Symposium S.EL01—Surfaces and Interfaces in Electronics and Photonics

The rapid growth of technologies such as machine learning and autonomous vehicles is fueling a need for more powerful electronic devices requiring complex nanoscale structures. With the growing demands on nanostructure fabrication, selective deposition is gaining attention as a process to enable pattern features at the sub-10 nm length scale. This talk will focus on area selective atomic layer deposition (AS-ALD), a bottom-up growth strategy in which material is added only where desired, without the need for subsequent lithography and etch steps. ALD is a good choice for selective deposition because its chemical specificity provides a means to achieve selectivity on a spatially patterned substrate. One popular approach to AS-ALD uses inhibitory layers such as self-assembled monolayers (SAMs) to alter the native surface reactivity, allowing ALD selectivity to be achieved. There have been a number of studies demonstrating that different types of SAMs, including organosilanes, phosphonic acids and thiols, can successfully block ALD processes. However, the efficacy of the blocking varies strongly across different ALD processes. For example, for the same SAM system, the blocking selectivity of Al₂O₃ is limited to only ~6 nm whereas ZnO can be blocked for over ~30 nm. The difference in blocking highlights the importance of precursor chemistry for AS-ALD. In this work, we describe our studies to elucidate the mechanism of AS-ALD based on a comparative study of ALD precursors. Several precursors are studied, including trichloroaluminum (TCA), dimethylaluminum chloride (DMACl), trimethylaluminum (TMA), and triethylaluminum (TEA), each used for Al₂O₃ ALD, allowing comparison between polarity and steric effects of the precursors. Results of quantum chemical calculations of the reaction pathways show product energetics that are correlated with experimental observations, and Monte Carlo simulations address the role of sterics on surface coverage. AS-ALD of Al₂O₃ using octadecyltrichlorosilane SAMs as an inhibitor is also investigated; the blocking properties of the ALD precursors will be compared and the results discussed based on the growth mechanism. By pursuing first principles design of selective ALD processes, this work may enable new methods for additive nanoscale patterning.

Etching of transition metals is one of the major challenges in magnetoresistive random-access memory (MRAM) device fabrication. In this work, atomic layer etching of iron and cobalt surfaces in sequential steps of exposure to halogen and an organic molecule was studied. We successfully performed etching of Fe Co, and FeCoB thin films through the formation of volatile metal complexes at low temperature with cyclic reactions of Cl₂and acetylacetone (acacH) or hexafluoroacetylacetone (hfacH). The etching reaction mechanism of acac and hfac reacting with chlorine-modified Fe, Co, and FeCoB surfaces was investigated: the surface was first activated with Cl₂gas, and subsequently, the top layer of chlorinated metal was removed by reaction with a diketone (acacH/hfacH) as shown in the photoemission spectra. The extent of Cl₂reaction determines the etching rate of the metal. At substrate temperatures lower than 135°C, acac could remove the chlorinated Fe metal layer from Fe surfaces, but not chlorinated Co from Co surfaces (Table 1 and 2). In-operandox-ray photoelectron spectroscopy (XPS) and density functional theory (DFT) simulation shows that the reaction of acacH or hfacH with chlorinated Fe or Co surfaces is likely following a complex reaction pathway instead of a simple diketone substitution for the metal chloride. Reactions of FeCoB with Cl₂results in the removal of B. The atomic layer etching cycle described above results in enrichment of the surface in Co. Diketone decomposition likely plays an important role in the etching process on all these surfaces. Moreover, the surface morphology becomes smoother as etching proceeds. The mechanisms for all of these observations will be discussed.

Passivation of electronic defects on surfaces and interfaces is vital for many (opto)electronic devices in the field of electronics, photonics and photovoltaics. In the last decade it has been demonstrated that a high level of passivation of surface and interfaces can be achieved by nanolayers of oxides prepared by atomic layer deposition (ALD). This passivation can be achieved by termination of bonds of the semiconductor surface but also by shielding of charges from the interface by fixed or trapped charges in the oxide. In this presentation the growth of nanolayers by ALD on a variety of semiconductors (Si, Ge, InP, …) will be discussed, both for planar wafers and for nanowires as used in the aforementioned devices. Results will be presented for various oxides such as Al₂O₃, TiO₂, Nb₂O₅, PO_x, MoO_x and ZnO and including data as obtained with a wide set of analytical techniques for electrical and structural interface characterization.

The atomic layer deposition of cobalt using Co(DAD)₂ and tertiary-butyl amine (TBA) has nearly infinite selectivity (>1000 cycles) on metal vs. insulator (SiO₂ or low-k SiCOH) planar samples. However, on patterned samples, selectivity under identical ALD conditions is limited, due to the diffusion of molecularly-adsorbed metal precursor from reactive to non-reactive surfaces. Three strategies have been found to improve Co ALD selectivity: adding a passivant to remove insulator defect sites, increasing the purge time, decreasing the precursor dose, and periodic annealing. The periodic annealing technique allows reabsorption of the Co nuclei from the insulator surface to the growth surface and is consistent with a low temperature reflow process. This technique has the added advantage of allowing a lower temperature for reflow, potentially allowing a scaling of the diffusion barrier between the Co and the SiCOH which is normally employed. Co ALD was performed using Co(DAD)₂ + TBA at 180°C on 85 nm wide Cu stripes on SiO₂. The planar structure of these stripes is used to demonstrate the effectiveness of passivation, as top-down SEM imagery and XPS quantification can be used to monitor growth and presence of unwanted Co nuclei on insulator. To control precursor dose, multiple precursor pulses were employed in each cycle to limit the maximum pressure. XPS is performed without breaking vacuum to prevent oxidation of Co. This work is supported by the Applications and Systems Driven Center for Energy-Efficient Integrated Nanotechnologies (ASCENT), a Semiconductor Research Corporation program sponsored by JUMP and DARPA.

Area-selective deposition (ASD)-driven processes gained recently a lot of attention from the microelectronics industry as a potential solution for the issues associated with top-down pattern formation at the nanoscale. However, the defectivity of typical ASD processes relying on delayed nucleation on a non-growth area is one of the key concerns for the adoption of this technology. Among the strategies addressing the defectivity issue, repetitive refreshment of the non-growth surface and selective etching of undesired nuclei are particularly promising.
This work is focused on the ASD of TiO₂ and TiN deposited by atomic layer deposition (ALD). In this respect, as an ALD inhibition route, we have been investigating both halogen plasma surface modification of amorphous carbon materials (non-growth area) and methyl-terminated monomolecular organic films on SiO₂ (non-growth area) substrates. On one side, on a-C non-growth area, we examined the possibility to combine the two defect-reduction strategies by employing low power Cl₂ or CF₄ plasmas for both surface functionalization/re-functionalization and for removal of nuclei. We employed prototypical water-based metal oxide process such as ALD TiO_x (TiCl₄/H₂O) to demonstrate area-selective deposition on top of silicon oxide using both blanket films and 45-60 nm half-pitch patterns of a-C. At first, the plasma parameters were optimized to minimize etching of a-C and silicon oxide. Then the selectivity of the ALD processes under study was checked on the plasma halogenated a-C films using a standard ALD sequence and a sequence interrupted by the appropriate plasma treatment steps. On the other side, self-assembled monolayers (SAMs) derived from octadecyltrimethoxysilanes allow ASD of >10nm of TiN (TiCl₄/NH₃) at temperatures as high as 325°C. Patterning of SAMs has been deployed to demonstrate successful selectivity and ASD surface dependency.
Special attention in this work is paid to the defectivity analysis of patterned a-C, which typically has surface composition deviating from that of the original coating as a result of employed dry etching. The efficiency of the proposed defect elimination strategy and its impact on resulting pattern quality are assessed using advanced wafer-level metrology including scatterometry and XPS.

From thin film solar cells to metal-oxide-semiconductor devices in leading edge integrated circuits, the electronic structure at and near the interfaces between component materials determines the most important fundamental operating characteristics of those devices such as turn-on voltage, power dissipation and off-state current leakage. Fermi level location at buried interfaces, semiconductor band-bending, charge transfer, oxide defects and work functions of the constituent materials all contribute to device performance. In this talk I will describe how these important parameters can be determined by employing femtosecond photovoltage spectroscopy (FPS)[1], an extension of ultraviolet photoelectron spectroscopy (UPS) using ultrafast lasers. While UPS is fundamentally a surface sensitive spectroscopy, I will describe how pump/probe techniques add a new dimension to this venerable spectroscopy, permitting the accurate extraction of the underlying band bending in Si. When combined with valence band edge location of the semiconductor and oxide, and determination of the system Fermi level, full characterization of the electronic structure of an MOS stack can be obtained providing key insights on device operating properties. I will describe the use of FPS to investigate device stacks ranging from Si and III-V based MOS structures to thin film solar cells formed from earth-abundant elements. In each case surprising new details were uncovered that led to performance optimization of these technologically important devices.
[1] Richard Haight and Adra V. Carr, "Industrial Applications of Ultrafast Lasers", World Scientific Press, 2018, ISBN:97898145690002

The drive for high performance and lower power consumption compels the semiconductor industry to consider novel nanoscale device structures based on non-silicon-based materials. Continued innovation and progress towards novel nanoelectronic materials and devices in turn requires metrology sensitive to electronic properties at these length scales. The electronic properties of a material at length scales relevant to the nanoscale electronic devices define how it will be used and integrated into device functional form. Angle-resolved photoemission (ARPES) is the traditional work horse to determine the electronic band structure of crystalline solids, but typically requires crystalline uniformity on the scale of 1 mm². Tip-based scanning probe techniques provide electronic contrast with sub-nanometer resolution, and therefore is a local technique. Photoemission electron microscopy (PEEM) can be employed to allow access to this vital information, providing full-field imaging capabilities sensitive to a variety of electronic contrast mechanisms at 10’s of nanometers length scales. The new generation of PEEM has energy filtering capability which now allows for ARPES measurements at smaller length scales (100 μm²; known as μ-ARPES). Here, we will discuss the commissioning of a recently installed PEEM at NIST, using results of graphene to demonstrate imaging capability and energy resolution of this instrument.

Graphene was grown epitaxially on 4H-SiC (0001) in a custom vacuum furnace at NIST [1] and transferred into the PEEM vacuum chamber. The PEEM instrument is equipped with continuous wave excitation sources in the deep ultraviolet (UV) regime, and a magnetic prism array for electron energy filtering. The terraces due to the underlying SiC substrate can easily be discerned since the PEEM provides imaging contrast due to topography and electronic contrast, and the terrace edges can be resolved to 50 nm. There are three primary regions in PEEM images that we assign as the (i) interfacial layer, (ii) 1-2 layer graphene, and (iii) “bulk” graphite regions. The sample surface is mostly dominated by regions with 1-2 layers of graphene for which μ-ARPES was performed at 8 μm length scales. The electronic structure of these regions is consistent with previous reports (e.g., [2]), with the π band and σ band dispersions clearly resolved. The Dirac point lies below the Fermi energy, consistent with n-doping present in graphene grown on SiC. We will also use confocal laser microscopy to correlate with the PEEM results. Finally, we will discuss our plans on advancing the lab-based PEEM capability with coherent, monochromatic laser-based photon sources.

[1] Y. Yang et al., Carbon 115, 229 (2017).
[2] T. Seyller et al., Surface Science 600, 3906 (2006).

Conducting polymers are an important class of Faradaic materials for energy storage, as they present opportunities to create energy dense supercapacitors. The vast majority of conducting polymers are p-type cathodes and already operating at or close to the upper voltage limit of non-aqueous electrolytes. On the other hand, it is generally safe for electrolytes to operate down to -2 V vs Standard Hydrogen Electrode. Therefore, extending the operational voltage in the negative range is an attractive route to increase energy density. Here we present promising n-type polymers that retained 90% of initial capacitance after 5000 charge-discharge cycles. The polymers are processed with solution based methods to ensure that the devices are scalable to large area. The electronic and ionic transport in the conducting polymers are monitored in operando by current-voltage and spectroscopic measurements, where spatial variations of film absorption would be mapped to discern electronic versus ionic degradation. We infer the mechanisms that lead to capacitance fade and suggests structural and electrochemical strategies to realize high-endurance energy storage devices for flexible printed systems.

In addition to energy storage, the narrow bandgap polymers are used to realize photodetectors that is relevant to a variety of applications including environmental monitoring and medical diagnosis. Here we show how the infrared polymers are processed into bulk heterojunction photodiodes with photoresponse up to wavelength of 1.7 micron. The performances of devices with different polymer structures were compared through metrics including detectivity, quantum efficiency, response time and rectification ratio. Example applications including hemodynamics and tissue composition measurements will be demonstrated.

Printed organic electronics is an expanding field with numerous advantages, including scalability, solution processability of the organic semiconductors, biocompatibility, mechanical flexibility, semi-transparency, color-tunability and low-cost manufacture.¹ Transitioning to eco-friendly printing inks for the large-scale manufacture of printed organic electronics is the next step on the horizon, as the use of halogenated solvents is environmentally hazardous, harmful to the health of workers, and will increase the cost of production.² Our recent work involves applying Hansen solubility theory, based on solution thermodynamics, to propose a Hansen solubility parameters (HSP)-based method that is a general platform for the rational design of ink formulations for solution-based organic electronics.³ Which will facilitate the green solvent transition of organic photovoltaics (OPV), in particular. A series of bulk heterojunction films were spin coated from green solvent ink formulations and X-ray spectromicroscopy was utilised to measure donor–acceptor domain size and domain composition. This seminar demonstrates that we can now engineer a nanostructured donor–acceptor thin film from green solvents, with donor and acceptor domain sizes closely matched to the exciton diffusion length in organic semiconductor systems.^3,4 The HSP approach is versatile for different organic semiconducting material systems, and we have detailed the pathway for using the HSP semi-empirical tool to predict morphologies, screen suitable ink formulations, and importantly design nanostructured donor–acceptor thin film morphologies for efficient OPV device production. By carefully considering HSP and choosing solvent blends with relative solubilities for the two materials, donor and acceptor can be driven to precipitate in the necessary order for forming a nano-structured interdigitated network morphology.

1. J. Bergqvist, et al. Flexible Electron. 2018, 2, 4.
2. C. Duan, et al., Energy Environ. Sci., 2013, 6, 3022–3034.
3. N.P. Holmes, et al., Green Chem., 2019, 21, 5090–5103.
4. N.P. Holmes, et al., Chem. Mater. 2018, 30, 6521−6531.

Metallic inks with superior conductivity and printability are necessary for high-throughput manufacturing of printed electronics. In particular, gallium-based liquid metal inks have shown great potential in creating soft, flexible and stretchable electronics. Despite their metallic composition, as-printed liquid metal nanoparticle films are non-conductive due to the surrounding metal oxide shells which are primarily Ga2O3, a wide-bandgap semiconductor. Hence, these films require a sintering process to recover their conductivity. For conventional solid metallic nanoparticles, thermal and laser processing are two commonly used sintering methods, and the sintering mechanism is well understood. Nevertheless, laser sintering of liquid metal nanoparticles was only recently demonstrated, and to date, the effect of thermal sintering has rarely been investigated. In this talk, eutectic gallium–indium nanoparticle films processed separately by laser or thermal sintering in an ambient environment will be presented. Laser and thermally sintered films will be compared with respect to electrical conductivity, elemental composition, crystallinity and surface morphology. Both methods impart thermal energy onto the films and generate thermal stress in the particles, resulting in rupture of the gallium oxide shells and achieving electrical conductivity across the film. For laser sintering, extensive oxide rupture allows liquid metal cores to flow out and coalesce into conductive pathways. For thermal sintering, due to less thermal stress and more oxidation, the oxide shells only rupture locally and extensive phase segregation occurs, leading to non-liquid particle films at room temperature. Electrical conductivity is instead attributed to segregated metal layers and gallium oxide which becomes crystalline and conductive at high temperatures. This comprehensive comparison confirms the necessity of oxidation suppression and significant thermal stress via instantaneous laser irradiation to achieve conductive liquid metal patterns.

