct photoluminescence around 0.85 eV at 8 K, indicating the formation of a new localized state. These provide a clue to the annealing conditions that lead to photoresponsivity reduction.

We develop a simplified approach for localized temperature measurement at the fiber tip, utilizing the standard configuration of Brillouin optical correlation-domain reflectometry (BOCDR) with sinusoidal modulation of optical frequency. This method eliminates the need for specialized modifications to the sensing fiber, thereby enhancing cost-efficiency. Compared to the noise-modulated low-coherence BOCDR, our approach allows for a more precise evaluation of theoretical spatial resolution. We successfully attain a theoretical spatial resolution of 5.7 cm with this method.

We measured and compared the I–V characteristics of needle- and junction-contact Ge Schottky barrier diodes (SBDs). The threshold voltage (V_th) of I–V characteristics in needle-contact SBDs was revealed to be lower than that of junction-contact SBDs, which resulted in higher current in the low-voltage region. This indicates that needle-contact SBDs are more suitable for low-power applications. The lower V_th in needle-contact SBDs indicates the lower Schottky barrier height. We proposed a band diagram of needle-contact SBDs, where the lower Schottky barrier height is assumed considering the effect of surface potential. We confirmed the validity of the model by fitting analysis.

We investigated the application of polycrystalline SiC coating on SiC ceramics using vapor from a Si melt as a Si source and C₂H₂ generated by decomposition of CH₄ gas as a C source. First, the temperature distribution and gas flow in the furnace, the decomposition behavior of CH₄ gas, and the production rate were calculated using the simulation software Virtual Reactor. Based on these numerical results, the structure of the furnace and its internal layout were decided. Next, SiC coating on SiC ceramics was performed with two different furnace layouts, and the generated SiC coating layer was evaluated by observation of the microstructure. With a CH₄ gas flow rate of 50 sccm and a pressure inside the furnace of 0.5 atm, SiC coating with a coating rate of 26 μm h⁻¹ was achieved.

The propagation loss in single-mode asymmetric waveguides due to interface nanometer-roughness was studied using amplified spontaneous emission (ASE) measurements. Poly(9,9-dioctylfluorene) (F8) was used as the organic gain medium and the structure of the asymmetric waveguide was quartz glass substrate/F8/air. The propagation loss was measured at the ASE wavelength (447 nm) of amorphous F8, and the surface roughness of the substrate and F8 was measured using an atomic force microscope. The propagation losses of F8 waveguides with different F8 slab thicknesses were in good agreement with those calculated using an analytical expression for single-mode asymmetric waveguides with nanometer-roughness interfaces. The results presented herein will be useful for the design of high-performance organic lasers and organic optoelectronic integration systems.

We have investigated an interference experiment of the two-slit Young experiment using an improved self-alignment (SA) liquid crystal optical control device structure. The newly developed SA process and optimization of slit structure and slit width were redesigned to micron size in order to make the interference pattern clear. As a result, we observed a clear NOT logic operation where the center peak intensity of the incident light is inverted.

Carbon nanotubes (CNTs) were successfully prepared on carbon fiber-based substrates via CVD using zirconia nanoparticles (nanozirconia) as additives. The dispersion of a moderate amount of nanozirconia on the substrate surfaces before the CVD process promoted the formation of Fe particles, which catalyzed the CNT growth and effectively promoted the formation of CNTs. The dispersion of nanozirconia on the substrates is thought to effectively suppress the surface diffusion of pyrolytically generated iron atoms from the precursor, leading to the formation of numerous fine iron nanoparticles with sizes suitable for CNT growth. The CNT/carbon fiber substrates are expected to be useful electrode materials.

ZnO has been utilized as the electron transport layer in organic optelectronic devices, however, the shortcoming of ZnO is that activation by UV light is essential and ZnO-based optoelectronic devices give poor photoresponse under UV-cut light irradiation. We developed 1,1-dicyano-2,2-bis(methylthio)ethylene (DBE)-doped ZnO which can be activated by visible light irradiation. The DBE-doped organic solar cells showed better photovoltaic performances under UV-cut light irradiation than those of undoped ZnO-based devices and DBE-doped photodetectors also showed better frequency response without applied voltage under monochronic laser irradiation (450 nm) than that of undoped ZnO-based ones.

We investigated the influence of the pre-application of reverse bias on the potential-induced degradation (PID) of n-type front-emitter (n-FE) crystalline Si (c-Si) photovoltaic modules. Applying a prior positive reverse bias to n-FE cells delays charge-accumulation-type PID (PID-1), decreases in short-circuit current density (J_sc) and open-circuit voltage (V_oc). The prior positive bias accumulation may accumulate negative charges in the SiN_x, which leads to an increase in a duration for the positive charge accumulation by the PID test applying a negative bias. We also found that sufficiently long prior positive bias application results in the delay of PID-2, a fill factor reduction by the incursion of Na ions into the depletion region of the p–n junction of n-FE cells.

A basic experiment on 2-bit quantum calculations related to the Deutsch–Jozsa algorithm on a linear optical quantum computer using a liquid crystal (LC) device as a linear optical element was investigated. By controlling the phase of each of the two split laser beams and checking the number of photons at the output, it is possible to determine if the input bits are “not equal” or “not uniform.” The phase of the laser light was controlled using an LC device and a half-wave plate, and changes in the interference state of the output light and the number of photons were observed. The voltage application to the LC device is set at the input bit. As a result, the interference state changed according to the input bit, and changes in the number of photons were observed.

Herein, we characterized vacuum-evaporated thin-films of several new naphthalene tetracarboxylic diimide derivatives (C12-NTCDI-R) as n-type organic thin-film transistor (TFT) materials. They have head and tail structure with dodecyl chain and phenylalkyl (benzyl or phenylethyl) groups. By introducing fluorine groups into the phenylalkyl groups, the electron transport properties were enhanced. In addition, modifying the substrate surfaces with a fluorine-rich polymer material and annealing at the appropriate temperature allowed the TFT devices with C12-NTCDI-F₂Bn (R = 3,5-difluorobenzyl) to reach a maximum electron mobility (μ_e) of 0.11 cm² V⁻¹ s⁻¹. Furthermore, the alteration of the number of methylene carbons in the side chain of phenylalkyl groups in C12-NTCDI-R (namely, phenylethyl was used as R) affected the thermal behavior of their molecules during the thin-film formation. For the TFT device with C12-NTCDI-PhEt (R = phenylethyl), the μ_e reached the maximum value of 0.16 cm² V⁻¹ s⁻¹ by just surface modification, without annealing.

We have observed a new optical transition in the photoreflectance spectra of indirect-gap BiOCl thin films, which were grown on SrTiO₃ substrates. The position of this transition is close in energy to its bulk critical point (CP) energy. Moreover, these are significantly lower than a higher-lying direct-type CP from an energetic point of view. The spectral line shape analysis for our observed signal suggests the presence of an excitonic effect of this compound. We determined its dependence of the optical anomaly on temperature ranging from 80 K to RT. We adopted the Varshni model for this analysis. At last, we compared photonic properties of BiOCl with those of an element and binary semiconductors.

Nowadays circularly polarized luminescence (CPL) gains much attraction because it can be applied to a variety of fields. An efficient method to achieve CPL is blending achiral emissive polymers with chiral small-molecule additives as a chiral inducer. In this work, we demonstrate that 1,1′-Bi-2-naphthol (BINOL) acts as a chiral-inducer for poly (9,9-dioctylfluorene-alt-benzothiadiazole) (F8BT) in a thin film. In the thin films of F8BT/BINOL before annealing, only circular dichroism (CD) signals of BINOL were observed. After the annealing at 120 °C, clear CD signals of F8BT were observed indicating that the chiroptical property was induced on the F8BT thin films. The CD signal intensity was enhanced by increasing the annealing temperature up to 200 °C. By optimizing the concentration of BINOL and the annealing temperature, we obtained a dissymmetry CD value of 0.012. Photoluminescence quantum yield of the F8BT/BINOL thin films was also increased after the annealing.

This paper particularly explains potential-induced degradation (PID) of wafer-based standard p-type crystalline silicon technology. We present a single-encapsulation method that is useful for simulating PID in the laboratory to facilitate microanalyses of module-level solar cells. The PID testing is performed for the module with single-encapsulation and conventional modules. Their current–voltage characteristics and electroluminescence images are investigated before and after the PID tests. As described herein, we infer that the single-encapsulation modules generate PID almost in the same way as conventional modules do, and that the degradation trend is almost identical to that of conventional modules. Results of PID recovery tests indicate that the time necessary for PID recovery is always less than that for PID generation, irrespective of the encapsulation method.

We demonstrated the low-cost, eco-friendly fabrication techniques for inverted perovskite solar cells (iPSCs) with multilayered electron transporting layers using fullerene derivative (PCBM) and ZnO nanoparticles (NPs) by combining the meniscus-coating, push-coating, and transfer-printing techniques. We fabricated the multilayered planar iPSCs by low-temperature printing process not higher than 120 °C. Using meniscus coating method by reciprocating the glass rod back and forth repeatedly for 2–10 times, the material usage of PCBM and toxic chlorobenzene for iPSCs to 1/15 ~ 1/20 compared with the conventional spin-coating method, and 1/10 for ZnO NPs without extending the tact time. The material utilization rate became about 50% and 100% for meniscus- and push-coating, and a pin-hole-free uniform film was obtained for meniscus coating. We also reduced the initial degradation of perovskite layer during the deposition of ZnO onto the perovskite layer and improved the photovoltaic properties by the transfer-printing of ZnO from “wet-PDMS” stamp onto the thin PCBM film coated perovskite layers.

As a two-dimensional material, graphene has outstanding properties, with high electron mobility and mechanical resistance. Therefore, it has good prospects to be used in nanoelectromechanical resonators for sensitive mass detection. Previously, a graphene resonator with graphene/metal contact able to zeptogram level sensing and a with quality factor of 42 at RT was presented. However, over-etching under the metal contacts degrades the quality factor, and resonators with metal/graphene/metal contacts were fabricated to overcome this issue. A quality factor of 350 at RT was obtained for a device with width = 0.8 μm and length = 1 μm. Tunning of the resonance frequency from 9.6 to 13.5 MHz with dc gate bias is demonstrated. Temperature reduction from 300 K to 30 K increased the resonance frequency by about 4–6 MHz. The temperature influence on nonlinearity is also discussed, with a resonator demonstrating softening and hardening nonlinearities at cryogenic temperatures.

We have theoretically calculated the performance of a Zeeman-type spin polarizer, which consists of ferromagnetic (FM) and metal–insulator–semiconductor (MIS) gate nanostructures on top of an InAs two-dimensional electron gas (2DEG) channel. For the calculations, we have taken a realistic electron concentration, electron effective mass and the effective g-factor of the InAs 2DEG into account. In addition, we have assumed realistic FM and MIS structure sizes by conventional electron beam lithography. In the calculations, we have demonstrated clear oscillation of spin polarization over ±85%. Furthermore, we have shown that it works not only as a spin-polarized current generator but also a detector. In addition, we have proposed a novel spin-filter device utilizing the Zeeman-type spin polarizers. We have found that its conductance characterization allows us to evaluate the operation of Zeeman-type spin polarizers. We expect the Zeeman-type spin polarizers and spin-filter devices to open up a new field of spintronics.