Screen-printing and fire through the anti-reflective coating of the front gridline for crystalline silicon solar cell is cost-effective due to the high throughput and simplicity. Silver is the metal of choice because of high conductivity and the understanding that has been gained in decades of research. However, the cost of silver is high and it is the second most expensive layer of a solar cell. In order to penetrate the solar electricity, it is essential to fabricate cheaper solar cells and one of the feasible ways is to replace the expensive Ag front contacts by Cu which has been studied. However, despite the oxidation of copper in the atmosphere, it is challenging to have Cu directly contact with Si because (1) Cu diffuses into Si under high temperature and forms recombination sites; (2) Cu reacts with Si and form micro-sized Cu3Si to shunt the p-n junction; Thus, a diffusion barrier to block Cu from diffusion into Si is required for Cu front contacts on Si. Although people have used electroplating or two-step printing to fabrication diffusion barriers before fabricating Cu contacts, these methods include extra manufacturing steps and are too expensive to be commercialized. Thus, there is a need to replace the costly silver with copper and use the low-cost and high throughput process that will not require additional equipment in production. In this case, the most cost-effective method is using Cu to fully replace the Ag in the paste to make screen-printable pure Cu paste but it requires the paste to prevent the Cu from diffusing into the Si during the high-temperature sintering. According to our previous work, extra Cu in commercial Ag paste with glass frits results in increased transition temperature of glass frits and causes poor etching of SiNx layer, leading to low Voc. In this work, antimony is added into pure Cu paste without glass frits. Here, we are reporting the successful printing and sintering of metal gridlines with pure copper paste. After the one-step atmospherical firing process, the pure Cu fingers formed effective contacts on the Si emitter with Voc larger than 600 mV. In addition to the electrical output parameters of the solar cell, the microstructure analysis to elucidate the contact quality is reported. The preliminary studies showed that uniform pure Cu contacts were formed under < 800 ^oC with a low ratio of Cu_xO and the antimony formed an effective diffusion barrier to prevent the Cu diffusion into the Si and ensured high Voc. The contact path should either through direct contact with Cu gridlines or tunneling through a thin glass layer.

Semiconducting metal oxides, particularly transparent conducting oxides, have attracted a great deal of attention for a variety of applications. Nevertheless, the solution chemistry and solution functionalization of these materials has been much less explored than that of traditional semiconductors, such as silicon.
One challenge in the stable functionalization of these materials is that their semiconducting nature enables efficient generation of long-lived photocarriers that can initiate chemical reactions that degrade the functionalization. This reactivity has typically been attributed to the undercoordinated metal cation sites at the surface, although direct evidence has been lacking. We have developed a new photochemical reaction that leads to the fluorination of all exposed metal cation sites on TiO₂. In contrast to approaches that use fluoride or HF-containing solutions, our method produces covalently bonded fluorine which is stable in aqueous solutions. To our surprise, these fluorine-terminated TiO₂ surfaces display higher photoreactivity than comparable “clean” surfaces. Possible origins of this behavior will be discussed.
We have also investigated the structure of solution-deposited monolayers on TiO₂ and demonstrated a rational approach to tuning intermolecular interactions and enabling long-range ordering. We show that simple electrostatic insights can be used to engineer away unfavorable intermolecular interactions, producing monolayers with exceptional long-range ordering. Quantitative measurements of the intermolecular interaction energies from molecularly resolved STM images are a factor of ~7 larger than those predicted by dispersion-corrected density functional theory (DFT). This finding suggests a new path to the production of highly ordered monolayers and superstructures of large molecules.

An interdigitated geometry of the interface between semiconductors remains a Graal in the photovoltaic field. Nanocylindrical solar cells based on a coaxial p-n or p-i-n junction have been generated in a small number of materials systems only so far, mostly among the classical semiconductors. The generalization of this interface geometry to alternative semiconductors, which are expected to profit of it most, has been hampered by the lack of appropriate deposition methods suited to coating complex three-dimensional structures in a conformal manner. Atomic layer deposition (ALD) provides this capability, but is limited by the necessity to use gaseous precursors in vacuum conditions. Modern classes of semiconductors such as the hybrid perovskites are not accessible by ALD due to their ionic nature.
In this perspective, we have transferred the principles of ALD to precursors dissolved in liquid solvents. The ‘solution ALD’ (sALD) method can be implemented in a variety of microfluidic or slot-die processing devices. We have demonstrated that the principles of classical, gas-based ALD —self-limiting surface reactions— are reproduced in sALD. Furthermore, sALD allows the experimentalist to design new, advantageous, reactions of known semiconductors (oxides, heavier chalcogenides), or to deposit materials otherwise inaccessible to ALD altogether, such as polymers, ionic solids, and metal-organic frameworks. They may be obtained in highly pure, crystalline form at room temperature, either as planar films or as conformal coatings of porous substrates. Based on these advances, we have been able to make the first sALD-derived hybrid perovskite cells.

Future nanoelectronics requires area-selective deposition to eliminate the alignment challenges in the fabrication of multilayered device structures. For example, area-selective atomic layer deposition (ALD) processes of both oxides and nitrides are currently being considered for various applications in interconnect technology, with especially the fabrication of fully self-aligned vias receiving much attention.¹ Most work in the field of area-selective ALD relies on local deactivation of the ALD growth by selective functionalization of the non-growth area. Previous research predominantly focused on surface functionalization using self-assembled monolayers (SAMs), which so far is the most successful approach for achieving a high selectivity.²
Our approach to area-selective ALD involves vapor-phase dosing of small molecule inhibitors in ABC-type (i.e. three-step) ALD cycles.³ One of the main merits of this alternative approach is that is compatible with industrial process flows, while it also allows for the use of a plasma as the co-reactant and therefore for area-selective ALD of a wider range of materials. It is however more challenging to block the ALD precursor adsorption using small molecules. In order to improve the selectivity of this approach, our current research aims at identifying and designing molecules that can effectively inhibit the ALD growth.
In this presentation, the requirements for the inhibitor molecules will be discussed, which will be illustrated with examples from our experimental and theoretical studies. Results for area-selective ALD of TiN, Al₂O₃, and SiO₂ will be presented, with a focus on the mechanisms of precursor blocking. Special attention will be given on our recent work on using aromatic molecules as inhibitor on metal surfaces.

(1) Mackus, A. J. M.; Merkx, M. J. M. Fully self-aligned vias: the killer application for area-selective ALD? – A discussion of the requirements for implementation in high volume manufacturing. www.AtomicLimits.com
(2) Mackus, A. J. M.; Merkx, M. J. M.; Kessels, W. M. M. From the Bottom-Up: Toward Area-Selective Atomic Layer Deposition with High Selectivity. Chem. Mater. 2019, 31, 2–12.
(3) Mameli, A.; Merkx, M. J. M.; Karasulu, B.; Roozeboom, F.; Kessels, W. M. M.; Mackus, A. J. M. Area-Selective Atomic Layer Deposition of SiO2 Using Acetylacetone as a Chemoselective Inhibitor in an ABC-Type Cycle. ACS Nano 2017, 11, 9303.

The amorphous oxide thin film transistor (TFT) had been widely applied not only to the display but other electronics such as sensors and semiconductor devices. To secure long term stability of TFT, minimizing charge trapping defects in the interface and gate insulator (GI) would be the most important concern. In the case of amorphous oxide semiconductor, oxygen vacancy (V_o), cation interstitial, ionized V_o, oxygen interstitial (O_i), hydrogen interstitial (H_i), hydrogen substitutional are well known defects. Among these, O_i, which is known as oxygen excess, weakly bonded oxygen, peroxide, or metal vacancy is known to be major source of charge trapping center. Meanwhile, H, which is known as a shallow donor in oxide semiconductor, can passivate the O_i charge trapping center as does the dangling bonding Si. Therefore, optimization of O_i and H is one of the most important processes to secure the stability of oxide TFT. The charge trapping centers in the active and/or interface between active and GI are formed during the process generating oxygen related reactive species such as sputtering for active and PECVD or PEALD process for GI. Comparing these processes, PECVD generates less defects in the channel of TFT thus, top gate TFTs yield better PBTS stability than that of bottom gate TFT. Meanwhile, it still needs careful and precise control of O and H to result in high bias stability with proper V_th for the high mobility oxide TFT of which semiconductor has high concentration of carrier amount. It is because V_th of high mobility oxide TFT is more sensitive to the addition of H into the active layer than that of typical IGZO TFT. For these reasons, deposition of GI in top gate oxide TFT is carried out at relatively low temperature to minimize the H incorporation into the active layer, giving rise to the degrading stability than that of IGZO TFT. Therefore, optimization of GI process and controlling interface are very important in materialization of high mobility top gate structured oxide TFT with highly stability and proper V_th. Here, we scrutinize the effect of O_i and H on the formation of the interface of top gate structured high mobility Al: ITZO TFT with double GI consisting of ALD grown alumina (AlO_x) and PECVD deposited SiN_x film. Oxygen plasma treatment was conducted on the active surface and diffusion of H into the interface and/or active layer during thermal annealing was induced from the AlO_x GI to inspect the effect of Oi and H, respectively. Here thermal ALD (T-ALD) AlO_x provides H which plays as a passivator of the charge trapping centers in the interface as well as the shallow donor. To control the amounts of H and Oi in the active and/or interface, we did O₂ and N₂O plasma treatment prior to the AlO_x deposition to yield TFTs with improved on/off characteristics. The TFTs with O₂ and N₂O plasma treatment showed 2.08V and 2.34V of V_th, respectively. We confirmed that both O₂ and N₂O plasma treatment increase the portion of M-O bonding and reduced the Vo. The O₂ plasma-treated TFTs with T-ALD GI yield high field-effect mobility over the 35.3 cm²/ Vs with improved bias stability compared with that of TFT of which AlO_x deposited by just PE-ALD. Therefore, we conclude that the T-ALD process for the thin GI deposition can be applied to fabricate even high mobility top-gate structured oxide TFTs by using O₂ or N₂O plasma treatment. The AlO_x deposited by T-ALD can passivate extra trap sites generated during the supply of the oxygen for the carrier control process of high mobility oxide TFT. This enabled us to obtain high mobility TFTs with high stability and proper Vth. The TFTs fabricated by oxygen plasma treatment, followed by the T-ALD GI deposition resulted in excellent bias stability because of both fine defect control of interface and the carrier concentration.

Gallium oxide has attracted renewed interest as an ultrawide band-gap transparent semiconductor. Monoclinic b-Ga₂O₃ is expected to be highly suitable for various applications such as high temperature and ultra-high voltage electronics, photon detectors, gas sensors as well as thin tunneling barriers in various devices. Amorphous gallium oxide has recently been shown to be an effective electron transport layer or an effective passivation layer in solar cells.
In this talk, we will discuss the growth and the characterization of amorphous gallium oxide thin films. The films were grown on (001) Si by plasma-enhanced atomic layer deposition (PE-ALD) using trimethyl gallium as the gallium precursor and oxygen plasma as the oxidizing agent. We will focus on the effects of the starting surface and of the oxygen plasma exposure time. In a temperature range of 100-360°C, the growth per cycle in the ALD growth regime is of ~ 0.67-0.70Å/cycle and the resulting films are amorphous. In situ characterization during the synthesis was performed by spectroscopic ellipsometry. By applying an expansion of the three-phase equation (air, thin film, substrate) in the thin film limit [1], we were able to determine unambiguously the optical properties (index and extinction coefficient) and the thickness of the thin film unlike with classic ellipsometry modelling that is limited by the non-uniqueness of the possible solutions. With this method, we could precisely determine the thickness evolution of each step of the ALD cycle. We evidenced the reduction of gallium oxide absorption for longer plasma times that can be attributed to a lower defect density. Finally, the leakage currents measured on MOS capacitors will be discussed as a function of the oxygen plasma exposure time. We will show that leakage current densities can be lowered by more than 3 orders of magnitude by adjusting this parameter.


[1] I.K. Kim, D.E. Aspnes, Analytic determination of n, κ, and d of an absorbing film from polarimetric data in the thin-film limit, J. Appl. Phys. 101, 033109 (2007)

Atomic layer deposition (ALD) utilizes self-limiting surface reactions to deposit approximately one atomic layer of material per precursor exposure. As such, ALD is a promising technique for the deposition of highly conformal thin films for a wide variety of inorganic materials. Molecular layer deposition (MLD) is a related technique that also relies on self-limiting surface reactions but aims to deposit entire organic molecules instead of single atoms. Hybrid ALD/MLD combines these techniques to deposit coordination polymers in which metal atoms are connected through organic linkers which are typically referred to as “metalcones”. In this talk, we will discuss the deposition of metalcones comprised of more than one metal. We will specifically discuss ternary metalcones formed from mixtures of trimethylaluminum (TMA) and diethylzinc (DEZ) as well as from DEZ and tetrakis(dimethylamido)hafnium (TDMAHf). The growth properties of the multi metalcone films will be discussed, with a specific focus on the challenges these ternary films face due to transmetalation reactions which occur during the deposition and result in non-ideal compositions of the two metals.

The MLD technique could be considered as the ultimate in additive manufacturing where it builds materials one molecular layer at the time from the gas phase. It is based on chemistry and relies on availability of building units from the gas phase. The current presentation will focus on growth of such materials with emphasis on building units containing carboxylic acids. Inorganic carboxylates are mostly stable, offering high saturation of the metal centre’s coordination sphere, leading up to the plethora of such structures as metalorganic frameworks (MOF). Such materials can be grown via the gas phase, and examples from growth of the UiO-66 series with variations in linker length and complexity will be given, in addition to alternative systems. This introduces concepts such as modulation of growth, reservoir effects, tuning of chemical reactivity, and amorphous porosity. The current contribution will provide highlights from such growth of organic- inorganic hybrid materials including MOF materials.

We demonstrate photovoltaic and photoconductive responses to near-infrared light in a new class of devices formed by depositing a film of gel permeation chromatography purified PbS quantum dots (QDs) on top of n-SiC epitaxial layers with natively grown, 10-15 monolayer thick epitaxial graphene (EG) Schottky contacts. The QD-film was removable by selective chemical etching, resetting the EG/SiC Schottky diode to its pristine state for multiple PbS-QD depositions. EG in these devices simultaneously forms Schottky contacts to SiC and ohmic contacts to PbS-QD, enabling electrical screening and isolation of these interfaces from each other. After PbS-QD deposition, the diodes exhibit photovoltaic and photoconductive response at photon energies far below the SiC bandgap, extending to the NIR gap of the QD film. Scanning photocurrent microscopy illustrates that this is due to charge transfer from the QD-film to the n-type 4H-SiC through a trap-limited, rectifying PbS-QD/SiC heterojunction with ideality n=2 in parallel with the EG/SiC Schottky diode. The photoconductive gain at this QD/SiC interface could be useful for IR detection in wide bandgap platforms. Response times as fast as 40 ms are suitable for imaging applications, although careful contact design is required to optimize work-function matching and spreading resistance.

CdSe_xTe_1-x alloys now used in state-of-the-art CdTe solar cell technology can be further characterized and optimized. Se alloying can significantly reduce the bandgap of the CdTe absorber due to the strong bandgap bowing between CdTe and CdSe. This has been demonstrated to the enhance quantum efficiency in the infrared and boost the device short-circuit current (J_SC). Interestingly, despite having a lower bandgap, devices using Se have achieved open-circuit voltages (V_OC) on par with (unalloyed) CdTe-based devices. Recent time-resolved photoluminescence and cathodoluminescence data suggest that the CdSe_xTe_1-x materials have lower recombination rates and improved grain-boundary passivation relative to CdTe, and this appears to be making up for bandgap-related V_OC losses[1]. Further characterization studies should reveal additional relationships between absorber processing and the CdSe_xTe_1-x composition, current collection, recombination, etc., that can be used to optimize device performance.
There are two methods of CdSe_xTe_1-x growth. The first method creates a CdSe window layer for a CdTe device then allows diffusion to create a dopant gradient. The second processing method deposits a CdSe_xTe_1-x alloy layer and then a CdTe layer. Previous characterization studies have focused on the CdSe/CdTe device structure and the effects of changing window layer thickness and composition. A study by Poplawsky, et al. [2], indicated that, as CdSe window thickness increased, device performance increased then quickly decreased. Quantum efficiencies and correlative SEM/TEM analysis indicated that if the Se concentration was above 40 at%, grains could change phases from zinc-blende to wurtzite and stop contributing to the device photocurrent. In the CdSe_xTe_1-x /CdTe structure, the Se concentration can be kept below 40 at% to prevent the formation of the wurtzite phase.
To better understand the impact of Selenium concentration on device photo-collection, an Electron Beam Induced Current (EBIC) study is presented which studies the impact of Selenium on device collection for the CdSe_xTe_1-x /CdTe device structure. EBIC is an SEM-based characterization technique which uses an electron beam to locally excite carriers, which are subsequently collected and analyzed to understand microstructural effects on recombination. A current collection map can thus be created with sub-micron resolution. EBIC has been previously used to identify local variation in collection current at microstructural features such as heterojunction interfaces and grain boundaries[2]. In this study, selenium concentration was varied in the alloy layer from 0 to 40 atomic percent while maintaining constant deposited layer thickness.
EBIC analyses of these alloy junctions indicate that the highest carrier collection was near the CdSe_xTe_1-x /CdTe interfaces with a second collection peak near the TCO/CdSe_xTe_1-x interface. This finding is indicative of a buried junction in the CdSe_xTe_1-x /CdTe device. Previous studies on the CdSe/CdTe structure have shown highest collection near the TCO contact with no buried junctions. The position and strength of this buried junction could decrease the effect of interface recombination at the TCO interface, but it could also decrease drift velocity, negatively impacting performance. These competing effects need to be better understood for the optimization of CdSe_xTe_1-x /CdTe device performance.
The presence of a buried junction in the CdSe_xTe_1-x /CdTe device implies a complex change in carrier concentrations and band structure within these devices. In order to better understand the implications of this observation, these results are correlated with a finite element device model. Using this model, it is possible to calculate the equilibrium band structure of the device and estimate the built-in potential between at each junction.

1.Zheng, X., et al. APL Materials 7(7).
2.Poplawsky, J. D., et al. Nat Commun 7: 12537.

Opto-electronic integration yields high density and high complexity electronic devices by improving speed and reducing interconnection bottle-necks. Heteroepitaxy and wafer bonding are the main technologies to achieve monolithic integration of heterogeneous materials. However, heteroepitaxy is limited by the mismatches between dissimilar lattice constants and between different coefficients of thermal expansion (CTE).

Direct Wafer Bonding (DWB) is the fusion of two mirror-polished, flat wafers, which overcomes the limitations of heteroepitaxy. DWB can be applied to many materials including bonding III–V, II-IV semiconductors and piezoelectrics to Si.

In this work, low-temperature DWB via NanoBondingTM is investigated to bond LiTaO₃ and LiNbO₃ to Si(100) and α-quartz SiO₂(100). Low-temperature NanoBonding™ of piezoelectrics such as LiTaO₃ and LiNbO₃ on Si-based materials can avoid issues arising from their CTE mismatch, since conventional DWB requires anneal temperature > 673 K.

The highly anisotropic CTE of LiTaO₃ ranges from ~ 4 x 10^-6 /K along the c-axis to up to ~ 16 x 10^-6/K along the a-axis. Thus, the CTE for LiTaO₃ is 2-10 times larger than the CTE for Si, 2.6 - 2.77 x 10^-6 /K, and 10-30 times larger than the CTE for fused SiO₂ and α-quartz SiO₂, which ranges 0.54 - 0.76 x 10^-6 /K. Thus even a low T anneal at 473 K is found to shatter a 6” LiTaO₃/Si bonded pair. Nano-Bonding™ at T < 473 K uses Surface Energy Engineering (SEE) instead of high temperature anneal to bond surfaces. In addition, processing temperatures for LiTaO₃ are found to be limited by thermal decomposition LiTaO₃ into Ta₂O₅ at T ≥ 453 K due to Li out-diffusion, as much as by LiTaO₃ fractures due to CTE mismatch.

SEE is guided by surface energy and hydro-affinity measurements, using Three Liquid Contact Angle Analysis (3LCAA) and the Van-Oss-Chaudhury-Good theory. 3LCAA is conducted with high statistics, by mapping wafers for three surface interactions, namely van der Waals interactions, electron donors, electron acceptors interactions. Multiple metered 10-µl drops of three liquids, water, glycerin, and α-bromo-naphthalene are used. The contact angles are extracted from a large set of drops and their reflection via an automated, fast, accurate algorithm, DROP^TM. The “as received” surfaces are characterized at equilibrium, to yield insights for SEE into modifying the initial surface energy and hydro-affinity to a far-from-equilibrium state.

As received 6” Si(100) wafers are initially hydrophilic with a uniform average water contact angle of 37.7 ± 0.6°, and an average surface energy γ^T of 57.6 ± 0.5 mJ/m². As-received α-quartz SiO₂(100) wafers surface were hydrophobic with an average water contact angle of 66.7 ±1.7° and an average γ^T of 33.2 ± 0.8 mJ/m². 6” wafers of LiTaO₃ are initially mostly hydrophobic with an average water contact angle of 64.4 ± 2.7°, while water contact angle average 54.2 ± 8.1° LiTaO₃ and LiNbO₃ surface energies γ^T are low and very close: 38.8 ± 4 mJ/m² and 39 ± 2.5 mJ/m² respectively.

SEE via aqueous HF creates far-from-equilibrium Si surfaces by decreasing the initial γ^T of Si(100) from 57.6 ± 0.5 mJ/m² to 35 ± 0.03 mJ/m² and increasing the γ^T of α-quartz SiO₂(100) was increased from 33.2 ± 0.8 mJ/m² to 48.7 ± 1.5 mJ/m². Likewise, LiTaO₃ and LiNbO₃ were made hydrophilic via SC1(H₂O:NH₄OH:H₂O₂) solution (4:1:1) followed by aqueous HF etching.

Nano-contacting of LiNbO₃, LiTaO₃ on Si(100) and α-quartz SiO₂(100) and subsequent anneal at T < 453 K was performed to create a bonded interfacial phase. However, nano-contacting was only observed for hydrophobic LiTaO₃ and LiNbO₃ with hydrophilic Si-based surfaces. Results show that bonding hydrophobic LiTaO₃ and LiNbO₃ to hydrophilic Si or SiO₂ without anneal is much more effective than trying to bond hydrophilic LiTaO₃ to hydrophobic Si or SiO₂. This can be attributed to the modification of surface polarity and will be discussed based on the surface crystallography of LiTaO₃ and LiNbO₃.

Metal induced crystallization is a well-known technique to reduce the crystallization temperature of amorphous semiconductors like a-Si using a catalytic metal. Crystallization temperatures even below 200 °C have been realized making it compatible with organic materials or other applications where only a limited temperature budget is acceptable. A combination of the wide band gap semiconductor 4H-SiC and such a low temperature process is barely investigated, although the Si/4H-SiC heterojunction offers huge potential in the field of e.g. heterojunction diodes, bipolar junction transistors and optical detectors. Conventional growth methods like chemical vapor deposition (CVD) need temperatures well above 800 °C to achieve epitaxial growth of Si on SiC. A low temperature method could make this heterojunction more practical, also in combination with other materials and temperature sensitive processes.

We present the first results to achieve partially epitaxial c-Si/4H-SiC interfaces using Al as catalytic metal. Sputter deposited amorphous silicon has been successfully crystallized on 4H-SiC at temperatures ranging from 350 to 700 °C. Annealing led to an almost complete layer exchange of Al and Si as well as to a heteroepitaxial interface of crystallized Si islands and the SiC surface. Si <111> was found to be the dominating crystallographic orientation of the re-crystallized Si. X-ray diffraction measurements show the change in Si crystallinity and the crystallographic impact of the underlying SiC substrate in this temperature range. TEM and SEM imaging was performed to investigate the film microstructure and the interface quality, thus revealing a clean and abrupt interface of large Si grains and the SiC substrate.

As the microelectronics manufacturing faces the challenges related to the Beyond Moore’s Scale adaptation, Chemical Mechanical Planarization (CMP) process control becomes much more critical yet difficult simultaneously. The need for planarizing larger scale wafers at an atomic scale precision with superior removal rate selectivity and zero-defect focus demands a nano-scale control on the process performance. This study reviews the atomic-scale chemical and mechanical interactions on CMP as it applies to traditional as well as the next generation materials in microelectronics manufacturing. Once the behavior of the wafer surface is understood at an atomic level, the process control metrics can be tuned accordingly.

In an attempt to come up with the design criteria for the slurry chemistry, chemically modified layer formation is evaluated on conducting, semiconducting and insulating materials as a function of evaluating the nano-scale surface topography and change in surface energy. Metallic films are studied in terms of their corrosion and passivation behavior through macro-scale electrochemical evaluations as well as surface-sensitive characterization techniques such as XRR, XRD and FTIR [1, 2]. In addition, a modeling approach is introduced through Cahn Hilliard Equation (CHE) approximation to the surface energy minimization on the metal-oxide formation by nucleation [3]. Furthermore, removal rate selectivity properties of the semiconductor materials are discussed on conventional and III-V semiconductors and insulators as a function of chemically modified film alteration [4-5]. Some high-end applications such as (i) tungsten T-gate transistors, (ii) new liner/barrier materials [6], (iii) germanium based high-speed, shallow trench isolation transistors and (iv) high-power transistors and LED applications where III/V semiconductors are deliberated. As a new challenge, 3-D implementations of the CMP process, where the chemically modified nano-films impact the performance of the materials are also introduced.
Ozdemir, Basim, G.B. "Effect of Chemical Mechanical Polishing on Surface Nature of Titanium Implants FT-IR and Wettability Data of Titanium Implants Surface After Chemical Mechanical Polishing Implementation," Data, in Brief, 10, P 20-25, 2017.
Karagoz, A., and Basim, G.B. “Controlling Germanium CMP Selectivity through Slurry Mediation by Surface-Active Agents," ECS Journal of Solid State Science and Technology, CMP Special Issue, 4 (11) P5097-P5104, 2015.
Karagoz, A., Sengul^, Y., Basim, G.B., “A Cahn Hilliard Modeling of Metal Oxide Thin Films for Advanced CMP Applications," 225^th ECS Meeting – Orlando, Florida, USA, May 11-15, 2014, ECS Transactions, 61 (17) 15-20 (2014).
Karagoz, A., Craciun, V., Basim, G.B., “Characterization of Nano-Scale Protective Oxide Films_ Application on Metal Chemical Mechanical Planarization”. ECS Journal of Solid State Science and Technology, 4 (2) P1-P8, 2015.
Ozbek, S., Akbar, W., Basim, G.B. “Optimized Process Design for GaN Chemical Mechanical Planarization” ECS Journal of Solid State Science and Technology, Special Issue on GaN-Based Electronics for Power, RF, and Rad-Hard Applications, 6-11, P S3084-S3092, 2017.
Yagan, R., Basim, G.B., “A Fundamental Approach to Electrochemical Analyses on Chemically Modified Thin Films for Barrier CMP Optimization," ECS J. Solid State Sci. Technol. 8-5, P3118-P3127, 2019.

Abilities to capture, guide, and direct light into pixelated detectors (CCDs, CMOS sensors, FPAs, etc.) using free-form trajectories would be transformative for a range of technologies that currently use traditional optics (lenses, mirrors, prisms, etc.) for light coupling. However, few manufacturing techniques provide the appropriate scale and surface quality to address this task. In this work, we consider “free-form waveguides” using the high-resolution 3D printing technique of multiphoton lithography (MPL; also known as direct laser writing, DLW). First, we printed hollow and solid core vertical waveguide arrays spanning millimeters with single (5 micron) and multimode (30 micron) compatible cross-sections. Unsupported 5-micron diameter pillars generally succumbed to capillary forces during development and drying and thus sheathed waveguides were designed by anchoring fibers to a support scaffold. In addition, we examined direct printing onto active devices (coupling a source to a sensor) using an air-clad multimode design. Efficient light coupling was observed with scattering losses primarily from fabrication artifacts such as stitch lines. To address these issues, we considered stitch-free cores using a clad-only design resulting in encapsulated photoresist in the interior that is subsequently cured (UV, thermal) following development. These initial results, combined with the limitless design space afforded the MPL technique, provide a promising path forward to develop ultra-wide-field and highly compact imaging devices.

Native oxides increase electrical resistance and inhibit epitaxy on semiconductors. Their removal is key in opto-electronic integration for devices such as tandem solar cells. Besides heteroepitaxy, Direct Wafer Bonding (DWB) is a viable integration technology for GaAs and Si. NanoBonding™ is DWB conducted at low T (<200°C) by modifying surface energy and hydro-affinity to reach far-from-equilibrium states and activate cross-bonding and electron exchange between the surfaces.
Hence, the surface energy has to be increased (if it is initially low), or decreased (if it is high). One of the surfaces in the pair has to be made hydrophilic if it is initially hydrophobic, and the other one hydrophilic if it is initially hydrophobic. Moreover, the surfaces have to be modified in a complementary way to so that electrons are transferred from electron donors from one surface to electron acceptors on the other.
To conduct Surface Energy Engineering (SEE) of GaAs and Si into a bonding pair, the initial surface energy, hydro-affinity, surface composition, and oxidation levels of as-received GaAs(100) and Si(100) are initially characterized. Next, the initial surface energies and hydro-affinities are modified not only by oxygen removal via chemical etching, but also to maximize cross-bonding. The characterization methods are Three Liquid Contact Angle Analysis (3LCAA) for surface energy and hydro-affinity, High Resolution Ion Beam Analysis (HR-IBA) for O coverage, and X-Ray Photoelectron Spectroscopy (XPS) for oxidation and C contamination.
In 3LCAA, the total surface energy, γT, is determined by the van Oss-Chaudhury-Good theory using 3 components: Lifshitz-Van der Waals, electron donors, and acceptors. GaAs(100) native oxides are hydrophobic with a low surface energy ( γT=33.4 ±1.1 mJ/m2) while Si(100) native oxides are hydrophilic with a high surface energy (59.7±1.7 mJ/m2 ). GaAs surfaces are then etched with NH4OH:H2O (1:10) to be highly hydrophilic, while their γT can be reproducibly increased by a factor of 2 to 65.5 ±1.4 mJ/m2. HF:CH3OH (1:20) etching makes Si(100) hydrophobic and decreases its γT by 20% to 48.2 ± 1 mJ/m2. Nano-bonding™ of GaAs to Si does not occur between these surfaces. However, Si(100) etched by aqueous HF becomes even more hydrophobic with a 25% decrease in γT to 44.4 ±2.3 mJ/m2. With this SEE of Si, GaAs nano-bonds to Si reproducibly.
O etching is measured by HR-IBA, while oxidation states and C-contamination are measured by XPS to correlate to nano-bonding. HR-IBA combines <111> channeling with O nuclear resonance to measure O coverage within ±0.2 Monolayers (ML) and analyzed via SIMNRA simulations. O coverage on GaAs is found to decrease by a factor 2 from 7.2±0.5 ML to 3.6±0.2 ML after etching.
The ratio of Ga and As is found to be unaffected by C-contamination, and remains ~54:46 within 1 atomic % by HR-IBA or ~1.18 within 10% by XPS. XPS shows that the relative ratio of Ga bound to GaAs versus Ga bound to Ga2O3 is initially 64:36 to within 1% with a slight increase in Ga bound to O. The relative ratio of As bound to GaAs versus As bound to O is much lower than Ga, 80:20 to within 1% and does not change with etching.
XPS analysis also shows that when C-contamination ranges between 12%-35% atomic %, the measured ratios of Ga and As bound to GaAs versus Ga and As bound to O vary less than 2%. After etching, Ga bound to Ga2O3 decreases with increasing carbon with a weak correlation factor of 0.4. C-contamination does not appear to significantly affect Ga oxidation. For As oxides, As2O5 shows a relative increase compared to As2O3 as C contamination increases with a strong correlation factor of 0.86 in native oxides before etching. After etching, the correlation factor is 0.55.
In summary, aqueous HF produces more hydrophobic Si(100) surfaces that can nano-bond to highly hydrophilic GaAs etched via dilute NH4OH while C contamination does not appear to modify GaAs surface composition nor oxidation.

Electrically conductive polymers are widely utilized in emerging electronic devices due to their excellent optoelectronic properties as a potential alternative to conventional conductors. Among them, Poly(3,4-ethylenedioxythiophene):Tosylate (PEDOT:Tos) thin films deposited by Vapor Phase Polymerization (VPP) exhibit electrical conductivity exceeding 3400 S/cm which is almost comparable to commercial ITO electrodes. [1] Systematic investigation on such films indicates a strong anisotropy in electrical and optical properties and this anisotropy is believed to be mostly caused by the preferential alignment of para-crystallines in the films. [2]
Here, I will present our recent study on the VPP PEDOT:Tos thin film system and reveal that vertical gradient could also be an important factor contributing to the electrical anisotropy. By examining the results from grazing-incidence wide-angle X-ray scattering and X-ray photo-electron spectroscopy, we observed significant differences of the top and bottom surfaces, including para-crystalline alignment preference and oxidation level (doping level). These show that the VPP PEDOT:Tos layer is not vertically homogeneous in terms of both “crystallization” fraction and charge carriers (doping level). To explain the origin of this inhomogeneity, we propose a growth mechanism for VPP based on diffusion-limiting transport of polymerization precursors. Conductive polymer thin films with certain gradient profile may find applications in functionally graded thermoelectrics, gradient-index optics, and optoelectronic devices with gradient doping.

References
[1] Fabretto, Manrico V., et al. "Polymeric material with metal-like conductivity for next generation organic electronic devices." Chemistry of Materials 24.20 (2012): 3998-4003.
[2] Chen, Shangzhi, et al. "On the anomalous optical conductivity dispersion of electrically conducting polymers: ultra-wide spectral range ellipsometry combined with a Drude–Lorentz model." Journal of Materials Chemistry C 7 (2019): 4350-4362.