In order to increase the photocurrent of tandem solar cells with a silicon (Si) solar cell as the bottom cell, we investigate the relationship between the localized surface plasmon resonance effect of gold nanoparticles (AuNPs) and the arrangement of AuNPs onto solar cells. AuNPs were deposited on P-type crystalline Si (c-Si) cells and Si heterojunction (SHJ) type cells. To obtain the localized surface plasmon resonance effect, the shape, size, and arrangement of the nanostructures are important. Especially, arrangement of AuNPs onto solar cells was inquired in this study because these parameters must be adjusted appropriately to produce a plasmon resonance in a specific wavelength range. It was found that the plasmon resonance effect is significant when deposited on the surface of c-Si cell rather than on top of the Ag electrode. AuNPs deposition on the c-Si cell surface leads to not only the localized surface plasmon resonance effect but also improvement in charge transport between the c-Si cell surface and Ag electrode. However, in the SHJ-type cells, the plasmon resonance effect was not observed because the surface layer was composed of a W-doped In₂O₃ (IWO) layer.

Laser annealing is an attractive process to form high-quality semiconductor films because of localized annealing area and short annealing time. In a previous study, a Cu₂ZnSnS₄ (CZTS) polycrystalline semiconductor film was realized using laser annealing in air as a light absorption layer for solar cells, although the crystallization was not sufficient in comparison with CZTS formed by the conventional thermal sulfurization process. In this study, we demonstrate a newly developed gas-atmosphere-controlled laser annealing system. A Cu–Zn–Sn–S-based precursor was formed, followed by laser annealing of the system. Laser annealing in air, Ar, and 5% H₂S/Ar gas was performed to investigate the influence of the gas species on the crystallization of the precursor. A 5% H₂S/Ar atmosphere promoted the crystallization of CZTS with the suppression of S desorption and Cu sulfide formation, while air and Ar atmospheres allowed the formation of Cu sulfide.

This study aims to develop a novel, efficient, and stable red-emitting material using benzoguanamine in which europium (Eu) is doped. We have demonstrated a simple, low-cost, gram-scale manufacturing process via an anneal synthesis method. By changing the Eu-doping concentration from 1 to 20 wt%, we found that the sample containing a Eu-doping concentration of 10 wt% shows the maximum photoluminescence quantum yield peak of 13.7%. Additionally, we confirmed the characteristic luminescence peaks of Eu ions. We also achieved a sharp red emission with a high color purity and a half-peak width of 13 nm. Furthermore, it exhibits high solubility in organic solvents, facilitating its application through wet processes. This new red-emitting material may present a promising solution for cost-effectively fulfilling the demand in flat panel displays and illumination light sources.

Yb mononitride (YbN) has a B1 (NaCl-type) structure similar to the structure of titanium nitride (TiN), which makes it a candidate for use in new hard coating materials via solid-solution hardening. Ternary system (Ti–Yb–N) thin films with various Yb contents were prepared on Si substrates by pulsed laser deposition, and compositional analysis confirmed that the resultant films contained the expected elements. Phase identification revealed that the ternary nitride (Ti_1−xYb_x)N thin films were all single-phase thin films that comprised YbN in TiN as a solid solution in the form of a B1- (Ti_1−xYb_x)N phase. Although the hardness of the Ti–Yb–N thin films could be increased to a certain degree compared with that of a pure TiN thin film, the improvement was not significant. A change in the optical properties of the (Ti_1−xYb_x)N thin films was indicated by a change in their color from gold to copper to silver with increasing x. This effect could make such thin films suitable for use as decorative coatings that are resistant to oxidation.

A cobalt sulfide (CoS_x) film compromising CoS₂ and Co₉S₈ nanograins was formed by sulfurizing the surface of a Co film to use for resistive switching (RS) memory. The work function and band gap of the CoS_x film were measured to be 4.78 eV and 2.18 eV, respectively. The CoS_x film was used as a resistive layer together with the Co film underneath as a bottom electrode, and a Ag or Cu film as the top electrode. Both Ag/CoS_x/Co and Cu/CoS_x/Co devices exhibited bipolar RS behavior with the capability of multi-level memory storage. The conduction of both devices in low resistive states was correlated with metallic filamentary paths following ohmic conduction, whereas Schottky emission originated at the Ag/CoS_x and Cu/CoS_x interfaces dominated in the high resistance state. The performance of Ag/CoS_x/Co and Cu/CoS_x/Co devices were compared and correlated with the properties of Ag and Cu electrodes.

The effects of soft X-ray irradiation and atomic hydrogen annealing on the reduction of graphene oxide (GO) to obtain graphene were investigated. To clarify the interaction between soft X-rays and GO, soft X-rays of 300 eV and 550 eV were used for C 1s and O 1s inner-shell electron excitation, respectively at the NewSUBARU synchrotron radiation facility. Low-temperature reduction of the GO film was achieved by using soft X-ray at temperatures below 150 °C at 300 eV, and 60 °C at 550 eV. O-related peaks in X-ray photoelectron spectroscopy, such as the C–O–C peak, were smaller at 550 eV than those at 300 eV. This result indicates that excitation of the core–shell electrons of O enhances the reduction of GO. Soft X-rays preferentially break C–C and C–O bonds at 300 and 550 eV, respectively.

A monovalent copper ion (Cu⁺) with the same valence as formamidinium is focused on the present work, and the effects of A-site inorganic cations on the electronic structures and device performance are discussed from the experiments and the first-principles calculations. The addition of inorganic cations increased the conversion efficiencies, and the copper-doped device showed the highest conversion efficiency. In particular, the hysteresis of current density–voltage characteristics was significantly suppressed by the addition of Cu⁺, which would be due to suppression of iodine ion (I⁻) diffusion by electrostatic interaction between Cu⁺ and I⁻. The addition of rubidium or cesium contributed to the increase in short-circuit current density by suppressing decomposition of perovskite crystals and formation of PbI₂.

Highly dispersed Pt nanoparticles supported on polyacrylonitrile-based carbon fiber substrates were prepared using a one-step liquid-phase process. Nanoparticles were grown on the fiber surface via electric resistance heating of the substrate in ethanol solutions containing a platinum-based organometallic complex as the nanoparticle precursor. After the synthesis, Pt nanoparticles were formed on the surface in highly dispersed states. Electrochemical characterization of the Pt nanoparticles was conducted using the substrates directly after synthesis. The Pt nanoparticles were electrochemically active and exhibited superior methanol oxidation properties.

We report the effect of substrate temperature on the vapor phase deposition of all-inorganic lead-free tin halide perovskite CsSnBr₃ thin films by co-evaporation of CsBr and SnBr₂. The amount of SnBr₂ supplied had little effect on the thin film when the substrate temperature exceeded the SnBr₂ effusion cell temperature (∼180 °C) indicating that the three-temperature growth was realized. The polycrystalline CsSnBr₃ thin films obtained by three-temperature growth showed micrometer-scale grains and no by-products. The average grain size increased for higher substrate temperatures following the Arrhenius law with an activation energy of 6.6 kJ mol⁻¹.

We have developed polymer-based inverted organic LEDs (iOLED) with a multilayered structure using orthogonal solvent, transfer-printing, improved meniscus-coating method that moves back and forth. Reciprocating the glass rod back and forth repeatedly, the material usage of electron-injection layers, polymeric light-emitting layer and polymeric hole-transporting layers were decreased to 1/10 for the film deposited onto the glass slide and to ca. 1/20 for the film deposited onto the elastomer stamp compared with the conventional spin-coating. We obtained higher external quantum efficiency (EQE) and lower operation voltage with solution processable metal oxide hole-injection layer and silver anode. The improvements in the threshold voltage and the maximum EQE were observed for the device with electron-transporting EIL, such as alcohol soluble phenanthroline derivative, which is an effective way to reduce the number of multilayers and the tact time for fabricating iOLEDs.

We used 3D micro-X-ray topography (3D μ-XRT) to construct a 3D structure of the threading screw dislocations (TSDs) in 4H-SiC. The 00012 diffraction condition, which can observe the strain field at a deep location in the sample, was employed in the 3D μ-XRT geometry. The dislocation core-lines of TSDs propagating from the surface to the interior of the sample were successfully captured spatially, and TSDs parallel to and inclined from the 〈0001〉 axis were both confirmed. Either the component 〈$\overline{1}\overline{1}20$〉 or 〈$\overline{1}100$〉 of the Burgers vector is considered to be mixed in the inclined TSD. To verify this suggestion, scanning electron microscopy (SEM) and cathodoluminescence (CL) spectroscopy were performed on the molten-alkali-etched TSDs. Both the SEM shapes and CL intensities of the TSD etch pits with inclination showed the features of the mixed component, supporting the correlation between the inclination and mixed component in TSD.

A single crystal strip of Ge, exhibiting a predominant crystal orientation of (001), was successfully grown within a 60 nm thick sputtered Ge film using the micro-chevron laser scanning method. The continuous lateral growth of the Ge strip was achieved through the strategic implementation of a thick cap layer and a SiO₂ interlayer between the Ge layer and cap layer. The thick cap layer was pivotal as a heat sink, effectively extending the period during which the Ge film remained molten, preventing unwanted nucleation. Yaw rotation of (001) crystal orientations, which typically trigger the formation of grain boundaries in Si, was found to be negligible in Ge. This observation offers compelling evidence for the potential for infinite lateral growth of (001)-oriented Ge strips. The Raman shift of the single crystal strip was measured at 297.4 cm⁻¹, indicating that the film exhibited tensile stress.

Terahertz (THz)wave wireless communication technologies have been developed to increase transmission capacity in 6G communications. Although the realization of integrated waveguides in THz wave transmitters and receivers is essential, bending loss in THz waveguides is important as well. This study investigated a THz waveguide based on a valley photonic crystal (VPC) structure. We numerically investigated a THz waveguide with sharp bending and Mux/DeMux filters based on the oblique rectangular resonator of a bearded-type VPC waveguide.

We demonstrate the reduction of the electron Schottky barrier height (SBH) to 0.48 eV at Mo/n-type Si junctions through the insertion of a semimetal Si-rich Mo silicide (MoSi_n, n = 7.9) film. Raman scattering measurements elucidated the persistence of the same amorphous structure within the MoSi_n film even when subjected to temperatures as high as 900 °C. This excellent thermal stability yielded a notable result: preservation of the SBH modulation effect even following annealing at 700 °C. Moreover, we investigated the capacitance−voltage characteristics of MOS capacitors, revealing that MoSi_n film has a remarkably low effective work function, measuring 4.1 eV when deposited onto SiO₂. The deposition of MoSi_n film was accomplished with an excellent coverage by using MoF₆ and SiH₄ gases. Thus, MoSi_n film is a promising contact material in advanced CMOS transistors.