Chalcopyrite materials based on CuInGaSe₂ is practically used for the commercially available high efficiency solar cells. The room temperature bandgap of AgGaTe₂ is about 1.3eV, hence the application of this material to the active layer of novel solar cells also attracted the attention. Among various film preparation methods, the closed space sublimation (CSS) method is suitable for the fabrication of mass production solar cells since it has various advantages toward the production cost reduction. The growth of crystallized AgGaTe₂ thin films utilizing this method on Si and Al₂O₃ substrates have performed in past, and it was discovered that the introduction of the two-step sequence was the key to realize high quality thin films. The Ag₂Te layer was deposited as the first step to avoid the melt-back etching related problems of Ga against substrate materials. Then, AgGaTe₂ thin films were formed using the mixed powder source of Ag₂Te and Ga₂Te₃. The AgGaTe₂ thin fims could be also prepared using the powder source of Ga₂Te₃. The first Ag₂Te layer was disappeared during the high temperature growth process of the second sequence, and a single phase AgGaTe₂ of the uniform layer has formed once the growth parameter was optimized.
It was clarified from the X-ray diffraction (XRD) measurement that the inclusion of AgGa₅Te₈ compounds occurred when the fraction of Ga₂Te₃ in the deposits was greater than that of Ag₂Te. On the other hand, mixed crystal of AgGaTe₂ and Ag₂Te were formed when the fraction of Ga₂Te₃ was less than that of Ag₂Te. These results corresponded to the phase diagram of the Ag₂Te–Ga₂Te₃ system. A pn-heterojunction solar cell structure was prepared using the AgGaTe₂ layer on the n-Si substrate, and the photovoltaic properties were confirmed. The AgGaTe₂ layer was also prepared on the Mo/quartz substrate using the same growth sequence.
In case of the CuInGaSe₂ layer growth on the Mo electrode layer, Se diffusion was observed in the Mo electrode layer. The advantage and disadvantage were still discussed for this structure. In this study, the interfacial diffusion of elements at AgGaTe₂/Mo was studied to understand how the diffusion of Ga could be suppressed. High resolution cross sectional TEM images of the interfacial region was analyzed using the Z-contrast images of the high angle annular dark field (HAADF) mode and EDX-mapping mode.
When AgGaTe₂ layer was directly grown on the Si substrate, Si surface was etched and became very rough, and it was considered that Ga has damaged the Si substrate surface. By introducing the two step procedure, the interface quality was substantially improved and the roughening was vanished.
When the layer was grown on the Mo layer using the two step procedure, sharp interface was also observed. By careful analysis using the EDX mapping, it was confirmed that Ga was diffused into the Mo layer when the growth temperature was above 650 ^oC. Different from the conventional interdiffusion, Ga accumulation was also observed at the diffusion front. The diffusion depth was fairly uniform across the specimen. The diffusion of Te was also observed for samples grown at high temperatures as well as samples post annealed at the high temperature. Those samples exhibited Mo and Te condensation region with the size of around 0.2 to 0.5 micron within the AgGaTe₂ layer. It probably suggested that a part of Mo was reacted with Te, and formed the volatile MoTe₂ material, and it was migrated into the AgGaTe₂ layer.
Further investigation is ongoing to identify the diffusion characteristics at the AgGaTe₂/Mo layers. It is reconfirmed that the two step growth procedure is a very useful approach to achieve the high quality layer, but processing temperatures including the growth temperature and post growth annealing temperature are very critical parameters to achieve the high quality interface with the abrupt profile.

This work was supported in part by Waseda University Grant for Special Research Projects.

Surfaces of materials such as lithiated nickel-manganese-cobalt oxide (NMC) can be characterized through experimental techniques such as photoelectron spectroscopy. These techniques take advantage of high-vacuum environments, though the vacuums are never perfect, and the materials are often exposed to ambient conditions prior to characterization or application, all of which could impact the surface composition. Hence, questions arise as to the data derived from periodic density functional theory (DFT) calculations that assume chemically clean surfaces in perfect vacuum and are used to describe these materials. To reconcile these differences, we investigate the surface chemistry of NMC through a combination of DFT and first-principles thermodynamics calculations. By examining the bulk structure, bare surfaces, and the adsorption of various atmospheric molecular and ionic species, we develop phase diagrams that reveal the expected surface chemistries as a function of various partial pressures, predicting the NMC surface chemistry as a function of the environment. Such predictions can be used to modify future computational studies of the material both in vacuum and in solution, providing additional tools for characterization.

Electrical contact resistance (ECR) directly influences the performance of micro-scale devices, where contact between components occurs through nanometer-scale patches. Consequently, developing the ability to accurately measure ECR as a function of true contact size at the nanometer scale is of crucial importance. To achieve this goal, we present an approach based on conductive atomic force microscopy, which is applied on a material system comprising atomically flat interfaces formed between gold nano islands and a graphite substrate [1]. Proof-of-principle experiments are complemented by atomistic simulations, which are used to predict the maximum island sizes below which the ECR at the island-graphite junction can be reliably extracted from the measurements. This approach has the potential to contribute to fundamental studies of electron conduction, e.g. through an investigation of the transition between the diffusive and ballistic transport regimes.

[1] Vazirisereshk, M.R., Sumaiya, S.A., Martini, A., Baykara, M.Z., Applied Physics Letters 115, 091602 (2019).

Colloidal quantum dots (QDs) have been widely used for light-emitting diodes (LEDs) because of their outstanding characteristics including high color purity, narrow emission peak, high luminescent efficiency, low operating voltage and adjustable emission wavelength that are tunable as a function of their size and composition. For the fabrication of QD-based LEDs, they have been applied to several processes such as spin-casting, inkjet printing and micro contact printing. However, these processes suffer from practical limitations such as high-resolution patterning, high cost, multiple-layer structures, and long processing time. In order to overcome the issues currently suggested, micro-contact printing method using the relief patterns on a master has been introduced and utilized extensively for QD-LEDs. In this way, the regular patterns of QDs layer is transferred on the hole transport layer (HTL) by simply physical contact between the master and the HTL surface. In spite of its efficiency, however, the cyclic process of the micro-contact printing inevitably gives rise to sever damages to the master patterns. Here, we demonstrate a unique method for directed patterning of colloidal quantum dots for the light emitting diodes from chemically modified surface. The chemically modified line-type patterns are initially fabricated on the substrate by using a conventional photolithography technique and selective chemical reaction of surface-modifying molecules such as the octadecyltrichlorosilane (OTS) sequentially. The width of line-type patterns produced is varying from 40 µm to a few µm. The quantum dots of CdSe(core)-ZnS(shell) that are covered with oleic acid as capping molecules are then spin-coated on the chemically patterned surface. To transfer the QDs layer to a hole transport layer, the polymethylmethacrylate (PMMA) is first spin-coated on the QDs layer followed by placing commercial adhesive tapes on the PMMA layer. Then the QDs/PMMA layer is detached from the original substrate by peeling the adhesive tapes and transferred to the target substrate. The target substrate consists of ITO(Anode), PEDOT: PSS(HIL), and TFB(HTL), respectively. It is noted that the only designed QD patterns are selectively transferred from the chemically modified surface due to surface energy difference during the transfer process. As the PMMA/tape layers are removed by rinsing with acetone, which is a selective solvent, the only QD patterns remain on the HTL layer of the target substrate. For the fabrication of QD-LEDs, ZnO(ETL) and Ag(Cathode) layers are deposited on the QDs layer sequentially. Consequently, the QD-LEDs fabricated in this study are characterized by measuring their electrical properties like luminance, current density and external quantum efficiency.

Interfaces govern the behaviour of all electronic devices. Herbert Kroemer coined the famous phrase “the interface is the device” in his 2000 Nobel Prize lecture, and we are still applying tremendous effort to understand interfaces in new material generations, with wide-bandgap materials being no exception. If anything, wide bandgap materials are more vulnerable to defect states purely due to their larger bandgap. Understanding gained for the bulk behaviour of semiconductors can often not be extended to the behaviour of materials in structured film stacks were interfaces play a vital role. SiC/SiO₂ is a prototypical wide-bandgap semiconductor/dielectric interface, which represents the challenges faced by many such material systems. A multitude of different defects leads to unacceptably large defect densities exceeding 10¹³ cm^-2 eV^-1 in the vicinity of the conduction band of 4H-SiC. The management of interfacial defects remains a topic of lively discussion and current interest.

Here, we present a systematic study of the 4H-SiC/SiO₂ interface in industrially manufactured samples with a particular focus on the effects of nitridation in N₂, NO, NH₃ and NO+NH₃ atmospheres. High temperature nitridation has been shown to compensate interface defects leading to an increase in device performance, however, information on the local chemistry at the interface after such processes is scarce, which limits the understanding of the interface and consequently the targeted improvement of device characteristics. Clear differences are found in both spectroscopy and electrical behaviour after high temperature treatments. The present work uses energy-dependent hard X-ray photoelectron spectroscopy (HAXPES) to systematically study the elemental distributions and chemical environments across the 4H-SiC/SiO₂ interface. We combine both laboratory- and synchrotron-based HAXPES results to obtain a complete picture of this important technological interface and to benchmark the capabilities of this spectroscopic technique for the exploration of buried interfaces in device heterostructures in general.

Alkali antimonides (such as Cs₃Sb, K₂CsSb and Na₂KSb ) constitute a rather interesting class of materials, exhibiting high quantum efficiency (>10%) and high brightness in visible light making them interesting materials for use in photon detectors and as photoemission based electron sources for large linear particle accelerators (used for X-ray Free Electron Lasers, Electron Colliders and Hadron Beam Coolers) and time resolved electron microscopes. Despite ongoing work on these extremely vacuum sensitive materials for the last several decades, there has been no conclusive understanding and little/ no consensus on its opto-electronic properties like band gap, electron affinity, optical constants etc. Lack of knowledge of these vital properties has stood in the way of understanding the process of photoemission in these materials near the photoemission threshold. The objective of our work is to grow high quality alkali-antimonide films with smooth surfaces and perform different spectroscopic characterizations like photoluminescence, photoconductivity and ellipsometry on them, in order to explore their opto- electronic properties in ultra high vacuum. It is believed that this detailed experimental study will elucidate the exact process responsible for efficient photoemission from these materials and allow us to develop state-of-the-art electron sources with maximized brightness.

The process of thermal atomic layer deposition (ALD) of metals on non-metal surfaces is sought for many applications, but often suffers from nucleation delays that result in poor-quality films. These growth delays can result from a lack of suitable chemisorption sites on the non-metal surface. Various strategies have been explored to improve nucleation in these systems. Vapor phase pre-treatments have proven to be an especially promising nucleation enhancement strategy due to the minimal substrate-modification induced, its robustness, and its uniformity.
In this work, we demonstrate that pre-functionalizing the surface with a sub-monolayer of small organometallic molecules such as ZnEt₂, AlMe₃ and AlMe₂Cl, from the vapor phase can lead to a significant increase in the surface coverage of the metal deposited by ALD at low cycle numbers. This process is demonstrated using Pt ALD as a model process, using (methylcyclopentadienyl)trimethylplatinum (MeCpPtMe₃) and O₂ as reactants. Pt coverage was enhanced by ~ 3-fold after 100 ALD cycles and even larger enhancements were observed at lower cycle numbers following the pre-treatment. We hypothesize that the high coverage of the small organometallic molecules provides an alternative chemisorption mechanism for the platinum precursor leading to increased MeCpPtMe₃ chemisorption during the first ALD cycle. This alternative chemisorption mechanism was investigated through in-situ X-ray photoelectron spectroscopy and mass spectrometry. Nanoparticle growth was monitored using grazing incidence small angle X-ray scattering and scanning electron microscopy. We determined that chemisorption of MeCpPtMe₃ on the treated substrate originates from a Brønsted acid-base reaction (i.e., proton transfer reaction), where the MeCpPtMe₃ acts as the Brønsted acid and the pre-treatment molecules serves as the Brønsted base. Growth analysis indicates that the pre-treatment leads to growth of highly dense, ordered Pt nanoparticles with improved wettability at low cycle numbers, leading to continuous and pinhole free Pt films.

For high switching speed HV Schottky diodes, with very high work function metal and extremely lightly doped epi, the built-in potential may be too high for thermionic emission to occur, when the applied external voltage is quite low (near the forward voltage: V_F = 0.07V). If the epi is lightly doped p type, the built-in potential (V_Built-in: potential difference between the metal and silicon Fermi levels) is 1.0V (measured with CV). If the external bias is 0.1V, near the measured V_F, it is not enough to overcome the built-in potential for thermionic emission as illustrated in the Figures. It is likely that in addition to thermionic emission, tunneling and diffusion currents also contribute to the total HV Schottky diode forward current. TCAD simulation of HV Schottky diodes with N⁺ guard bands suggests the potential barrier and electric fields at the Schottky junction are relatively high for thermionic emission to occur, when external bias V » V_F.

In the Advanced Technology Laboratory (ATL) at Northrop Grumman Corporation, high performance RF ASICs and power devices in silicon, GaAs, GaN and 4H SiC are simulated, designed, fabricated, and packaged. Extensive DC and RF tests are conducted. The performance of HV Schottky diode is superior to conventional p-n junction diode due to higher BV, lower reverse leakage, lower V_F and higher output current. We have fabricated HV Schottky Diodes. BV > 150V. V_F = 0.07V. High breakdown voltage, low forward voltage drop, and low reverse leakage are achieved with very high work function metal alloys and low doping epi films.

Lightly doped p (100) or n (100) type silicon films are deposited with epitaxy. Metal or metal alloys are deposited with sputtering. Figure 1 shows the measured I-V’s for the Schottky Diodes. Avalanche breakdown voltage > 150V is recorded with low reverse leakage current. The voltage required to turn the diode on is defined as V_F – when the current reaches 0.1uA.

Forward bias characteristics of Ti, Au, and NiCr are measured. These data indicate that in order to achieve low V_F and high current, very high work function metals are needed.

High frequency CV tests are conducted using the HV Schottky Diodes. These CV data are used to extract the doping profiles .

Metal work functons are extracted with the 1/C² vs. V plots. Interception with the x axis is DV, which is the potential difference between metal and silicon Fermi levels, or built-in potential V_Built-in. Metal work function is extracted with DV. One of the Figures shows the measured doping profiles and 1/C² vs. V plots. From the measured I-V and C-V data we may calcuate the avalanche breakdown critical electric fields and metal work functions. A summary of HV Schottky diodes fabricated with NiCr, Ti, Au in Si n and p epi films will be presented.

Energy band diagram is analyzed for HV Schottky diodes: NiCr in p epi (external bias = 0V) V_Built-in = 1.00V.

Another energy band is analyzed for HV Schottky diodes: NiCr in n epi (external bias = +0.50V). V_Built-in = 1.56V.

In conclusion: we have fabricated HV Schottky Diodes. BV > 150V. V_F = 0.07V. High breakdown voltage, low forward voltage drop, and low reverse leakage are achieved with very high work function metal alloys and low doping epi films. Diffusion and tunneling currents likely contribute to the total Schottky junction current.

Metal halide perovskite nanocrystals (NCs) are attractive materials for optoelectronics because of their optical and electronic properties, such as tunable direct band gaps, high absorption coefficients, low exciton binding energies, relatively high electron and hole mobilities, narrow emission line-widths, and high photoluminescence quantum yields (Φ_PL). However, further improvements in stability and reproducibility in Φ_PL of NCs are essential for enabling future optoelectronic applications. Inadequate surface passivation due to labile parent ligands, such as oleate and oleylammonium ions is a major cause of instability and less than unity Φ_PL. Herein, we probe the influence of multiple molecular and ionic ligand binding groups including thiols, thioethers, phosphine, phosphine oxide, sulfobetaine, thiocyanate, and sulfate salts on the colloidal stability and Φ_PL of CsPbBr₃ NCs. Among all the binding ligands, post-synthetic treatment with dodecanethiol reproducibly yields highly stable NCs with near unity Φ_PL for a range of synthetic conditions and initial Φ_PL of the as-synthesized NCs. A few of the other ligands can also increase the Φ_PL, but in ambient atmosphere they are unable to maintain stability more than 5 days. A mechanistic investigation shows that thiol addition to the as-synthesized NC dispersion leads to the thiol-ene reaction with octadecene, oleic acid, & oleylamine and formation of thiolate ions and thioethers. Both thiolates and thioethers are suspected to bind to undercoordinated Pb atoms and consequently eliminate metallic Pb particles commonly observed with transmission electron microscopy on the NC surfaces. The process of enhancing PL intensity and stability can be rapidly accelerated through exposure to blue or UV light.

Developing a scalable method to fabricate atomic wires is an important step for building solid-state semiconductor quantum computers. In recent years, self-assembled molecular monolayer doping (MLD) has attracted intensive research interests due to its capability of facilitating mass production and forming ultra-shallow junctions with atomic precision^[1]. MLD is believed to have a great potential in controllable dopant manipulation at sub-10 nm scale^[2]. However, there is still no general method for effectively patterning dopant-containing monolayer at nanoscale. In this work, a novel selective doping strategy is proposed by patterning self-assembled monolayer to a few nanometers using standard nanofabrication processes, which significantly improves the lateral doping resolution of monolayer doping from microscale to nanoscale.
We first found an effective way to remove the self-assembled monolayer via oxygen plasma followed by HF etching, which is confirmed by X-Ray photoelectron spectroscopy and Van der Pauw measurements. Electron beam resist HSQ (XR1541-002, Dow Corning) was used to create patterns that protect part of the self-assembled DVP monolayers from being oxidized by oxygen plasma, since HSQ will turn into amorphous SiO₂ after electron beam exposure. To increase the resistance of the HSQ SiO₂ against HF etching, a second time of high dosage exposure was performed on the HSQ structure. The high dosage electron beam exposure might induce a thin amorphous carbon layer covering the HSQ structures and dramatically improved the resistance of HSQ against HF etching.
Using this method, we further explore the possibility to fabricate phosphorus atomic wires in silicon by patterning self-assembled diethyl vinylphosphonate monolayers into lines with a width ranging from 500 nm to 10 nm. The phosphorus dopants are driven into silicon by rapid thermal annealing, forming dopant wires. Four-probe and Hall effect measurements are employed to characterize the dopant wires. The results show that the conductance is linear with the width for the wires, showing the success of the monolayer patterning process to nanoscale. To fabricate atomic wires made of one or a few lines of phosphorus atoms, we need to shorten the thermal diffusion length and increase the dopant incorporation rate at the same time. Pulsed laser annealing may be a promising solution. The present work provides a promising pathway for mass fabrication of atomic wires in silicon that may find important applications in quantum computing.

References
[1] a) J. C. Ho, R. Yerushalmi, Z. A. Jacobson, Z. Fan, R. L. Alley, A. Javey, Nature Materials 2007, 7, 62; b) J. C. Ho, R. Yerushalmi, G. Smith, P. Majhi, J. Bennett, J. Halim, V. N. Faifer, A. Javey, Nano Letters 2009, 9, 725; c) X. Gao, B. Guan, A. Mesli, K. Chen, Y. Dan, Nature Communications 2018, 9, 118.
[2] a) P. W. Voorthuijzen, D. M. Yilmaz, W. J. M. Naber, J. Huskens, W. G. van der Wiel, Advanced Materials 2011, 23, 1346; b) M. Perego, G. Seguini, E. Arduca, A. Nomellini, K. Sparnacci, D. Antonioli, V. Gianotti, M. Laus, ACS Nano 2018, 12, 178.

Fast synthesis of copper sulfide nanoparticles through two different methods is described in the present investigation. The first reaction was carried out by applying of heating in constant temperature water bath while in the second case the mixture was sonicated. As copper and sulfur sources, copper nitrate and thiourea solutions were used respectively. Main physicochemical properties of the synthesized samples were studied using several characterization techniques. Formation of copper sulfide with hexagonal crystalline structure was corroborated, coinciding with previously reported values for covellite. The most important results corresponded to the un-sonicated sample, which exhibited spiky-like nanoballs morphology and better size distribution. Optical tests revealed band gaps of 1.69 and 2.05 eV, while photoluminescence emissions at 421 and 430 nm, showing promising possibilities for applications in solar cells, catalysis and optoelectronic devices.

The continuous innovation in nano materials has allowed the development of new antireflective coats (ARCs) for optical lenses, solar cells, data storage, displays and others that require a reduced reflectance to increase efficiency. Multi-layered ARCs particularly have attracted attention lately for demonstrating low and tunable spectral reflectance.

This paper presents a simple approach for forming anti-reflective coating stacks on bulk Ge and PECVD Ge grown on Si employing colloidal dispersions of Si nanoparticles (NPs).

The Si NPs have a diameter of 2.85 nm and are prepared by chemical etching of crystalline Si wafers in HF and H₂O₂ using electrical or hexachloroplatinic acid catalyst. The nanoparticles are recovered by sonicating the wafers in isopropyl alcohol, forming a stable colloid with concentration of 10 µM [1].

Each antireflective coat is applied by the drop casting of monodispersed Si NPs on a 3 by 3 cm pieces of both Ge bulk and PECVD Ge grown on Si. A total of 5 drops are applied with each drop consisting of 100 µL. After each drop, the samples are left to dry for an hour and a half.

For bulk Ge, (110) Sb doped (n-type) Ge with a resistivity of 0.1–0.5 Ωcm were used, while for the PECVD Ge growth, (100) P doped (n-type) Si with a resistivity of 0.01–0.02 Ωcm are used as starting substrates. An Oxford Instrument System 100 PECVD activated by 13.56 MHz radio frequency signal is used to grow Ge layers on Si substrates. High purity germane (GeH₄), H₂ and Ar gases are used as precursors for Ge growth. The undoped Ge film was grown using the two-step approach, which has been described in detail in our previous work [2].
The impact of multiple drop casting of Si NPs on the antireflective properties of the Ge surfaces were investigated by the analysis of their reflection spectra. The reflectance of the coating was measured at four locations on each sample by a UV–VIS spectrophotometer (Lambda 850, PerkinElmer). The samples were qualified by controlling the reflectance at the four points within an error of 1.2%. The corresponding reflection spectra were captured as a function of the incident wavelength, ranging from 250 nm to 1500 nm.

For PECVD Ge grown on Si substrates, the average reduction in reflectance has shown to increase as the incident wavelength decreases, with a decrease of ~2% after each 100 μL coating. After 5 drops (100 μL ×5), the overall reflectance across the whole spectra is on average 9.35% lower compared to the uncoated samples. The minimum decrease in reflectance was of 2.49% at 1290 nm and the maximum of 34% at 250 nm.

For Bulk Ge, the average reduction in reflectance has shown to increase as the incident wavelength decreases from 1500 nm to 320 nm. At 500 μL, the reduction in reflectance had an average of 6.16% (Min.: 2.73% at 1500 nm, Max.: 13.75% at 320 nm and Median: 5.1%). At lower wavelengths range from 250 nm to 320 nm), the amount of reflectance reduction is less compared to wavelengths above 320 nm.

After 5 drops (100 μL ×5) on PECVD Ge samples, the incident wavelength where the minimum reflectance occurred (λ_min), shifted from 1050 nm where reflectance is lowered by 34.5%, to 410 nm with 25.2% reduction in reflectance. While for bulk Ge sample, λ_min shift occurs from ~1480 nm to ~420 nm with a decrease in reflectance of 37.9% and 28%, respectively.

Results demonstrate that it is feasible to deposit a coating of Si NPs with tunable reflectivity over a wide range by drop casting multiple layers of the nanoparticles.

[1] Akcakir, O., Therrien, J., Belomoin, G., Barry, N., Muller, J.D., Gratton, E., Nayfeh, M., Appl. Phys. Lett. 76, 1857 (2000); https://doi.org/10.1063/1.126191
[2] Dushaq G., Rasras M. Nayfeh A., Thin Solid Films 636, 585–92 (2017); https://doi.org/10.1016/j.tsf.2017.07.009

As environmentally benign and highly-efficient thin-film solar cells, low-cost Cu₂ZnSn(S,Se)₄ (or kesterite) solar cells, herein CZT(S,Se) have been steadily emerging as an alternative for the 2^nd generation Cu(In,Ga)Se₂ (CIGS) and CdTe solar cells in the past several years. CZT(S,Se) solar cells have several advantages such as high absorptions coefficient (>10⁴ cm^-1), tunable direct band gap energy (1 to 1.4 eV), and usage of abundant elements in earth's crust, becoming an ideal platform to harvest solar energy. However, the efficiency improvement and understanding of emerging CZT(S,Se) is still in an early stage compared to the 2^nd generation counterparts. In recent progress, our team reported that Germanium (Ge) incorporation into CZTSe solar cells has improved device performance, which contributed to deepening understanding of Ge-alloyed CZT(S,Se) solar cells, in particular, on the Schottky-type back-contact barrier.
We present the complete set of models about the Schottky barrier with the MoSe₂ interlayer between the absorber and Molybdenum (Mo) back-contact. Developed analytical models with in-house MATLAB simulators agree well with TCAD numerical simulation and experimental results. A new concept of an effective back-contact barrier is introduced, which provides the mechanism of the improved device performance with lower back-contact barrier for CZTSe Ge bi-layers devices. By incorporating nanoscale Ge bi-layers below and above the CZTSe absorber, the back-contact barrier is improved by 33 %, and power conversion efficiency is improved by 1.1 % than Ge single-layer devices. Based on the back-contact model, defect concentration of Ge reference devices is estimated to be 5.4 x 10¹⁶ cm^-3, whereas Ge bi-layers devices show an estimated <9.8 x 10¹⁵ cm^-3 at the back-contact interface. To compare models with experimental results, we fabricated three different types of CZTSe: nanoscale Ge devices. The property of the back-contact barrier is assessed by characterizing a temperature-dependent behavior of CZTSe alloyed devices.
Collectively, the reduced defect concentration by Ge incorporation in the MoSe₂ interlayer improves the barrier height of back-contact. As the defect concentration of the MoSe₂ interlayer increases, the benefit from the back-contact barrier lowering by the valence bands offset is reduced and essentially eliminated. Moreover, the increased effective back-contact barrier degrades the series resistance and the overall device performance. However, the barrier at the interface between MoSe₂ and Mo metal contact remains the same, even with MoSe₂ defect concentration increase. Incorporating thin Ge nanolayer at the interface between the absorber and MoSe₂ could positively influence and possibly reduce the defect states, which lowers the effective back-contact barrier from modeling and simulation results. Hence, the Ge layer at the back-contact interface successfully improves the conversion efficiency up to 8.3 % with the improved back-contact.

The replacement of traditional CdS buffer layer with magnesium doped zinc oxide (ZMO) has been demonstrated helpful to boost higher CdTe solar cell performance exceeding 18% due to the reduced interface recombination and parasitic light absorption by buffer layer .¹ However, due to the atmosphere sensitivity of ZMO film, The post-treatments of ZMO/CdTe stacks, including CdCl2 treatment, back contact deposition et. al, which are critical for high performance CdTe solar cells, became crucial challenge. In order to maximize the expected benefit of ZMO cooperation, plenty of investigation need to be accomplished in the future. Here, copper thiocyanate (CuSCN) is demonstrated to be a powerful back-contact material with multi-advantages without degrading ZMO layer in CdTe solar cells, even with ammonium hydroxide as the solvent. The thickness of CuSCN film as well as the subsequent annealing temperature are systematically investigated and the highest efficiency 16.5% has been achieved with open-circuited voltage 850 mV, short-circuited current 26.2 mA/cm² and fill factor 74.0%.


1 Munshi, A. H. et al. Polycrystalline CdSeTe/CdTe Absorber Cells With 28 mA/cm² Short-Circuit Current. IEEE J. Photovolt. 8, 310-314, DOI:10.1109/JPHOTOV.2017.2775139 (2018).

Tin sulfide (SnS) is a desirable absorber layer for thin film solar cells. Its optical bandgap of 1.1 eV, absorption coefficient of 10⁵ cm^-1 in the visible region, inherent p-type electrical conductivity, non-toxicity, earth abundance of the constituent elements, and stability in humid and oxygen ambient are all attractive toward developing alternative solar cell technology. Nevertheless, solar cells of SnS have reached a photovoltaic conversion efficiency of only 4.8 %. The presence of secondary phases (SnS₂, Sn₂S₃) and small grains is among the efficiency limiting factors. Here, we present methods to enhance the quality of chemically deposited SnS films through using thin layers of Sn and SnCl₂ as flux materials produced by thermal evaporation and/or sulfur vapor to play a key role in grain growth during heating at 450 °C, while conserving its p-type electrical conductivity.
We prepared SnS thin films of 300 nm in thickness in an autoclave at 2.5 atmospheres and 55 °C with duration of deposition of 6 h. The deposition solution contained tin chloride and sodium thiosulfate. The as-deposited SnS thin films were of orthorhombic crystal structure, with an optical bandgap of 1.33 eV (indirect) and electrical conductivity of 1×10^-4 Ω^-1 cm^-1 in the dark and 1×10^-3 Ω^-1 cm^-1 under illumination. The films presented compact surface and dense cross-sectional morphology. We heated this film for 30 min, in a tubular oven in nitrogen atmosphere at 450 °C. Different flux materials were added during the heating to promote grain growth: sulfur, evaporated thin films of Sn metal, or SnCl₂. The film heated with sulfur powder was p-type, with electrical conductivity, 10 ^{– 4} Ω^{- 1} cm ^{– 1}. SnS films heated under the same conditions using Sn as well as SnCl₂ as flux materials retained the single phase of SnS with the formation of well-defined columnar grains of size equivalent to film thickness. They have a bandgap of 1.1 eV (indirect) and p-type electrical conductivity, 10^-3 Ω^-1 cm^-1 in the dark, which increases to 1.5 times, under illumination. Thus, heating SnS absorber films deposited at a low temperature using chemical methods, with Sn or SnCl₂ as flux materials is a promising methodology to achieve grain growth. We shall illustrate the effect of this in solar cells using SnS as absorbers.

Multijunction photovoltaics employed in solar concentrator architectures can have theoretical efficiencies as high as 65% for a 4-junction cell [1]. For these devices, sub-cell band gaps of 0.96-1.18 eV are required. Epitaxially grown GaAsBi is of interest for this application due to (1) less lattice mismatch to GaAs than InGaAs for comparable band gaps [2] and (2) a proposed temperature-insensitive band gap and reduction in Auger recombination [3]. However, these materials are challenging to grow due to the surfactant-like nature and low solubility of bismuth (Bi) in this system. The low growth temperatures necessary for good bismuth incorporation is often at odds with good material quality. The need for depleted arsenic overpressure can often lead to challenging dopant incorporation [4], a critical requirement if these devices are to be employed in PN junctions.

In this work, we explore methods to improve dopant incorporation and carrier mobility in GaAsBi_0.02 and GaAsBi_0.06 for future photovoltaic devices. Samples were grown on a Veeco GENxplor MBE using a valved As₄ source and a solid source effusion cell for group-III elements and Bi. Bismuth fraction and strain state were determined by high-resolution x-ray diffraction and absorption edge was measured by transmission spectroscopy. Select samples are grown lattice matched to InGaAs virtual substrates to decrease compressive strain which delays both Ga-Bi droplet formation and defect formation from strain relaxation [5]. Samples grown on GaAs and InGaAs substrates are doped with silicon or tellurium for n-type doping and beryllium for p-type doping. The benefits and challenges of using a group-IV n-type dopant verses using a group-VI n-type dopant in this materials system will be discussed. Mobility of GaAsBi as a function of dopant concentration was measured using Hall effect measurements and compared for the different choice of dopants. Temperature-dependent Hall measurements are utilized to explore scattering mechanisms. Additionally, the effects of ex situ thermal annealing at 500 °C on optical and electrical parameters will be discussed. Finally, mobility data will be fed into device simulations using Silvaco to evaluate best device architectures to use bulk GaAsBi films in photovoltaics.

[1] S.P. Phillips, F. Dimroth, A.W. Bett, ‘Chapter I-4-B – High Efficiency III-V Multijunction Solar Cells’ in McEvoy’s Handbook of Photovoltaics, Academic Press, 2018, p. 439-472
[2] H. Kim et al Appl. Phys. Lett. 112, 251105 (2018)
[3] S. J. Sweeney, Z. Batool, K. Hild, S. R. Jin, and T. J. C. Hosea, International Conference on Transparent Optical Networks, Mo.C6.3 (2011).
[4] R.L Field III et al Appl. Phys. Lett. 109, 252105 (2016)
[5] M.A. Stevens, K.A. Grossklaus, and T.E. Vandervelde Journal of Crystal Growth 527 125216 (2019)

Metastable germanium-tin has attracted significant interest as a potential group IV direct bandgap material for CMOS compatible electronics and photonics. With sufficient tin incorporation, there is experimental evidence of a direct bandgap germanium-tin material.¹ Although this is a promising step for germanium-tin devices, there are further considerations for incorporating germanium-tin into a full device. Namely, the evolution of the germanium-tin surface during subsequent annealing and growth of additional device components. Previous annealing results on strained germanium-tin thin films have shown surface tin segregation and island formation.² In this study, we report results from in-situ X-ray photoelectron spectroscopy during annealing of vertical core/shell Ge/Ge_1-xSn_x nanowires grown via chemical vapor deposition. We have previously shown the capability of the core/shell structure to produce a largely strain-free Ge_1-xSn_x shell material on a highly tensile-strained Ge core.³ This structure is utilized here to investigate the annealing characteristics of the germanium-tin surface and native oxide for tin contents in the range of 2 at% - 12 at% in air-exposed nanowires. Our results show the presence of a tin rich oxide with a composition dependent decomposition temperature. After in-situ decomposition of the native oxide, the tin content of the germanium-tin surface is observed to remain constant throughout annealing at temperatures exceeding 500 °C and for anneal durations greater than one hour. Additionally, we have measured the photoluminescence at room temperature, characterized strain by X-ray diffraction, and imaged the nanowire sidewall morphology before and after annealing to examine its effects on material performance, bulk tin distribution, and local surface structure.