Improving the photodetection performance of thin-film transistor (TFT)-based UV photodetectors (UVPDs), using thick channel layers to promote photocurrent (I_ph) or using thin channel layers to suppress dark current (I_dark) is typically a trade-off. In this work, UVPDs based on oxygen-doped Si-Zn-Sn-O (SZTO) TFT with a stack of Pt/NiO capping layers (CLs) to release the trade-off between I_dark and I_ph are demonstrated. The Pt CL creates a wide depletion region in the channel layer to allow the use of thick channels, but still maintains low I_dark, while the NiO CL forms a pn heterojunction to provide additional photogenerated carriers and enhance I_ph under UV irradiation. Experimental results show that the proposed 95 nm-thick oxygen-doped SZTO TFT with a stack of Pt/NiO dual CLs exhibits an excellent photoresponsivity of 2026 A W⁻¹ and photosensitivity of 9.3 × 10⁷ A A⁻¹, which are about 76× and 82.5× higher than a conventional 45 nm-thick SZTO TFT under 275 nm UV irradiation.

Rocksalt-structured (RS) Mg_xZn_1−xO films with x = 0.65–1.0 were grown on MgO (100) substrate using the mist CVD method. A comparative study of the RS-Mg_0.92Zn_0.08O films grown under slow and rapid-cooling rates apparently showed simultaneous reductions in the surface pit density, FWHM values for the X-ray diffraction peak, and defect-related cathodoluminescence (CL) for the film grown under the slow-cooling rate. CL spectra for the RS-Mg_xZn_1−xO films grown under the slow-cooling rate eventually showed near-band-edge emission peaks in the 180–190 nm spectral range for MgO molar fraction x ≥ 0.92 at RT.

Threshold voltage instability (shift) due to positive bias stress in GaN planar-gate MOSFETs was investigated. Gate dielectric (SiO₂) was formed by remote-plasma-assisted CVD on homoepitaxial Mg-doped p-type GaN layers with Si-implanted n-type source and drain regions. The threshold voltage shift of 5.8 V was observed after a stress voltage of 30 V for a sample without post-deposition annealing (PDA). The threshold voltage shift was significantly reduced to 1.4 V for a sample with PDA (800 °C for 30 min). Stress time dependences up to 6000 s were measured, revealing that the main origin of the threshold voltage shift is electron trapping into near interface traps (NITs). These results suggest that PDA is effective for the reduction of the NITs.

In this work, we present the biquadratic field H_bq contribution to increase a frequency of spin-torque oscillation (STO) in an orthogonal magnetization structure in simulation, and realize such an orthogonal structure by preparing Co/Pt lamination as the bottom perpendicular magnetic anisotropy layer, Cr or Cu as the spacer, and experimentally realize Fe₉₀Co₁₀ as the top free layer. Our observations of the Cr-spacer sample reveal a notable challenge in achieving magnetic saturation, underscoring the role of H_bq in suppressing magnetization reversal and its potential to broaden the STO current range and increase the STO frequency. This leads to the manifestation of spin-transfer-torque oscillations in an orthogonal structure, bolstered by robust biquadratic magnetic coupling, thus attaining high and stable STOs in the simulations.

We conducted thermodynamic analysis and kinetics calculations for hexachlorodisilane (Si₂Cl₆), which is used as source gas for the CVD of silicon nitride films. Thermodynamic analysis clarified Si₂Cl₆ decomposes almost completely to SiCl₄ and SiCl₂ in the range of 600 °C–1100 °C under equilibrium condition. Therefore, it is estimated that the main gas-phase reaction of Si₂Cl₆ is represented by the reaction Si₂Cl₆ → SiCl₄ + SiCl₂. Thermodynamic analysis also shows that the Si₂Cl₆ system has a larger equilibrium partial pressure of SiCl₂ than SiH_xCl_4-x (x = 1 ~ 3) systems. Kinetics calculations revealed Si₂Cl₆ decomposes by 90% in 0.11 s at 600 °C, and 0.55 × 10⁻³ s at 800 °C, respectively. The time-dependent pyrolysis ratio of Si₂Cl₆ becomes larger as the total pressure decreases at 600 °C. On the other hand, the ratio is almost the same regardless of total pressure at 800 °C. These results will help optimize CVD process conditions using Si₂Cl₆.

This research presents a gas sensor system featuring a selenized two-dimensional (2D) film as its primary sensing material, integrated with metal-gate-coupled floating gate devices to enable self-convergent calibration. The inherent variability in resistance levels of 2D gas-sensing materials across different devices has been a significant challenge, resulting in substantial deviations of the output signal within the sensing circuit. To address this issue, we introduce a novel self-convergent operational technique, which effectively mitigates the impact of resistance variations thereby enhancing the precision and reliability of gas-sensing outcomes. The proposed gas sensor system promises to deliver consistent and accurate results, even with device-to-device resistance variations, making it a valuable contribution to gas-sensing technology. This work holds substantial potential for various applications requiring highly precise gas detection and quantification.

The effects of strain on the carrier transport in gallium nitride (GaN) are investigated using a full-band Monte Carlo method combined with an empirical tight-binding method. The impacts on the carrier mobility, carrier drift velocity, and breakdown characteristics are discussed. Compressive uniaxial or tensile biaxial strain is beneficial for achieving higher hole mobility in vertical GaN devices due to the light-hole band being lifted above the heavy-hole band. Analysis of the breakdown phenomena indicates that strain does not degrade the breakdown characteristics in terms of its effect on the band structure.

Surge voltage at insulated gate bipolar transistor turn-off switching was analyzed with dependence on cell design parameters. Although drift layer thinning is effective in improving trade-off characteristics between turn-off loss E_off and on-state voltage V_on, voltage surge is induced due to the quickly expanding depletion layer. Therefore, the surge voltage V_surge has also a trade-off relationship with V_on at the same E_off condition. The origin of the voltage surge was analyzed using TCAD simulation, and the total amount of remaining holes in the drift layer during turn-off switching is a key factor for the V_surge. A narrow mesa structure and thick buffer layer are effective for the improvement of trade-off characteristics between V_surge and V_on. However, the optimum buffer layer thickness depends on the voltage-class due to the speed for punch through.

We propose a polarization-controlled ultrasmall single-photon emitter that combines a single photon source and an elliptical split ring (SR) type metamaterial element. Simulations using the finite difference time domain method showed that in elliptical metamaterial elements, the annular mode is suppressed and axial electric field oscillations occur preferentially, resulting in upward light emission from the photon source with controlled polarization. We fabricated prototype devices by focused ion beam lithography and PbS quantum dots. Evaluation of the optical properties of the devices revealed that the emission spectrum width was narrower and the emission lifetime was shorter than those of quantum dot ensemble, and that the polarization was controlled along the minor axis of ellipse. These suggested that the quantum dot emission and the metamaterial element resonated as expected. The combination of a single quantum dot and a single elliptical SR-type metamaterial element was shown to have the potential to operate as a polarization-controlled single-photon emitter.

We demonstrate the scalability of cavity-free planar integrated thermoelectric (TE) devices using silicon nanowires (Si-NWs), where miniaturizing the thermoelement by shortening the Si-NWs improves the areal power density. Shortening the Si-NW length decreases the temperature difference between the Si-NW ends and the open-circuit voltage. Meanwhile, the integrated number density of the thermoelement is increased by shortening the Si-NW length, thereby preserving the total electrical resistance. Bileg devices comprising both n- and p-type Si-NWs exhibited superior performance compared to unileg devices comprising only n-type Si-NWs. In unileg devices, the hot and cold electrodes in the adjacent thermoelements are interconnected through metal wiring, which leaks heat, resulting in a lowering of the temperature difference between the Si-NW ends. This study yields structural design guidelines for planar-integrated TE devices using Si-NWs.

This paper reports on the newly developed silicon insulated gate bipolar transistor (IGBT) for Toyota’s fifth generation power devices. The designed IGBTs with a grid-patterned trench provide a lower on-state voltage (V_ON) while maintaining a wide mesa width. A successful 15% reduction of V_ON was accomplished by enhancement of the two-dimensional hole storage effect. Furthermore, the grid-patterned trench IGBT merged with Schottky and multi-layered anode technology demonstrated a significant improvement in the trade-off curve between the reverse recovery current (I_rr) and V_ON in a reverse conducting IGBT without a lifetime control process. As a result, the grid-patterned trench IGBT has contributed to reduce power loss and volume in the new power control units.

Defect formation in epitaxial Mg₂Sn_1–xGe_x thermoelectric (TE) thin films grown via MBE was studied. We examined the defect formations and structures using cross-sectional transmission electron microscopy and positron annihilation spectroscopy. The defect formation tends to be influenced by Ge incorporation into the Mg₂Sn matrix phase of epitaxial thin films. Mg vacancies (V_Mg) were identified as point defects, primarily concentrated in the film’s mid-layer. In films with higher Ge composition, stacking faults were observed. The concentration of vacancy-type point defects decreased as the Ge concentration increased. This implies that the vacancy atoms, which would have otherwise been created by increasing chemical pressure due to the higher Ge content, might have played a role in the formation of stacking faults. The high concentration of vacancy-type defects resulted in the lowest thermal conductivity, demonstrating their significance as effective phonon scattering centers in epitaxial TE films.

Molybdenum (Mo) doping is a pivotal strategy to enhance the performance of bismuth vanadate (BiVO₄) photoanodes in photoelectrochemical (PEC) devices. This research explores the effects of Mo-doping on BiVO₄’s electronic properties, uncovering mechanisms behind improved PEC behavior. Mo-doped BiVO₄ was produced via single-target RF sputtering, leading to films with increased photocurrent density. Optimal results were achieved with a 3% atomic ratio of Mo and 15% oxygen partial pressure during deposition. Analysis of the local structure revealed Mo⁶⁺ substituting V in the BiVO₄ host. Mo doping introduced defect states within the VB, partially occupying the d-band of V⁴⁺ and creating additional electron states, causing the fermi level to shift from 1.75 to 2.19 eV from the VB edge. This study underscores the adaptability of Mo-doping in shaping BiVO₄’s electronic characteristics, opening new pathways in advanced energy conversion technologies.

We fabricated an inversion mode n-channel thin-film-transistor (TFT) on polycrystalline (poly-) Ge at low temperatures for monolithic three-dimensional large-scale IC (3D-LSI) and flexible electronics applications. Based on our previously reported solid-phase crystallization (SPC) method, we designed an n-channel TFT fabrication process with phosphorous ion implantation to provide the source/drain (S/D). We succeeded in fabricating an n-channel TFT with typical electrical characteristics on poly-Ge and confirmed its operation mode to be inversion mode. However, the fabrication process included a high temperature (500 °C) step for S/D activation. To reduce the process temperature, we used a metal-induced dopant activation method and successfully reduced the activation temperature to 360 °C. This combination is expected to pave the way for high-performance 3D-LSI and flexible electronic devices based on SPC-Ge.

A detailed understanding of cycling induced degradation of interfacial SiO₂ in HfO₂-FeFET devices is presented. By direct observation of electron trapping to cycling-generated trap sites in various time domains, we found that trap sites disappear over time. This recovery of the degradation has a non-negligible impact on the retention characteristics after cycling.

This work presents a new serially connected strip-type acoustic resonator using X40°Y-LiNbO₃ (LN) (XSAR). The proposed design is composed of a serial of strip-type bulk acoustic wave resonators connected together. The volume between two strips is penetrated through to eliminate the spurious response and enhance quality (Q) factor. The Euler angle (90°, 90°, 40°) of LN is optimized to obtain the maximal piezoelectric constant e₃₄, which determines the coupling factor of shear mode of the resonator. Simulation results show that the new design is superior in terms of impedance ratio (Z-ratio) and bandwidth (BW) and exhibits the largest BW of 32% ever reported. The slow shear mode spurious response is successfully removed by the proposed penetrated structure combined with narrow strip. Furthermore, the device was prototyped in macro scale using 350 μm thick LN, by which the ultra-wide BW and potentially good performance in HF were confirmed.

We investigated high-k polymers as gate dielectrics to develop an atmospheric-pressure process for organic transistors. A high-k terpolymer, p(VDF-TrFE-CFE), was formed by drop-casting. Self-assembled molecules (SAM) of triptycene derivatives have been utilized as surface modifiers for high-k gate dielectrics to improve the transistor characteristics. High-k terpolymers reduce the operating voltage of organic thin-film transistors (TFTs) by up to 2.5 V to their high capacities. Introducing a SAM onto the gate dielectric surface improved the organic TFTs’ mobility by more than three orders of magnitude. In addition, a terpolymer and a SAM can be formed simultaneously by drop-casting. This atmospheric processing of the gate dielectric is expected to contribute to organic electronics in large-area or mass production, such as roll-to-roll processes.

Simulations based on the sp³d⁵s^* empirical tight-binding method were performed to provide a statistical understanding of the electronic structures and bandgap distributions of III–V (InAs, InP, GaSb) and IV (Si) semiconductor quantum dots (QDs) with surface roughness. The electronic states and wavefunctions of QDs with surface roughness of different sizes, shapes, and materials were computed. The effects of surface roughness on the electronic structures and the bandgap distributions of QDs were investigated. The results show that the bandgaps of QDs of considered materials/sizes/shapes increase on average when introducing surface roughness. It is shown that the simulated bandgap distributions of QDs with surface roughness can be reproduced by a simple model formula, which can be applied to different materials, sizes, and shapes. The model formula was derived by assuming that removing and adding of one atom procedures are independent random processes.

Ferroelectric FETs (FeFETs) are nonvolatile memory devices in which the threshold voltage is programmed by the polarization state of the ferroelectric gate insulator. In this study, we theoretically investigate the impact of charge trapping on the ferroelectric memory window by establishing an accurate picture of mechanism determining the threshold voltage. Under the co-existence of ferroelectric polarization and charge trapping, we show that the threshold voltage shift cannot be described by the linear superposition of the ferroelectric-polarization component and the charge-trapping component. We point out that the misuse of the linear superposition can significantly underestimate the trapped charge density in FeFETs, sometimes even by an order of magnitude. The concept of effective dielectric constant determined by both the original dielectric constant and the polarization state is discussed to describe the relation between the threshold voltage shift and trapped charge density in FeFETs.

Nanoscale short-channel oxide thin film transistors (TFTs) have attracted widespread research interest due to their potential applications in advanced display and memory devices. In this work, we fabricate 10 μm channel length indium-tin-oxide (ITO) TFTs and analyze the uniformity and repeatability of ITO TFTs at the long channel range. Then we fabricate ITO TFTs with a series of channel lengths ranging from 10 μm to 150 nm and through an optimized process we finally fabricate 130 nm channel length high-performance ITO TFTs with an on-state current of 93 (μA/μm), a subthreshold swing of 102 (mV decade^–1), and on/off ratio over 10⁷ at a drain voltage of 3 V.

The Fermi level pinning effect significantly limits the application of electrical devices based on two-dimensional materials like transition metal dichalcogenide. Here, a CMOS inverter, which is comprised of an n- and a p-MoTe₂ FET with optimized properties, has been successfully fabricated by using contact doping and channel encapsulation methods. Contact doping is to control the polarity of MoTe₂-FET and improve contact properties, which is achieved by laser irradiation in different environmental conditions. The channel of two MoTe₂-FETs was encapsulated by hexagonal boron nitride (h-BN) to enhance carrier mobility and device stability. The fabricated CMOS inverter showed a very high gain value of 31 at V_dd = 4 V at RT.

Magnesium (Mg)-induced lateral crystallization (Mg-ILC) of amorphous germanium (Ge) on a SiO₂ stacked structure was investigated. From Raman mapping images, the critical annealing temperature necessary to induce Mg-ILC of amorphous Ge was estimated to be about 350 °C. Furthermore, the Mg-ILC length was truly narrow (∼2 μm) compared with other metal catalysts after annealing at 350 °C for 1 h. To enhance the Mg-ILC, we have examined a two-step annealing method for Mg-ILC of amorphous Ge on SiO₂. The Mg-ILC length is significantly enhanced (∼4.5 times) by using a two-step annealing process, which is due to the enhancement of Mg diffusion into amorphous Ge during first-stage low-temperature annealing.

Low-temperature poly-Si thin film transistors (TFTs) with a metal/oxide/nitride/oxide/silicon structure were fabricated to investigate the feasibility of suppressing threshold voltage variations between TFTs. By applying relatively high positive and negative gate bias voltages, threshold voltage could be tuned positively and negatively, respectively, by injecting charges into the charge trap layer. Stability of threshold voltage against positive gate bias at a level close to the actual circuit operation was not degraded compared with the case of an as-fabricated TFT. A uniform threshold voltage distribution could be achieved in a preliminary test using 16 TFTs by converging threshold voltages with a target value by tuning the threshold voltage for each TFT.

We examine the influence of doping profile optimization on the trade-off relationship between modulation bandwidth and optical loss in an InP-organic hybrid (IOH) optical modulator, comparing it with a Si-organic hybrid (SOH) optical modulator. By incorporating the RF transmission line model, which enables a more precise modulation bandwidth analysis than the RC constant model, we demonstrate that the IOH modulator can achieve a modulation bandwidth of over 500 GHz with a 2 dB loss, capitalizing on the higher electron mobility of InP. In contrast, the SOH modulator cannot attain a 200 GHz modulation bandwidth with acceptable optical loss. Furthermore, we explore the potential for further enhancing the modulation bandwidth of the IOH modulator by shortening its length, making the IOH modulator a promising candidate for future ultra-high-speed optical modulation.

Locally dislocation-free SiGe-on-insulator (SGOI) is fabricated by CVD. Lateral selective SiGe growth of ∼30%, ∼45% and ∼55% of Ge content is performed around ∼1 μm square Si(001) pillar located under the center of a 6.3 μm square SiO₂ on Si-on-insulator substrate which is formed by H₂-HCl vapor-phase etching. In the deposited SiGe layer, tensile strain is observed by top-view. The degree of strain is slightly increased at the corner of the SiGe. The tensile strain is caused by the partial compressive strain of SiGe in lateral direction and thermal expansion difference between Si and SiGe. Slightly higher Ge incorporation is observed in higher tensile strain region. At the peaks formed between the facets of growth front, Ge incorporation is reduced. These phenomena are pronounced for SiGe with higher Ge contents. Locally dislocation-free SGOI, which is beneficial for emerging device integration, is formed along 〈010〉 from the Si pillar by lateral aspect-ratio-trapping.

Ring resonators are traditionally popular optical devices that apply to various components in photonic ICs. They also play an important role in the on-chip generation of many novel optical states in topological systems and non-Hermitian systems. Unidirectional lasing of ring resonators is used in many such systems to create exotic states of light including optical vortexes and optical skyrmions, but the unidirectional behavior has not been fully understood. Previous research has constructed a simplified model to explain the steady state behaviors of unidirectional ring resonators, but the carrier dynamics and spontaneous emission were omitted. In this work, we give a numerical analysis of unidirectional ring resonators with an S-shaped coupler. We identified the importance of the gain saturation to robustness against backscattering and high unidirectionality by comparing to the model without saturation. We also discuss the effect of asymmetrical coupling on the deterministic realization of unidirectionality.

In this study, a novel self-rectifying twin-bit via resistive random-access memory (Via RRAM) has been implemented in a cross-bar memory array using via and metal layers within the standard FinFET Back End of Line processes. By using SiO_x and TaO_x, this RRAM integrates well with advanced CMOS logic circuits. Additionally, this device can be put into a unique diode state (DS) with asymmetric IV characteristics after specific operations. By placing the device in its DS, sneak current paths can be effectively suppress to allow the realization of 1 R cross-bar array. Hence, the newly proposed cell provides a distinct advantage of integrating RRAM cross-bar array by top metal structures. This high-density array structure, featuring compact cell sizes, can be produced without the requirement for additional masks or processes. Moreover, as the density increases, it demonstrates reduced operational voltages and accelerated operating speeds.

The distance between memory and central processing unit has led to a memory wall. To solve it, an in-memory technology that performs both memory and computation has been studied. To realize an ideal in-memory computing, we propose a positive feedback FET based on vertical NAND flash structure that can act as a memory and perform computation. The device can reconfigure the processing operations into AND or OR operations depending on the control gate bias. It performs memory by accumulating charge in the body, and logic operations can be performed by reading data stored in the charge trap layer. After this, it can also perform a writing operation. This component enables memory and read-compute-write operations, making it capable of implementing intrinsic in-memory computing. As a result, in this study, we designed and verified a structure that implements the core principles of in-memory computing.

Robustness under high-temperature clamped inductive turn-off has been compared systematically among 3300 V scaled insulated gate bipolar transistors (IGBTs) with scaling factor k from 1 to 10 by technology computer-aided design simulations. Degradation of the reverse bias safe operating area (RBSOA) has been observed in proportionally scaled IGBTs, especially at a high turn-off current density. A non-proportional scaling method has been proven to be able to restore the robustness degradation with RBSOA close to the original k = 1 case. Moreover, the method for adjusting R_pf (P-floating connecting resistor) adds more flexibility to device design and also improves the overall electrical performance of non-proportionally scaled IGBTs. The adjustment of R_pf has been found to have a minimal effect on the RBSOA.

Electrostatic discharge (ESD) and electromigration are critical issues that significantly impact the reliability of ICs. While both of these phenomena have been studied independently, the combination of the two, ESD-induced electromigration, has received less attention, potentially compromising IC reliability. This work analyzes various types of metal with different lengths, widths, and angles commonly used in ESD protection circuits in the CMOS process. The objective is to observe their behavior under continuous ESD zapping. The ESD-induced electromigration of metallization in the CMOS process has been analyzed, and metal sensitivity to system-level ESD events has also been identified. It is also analyzed from the perspective of energy that the ESD energy that metal can withstand will decrease as the ESD voltage increases, which will be even more detrimental to the ESD reliability of ICs. The findings from this study aim to provide valuable insights for designing metal lines in ICs to enhance ESD protection.

A 28 nm 512 Kb NanoBridge (NB) non-volatile memory is developed for an energy-efficient microcontroller unit. 0.11 pJ/bit read energy is achieved by utilizing an inverter sense scheme thanks to large ON/OFF conductance ratio of a split-electrode NB. The read energy is 71% and 54% less than those of a ReRAM and a silicon oxide nitride oxide silicon commercial embedded NOR flash at the same technology node, respectively. Moreover, a 28 nm 32 bit RISC-V microcontroller unit embedded with a 2 Mb NB non-voltage memory is fabricated and achieves 80 MHz operation frequency.