¹M. Seifner, A. Dijkstra, J. Bernardi, A. Steiger-Thirsfield, M. Sistani, A. Lugstein, J. Haverkort, S. Barth, “Epitaxial Ge_0.81Sn_0.19 Nanowires for Nanoscale Mid-Infrared Emitters,” ACS Nano 13, 7 (2019), 8047-8054.
²W. Wang, L. Li, Q. Zhou, J. Pan, Z. Zhang, E. Tok, Y. Yeo, “Tin surface segregation, desorption, and island formation during post-growth annealing of strained epitaxial Ge_1-xSn_x layer on Ge(001) substrate,” Applied Surface Science 321 (2014), 240-244.
³A. Meng, M. Braun, Y. Wang, C. Fenrich, M. Xue, D. Diercks, B. Gorman, M. Richard, A. Marshall, W. Cai, J. Harris, P. McIntyre, “Coupling of coherent misfit strain and composition distributions in core-shell Ge/Ge_1-xSn_x nanowire light emitters,” Materials Today Nano 5 (2019), 100026.

Since the development of display technology, a new type of industry, such as augmented reality (AR), virtual reality (VR), and wide screen displays, has arrived. Requirements of performance standards on new type of displays have become more severe. Therefore, improving the mobility and stability of thin-film transistors (TFTs) for using in such devices have been actively conducted. To date, low temperature polycrystalline silicon (LTPS) with high field-effect mobility of about 100 cm²/Vs has been widely used as channel material for TFTs, but some problems remain, such as poor uniformity for large area, high leakage current, and high fabrication costs. To resolve the issues mentioned above, amorphous oxide semiconductor (AOS) has been attracted attention due to low off current, large area uniformity, and relatively low fabrication cost. Especially, amorphous Indium-Gallium-Zinc-Oxide (a-IGZO) is most used semiconductor among AOS materials. However, for AOS materials to be applied to the above-mentioned next-generation displays, low field-effect mobility and low stability under bias stresses, which are lower than LTPS, must be overcome.
In this sense, we propose the fabrication of a-IGZO TFTs with high mobility and high stability through insertion of a carrier induced interlayer (CII). CII is a layer introduced between the channel layer and the gate insulator layer, which is thinly deposited to absorb oxygen in the effective channel of a-IGZO through oxidation. Consequently, mobility of a-IGZO TFTs can be improved by oxidation of CII increasing carrier concentration of the a-IGZO front channel.
In this study, thin aluminum layer was deposited for CII by RF sputtering process at 80 W with various deposition time 60, 90, and 120 s. Then, a-IGZO was deposited on the CII for channel layer while changing the oxygen partial pressure from 1.7 to 8.3%. After the activation process through thermal annealing at 300^oC for 1 hour, Al was deposited by RF sputtering and used as source and drain electrodes. When the a-IGZO channel was deposited with oxygen partial pressures of 1.7, 3.3, and 5% without CII, the saturation mobility was 7.36, 5.45, and 2.83 cm²/Vs, respectively. If it exceeds 5%, it is insulated. The above phenomenon caused by the oxygen vacancy concentration in the a-IGZO thin film decreases with increasing oxygen partial pressure. If the partial pressure of oxygen is less than 5%, Al for CII cannot be completely oxidized, resulting normally conductive property. On the other hand, if the partial pressure of oxygen is more than 5%, the characteristics of the TFTs deteriorate. As a result, 5% condition achieves the best characteristics and determined as the optimize condition.
In the transfer characteristics of the TFTs with various CII thickness, the saturation mobility of the TFTs with CII of 60 s and 90 s deposition were calculated as 11.30 and 13.96 cm²/Vs respectively. As the thickness of CII increases, the saturation mobility also increases. However, switching characteristics of the TFTs with CII are gradually degraded due to insufficient oxidation of CII with metal Al over 120 s of the CII deposition time. The CII with insufficient oxidation acts as current path at the interface between the gate insulator and the channel layer, resulting in a drastically increased conductivity. Consequently, the optimize condition was determined to be CII deposition time of 90 s.
Lastly, bias stress tests were performed to determine the stability of the TFTs. Negative bias temperature illumination stress (NBTIS) test under -20 V, 50^oC and 1000 lux for 3600 s was conducted on the TFT fabricated with oxygen partial pressure of 5% with CII deposition time of 90 s. When the result was compared with the a-IGZO TFT without CII (pristine), the stability characteristic was improved as the threshold voltage shifts of the pristine showed -12.05V and the device with oxygen partial pressure of 5% and with CII showed -6.73V.

Electronic devices, such as electronic control unit, power module, mobile equipment, etc., are required downsizing and higher power density. To achieve the downsizing and higher power density, electronic parts are closely mounted together in a small external package, which causes an increase of heat density. Thermal-radiation is one of the heat dissipation methods. However, conventional thermal-radiation is not useful in the electronic devices, since most of the radiated heat is absorbed by an external package such as resin.
The purpose of this study is to newly develop a thermal-radiation material, which generated heat is converted into infrared light with spectrum transmitting through the resin. The material is so-called spectrally selective thermal-radiation material (SSTRM). The studied SSTRM is composed of metallic spheres arrayed on the surface of a heating element, and resin film covering the metallic sphere array. The film is formed by applying a coating solution including metallic spheres. The optimum diameter and arrangement of the metallic sphere are founded numerically and experimentally.
The thermal-radiation spectrum was calculated by using an optical simulation tool based on Rigorous Coupled Wave Analysis. The calculation results clearly showed the localized plasmon resonance excited in the metallic spheres array. The localized plasmon resonance enhanced the emission at the resonance wavelength. The resonance wavelength was controlled by changing the diameter and arrangement of the metallic spheres. It was revealed that infrared light with spectrum transmitting through the resin film is emitted, when the spheres with a diameter of 1.5 um are densely arrayed.
The effectiveness of the SSTRM was experimentally demonstrated by evaluating thermal-radiation characteristics under the stationary atmosphere with a temperature of 298 K. Input power to a heating element was fixed at a constant value, so that the surface temperature of the heating element was 381.1 K. Three types of coating materials were prepared as follows; A. only resin film, B. metallic spheres dispersed in resin film, and C. optimized metallic sphere array covered with resin film (SSTRM). The surface temperatures of the heating elements coated with the films A, B, and C were evaluated to be 358.3K, 355.8K, and 348.1K, respectively. The film C achieved drastical temperature reduction of 33K compared with the heating element without the coating materials. The temperature of the film C was lower than that of B, due to the effect of the metallic sphere array formation. Consequently, the optimized metallic spheres array acted as the spectrally selective emitter that radiates infrared light transmitting through the resin. The SSTRM will be powerful material for heat dissipation from electronic devices packed in resin case or mold.

Zinc oxide (ZnO) is a versatile semiconductor material used for a wide range of applications in electronics, photonics, and optics. Bulk ZnO is a wide band gap (3.37 eV) semiconductor that generates electrons under an electromagnetic wavelength shorter than the ultraviolet (UV) range. ZnO nanowires (ZnO NWs) exhibit enhanced electrical properties over ZnO thin film via a nano size effect based on the electrical resistivity of chemical under UV irradiation. The NW structure induces movement of electrons along an axial direction due to the electron confinement effect by structural features. In order to utilize the advantages of nanostructured film in commercial sensors, mass production should be achieved with simple manufacturing processes. Electrospinning with a continuous process is possible in mass production with a roll-to-roll system. The use of a high porous electrospun nanofiber (ES NF) film is advantageous in terms of fabricating dense ZnO NWs. Additionally, a highly aligned state of ES NF films is possible to fabricate via controlling the electrospinning condition without entailing additional processes. The aligned nanofiber film aids in electron movement along the direction of the fiber, thereby revealing possibilities for its application to high responsibility in UV detector. Under the UV irradiation condition, ZnO NWs on ES NF film generate electrons under light energy absorption conditions that exceed the energy required to excite electrons based on Planck’s equation. Under continuous UV irradiation on the ZnO NWs surface, electrons in the valence band are excited to the conduction band when the electrons gain sufficient energy. The excited electrons on the ZnO NWs surface along the ES NF support have a strong directionality that tends to flow into the low current density region. The electrical conductivity on ZnO NWs increases by using aligned ES NF when compared to that of random ES NF by large gradient of current density. Linearity error of the ZnO NWs on aligned and random ES NF films are less than ±0.58% and ±1.60%, respectively. Under UV irradiation conditions, the sheet resistance of ZnO NWs on the aligned ES NF film (0.17MΩ/,) is 9.88 times lower than that for ZnO NWs on random ES NF film (1.68MΩ/,). The results for ZnO NWs on ES NF films under the UV irradiation and dark conditions indicate sheet resistance ratios (Rsq,off/Rsq,on) of 551.0 and 47.8 for ZnO NWs on the aligned ES NF film and ZnO NWs on the random ES NF film, respectively. The sheet resistance ratio of ZnO NWs on the aligned ES NF film is 11.5 times than that of ZnO NWs on the random ES NF film. The electric current on aligned ES NF surface shows higher flow rate than electric current on random ES NF surface. Even though under the identical hydrothermal conditions to fabricate ZnO NWs, a geometry of ES NF film substrate is an important factor for improving the photo detecting performance of ZnO NWs on ES NF films.

Among various nanostructured metal oxides, molybdenum oxide (MoO₃), a wide band-gap n-type semiconductor, has attracted much interest as a promising gas sensor because of its excellent sensitivity, good reproducibility, low cost, high compatibility with microelectronic processing, low electron recombination rate, and high stability. The MoO₃ thin films have promising electrical properties for gas sensing applications.
To synthesize two-dimensional materials under ultra-high vacuum is a key advantage for enhancing the properties of nanostructured materials. In this presentation, we demonstrate the synthesis, characterization, and gas sensing properties of MoO₃ films, which were grown using Krypton-Fluoride (248 nm) excimer laser. Deposition was done under 200 mTorr Oxygen pressure at the substrate temperature of 600^oC. The targets used for deposition were synthesised using hydrothermal method. The deposition was done with the optimized parameters, to ensure good polycrystalline films on the Si (001) substrate. The optical, morphological and compositional properties of MoO₃ films were studied by using UV-Visible spectroscopy, Scanning Electron Microscopy and Raman Spectroscopy. The formation of MoO₃ was also confirmed using X-Ray Diffraction technique. The MoO₃ films showed excellent response to NO_X and NO_X gas moieties, at even low gas concentrations from 100 ppm onwards. The sensor not only exhibits high sensitivity but also good selectivity towards NO₂ gas.

As devices continue to trend towards the micro- and nanoscale, the influence of individual crystal defects on their optical, electronic, and thermal properties becomes significant. As an example, the thermoelectric properties of these nanodevices are closely related to the presence of crystal defects, grain boundaries, and strain. While the effects of interfaces and grain boundaries on thermoelectric performance have been widely studied in semiconductor materials, there remains an incomplete understanding of the role that crystal defects play in metals. A major challenge in pinpointing the origins of thermoelectric variation in metallic nanodevices is the difficulty in preparing high quality crystalline thin-film metal microstructures on thermally and electrically insulating amorphous substrates.

Controlling the micro- and nanostructure of metals offers the opportunity to investigate and engineer the properties of metal nanotechnologies. Here, we report a high-throughput method for preparing patterned crystalline gold microstructures grown directly on amorphous insulating oxide via rapid melt growth. In this growth process, thin-film gold-platinum metal-seed microstructures are encapsulated in an insulating crucible and annealed at temperatures above gold’s melting point. During cooling, liquid epitaxial growth directs single crystal solidification [1]. We can extend this platform to controllably prepare single grain boundaries [2], enabling the systematic study of the role of defects in governing local thermoelectric behavior. A scanning photothermoelectric (PTE) setup to allows us to probe the microscopic variations in photovoltage across the gold single crystal and bicrystal wires. Surprisingly, we observe a photovoltage distribution where local changes in Seebeck coefficient are highly influenced by long-range strain fields and concentrated dislocations and uncorrelated with plane grain boundary defects. Our results showcase the link between the presence of elastic strain and dislocations to the PTE effect using a model material system.

[1] K. Zhang, X. B. Pitner, R. Yang, W. D. Nix, J. D. Plummer, J. A. Fan, Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 685.
[2] L. T. Gan, R. Yang, R. Traylor, W. Cai, W. D. Nix, J. A. Fan, Adv. Mater. 2019, 1902189.

Amorphous oxide semiconductor (AOS) based thin-film transistors (TFTs) have been rapidly evolved from academic curiosity to the display industry because of their many advantages such as low off-current, high uniformity over large area fabrication, and high visible light transparency compared to conventional amorphous silicon (Si) TFTs. Despite these advantages, the poor electrical performances such as low field-effect mobility compared to low-temperature polycrystalline Si TFTs have been an issue for high resolution and high frame rate displays. Thus, a wide range of research has been conducted to enhance the field-effect mobility in AOS TFTs: doping a nano-conductor in the back channel, adoption of high-k gate insulators, and capping reduction layer. Among such researches, one of the most prominent methods was applying heterostructure with different oxide semiconductors to the channel layers of TFTs; due to large on-current and high field-effect mobility. The large difference in the work function of oxide semiconductor layers results in bandgap mismatches, forming potential wells. Therefore, electrons are confined in the potential well forming a 2-dimensional electron gas (2DEG). However, to apply 2DEG-based TFTs to the actual display industry, accurate 2DEG control and formation mechanisms need to be further studied.
In this paper, high performance oxide TFTs were fabricated by sodium hydrosulfite (Na₂S₂O₄) reducing agent treated indium-zinc-oxide (r-IZO) as the front channel of zinc-tin-oxide (ZTO)/r-IZO heterostructure. Front channel IZO film was deposited by RF sputtering at room temperature. For reducing the IZO film, the IZO film was dipped into Na₂S₂O₄ solution. Applying the Na₂S₂O₄ reducing agent treatment on the front channel allows controlling the amount of confined electron at the interface between the front and the back channels for heterostructure oxide TFTs. For the back channel, ZTO film was deposited by RF sputtering at room temperature. ZTO/r-IZO heterostructure is suitable for the formation of 2DEG due to the large difference in work function, so it could produce high field-effect mobility even when fabricated at room temperature. Also, the large on-current could be achieved by precisely controlling the formation of 2DEG through Na₂S₂O₄ reducing agent treatment on the IZO front channel. To confirm the formation of 2DEG at the interface between the ZTO film and the r-IZO film, we performed a UV-vis spectrophotometer and ultraviolet photoelectron spectroscopy (UPS) measurements. The 2DEG was successfully formed by showing the difference of 170 meV work function at the heterointerface. However, conventional 2DEG-based oxide TFTs have high off-current because 2DEG free electrons remained at the heterointerface regions even when the TFT is in the off-state. To solve this problem, the ZTO back channel was applied to different oxygen partial pressures to the lower ZTO layer (20%) and the upper ZTO layer (30%) when deposited via sputtering. As a result, the high off-current could be adjusted to near the 10^-12 A range and ZTO/r-IZO TFTs exhibit field-effect mobility of 29.55 cm²/Vs, a subthreshold swing of 0.48 V/decade, and an on/off current ratio of 5.02×10⁸.
The results show that room temperature fabricated ZTO/r-IZO heterostructure TFTs can be used as a general route for high field-effect mobility TFTs and controllable oxide semiconductor devices for offering compatibility of the flexible and high-end display technology.

For a long time, the etching technique played an important role in the electronics industry, and had a critical influence on the development of modern manufacturing technologies. Regarding to the operation of electronics devices, surfaces and interfaces significantly affect the functionality and performance of the devices. To investigate the formation mechanisms for characteristic surface morphologies during chemical etching, in this study an anisotropic chemical etching model was established and the formation and evolution of surface structures with resist-masks were simulated numerically. In this theoretical model, two major crystallographic orientations in a cubic system, such as <100> and <111> directions, were taken into consideration and included in the reactions of surface diffusion and anisotropic etching. By adjusting the numerical parameters corresponded to the different experimental conditions, various surface profiles could be created and the influence of the experimental parameters on the surface morphologies would be revealed. From the calculations of anisotropic etching along <100> preferred-directions, a V-shape profile with faceted surface morphologies appeared. A larger loss of linewidth or a higher etch bias was displayed with the increase of the etching rate. While an anisotropic etching along <111> preferred-direction was examined, etch pits were firstly produced in the early stage of etching process. A different surface evolution during a chemical etching was distinctly demonstrated due to the crystallographic structure. In addition, since the resist-mask was included in the numerical model, the phenomenon of undercutting in an anisotropic chemical etching could be also discussed in this study. These numerical simulations could provide a fundamental information to understand the formation mechanism of etched surface, and broaden the applications of chemical etching in the advanced manufacturing technologies.