There is a growing demand for physical reservoirs that operate with low power consumption and low computational cost. We have conducted research on the basic properties of Ag₂S reservoirs, which are a type of physical reservoir. However, little research has been conducted on their applications. In this study, as a first step toward the practical application of Ag₂S reservoirs, we implemented two types of rock-paper-scissors judgment systems using Ag₂S reservoirs. In these experiments, we were able to demonstrate fast learning in the reservoir by comparing the results with methods using a single-layer perceptron and a convolutional neural network. In addition, we could obtain a maximum accuracy rate of about 98%.

This paper reports nanobeam X-ray diffraction (nanoXRD) measurement results for strained SiGe spintronics devices grown by MBE. A quantitative nanoXRD analysis verifies that in-plane strain is properly exerted on the SiGe spin channel layer in the device showing enhanced spin diffusion length, whereas overall strain relaxation and local change in crystallinity occur in the sample with unclear spin signals. Crystal defects such as dislocations and stacking faults found in cross-sectional transmission electron microscopy observations are correlated with the results of nanoXRD analysis and their influence on spin transport properties are discussed. The present results demonstrate the capability of the nanoXRD based method for quantitative nondestructive analysis of strain distribution and crystallinity in real device structures.

The electronic states in 4H-SiC MOS inversion layers are theoretically analyzed using the empirical pseudopotential method (EPM). The analysis shows that the Wannier–Stark localization occurs, which is absent in an effective mass approximation (EMA). The Wannier–Stark localization modifies the electronic states in the MOS inversion layers. A model is proposed to describe the in-plane dispersion of subbands affected by the Wannier–Stark localization. The differences between the EPM and EMA results for the subband energy levels and the in-plane effective masses are discussed.

We propose a novel training method named hardware-conscious software training (HCST) for deep neural network inference accelerators to recover the accuracy degradation due to their hardware imperfections. Existing approaches to the issue, such as the on-chip training and the in situ training, utilize the forward inference data that are obtained by the inference accelerators for the backpropagation. In the approaches, since the memory devices that are used for the weights and the biases are to be switched after each epoch, the total number of the switching in the training process grows too large to avoid the problems of endurance limitation, nonlinearity and asymmetry in the switching of the nonvolatile memories used for the weights and the biases. The proposed training method is totally conducted by software whose forward inference path and backpropagation reflect the hardware imperfections, overcoming all the above problems. The HCST reformulates the mathematical expressions in the forward propagation and the gradient calculation with the backpropagation so that it replicates the hardware structure under the influence of variations in the chip fabrication process. The effectiveness of this approach is validated through the MNIST dataset experiments to manifest its capability to restore the accuracies. A circuit design is also disclosed for measuring the offset voltages and the open loop gains of the operational amplifiers used in the accelerator, showing that the chip area overhead is minor.

This paper reports a stacked Dynamic random-access transistor (DRAM) memory structure that is based on gate all-around nanosheet access transistor. A TCAD study is done to compare nanosheet DRAM and conventional saddle fin recessed channel access transistor (SRCAT) in terms of DRAM electrical characteristics and its row hammer-induced leakage. The nanosheet DRAM shows superior characteristics in terms of current driving capability, speed, and refresh than SRCAT. The nanosheet DRAM also shows significantly lower hammer-induced failure as compared to SRCAT because the original leakage path from the cell to the neighboring cell gets blocked due to the nanosheet device structure. We also investigate the effect of spacer length on nanosheet DRAM characteristics and show that extended spacer length is favorable for having better DRAM characteristics due to the floating body effect. Our study demonstrates the potential, and advantages of nanosheet DRAM architecture compared to the conventional SRCAT DRAM.

The accumulation of photoinduced carriers at the SiO₂/Si interface was observed via graphene transport. CVD graphene was transferred to a lightly p-doped silicon substrate with a SiO₂ dielectric layer and served as a charge sensor for detecting the accumulation of photoinduced carriers at the SiO₂/Si interface. The sample was cooled to 4.2 K to realize an undoped silicon substrate without intrinsic carriers. Photoexcited carriers in the undoped silicon substrate were collected at the SiO₂/Si interface via a gate voltage and the carrier polarity was controlled well by the polarity of the gate voltage set during the light illumination. The photoinduced carrier density was determined by the number of photons incident on the silicon substrate with a photon–electron conversion efficiency of about 0.036. These results may provide polarity control of the conduction channel (n- or p-type) in standard Si-MOS structures, paving the way for optically programmable Si-CMOS.

The precision design of beta-phase gallium oxide (β-Ga₂O₃) power devices requires appropriate physical properties, including a breakdown electric field intensity. To clarify the breakdown electric field in β-Ga₂O₃, we fabricated vertical Schottky barrier diodes (SBDs) with a field plate (FP) on a β-Ga₂O₃ (001) epitaxial layer and simulated their breakdown situations under reverse bias. The calculated breakdown voltages of the SBD with a single FP were consistent with the experimental values when assuming an electric field criterion of 5 MV cm⁻¹ in β-Ga₂O₃. By designing a SBD with a double FP according to the electric field criterion above, a breakdown voltage over 2 kV and specific on-resistance of 6.9 mΩ cm² were realized simultaneously. This results in an electric field criterion that is useful for the design parameter of β-Ga₂O₃ devices to balance a high breakdown voltage and low on-resistance.

In this paper, we report achieving extremely high-density packing in high-voltage vertical gallium nitride (GaN) nanocolumn Schottky barrier diodes (NC-SBDs) through the adoption of a bottom-up process. The NC-SBDs were formed via epitaxial growth using Titanium-mask selective area growth (Ti-SAG) by rf-plasma-assisted MBE (rf-MBE), realizing a packing density equivalent to exceeding 10 million columns/mm². Our fabricated NC-SBDs with a period of 300 nm, a diameter of 250 nm, and a drift length of 1.3 μm demonstrated a breakdown voltage (BV) of 260 V with an on-resistance of 2.0 mΩcm², yielding an excellent figure of merit of 33.8 MW/cm² for nanocolumn-based high-voltage devices. We also discuss dielectric reduced surface field effect and impurities within the nanocolumns as potential factors contributing to the achievement of higher BV devices.

Using laser power sweep Raman spectroscopy, this research reports that the thermal transport changes with the wire width of silicon-germanium (SiGe) nanowires (NWs). The temperature in SiGe NWs was calculated using the relationship between Raman shift ω and temperature T (dω/dT) to evaluate the correlation between the thermal transport mechanism and SiGe NWs structure. We clarified that the thermal conductivity of the SiGe NWs decreases as the wire width becomes narrower. Also, a positional dependence of the thermal conductivity properties of the SiGe NWs was observed by laser power sweep Raman spectroscopy.

The orientation dependence of band alignments and the formation of dipoles at the 4H-SiC/SiO₂ interface are theoretically investigated on the basis of first-principles calculations. The calculations demonstrate that the offsets of valence and conduction bands depend on the surface orientation and chemical bonds at the 4H-SiC/SiO₂ interface. When we exclude the interfaces with C–O bonds which result in CO desorption, the calculated conduction band offset (CBO) on the Si-face with Si-O bonds is larger than those on the C-face with C–Si bonds and m-face with both Si-O and C–Si bonds. Furthermore, it is found the atomic configurations at the 4H-SiC/SiO₂ interface result in the formation of dipoles, whose magnitude is large for Si–O and C–O bonds. The formation of large dipoles significantly changes the band structure of 4H-SiC, resulting in large conduction bands offset. Therefore, the formation of a Si-O bond with large dipoles at the interface is of importance in order to obtain a large CBO. The calculated results give insights into improving the reliability of SiC MOSFETs.

In this work, behaviors of higher-order modes in an I.H.P. SAW device with a layered LiTaO₃/SiO₂/SiN/Si substrate are investigated, and techniques for their suppression are discussed. First, variation of the higher-order mode responses with structural parameters are analyzed using two-dimensional FEM. Then, on the basis of the calculated data, structural parameters are determined for higher-order modes suppression while retaining the main mode characteristics. In addition, the effectiveness of the optimized structure is confirmed using measured data from a fabricated one-port resonator and filters. In the measured data, it is shown that higher-order modes in a specific frequency range are successfully suppressed, while the main made characteristics, such as Bode-Q, bandwidth, and TCF, are almost equivalent to those of conventional I.H.P. SAW resonators and filters.

We present our systematic theoretical study by performing ab initio calculations to clarify the behavior of adsorption for constituent atoms such as Al and N on a vicinal AlN(0001) surface with step edges and kinks during metal-organic vapor-phase epitaxy (MOVPE). The calculations reveal that the surface reconstruction affects the adsorption of Al and N adatoms near the kinks and step edges. Furthermore, we find the incorporation of an Al adatom at the kink and that of N adatoms not only at the kink but also in the terrace regions. The calculated results give some insights for an atomic-scale understanding of the step-flow growth during the MOVPE growth of AlN.

We have demonstrated the formation of ultrathin Ni silicide on SiO₂ by annealing Si/Ni/Si structures and have systematically evaluated the impacts of the Si layer thickness on oxidation, surface roughening and the silicidation reaction. X-ray photoelectron spectroscopy analyses revealed that suppression of Ni oxidation due to the top Si layer makes it possible to form an ultrathin Ni silicide layer with a thickness of around 2 nm. Then, it turned out that the composition ratio of Ni and Si depends on not only the annealing temperature but also the initial thickness ratio of the top and bottom Si layers. Furthermore, this work clarified that the ultra-thin top Si layer has a large impact on the surface morphology during Ni silicide formation with diffusion and preferential oxidation.

This work demonstrates the applicability of periodically slotted SiO₂ (PSS) structures (PSSs) for controlling the urface acoustic wave (SAW) slowness shape and eliminating the transverse modes in layered SAW structures represented by the Incredible High Performance SAW. The use of PSS comprising SiO₂ blocks and air gaps provides larger acoustic anisotropy for slowness shape manipulation. The effect of PSSs on transverse mode suppression is revealed. The slowness curves and frequency responses with different structural parameters of PSS are calculated by FEM analysis. Then, an electrode is designed for further spurious resonance suppression. A hammer-head configuration and Si₃N₄ overlay are applied. It is shown that higher-order transverse modes in the passband were well suppressed without reduction of the electromechanical coupling factor.

The performance of a cavity-free micro thermoelectric generator (TEG) consisting of Si nanowires (Si-NWs) is investigated in this research. In the cavity free structure, one side of TEG is heated by a structure of a metal overlayer, called the Heat Guide (HG), to supply heat selectively to specific microregions within the device. Thus, heat energy flows in the perpendicular direction, which forms steep a temperature gradient in the NWs. However, the performance can be varied by varying the thickness of the HG. In this work, the impact of the HG structure of an integrated TEG upon the power generation performance was experimentally demonstrated. Higher metal HG thickness and thick interlayer dielectric (ILD) thickness show higher power generation performance with the integrated device.

A computational model to simulate the electron scattering by step-terrace structures is proposed. Using the proposed model, the transmission function, conductance, and mean free path are calculated. It is shown that the dependence of the mean free path on the perpendicular electric field, step position fluctuation, and step height is consistent with a surface roughness scattering theory based on the disordered periodicity of the step-terrace structures. It is suggested that the proposed model could be used to simulate the surface roughness scattering due to step-terrace structures in SiC MOS inversion layers.

A comprehensive comparison of the electrical characteristics between the planar (P-FTJ) and cylindrical ferroelectric tunnel junction (C-FTJ) is conducted based on physical modeling and simulation. The FTJ architecture consists of metal-ferroelectric–dielectric-metal stacks. Two configurations of C-FTJ are considered depending on whether the position of the ferroelectric (FE) layer is close or away from the inner electrode. The differences between the P-FTJ and C-FTJs in the distributions of the electric field and FE polarization are analyzed. The resultant tunneling electroresistance (TER) is explored as a function of the inner radius, FE thickness, dielectric thickness, and remnant polarization. These simulation results offer physical insights into achieving highly integrated three-dimensional storage structures by improving the TER ratio.

A probabilistic-bit (p-bit) is the fundamental building block in the circuit network of probabilistic computing (PC), producing a continuous random bitstream with tunable probability. Among the various p-bit designs, the recently proposed ferroelectric FET (FeFET)-based p-bit is found to possess the advantages of both ultralow hardware cost and CMOS compatibility. In this work, we perform a comprehensive analysis of the stochasticity in ferroelectric (FE) material, revealing the mechanisms of domain dynamics and temperature on FE stochasticity. The device-to-device variations of several device/material properties are then evaluated, which affect the probabilistic-curves to different extents. Furthermore, the integer factorization is performed based on the invertible logic circuits comprising of FE p-bits to verify its functionality. The accuracy of integer factorization is found to highly depend on FE p-bits’ stochasticity, and various methods have been implemented to optimize its performance, providing unique insights for future large-scale PC applications.

Threading dislocation density (TDD) in a Ge heteroepitaxial film on Si is reduced based on a dislocation trapping mechanism in a patterned substrate. An array of V-shaped grooves on the submicron scale is patterned in the [110] direction on a (001) Si substrate. Epitaxial growth of Ge with a thickness of 1 μm by CVD realizes a reasonable flat surface despite the non-flat starting surface. The TDD in Ge for a V-shaped groove pattern of 0.5 μm in width with an inter-groove distance of 0.3 μm is obtained as low as 4 × 10⁷ cm^–2, which is lower than about 6 × 10⁷ cm^–2 for the rectangular one with the same groove width and inter-groove distance and about 22 × 10⁷ cm^–2 for the unpatterned one. The reduction is attributed to the dislocation trapping at the groove regions, as observed by cross-sectional transmission electron microscopy.

The design proposes a low-dropout regulator incorporating dynamic loop gain control to accommodate a wide load current range. It introduces a multi-phase error detection mechanism, a tri-state damping selector, and a dynamic integration shift register. These elements predict load variations and dynamically increase loop gain when the load variation is significant, all without increasing the input clock. This approach achieves a fast response and low power consumption simultaneously. The chip was implemented in 0.18 μm CMOS, with an input operating voltage of 1.2 V and an output voltage of 1 V. The measured results revealed that it could provide output currents from 0.5 to 150 mA, indicating that it had a 300 times wider current range. The maximum current efficiency was 99.93%, and the core area was 0.198 mm².

In this study, we systematically studied the polarization properties of thin ferroelectric Hf_0.5Zr_0.5O₂ (HZO) films annealed by flash lamp annealing (FLA). We have recently proposed a unique annealing method, two-step FLA, which features high temperature annealing with a minimal increase in thermal budget. Using this technique, we observed improvements in the polarization properties in 5 nm HZO films. These were an increase in the remanent polarization (2Pr) to 24.2 μC cm⁻² and better durability compared with other more conventional annealing techniques. In addition, we confirmed that there was no clear degradation in 2Pr under thermal stress. Two-step FLA is one of the effective ways to obtain ferroelectricity with high values of 2Pr and good durability in HZO films thinner than 5 nm.

Accurate estimation of internal quantum efficiency (IQE) is essential for a comprehensive understanding of carrier dynamics in light-emitting materials in optical devices. We proposed simultaneous photoacoustic (PA) and photoluminescence measurements as a method for estimating accurate IQE values. This method detects the heat generated by non-radiative recombination as well as the light generated by radiative recombination, and this reduces the need to use theoretical models and assumptions for IQE estimation, unlike conventional methods that measure only radiative recombination processes. In some cases, however, there is a problem of heat generated outside of the measurement point being mixed into the PA signal. HF measurement reduces the mixing of false signals, but the PA signal intensity itself deceases with frequency, resulting in a poor S/N ratio. In this study, we have significantly improved the S/N ratio of PA measurements at HF by two measures in the microphone method we use for PA measurements: increasing the air pressure in the cell and utilizing Helmholtz resonance.

Perovskite solar cells (PSCs) have garnered attention as novel photovoltaic devices due to their cost-effectiveness, lightweight nature, and high photoconversion efficiency. To facilitate their commercialization, developing PSCs with enhanced environmental stability with respect to thermal and light resilience, making them suitable for outdoor applications, is imperative. However, despite ongoing research and development efforts, PSCs exhibit stability issues, including thermal- and light-induced degradation. In conventional PSCs, lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) as a dopant is essential to enhance the conductivity of the hole transport layer (HTL), such as 2,2′,7,7′-tetrakis[N,N-di(4-methoxyphenyl)amino]-9,9′-spirobifluorene (spiro-OMeTAD). However, LiTSFI incorporation into spiro-OMeTAD renders it unsuitable for applications requiring thermal stability owing to the Li⁺ diffusion within the PSCs. Since LiTFSI is a hydrophilic salt, we utilized organic TFSI salts to improve PSC thermal stability. By optimizing HTL using organic TFSI dopants, we achieved a remarkable two-fold enhancement in thermal stability compared to non-optimized PSCs.

This work demonstrated leaky-integrate-and-fire (LIF) behavior using a gate-controlled carrier-injection silicon-on-insulator transistor (GCCI SOI-Tr) and confirmed its signal amplification capability. We believe that a large-scale spiking neural network requires neural devices capable of signal amplification to effectively drive multiple post-synaptic neurons. The GCCI SOI-Tr could realize a capacitorless LIF behavior with a gate-controlled weight parameter. In addition, significant signal amplification might be achieved by scaling the GCCI SOI-Tr.

When observing the ultrasonic signal transmitted through a suspension of particles dispersed in a liquid, a peak is observed in the frequency spectrum of the attenuation coefficient of the ultrasonic energy. This peak mainly reflects shear elasticity, even though longitudinal ultrasound is used. Hence, the shear modulus of a single particle dispersed in a liquid can be obtained in a non-contact manner. In this study, we used ultrasound to analyze the process of polymer particle formation from an oil-in-water emulsion, in which a polymer solution containing an organic solvent is used as the oil phase, and the organic solvent is evaporated in water. Furthermore, the liquid-to-solid transition was quantified for microparticles consisting of various polymeric materials (liquid, rubber, and glass), and the relationship between the glass transition temperature, peak behavior of ultrasonic attenuation coefficient, and shear modulus was clarified.

A new polymeric microneedle (MN) fabrication technique is described in order to facilitate both higher mechanical stability and continuous drug release capability, a well-recognized challenge in the community. The technique involves filling the pores of a porous MN (PMN) array with a hydrogel. Cellulose acetate (CA) was used to prepare PMN, the interconnected cavity of which was then occupied by a crosslinked poly(N-isopropylacrylamide) hydrogel. Alkali treatment of the PMN array resulted in deacetylation of CA and improved the hydrophilicity on the surface. The hydrogel was readily incorporated by thermal polymerization of the monomers soaked to the PMN array. Mechanical strength tests demonstrated that pore filling enhanced the PMN stability by up to 50%, which was well-above the threshold required for skin penetration. The permeability of the hydrogel remained after pore filling and the drug release rate could be varied by alkali treatment process.

The piezoelectric signals generated in cancellous bone by ultrasound irradiation were numerically simulated using a piezoelectric finite-difference time-domain method. The changes of the piezoelectric signals with the cancellous bone thickness were investigated with the changes of the transmitted ultrasound signals. In the ultrasound direction of the strong trabecular orientation, two piezoelectric signals could be observed and was associated with the fast and slow waves in the ultrasound signal. The former wave amplitude did not vary with the thickness. The later wave amplitude randomly varied, but not significantly. In the ultrasound direction of the weak trabecular orientation, the piezoelectric signal due to the overlapped fast and slow waves and its multiple reflected wave could be observed. The former wave amplitude did not vary with the thickness. In both ultrasound directions, the local piezoelectric signal generated at the shallow depth of cancellous bone was dominant, which was because of the high ultrasound attenuation.

This paper discusses the applicability of periodically slotted SiO₂ as a coupler for ultra-wideband longitudinally coupled-resonator filters. It is shown that a −3 dB fractional bandwidth of more than 31% can be achieved by using lower-order three resonance modes. It is also shown that a periodically slotted SiO₂ structure can also control the frequency dispersion of wave propagation along the surface, and transverse modes are suppressed well for one particular mode under a proper setting. Then the piston design is applied for the remaining two modes, and the spurious-free and flat-passband triple-mode filter is synthesized.

We have prepared a-axis oriented CeO_2−δ thin films prepared on yttrium-stabilized zirconium (200) single crystal substrates by RF magnetron sputtering using a ceramic target. The amount of oxygen vacancies in the CeO_2−δ thin films was quantitatively evaluated by X-ray absorption spectroscopy and defect chemical analysis. In terms of its electronic structure, the O 1s photoemission spectrum of the wet-annealed film shows an O–H bond peak, which suggests proton conduction. The a-axis oriented CeO_2−δ thin film annealed at 600 °C in a wet atmosphere exhibit high proton conductivity of more than 10¹ Scm⁻¹K at medium temperature range from 300 °C to 550 °C by thermal activation process and surface proton conductivity of σT = ∼7.0 × 10⁻¹ Scm⁻¹K by Grotthuss mechanism below 50 °C. The enhancements of proton conductivities are considered to be due to large amount of oxygen vacancies of ∼3.3% at the surface of wet-annealed CeO_2−δ thin film. These results indicate that the wet-annealed a-axis oriented CeO_2−δ thin films can be applied as an electrolyte for solid oxide fuel cell operating at medium temperature range or electric double layer transistor operating below 50 °C.

Euglena gracilis has attracted attention because it contains the polysaccharide paramylon. In this study, we aimed to destruct E. gracilis by applying ultrasonic cavitation and to elucidate the mechanism. We also examined the breakdown of paramylon particles and attempted to extract paramylon nanofibers. It was suggested that the damage caused by ultrasonic waves was frequency dependent and influenced by the size ratio of the cell to cavitation bubbles, yield strength, and inhibition of cavitation bubble growth in suspension. It is also assumed that the cell destruction rate decreased because it was also dependent on the initial cell density, and an increase in the initial cell density resulted in a decrease in acoustic pressure. The fracture strength of the paramylon particles was much greater than the microjet stress at the acoustic power used in this study, and the paramylon particles did not fracture.

This paper discusses the suppression of a spurious resonance called the gap mode which appears when the double busbar configuration is applied to multi-layered surface acoustic wave resonators. Two kinds of overlay structures are proposed for the gap mode mitigation. First, the use of lossy ones is studied. The results indicate that the lossy overlay can effectively mitigate the gap mode response when the viscosity is adequate. The layer thickness has a negligible impact on the mitigation. Then, the discussion is extended to the application of high-acoustic-velocity materials as the overlay. The resonance frequency of the gap mode can be moved to the upper-frequency side without affecting the main resonance by simply depositing a high-velocity overlay to the gap region. Finally, the comparison and choice of overlay materials are also studied. The results point out that Si₃N₄ is the optimum choice for industrial use.