GaN based high electron mobility transistors (HEMTs) are considered of great potential to be the building blocks in the next-generation microwave power amplifiers due to their excellent power handling capabilities Ohmic contact and passivation technology are the main factors of device power and efficiency. Improving device efficiency is the premise of improving application system efficiency. Nowadays, More and more efforts are made to improve the efficiency of GaN HEMTs. In this paper, we report the AlGaN/GaN HEMT with improved ohmic contact and double layer pssivasion. the device show an excellent power characteristics.
The epitaxial layers consists of a 1.5nm GaN cap layer,22nm Al0.25GaN, 2nm AlN and GaN buffer layer. The process for GaN HEMT devices is as followed the steps: a)maker, b) ohmic contact etching ion; c) isolation; d) gate; e)doubel passivation by Al/SiN;f) via and metal. The source-drain and gate length are 2.40µm and 0.25µm.the TEM result shows the cross section of recessed TiN/Ti/Al/Ni/Au. For comparison. Conventional structures of MOCVD HEMT is also fabricated simultaneously.
Measurements are performed in continuous wave (CW) mode, load-pull power measurements of this 4×50μm device at 10 GHz is carried out. The device is biased at a voltage of 35V.the output power reached 31.50dBm (7.0W/mm) with 70%
power-added efficiency (PAE) and 14dB gain. Such a good RF power performance comes from the improved ohmic contact and new double passivasion layer.
The interface of ohmic contact is analyzed by means of TEM. It is found that tin thin layer combined with recessed process is more conducive to the formation of low contact resistance.

The use of halogen chemistries on silicon surfaces offers a promising path towards atomic-scale manufacturing of acceptor-based devices. Atomically-precise fabrication of electronic devices in silicon has been widely demonstrated using a scanning tunneling microscope (STM) and hydrogen-based surface chemistries to precisely place phosphorous donor atoms on the surface. A lithographically patterned hydrogen resist is used to selectively incorporate PH₃ to produce metallic wires, quantum dots defined by electrostatic gates, and single donor atom qubits for quantum information (QI) research. However, interest in acceptor dopants and hole-based devices necessitates the development of alternate precursors and/or resist chemistries for device fabrication. To incorporate acceptors into silicon with atomic placement, halogen chemistry is potentially more favorable than hydrogen. To that end, we explore halogen-based resists including chlorine, bromine, and iodine in conjunction with metal-halide precursors to achieve area-selective deposition of acceptor dopants in silicon. Here, we present results on the complete surface passivation of Si(100) using Cl, Br, and I. We study the stability of halogen-terminated Si(100) surfaces in ambient environments using STM and x-ray photoelectron spectroscopy (XPS) to facilitate transport of samples outside of ultra-high vacuum environments. Finally, we present STM and secondary ion mass spectroscopy (SIMS) results on the adsorption and incorporation of metal-halide precursors on Si(100) as a viable path towards acceptor doping.

Superconducting resonators and qubits are essential for the development of quantum-computing architectures. These systems depend on high quality superconducting materials and interfaces, in which the two-level systems loss is minimal. Our previous work focused on understanding of the effects of plasma modification of vacuum/silicon and vacuum/superconductor interfaces on the performance of TiN/Si superconducting microwave resonators. In this paper, we discuss plasma-based modification of TiN/Si interface. We provide a systematic study of every step in the preparation of the TiN/Si heterostructures, namely the chemical treatment with hydrofluoric acid of SiO₂/Si surface, the plasma treatment of H-terminated Si and the TiN film deposition processes. We find that both the gas environment and the plasma processing conditions affect the modification of Si, which in turn affects the composition of the TiN/Si interface. The impact of this work lay in providing new approaches to TiN/Si interface preparation which may enable reduction of metal/dielectric interface losses thus improving the superconducting microwave resonators performance. This work was partially supported by the office of Naval Research through Naval Research Laboratory Base Program.

In this talk, we report the use of surface-enhanced Raman scattering (SERS) to investigate the adsorption of thiol and isocyanide molecules on colloidal Ag nanocubes in situ. In a typical process, we mixed Ag nanocubes with ligand molecule solutions at different concentrations and then collected SERS spectra at different time points. We demonstrate that 4−nitrothiophenol (4-NTP) binds to the Ag surface rapidly because of the strong Ag−S chemical bonding. By decreasing the concentration of 4−NTP, we observe red shifts of the ν_NO2 band, which suggests the change to the orientation of ligands on the surface. Different from thiols, the 1,4−phenylene diisocyanide (1,4−PDI) binds to the Ag surface through σ−donation with weaker binding strength. By increasing the concentration of 1,4−PDI, our results indicate that the molecules tend to adopt an orientation that tilts away from the surface. More interestingly, we investigate the competitive binding behavior of these two ligands by introducing Ag nanocubes into a solution containing both 4−NTP and 1,4−PDI. Our results show that thiol is more preferential to bind onto the surface when the concentrations of the two ligands are comparable in the solution.

Conventional thermal management solutions in microelectronics packaging suffer from an inability to effectively couple thermal conduction and mechanical compliance into a single material, routinely necessitating a sacrifice in thermal, and therefore device, performance in favor of mechanical reliability. For the past few decades, compliant thermal interface materials (TIMs) have been limited to polymers loaded with dispersions of unaligned or randomly distributed conductive elements which can only achieve thermal conductivities up to ~5 W m⁻¹K⁻¹. Here, we present the processing and characterization of a compliant, vertically-aligned metal nanowire composite ideal for achieving thin, flexible interfaces with low thermal resistance. The aligned nanowire composite is fabricated by electroplating copper into a 25-45% porous, sacrificial ceramic template, liberating the nanowires by chemically dissolving the template, and infusing the resulting nanowire array with a thermally stable polymer by capillary-driven wicking. An integrated bonding layer is added by sub-filling the template with copper and sequentially electroplating metallic solder to form localized solder tips on the nanowires. The entire fabrication process can be completed at wafer-scale with a lateral area up to 12.5 inches². The thermal performance of the composite TIM was evaluated with a modified thermal resistance measurement designed to mimic typical operating conditions. The double-sided construction of the composite and integrated bonding layers facilitate integration between any two arbitrarily-chosen surfaces by enabling the absorption of interfacial strain, mismatch, and roughness while achieving thermal conductivities up to 200 W m⁻¹K⁻¹, significantly advancing the state-of-the-art in TIMs.

The need to increase coherence times in quantum devices simulates the search for materials systems that can exhibit electronic states not affected by structural and chemical inhomogeneities. Owing to the advances in synthesis of III-V semiconductors, concentrations of point defects and impurities in these materials can be well controlled, while stacking faults (SFs) provide asymmetric quasi-2-dimensional potential wells that behave as internal interfaces capable of trapping excitons [1]. Here we use ab initio simulations to examine how the stability and electronic properties of SFs change as a function of the distance between them and how the details of their configuration affect spatial distribution of the electron and hole components of excitons and their effective masses. Our simulations suggest that a potential drop of the order of 10 meV is associated with an isolated SF, which is sufficient for the hole to localize predominantly at the SF and to induce an asymmetric distribution of the excited electron density in the vicinity of the SF. For interacting SFs, we find that the electron distribution is sensitive to both the type of the SFs and the distance between them. This coupling provides a pathway to the design of materials with spatially resolved excitonic properties and photo-emission energies by controlling the density and relative arrangements of the stacking faults.

[1] T. Karin, X. Linpeng, M. M. Glazov, M. V. Durnev, E. L. Ivchenko, S. Harvey, A. K. Rai, A. Ludwig, A. D. Wieck, K.-M. C. Fu, Phys. Rev. B 94, 041201(R) (2016)

4H-SiC metal-oxide-semiconductor field-effect transistors (MOSFETs) have great potential in high power and high temperature applications. However, near-interface defects degrade the effective channel mobility and thus limit the performance of 4H-SiC MOSFETs. In this work, we utilize electrically detected magnetic resonance (EDMR) to study hyperfine interactions with spin-dependent recombination (SDR) of silicon vacancy defects at the 4H-SiC interface in 4H-SiC MOSFETs. The physical foundation of EDMR is electron paramagnetic resonance (EPR). In EPR and EDMR, a sample is subjected to a large, slowly varying magnetic field and an RF or microwave oscillating field at a frequency ν. In the simplest case, EPR is achieved when hν=g_eβH.¹ Here, h is Planck’s constant, ν is the electromagnetic radiation frequency, g_e is the free electron g factor (g_e ≈ 2.0023), β is the Bohr magneton, and H is a large externally applied magnetic field. At resonance, the electron transitions from its +1/2 to -1/2 spin state (or vice versa). The analytical power of EPR and EDMR comes from deviations from this simple case caused by the paramagnetic site environment. These perturbations are, for the most part, caused by hyperfine interactions, interactions of the electron with magnetic nuclei, and spin-orbit coupling. EPR detected via SDR allows for measurements of small numbers of defects via spin-dependent changes in device current. The sensitivity of SDR/EDMR is very nearly independent of the field or frequency at which the measurement is made.²
Prior continuous wave EDMR (cwEDMR) studies have linked interface and near-interface defects to carbon dangling bond defects³ and silicon vacancy defects.⁴ The identification of silicon vacancy defects in 4H-SiC MOSFETs was based on measuring nearest neighbor carbon 13 hyperfine interactions.^{4 29}Si next-nearest neighbor (NNN) hyperfine interactions are too small to resolve in an EDMR measurement due to the width of the center part of the resonance signal. The interactions of NNN ²⁹Si are isotropic with a value 2.96 G determined from conventional EPR. Broadening in EDMR obscures these ²⁹Si NNN hyperfine interactions. We have overcome the EDMR broadening problem using the Breit-Rabi effect,⁵ a shift in the hyperfine side peaks towards lower fields and frequencies. This effect is sufficiently large so that it may be detected in ultra-low field and frequency measurements. We utilize frequency swept EDMR (fsEDMR). With fsEDMR, tuning the resonance condition is performed by varying a low static field. In fsEDMR, the magnetic field H is held constant as frequency is swept. Our experiments were carried out at fields less than 10 G and frequencies less than 30 MHz. We observe a centerline with a sharp peak on the low field side only at these low fields and frequencies. We attribute the peak to otherwise unresolved hyperfine interactions made observable by shifting due to Breit-Rabi effects. The peak has fairly large amplitude. EasySpin⁶ was utilized to simulate the results accounting for the presence of one ²⁹Si NNN. The results of the simulation are in good agreement with our experimental observation.
¹ W. Gordy, Theory and Applications of Electron Spin Resonance (John Wiley & Sons, 1980).
² D. Kaplan, I. Solomon, and N.F. Mott, J. Phys. Lettres 39, 51 (1978).
³ J. Cottom, G. Gruber, G. Pobegen, T. Aichinger, and A.L. Shluger, J. Appl. Phys. 124, (2018).
⁴ C.J. Cochrane, P.M. Lenahan, and A.J. Lelis, Appl. Phys. Lett. 100, 023509 (2012).
⁵ G. Breit and I.I. Rabi, Phys. Rev. 38, 2082 (1931).
⁶ S. Stoll and A. Schweiger, J. Magn. Reson. 178, 42 (2006).</p>

Color image sensing by a smartphone or digital camera is usually performed by an array of sensor elements covered by color filters. The optical color filters cause color errors like the color aliasing or Color Moiré effect. Furthermore, the quantum efficiency of typical conventional optical color sensor consisted of side-by-side arranged color filters is limited to 1/N, where N is the number of color channels per pixels. Hence, the efficiency of the color detection process is limited to max. 33% by the side-by-side arrangement of sensor channels. In this study, color sensors consist of vertically stacked diodes using perovskite alloys is described with maximum quantum efficiency approaching 100%. The described sensor structure is color aliasing or Color Moiré error free, while conventional sensors using optical filters are limited by this error. The optical constants of the required perovskite alloys were determined by applying the energy shift model to experimentally determined optical constants of pure perovskites. The spectral sensitivities of the color sensors were determined by electromagnetically modelling. The spectral sensitivities of the sensors were colorimetrically characterized and compared to sensors in literature including conventional sensors using optical filters. Up to our knowledge for the first time it could be shown that a vertically stacked three color sensor exhibits a color error equal to, or smaller than errors of conventional sensors using optical filters as well as silicon based Foveon stacked color sensor. Details on the used materials, the device design, and the colorimetric analysis are provided.

Thin crystalline silicon (less than 50 µm thickness) with outstanding flexibility has been considered as a means for developing flexible crystalline silicon (c-Si) solar cells. However, the optical management of the thin c-Si is required for developing the highly efficient c-Si flexible solar cells due to the high surface reflection (more than 30%) and absorption loss over 900 nm wavelength region. Here, we introduce novel high efficiency c-Si flexible solar cells which include random inverted pyramidal polydimethylsiloxane (RIP-PDMS) optical sticker on the 20-µm-thick thin c-Si solar cells. Due to the enhanced light absorption at the entire wavelength region, the thin c-Si with RIP-PDMS optical stickers showed the improved integrated photon flux absorption up to 85.4% (75.0% to 85.4%). The RIP-PDMS sticker enhances the light absorption capability of the thin c-Si by increasing the optical path length in the thin c-Si up to 10%. Further, we analyzed the optical enhancement of the thin c-Si with the RIP-PDMS optical sticker through the finite-difference time-domain (FDTD) simulation. From the optical simulation, we confirmed that the optical path length in the thin c-Si increased because the angles of incident light at the front and rear surface of the thin c-Si tilted up to 10° and 15°, respectively, by attaching the RIP-PDMS optical stickers. Notably, the thin c-Si solar cells with the RIP-PDMS optical stickers showed 18.4% power conversion efficiency while maintaining superior flexibility (bending radius of 10 mm). In addition, our flexible thin c-Si solar cells with the RIP-PDMS optical stickers showed high bending stability without any change in efficiency, even with 1000 bending cycles with a bending radius of 10 mm. Thus, we believe that the thin c-Si solar cells with RIP-PDMS optical stickers can be a promising candidate for the highly efficient flexible solar cells.

The process by which strongly-photoexcited semiconductors return to their ground state consists of a series of complex, strongly-correlated many-body interactions which overlap in space and time, resulting in an evolution of behaviors that is both difficult to study and valuable, both for fundamental insights into the quantum nature of matter and for control of device and material behaviors in electronic and optoelectronic applications. Extensive work has been done in this regard, illuminating the sequence of hot carrier excitation, carrier-carrier scattering, carrier-lattice scattering and thermalization, recombination, and lattice heat diffusion. After strong photoexcitation creates a high-energy bath of “hot” carriers, electron-phonon coupling results in the emission of optical phonons, which decay into lower-energy phonons, either optical or acoustic. The resulting nonequilibrium phonon populations can in turn influence hot carrier relaxation and transport, an inextricability which has been demonstrated through transient Raman scattering [1-3]. The inseparable nature of structural behavior from that of excited electron-hole plasmas is further demonstrated by the result that when not in equilibrium, acoustic mode phase velocities change as a function of time due to residual Coulomb interaction caused by imperfect screening by electrons [4]. Thus, a thorough understanding of ultrafast post-excitation structural dynamics is essential to a full comprehension of semiconductor behavior.
As electronic devices continue to shrink, the associated increased importance of device surfaces, interfaces, and defects add an additional layer of complexity to the structural behavior. Techniques typically employed to investigate post-excitation responses operate in reciprocal or momentum space and require spatial averaging, reducing real-space resolution. It is known that interfaces change the electronic structure in their surrounding regions, strongly influencing carrier dynamics, and it has been shown in our previous work that phonons are structure-directed, propagating specifically from interfaces and repeatably traveling perpendicularly to the launching interface [5]. In sum, electron-phonon coupling behaviors are known to be influenced by structure, making real-space information relevant to a complete picture of post-excitation behavior.
Here we use ultrafast electron microscopy (UEM) to image phonons in real space and thereby demonstrate the influence of both the meso-scale and atomic structure of GaAs on the post-excitation phonon behavior. Making use of combined picosecond-nanometer real-space resolutions, we have directly imaged excitation of GHz coherent strain waves propagating exclusively from specimen-vacuum interfaces with initial hypersonic velocities which decay over time to the speed of sound in GaAs. We also report the change in the properties of these waves upon laser annealing of the specimen, epitaxially crystallizing the focused-ion-beam-induced amorphous surface layer and resulting in measurable hardening of the specimen. This appears in the phonon behavior as an increase in the velocity of the phonons. By imaging the phonon interactions with both the specimen-vacuum interface and the amorphous-crystalline interface within the specimen, we demonstrate the importance of interfaces in fully understanding the phonon dynamics in photoexcited GaAs.