This paper discusses the basic properties of a surface acoustic wave (SAW) mode like the zeroth-order symmetric (S₀) Lamb mode in the LiNbO₃/SiO₂/SiC configuration in detail. At first, the influence of the SiO₂ insertion is discussed, and it is shown that SiO₂ insertion offers significant enhancement of the electromechanical coupling coefficient K² for the S₀-like SAW mode without significant reduction of the phase velocity. Then, the discussion is extended to spurious resonance characteristics. It is shown that the dummy electrodes are not necessary in this case. Then, two kinds of piston design are applied for the transverse suppression, and their effectiveness is discussed.

This paper discusses the impact of SiO₂/SiOF on second harmonic (H2) generation in temperature-compensated (TC) film bulk acoustic resonators (FBARs). First, two types of TC-FBARs are fabricated and their H2 behavior is compared with that of the standard FBAR. Next, the H2 characteristics of the TC layer are investigated by the use of a series of TC-FBARs where SiOF and Cr layers are placed on the outside of electrodes. The simulations agree very well with the measurements and can explain well how the H2 response changes with Cr thickness. Not only AlN but also SiOF contributes to the generation of H2 response, and H2 becomes very small at an appropriate Cr thickness. Finally, the TC-FBAR filter operating at 5.9 GHz is fabricated and its H2 response is compared with the simulation. The simulation agrees well with the experiment. It is also demonstrated that the insertion of the TC layer is effective in suppressing H2 generation.

White balance (WB) is a critical back-end processing function in image sensors to keep color constancy under various lighting conditions by adjusting the RGB color channel gain values. In this work, we proposed a novel gate-controlled gain-tuning method for a fully depleted silicon-on-insulator (FDSOI) one-transistor (1 T) pixel to achieve WB inside the sensor. Based on the pixel structure of a p-well under the buried oxide for sensing and an n-type nMOSFET on the top for readout, the 1 T pixel output gain is modulated by the MOSFET gate according to the transistor transfer characteristics. About 5x gain modulation range in RGB spectrum photoresponse (nonlinearity < 3%) is experimentally demonstrated in the devices fabricated by 22 nm FDSOI-based technology. The scheme for in-sensor WB demonstration is provided with a novel 1 T pixel array design, and the evaluation result shows in-sensor WB achieving an almost equivalent performance (Delta-E deviation < 1) compared with using conventional back-end WB.

A silicon carbide p-channel insulated gate bipolar transistor (IGBT) with higher breakdown voltage (BV) and low V_F·C_res figure of merit (FOM) has been simulated, fabricated, and characterized successfully. The proposed IGBT adds two n-type implant regions in the junction FET (JFET) area and increases the gate oxide thickness above the JFET area to reduce the reverse transfer capacitance (C_res) and gate oxide electric field (E_ox). The proposed structure notably lowers E_ox below 3 MV cm⁻¹ while elevating the BV to 16.6 kV. A new FOM of V_F·C_res is defined to evaluate the trade-off between the on-state and the C_res characteristics. The experimental results demonstrate that a lower V_F·C_res FOM of 0.369 V·pF is achieved from the proposed IGBT with a reduction of 66.4%, compared to the conventional current spreading layer IGBT. Meanwhile, the simulated turn-on and turn-off times of the proposed IGBT are reduced by 29.4% and 20%, respectively.

This paper discusses the use of periodic trenches to the top SiO₂ layer of the Bragg reflector in solidly-mounted-resonator type laterally-excited-bulk-wave-resonators for the slowness manipulation to suppress both transverse mode resonances and lateral leakage. First, it is shown that the SAW slowness surface can be flattened by the application of periodic trenches, and this is owing to the coupling of the propagating SAW mode with resonances of periodic trench walls. It is also shown that flattening the SAW slowness surface enables us to suppress transverse mode resonances well. However, its impact on lateral leakage is not obvious. Then, the broadband piston mode design is applied, and its effectiveness is demonstrated for the suppression of both transverse mode resonances and lateral leakage. Deposition of a dielectric layer to the gap regions is also proposed as an alternative for leakage suppression, and it is shown that Si₃N₄ is appropriate as a dielectric material.

This paper presents an ultra-low power low-dropout (LDO) linear regulator with a load current tracking bias current generator (BCG). The proposed LDO employs the load current tracking BCG to achieve stable and ultra-low power operation. The measurement results demonstrated that the proposed LDO produced a 1.2 V output and achieved 78 nA quiescent current, 145 mV  mA⁻¹ load regulation, and 2.2 mV V⁻¹ line regulation.

Sc-based contacts to Si:P have shown great potential for NMOS devices. However, the promising properties of this material system are not yet fully understood. This work provides new insights into the crystallinity and composition of annealed TiN/Sc/Si:P stacks. After silicidation, two distinct phases are evidenced, with orthorhombic ScSi lying atop a thin Sc_1−x−ySi_xP_y interfacial layer that shares a commensurate interface with the underlaid Si:P, hypothetically resulting in a low interface defectivity. The formed ScSi phase is observed to be thermally stable between ∼450 °C and 700 °C, which is suitable for most device applications. The impact of additional thermal budgets within this temperature range is investigated, revealing potential origins for thermally induced degradation of the contact properties.

We investigated the effects of post-annealing on the magnetic properties and spin orbit torque (SOT) in CoPt/YPtBi heterostructures, where YPtBi is a topological semimetal with a large spin Hall angle. From crystal structure analysis by X-ray diffraction and transmission electron microscopy, magnetization, anomalous Hall resistance, and second harmonic Hall measurements, we found that the magnetic properties and SOT performance in CoPt/YPtBi are improved by post-growth annealing up to 300 °C, while they are degraded after 350 °C annealing due to intermixing and alloying. By comparative experiments using CoPt/W, we show that YPtBi is more resistive to intermixing and alloying with CoPt than W.

We have demonstrated the formation of β-FeSi₂ nanodots (NDs) with an areal density as high as ∼10¹¹ cm⁻² on SiO₂ by exposing Fe NDs to SiH₄ at 400 °C and characterized their RT light-emission properties. After SiH₄ exposure, even at 1.0 Pa for 60 s, stable photoluminescence (PL) signals, being characteristic of the semiconducting phase of β-FeSi₂, were observed in the energy region from 0.7 to 0.85 eV. With an increase in the amount of SiH₄ exposure from 60 to 600 Pa·sec, PL intensity increased by a factor of ∼13. Note that, with a further increase in the amount of SiH₄ over 600 Pa·sec, the PL intensity is weakened slightly. The observed decrease in the PL intensity is attributable to the selective growth of Si onto the NDs after the formation of the β-FeSi₂ phase.

Herein, we report the photoluminescence (PL) properties of Cr³⁺-doped Mg₄Nb₂O₉ phosphors for applications in NIR phosphor-converted LEDs. Trigonal Mg₄Nb₂O₉ crystals with Cr³⁺ substitution at three sites were synthesized as the main phase using a solid-state reaction. X-ray photoelectron spectrometry and diffuse reflectance spectroscopy indicated that only trivalent chromium ions were present in the samples. Broadband NIR emission of Cr³⁺ suitable for combination with Si photodetectors was confirmed at approximately 850 nm. The maximum PL quantum yield (PLQY) was 0.62, and concentration quenching explained the Cr concentration dependence of the PLQY and PL decay time. Low-temperature PL measurements suggested that the PL spectrum comprised three distinct emission bands. The results of the phonon line simulation, low-temperature PL excitation measurements, and crystal-field parameter estimation clarified the correlation between these emission bands and the Cr³⁺ substitution sites.

K₂Cu(Cl,Br)₃ crystals were prepared by the slow cooling method and evaluated for photoluminescence (PL) and scintillation properties. PL emission and scintillation peaks due to the recombination of excitons were observed at 400 nm. The PL quantum yields were estimated to be 61.0% for K₂CuCl₃, 72.7% for K₂Cu(Cl_0.5,Br_0.5)₃, and 75.8% for K₂CuBr₃, respectively. The scintillation decay curves were approximated by a single exponential function model, and the calculated decay time constants were 14–60 μs. The afterglow levels at 20 ms passed after X-ray irradiation were obtained to be 5–1225 ppm. The light yields of K₂CuCl₃ and K₂Cu(Cl_0.5,Br_0.5)₃ were 7700 and 9900 photons MeV⁻¹ when calculated from the pulse height spectra under ¹³⁷Cs γ-ray (662 keV) irradiation.

Aiming at clarifying the structures and magnetic anisotropies of kagome Mn_3−xFe_xSn crystal over the wide range of x, we theoretically investigate the x-dependence of lattice parameters and bond lengths of their six magnetic phases by the first-principles calculations, and discuss their stabilities. Under the virtual crystal approximation at the Mn/Fe sites with a composition increment of Δx = 0.25, we find the following points. First, the lattice parameters and intra-plane first-neighbor bond lengths highly depend on the magnetic structures. Second, at x ≤ 0.25, the in-plane noncollinear antiferromagnetic (AFM) phase is most stable. Third, at x = 0.50, magnetic structures such as the out-of-plane spin canting structure, which are like superpositions of the AFM and ferromagnetic phases, are stable, and the magnetization easy axis is the c-axis. Finally, at x ≥ 1.00, the in-plane ferromagnetic phase is most stable, indicating that the easy axis is in the ab-plane, and the magnetocrystalline anisotropy energy reaches −1.60 MJ m⁻³ at x = 3.00.

Multi-dot single-electron devices (SEDs) have been fabricated using very thin Fe films by vacuum deposition on thermally oxidized or sputter-deposited SiO₂ substrates. Although the SEDs fabricated on the two substrates showed very different conductance, Coulomb blockade (CB) oscillation clearly appeared in certain Fe thickness ranges for both cases. The CB oscillation changed from complex to simple with increasing Fe thickness, indicating that the decrease of the number of dots contributed to the CB oscillation. While the simple CB monotonically disappeared by the drain voltage (V_D), the complex CB was robust against V_D because V_D distributed over the array composed of plural dots. The CB property change from complex to simple appeared in different thickness ranges for the two substrates, but in similar conductance ranges. This demonstrates that the conductance influenced by the inter-dot distance is an important factor for the CB characteristics of randomly distributed multi-dot SEDs.

We investigated the magnetic structure of HfMnSb₂ and TiMnSb₂ with first-principles calculations. We assumed ordered and disordered crystal structures for both compounds and investigated the transition of the magnetic structure. We theoretically confirmed that the helical magnetic order is stable in HfMnSb₂, while the conical magnetic order was observed experimentally. Additionally, assuming a chemically ordered crystal structure in TiMnSb₂ stabilizes the antiferromagnetic order. However, when the degree of order gradually decreases, the system exhibits spin-spiral density wave states and finally attains ferromagnetic order when the degree of order is 0. This trend is consistent with the experimentally measured results. We also found that the distortion of the lattice constants in TiMnSb₂ compared to that of HfMnSb₂ also plays an important role for the ferromagnetic order.

In this work, heteroepitaxial growth of copper phthalocyanine (CuPc) on a large single-grain of thin film N,N’-Di-n-octyl-3,4,9,10-perylenetetracarboxylic diimide (PTCDI-C8) was investigated. Vacuum-deposited PTCDI-C8 thin films at a growth temperature of ∼180 °C exhibited large grain growth of several hundred micrometers or more in size, and their surface consisted of stripe islands with molecular steps. CuPc deposited at 180 °C on this PTCDI-C8 large grain underlayer was found to grow epitaxially with its needle-like crystals dominantly oriented in one unique direction. The mechanism of the observed epitaxy is discussed based on the results of in-plane X-ray diffraction and the initial growth morphology.

The use of two-dimensional material like graphene to alleviate lattice mismatch has been an effective way to realize high-quality GaN on heterogeneous substrates. The lack of hanging bonds on the graphene surface provides a new attempt for epitaxial lateral overgrowth (ELOG). In this study, a hexagonal graphene mask was used for the growth of GaN, the graphene mask disappeared during the GaN growth process, but GaN still maintained the ELOG mode, and the threading dislocation density was significantly reduced. Raman and PL spectra demonstrated the stress relaxation in ELOG GaN and showed a stress relaxation of 0.157 GPa at the interface between the substrate and ELOG GaN. This study demonstrates the feasibility and advantages of graphene masks for nitrides and extends the research on stress relaxation of ELOG GaN using a graphene mask.

The control of the electrical properties and the growth mechanism of the NiO films were investigated by Li addition using the spray pyrolysis method. In addition, to enhance the uniformity and flatness of the NiO films, the electric field applied spray pyrolysis named electrostatic spray deposition (ESD) process was applied. Experimental results suggest that a certain amount of Li acts as an interstitial on the Ni sites in the NiO film and enhances crystal growth. Moreover, excessive Li additions resulted in Li segregation into NiO crystal defects, whereas the resistivity decreases and the promotion of crystal growth was not inhibited regardless of the presence of Li on the NiO film up to a Li concentration of 10 at.%. Furthermore, ESD was confirmed to deposit highly flat NiO films. These results represent the initial step toward the practical application of visible-light-transparent devices using spray deposition.

We studied the surface structure of SrF₂(111) using low-energy atom scattering spectroscopy. For the spectroscopy, we used pulsed 3 keV ⁴He° and 3 keV ²⁰Ne⁰ beams as incident particles and detected 180° backscattered particles using a microchannel plate. We compared the experimental results with simulations employing three types of surface atomic structural models: F–Sr–F, Sr–F–F, and F–F–Sr Our results demonstrate that the topmost layer of SrF₂(111) has approximately 60% and 40% of F–Sr–F and Sr–F–F, respectively.

Microwave surface-wave plasma (SWP) provides relatively higher ion density and radical species density than RF plasma, making it a promising plasma source for semiconductor fabrication. However, its plasma uniformity is compromised due to the non-uniform field distribution as the microwave wavelength is typically shorter than chamber dimensions. This could be a bottleneck as the practical application of SWP in semiconductor fabrication demands a large-area plasma with a high uniformity. In this work, we propose field agitation by applying the phase difference of 0° and 180° between the two input microwaves in the chamber to enhance the plasma uniformity. The chamber, optimized for the ${{\rm{TM}}}_{130}$ mode at 2.458 GHz, successfully produces a large-area SWP with a diameter of 450 mm under 800 W absorption power and 200 mTorr argon pressure. The measurement results show that the system is capable of yielding enhanced uniformity of large-area SWP while maintaining high ion density.

A microwave probe plays a critical role in near-field imaging, and there is a continuous effort to develop them through straightforward methods. This study designed and fabricated an oscillating metal probe and used it for scanning microwave imaging of micro-nano structures. The surface smoothness of the cantilever is approximately 19.3 nm after polishing with diamond abrasive paper, and the tip radius is less than 20 nm using electrochemical etching. The impact of metal electrode materials on microwave signals was assessed in the frequency range of 1–20 GHz. The microwave imaging capability of the devised probe was explored through the imaging of a micro-nano structure. The spatial resolution of microwave imaging reached 0.5 μm over a scanning area of 50 μm × 50 μm. This study has far-reaching significance for developing higher-performance microwave probes and advancing scanning microwave microscopy.

In this paper, we have examined the dopant activation levels in various materials (Si, SiGe, and Ge) annealed using two different microwave annealing (MWA) frequencies, 2.45 GHz and 5.8 GHz, with rapid thermal annealing serving as a comparative counterpart. The aim is to identify a material-specific annealing method that minimizes the high thermal budget typically required in the annealing process. For the dopant activation of Si wafer with B doping, there are two different frequencies (2.45 GHz and 5.8 GHz) MWA methods were compared. As the results of electrical measurement, the high activation level by 5.8 GHz MWA shows the material-targeted annealing method to reduce the thermal budget of the device. Additionally, we fabricated a high-k/metal gate (HKMG) MOSCAP structure, which was annealed using 2.45 GHz and 5.8 GHz MWA as part of the post-metallization annealing process. The results show that MWA at 2.45 GHz is more efficient than HF 5.8 GHz MWA at 3000 W in minimizing the interface trapped charge and the reduction in leakage current density. The highest capacitance (1.73 μF cm⁻²) was observed for MWA 2700 W at 2.45 GHz. However, there is a slight clear reduction in capacitance as the frequency increased to 5.8 GHz increasing in interfacial layer thickness. The post-metallization annealing of gate-last MOSCAPs using MWA at 2.45 GHz not only inhibits the growth of the equivalent oxide thickness and Al diffusion but also preserves the effects of pre-thermal annealing on doping activation at 5.8 GHz. Due to undesirable effects such as Al diffusion into the dielectric layer, the MWA at 5.8 GHz and 2.45 GHz demonstrates great a potential candidate, respectively as doping activation and as a post-metallization annealing method for HKMG structures.

During the treatment of leukemia, side effects create a great burden for patients. Therefore, new therapies to replace conventional chemotherapy and hematopoietic stem cell transplantation are highly desirable. In a previous study, I found that UV irradiation at 365 nm induced apoptosis in 98% of leukemia (Jurkat T) cells. To determine the most suitable wavelength for such treatments, it is necessary to measure the UV light absorption of cancer cells. In this study, a biochip was fabricated and evaluated for this purpose. Three types of leukemic cells (human acute myelogenous leukemia cells, human acute lymphoblastic leukemia cells, and human chronic myelogenous leukemia cells) were injected into the biochip, and light absorption was measured, revealing that individual leukemic cells have different light absorption characteristics. This variation is important in order to develop an implant care device to treat leukemia more effectively.

Resist materials and their lithography processes are essential for achieving further miniaturization in semiconductor microfabrication. In the sub-20 nm half-pitch region, the development process becomes particularly important due to the occurrence of stochastic defects (pinching and bridging). However, the details of dissolution dynamics are still unclear. In this study, we investigated the dissolution dynamics of poly(4-hydroxystyrene-co-methacrylic acid) (PHSMA) films in tetraalkylammonium hydroxide (TAAH) aqueous solutions using a quartz crystal microbalance method. PHSMA is a typical backbone polymer used in extreme UV lithography. For poly(4-hydroxystyrene) films, it has been reported that the dissolution mode changed when the alkyl chain length of TAAH increased. However, the mode change was not observed for PHSMA films. This difference is considered to be caused by the formation of thick transient swelling layers of PHSMA films owing to the fast water intake and the molecular interaction of the hydroxyl groups of hydroxystyrene units.

This paper presents the frequency characteristics of a barometric pressure-sensing wave height sensor employing a MEMS piezoresistive cantilever and a micromesh waterproof film. The proposed sensor exhibited mechanical bandpass filter characteristics owing to air leakage around the cantilever and film. Consequently, high robustness can be achieved if the frequency is designed to match the wave range. We theoretically and experimentally verified that the frequency response is influenced by the film’s chamber volume and permeability constant. We demonstrated that the proposed frequency-response model helps design a waterproof wave height sensor using a cantilever-type differential pressure sensor.

Bioactive glass has been of interest for applications in bone regeneration. Floating bioactive glass particles were observed to sink in ultrasound. The purpose of this study was to qualify and quantify bubble formation from floating bioactive glass particles. Water droplets containing borosilicate glass 13-93B20 particles, where 20% of the SiO₂ was replaced with B₂O₃, of dimensions <38 μm were subjected to pulsed ultrasound, whilst being video-recorded at high speed. Measured radial expansions >20 μm corresponded to cavitation nuclei of initial radius 0.6 μm. This study provides experimental evidence that gas trapped inside bioactive glass may be released using high-amplitude ultrasound pulses.

Small, thin devices that can realize high-speed light modulation are important in the imaging and sensing technology fields. In this paper, a combination of iron(III) chloride and ultrasound was proposed to modulate the spatial distribution of the iron(III) chloride layer’s refractive index. An ultrasound cell containing an iron(III) chloride layer and two ultrasound transducers was fabricated to investigate the variations in the refractive index under ultrasonication. The maximum change in the material’s birefringence was observed at the center of the cell, indicating that thermal generation caused by the ultrasound vibration induced this phenomenon.

This study explores the impact of a micro dry-etching technique on strain sensing based on multimode interference in a perfluorinated polymer optical fiber (POF). We demonstrate a drastic shift in strain sensitivity, from +34 nm/% in a non-etched POF to −981 nm/% in the etched POF. This notable change may be attributed to the excitation of the axially asymmetrical modes in the etched POF, which is confirmed through simulation on the stress distribution across the core. In addition, a critical wavelength for multimode interference is observed in a POF.

In this research, we employed AlOx and AlON thin films as interfacial layers (IL) in ferroelectric capacitors to achieve a significantly large memory window (MW) of 4.5 V and endurance up to 10⁸ cycles. By manipulating the IL materials, we obtain different intensity of voltage drop across the IL, thus changes the depolarization field in HZO, leading to increased MW. Furthermore, we integrated AlO_x/AlON into ferroelectric FETs and confirmed that AlO_x/AlON indeed enhances the threshold voltage difference (∆V_T), while maintaining an endurance of 10⁹ cycles. this study shed light on design guidelines for large MW memory devices.

In prior work, rough handling of oral tablets had been observed to drastically speed up their disintegration in water. The purpose of this study was to confirm or refute that the formation of internal microscopic fractures during rough handling is the underlying mechanism. Impacted and control tablets were subjected to micro-computed tomography and to brightness-mode ultrasound. The former revealed fracturing with a maximum crack width of 14 μm. The latter revealed strong acoustic response from the internal structure of the impacted tablets. These results confirm the hypothesis. Disintegration speed is used as a quality control mechanism after tablet manufacturing and transportation.

For HF operation of surface acoustic wave (SAW) devices, the resonance properties and temperature coefficient of frequency (TCF) of the longitudinal leaky SAW (LLSAW) third harmonic on a bonded structure comprising a LiNbO₃ (LN) thin plate and quartz support substrate were simulated using a finite element method. The theoretical results showed that, by utilizing the bonded structure, the third harmonic can be excited more strongly than when using a single LN substrate. The TCF of the LLSAW fundamental wave had larger absolute values than that of a typical SAW device.