[1] A. Othonos, Journal of Applied Physics 83, 1789 (1998), p. 1789.
[2] D. Von Der Linde, et al., Physical Review Letters 44, 1505 (1980), p. 1505.
[3] J. A. Kash, et al., Physical Review Letters 54, 2151 (1985), p. 2151.
[4] B. Y.-K. Hu, et al., Physical Review B 44, 11067 (1991), p. 11067.
[5] D. R. Cremons, et al., Physical Review Materials 1, 073801 (2017), p. 073801.

Infrared detector technology has shown remarkable progress towards the selection of detector materials and device architectures. Infrared detectors are an integral part of various surveillance devices, remote sensing system, night vision, all weather surveillance, missile tracking, thermography, enhanced vision systems for aeroplanes, automobiles and gas sensing.
Current InSb and HgCdTe based IR technologies have shown high performance because of excellent material properties, covering near infrared (NIR 0.75-1.1mm), short-wavelength infrared (SWIR 1-3mm), mid-wavelength IR (MWIR 3-5mm), long-wavelength IR (LWIR 8-14mm) and very long-wavelength IR (VLWIR 14-30 mm) range. However, these need to be cooled to cryogenic temperatures making them bulky and power hungry. Thus future infrared photodetector systems must possess reduced cooling requirements, consume less power, exhibit longer lifetimes, and have improved manufacturability. These targets have been partly addressed by the development of nBn detector architecture raising the operation temperature , reliability, higher sensitivity and low SWaP (Small Weight and Power ) .
Type2Superlattice (T2SL) detectors combine key advantages of InSb and Mercury-CadmiumTelluride (MCT)-based technologies namely scalability of focal plane arrays (FPAs), high sensitivity, lower tunneling currents and suppressed Auger recombination rates nad also facilitates a higher operating temperature. A super lattice is made up of a repeated series of very thin epilayers of different materials that leads to creation of sub-bands. The corresponding bandgap energies of these subbands determine the cut-off wavelength of the sensor. The III-V InAs –GaSb materials, based on nBn unipolar photodetector architecture, with an n-type absorption layer, a barrier layer, and an n-type contact layer have been fabricated and engineered for bandgap tunability. The barrier layer is constrained to have both a negligible valence band energy offset with the absorption layer, that suppresses the Shockley-Read-Hall (SRH) generation current thereby decreasing the primary noise In addition, detector is passivated to minimise the surface leakage current The nBn photodetector, through the near elimination of SRH and surface leakage currents, requires less cooling to operate optimally than the p-n junction photodiodes. Bandgap engineering and optimization of the material composition to achieve maximum quantum efficiency, low dark current, maximum absorption strength and higher operating temperature.
Appropriate tools suitable for the design of T2SL material systems are not yet fully developed due to the intricacies of T2SL material system and paucity of experimental data for input and for the validation of mathematical formulations. A holistic approach has been followed involving design of graded energy-gap T2SL structure, growth of material system, characterization of material, fabrication and characterization. Our design is aimed to grow T2SL epilayers with optimized thickness, material composition, energy gap grading and varied doping concentrations by Molecular Beam Epitaxy and detector fabrication with a combination of wet and dry etching. A favorable impact ionization coefficient leads to higher bandwidth range with low excess noise.to achieve maximum quantum efficiency, low dark current, high avalanche gain, low excess noise, maximum absorption strength and higher operating temperature. Design optimisation has resulted in high quantum efficiency of 75% validated with FTIR measurements and low NETD of 18mK at an elevated operation temperature of 140K. Current-Voltage measurements, quantum efficiency, noise equivalent temperature difference (NETD) and issues related with growth of graded energy-gap T2SL material system will form part of the full paper.

Wearable wireless communication systems based on Electronic textile (E-textile) have become significant interests owing to their potential applications in sportswear, military, firefighting, health care, and tracking systems. Among the fabrication methods of conductive textile, screen printing which prints metal nanoparticle based conductive inks on a fabric substrate has been known to be a promising approach in the field of wearable electronic applications. However, the morphological characteristics of textile materials such as high surface roughness, high porosity, and capillary force in the woven fabric structures and the lack of manufacturing technology to overcome the issues are obstacles to commercialization of electronic textiles (E-textile). Herein, we compare the flat and round yarn based polyethylene terephthalate (PET) fabrics as the substrates for screen printing. We also investigate the effect of the parameters of the screen printing process such as the screen mesh size (70, 120 pixel/inch) and the printing counts. The uniformity of the screen-printed layers and their electrical properties are directly related with the shape of the yarn, the roughness of fabric substrates, and the printing processing conditions. A minimum average sheet resistance of 16 ± 3 mΩ/sq. was achieved on the flat yarn PET fabrics, and there was little change in electrical performance after the bending test (r=3mm) was carried out 1,000 times. To demonstrate, we show textile antennas operating at 2.4GHz which is most widely used for public wireless LAN service, Bluetooth, radio frequency identification (RFID). We analyzed the effects of the uniformity and the conductivity of the printed layers on the antenna performance. The field tests of the textile antennas in open areas showed better performance than commercial antennas. This study aims to advance E-textile devices with excellent electrical properties and mechanical durability for wearable wireless communication applications.

Binary manganese oxide/cobalt oxide (MnO_2^/Co₃O₄) composites on the well-aligned electrospun carbon nanofibers (MnO₂/Co₃O₄@ECNFs) were synthesized via a simple and economical one-step template-free electrodeposition technique with a constant low-current (60μA) for 3h, and their electrocatalytic activity towards non-enzymatic glucose sensors was evaluated in detail. The MnO_2^/Co₃O₄ nanostructures were uniformly deposited on the well-aligned ECNFs with high porosity, effective interconnectivity and large amounts of extended number of conducting channels. The electrocatalytic performances of as-prepared MnO₂/Co₃O₄@ECNFs towards the oxidation of glucose were carried out by cyclic voltammetry (CV) and chronoamperometry (CA), and the experimental results show that the MnO₂/Co₃O₄@ECNFs exhibited a superior sensing performance towards glucose including a rapid glucose response within 5 seconds, a wide range (5 μM to 10.89 mM), excellent sensitivity of 973.91 μA mM^-1 cm^-2 and a low detection limit of 0.02 μM (S/N=3) with satisfactory selectivity, great reproducibility and stability. These excellent electrochemical performances were attributed to the synergistic effect of binary metal oxides and the unique porous architecture of the electrode. These features indicate that as-synthesized MnO₂/Co₃O₄ binary nanostructure decorated well-aligned ECNFs hold great potential in the development of a commercialized glucose sensor.

Colloidal nanoparticles of organic semiconductors have attracted increasing interest in recent years due to the eco-friendly fabrication of printed electronics afforded by colloidal ink technology,¹ including solar modules. Utilising the colloidal nanoparticle approach, it is now possible to engineer the microstructure of the light absorbing/charge generating layer of organic photovoltaics; decoupling film morphology formation from film deposition.² This seminar presents recent work on core-shell nanoparticle synthesis and single-component nanoparticle synthesis for printed organic electronics applications, via a miniemulsion synthesis route.^2,3 Single-component nanoparticles of donor and acceptor materials are utilised to engineer two-phase structured films with control over domain size.³ Engineering a three-phase structured film is either via the synthesis of core-shell nanoparticles or via thermally annealing a two-phase structured film above the glass transition temperature (T_g) of the materials promoting the diffusion of the fullerene acceptor into amorphous polymer donor domains. Synchrotron-based X-ray spectromicroscopy and electron microscopy were used to characterize donor–acceptor domain size and domain composition. Donor-acceptor material systems studied span traditional polymer donor:fullerene acceptor systems and new polymer donor:non-fullerene acceptor systems, including poly(3-hexylthiophene) (P3HT) and poly[2,3-bis-(3-octyloxyphenyl)quinoxaline-5,8-diyl-alt-thiophene-2,5-diyl] (TQ1) donor materials and phenyl-C61 butyric acid methyl ester (PC₆₁BM) and poly{[N,N'-bis(2-octyldodecyl)naphthalene-1,4,5,8-bis(dicarboximide)-2,6-diyl]-alt-5,5'-(2,2'-bithiophene)} (N2200) acceptor materials. This seminar demonstrates the high level of control over donor−acceptor film morphology enabled by customizing nanoparticulate colloidal inks, where both two-phase and three-phase film architectures can be designed and engineered for new age printed electronics.⁴

1. C. Xie, et al., Nat. Commun., 2018, 9, 5335.
2. M. Marks, N.P. Holmes, et al., Phys. Chem. Chem. Phys. 2019, 21, 5705−5715.
3. N.P. Holmes, et al., Chem. Mater., 2018, 30, 6521−6531.
4. N.P. Holmes, et al., Nano Energy, 2016, 19, 495−510.

Luminescent solar concentrators (LSCs) are luminophore embedded waveguides capable of collecting sunlight across a wide area and concentrating it onto smaller cells for energy conversion. Unlike conventional solar cells, LSCs do not require solar tracking as they have omnidirectional absorption and demonstrate similar performance under diffuse and direct light conditions. However, LSCs with luminescence in the near infrared (NIR) have yet to realize their full potential. This is in part due to non-ideal luminophores; low photostability, large full width half maxes, poor quantum yields, and parasitic self-absorption are among the challenges that have plagued luminophore performance in LSCs over the past decades. This work explores the synthetic routes and post-synthetic improvement of highly emissive, NIR emissive quantum dot heterostructures. CsSnI₃ and InAs quantum dot based LSCs in particular have excellent potential in Si systems due to their narrow emission near the band edge of a Si photovoltaic cell. Multishell luminophore systems with CsSnI₃ and InAs cores are synthesized to improve luminophore quantum yield and to reduce parasitic self-absorption. The advantage of incorporation of these NIR emissive cores into shelled systems is multifaceted: (1) wider bandgap protective shells improve quantum yields through passivation of surface traps, (2) large shell to core volume ratios reduce parasitic self-absorption by separating the energies of absorbance by the shell and emission from the core, and (3) the surface of the outer shell can be post-synthetically modified to tune the luminophore’s aggregation properties and interactions with polymer waveguides without damaging the emissive properties of the core. Despite their clear potential role in LSCs, NIR emissive quantum dots have lagged behind their visible region fluorescent counterparts due to lower quantum yields and more limited precursor availability. This work develops synthetic and post-synthetic strategies for obtaining bright, near infrared emissive heterostructure quantum dots with a specific interest to incorporate them in LSCs with Si microcells.

Anisotropic emitters such as gold nanorods[1], quantum rods[2] and organic dye-doped rods[3] have been exploited as orientation probes. Their orientation analysis is based on comparing the emitted light intensities measured with different excitation and emission polarizations. However, the practical use of this technique requires overcoming several problems: (i) various sources of signal fluctuation: blinking and bleaching of emitters, unstable excitation power, instrumental noise… (ii) Polydispersity of emitter particles affecting the signal’s degree of polarization (DOP), (iii) complicated optical setup with multiple directions of excitation and emission.
Here we present a straightforward and accurate orientation analysis through a purely spectroscopic method using lanthanide luminescence. Nanocrystals doped with europium ions (Eu³⁺) exhibit unique photoluminescence (PL) spectra composed of a number of sharp peaks originating from the multiple 4f transitions and degenerate sub-levels due to the crystal-field splitting. Each of those sub-level peaks is polarized along a certain direction independent to others. Therefore, the orientation of a nanocrystal can be represented simply by the line shape of the polarized emission spectra regardless of its global intensity that can fluctuate due to other causes. This spectroscopic method enables three-dimensional (3D) orientation analysis on a single measurement plane not even taking into account the excitation polarization[4]. The polarization of Eu³⁺ emission is determined only by the crystal structure and not by the particle morphology, which makes the analysis reliable even with polydisperse particles. Moreover, the Eu³⁺ emission is highly photo-stable and free from photo-bleaching or blinking.
We used Eu³⁺-doped monocrystalline NaYF₄ nanorods as a proof of concept 3D orientation probes. The nanorods were solvothermally synthesized and randomly suspended in a polymer matrix (PVA - Polyvinylalcohol). By performing the conventional scanning confocal microscopy, polarized fluorescence spectra of single nanorods were obtained, from which the spherical angles (θ, φ) of rod orientation were calculated with high accuracy.
This new spectroscopic method to determine 3D orientation of nanoprobes offers a unique opportunity to unveil the complex dynamics of bio-molecular systems (e.g. rotating proteins) and also the ensemble behavior of nanoparticles (e.g. self-assembly).

[1] Marchuk, K. et al, Nano Letters 13, 1245-1250, (2013).
[2] M. Ohmachi et al, Proc Natl Acad Sci, 109, 5294 (2012)
[3] T.H. Besseling et al, J. Phys. Condens. Matter, 27, 194109 (2015)
[4] J. Kim et al, Nat. Nanotechnol, 12, 914 (2017)

Two-dimensional metal halide perovskite nanoparticles show great potential for use in optoelectronic devices because of blue-shifted absorption relative to their 3D counterparts, narrow emission line widths, high absorption cross sections, and high radiative recombination. Two-dimensional CsPbX₃ nanoparticles have been synthesized using hot injection methods, solid state crystallization methods, microwave-assisted methods, as well as high temperature-and-pressure methods. Herein, we show a different approach relying on rapid anion exchange between CsPbCl₃ nanocubes and a mixture of dodecanethiol (DDT) & AlX₃ (X = Cl, Br, I) to transform CsPbCl₃ nanocubes into CsPbX₃ nanoplates at room temperature. The combination of both DDT and AlX₃ is necessary to facilitate nanoplate formation and maximize the photoluminescence quantum yield, Φ_PL, and stability. Either DDT or AlX₃ addition alone will result in improved Φ_PL, but without the formation of nanoplates and with lower Φ_PL and stability than when added together. The anion exchanged and surface modified nanoplates are highly emissive and stable due to the introduction of both an anion source and surface ligand. Nanoplates synthesized through DDT-AlX₃ ligand mediated anion exchange show photoluminescence quantum yields of 65% for violet-blue emitters (CsPbCl₃), near 100% for green emitters (CsPbBr₃), and 81% for red emitters (CsPbI₃), while retaining 80-90% of their maximum Φ_PL after 15-18 days of storage in ambient atmosphere. This combined approach of simultaneously adding a halide source and surface ligand provides an attractive route to simultaneously manipulate the optical properties and nanocrystal morphology of metal halide nanocrystals.

Lithium ion batteries (LIBs) are key components of modern electronics and are regularly utilized as their primary power source. LIBs have become ubiquitous in a variety of applications ranging from portable devices to electric vehicles because of their high energy density, flexible and lightweight design, lower self-discharge, low cost and longer lifespan compared to other battery technologies. Despite these advantages, the reliability and performance of LIBs still need to be upgraded to meet the requirements of applications such as large-scale energy storage and hybrid electric vehicles. Extensive research has focused on the development of four cell materials in particular to achieve better performance: positive and negative electrode active materials as well as separators and electrolytes. Understanding the electrical and mechanical properties of electrode materials plays a major role in the performance improvement of LIBs. It has been shown that improved adhesion between electrode particles, electrode films and current collectors leads to better retention of discharge capacity during cycling, especially when electrode materials exhibit faster and/or larger volume expansion. Moreover, enhanced electronic conductivities and ionic diffusivities in electrodes also lead to LIB capacity improvements. As devices are becoming more compact, optimizing electrical and mechanical properties on a nanometer scale becomes more relevant and leads to improved interfaces.
There are several methods which can measure these local properties; the more common methods include impedance spectroscopy and nanoindentation. However, even using both methods, one cannot get the full local information about the aforementioned properties. Impedance spectroscopy needs an exact model and only distinguishes between interfaces, and therefore does not have local information for each interface. Nanoindentation is destructive and does not provide any electrical information. One of the more effective tools used to overcome the problems encountered in electrical property measurement is Scanning Spreading Resistance Microscopy (SSRM). This technique simultaneously measures both electrical properties and topography. However, SSRM has some disadvantages, such as rapid wearing of the tip, degradation of image resolution and low signal-to-noise ratio. These disadvantages stem from the high frictional force arising from continuous tip-sample contact during scanning. Recently, a new operational AFM mode developed by Park Systems called PinPoint™ can be coupled with SSRM to offer scientists and engineers an innovative solution to avoid the troublesome frictional forces during a scan. This method operates in an approach-retract manner ensuring a frictionless operation, which eliminates the lateral force from continuous tip-sample contact. In addition, the integration of these methods with AFM enables acquiring topography as well as electrical/mechanical property data simultaneously without changing the sample or tip. Here, we demonstrate that PinPoint™ SSRM effectively measures both electrical and mechanical properties of LIBs electrode surfaces at a much higher quality in high vacuum environment than in ambient condition using a Park NX-Hivac AFM system.

Most efforts to directly write conductive metal traces have focused on nanoparticle ink suspensions that require aggressive sintering (>200°C) yet exhibit low density, small grains, and a maximum electrical conductivity of ~25% of bulk silver [1]. Here, we demonstrate an alternative approach to printing high conductivity metals using a reactive ink solution, where a molecular precursor reacts into a highly dense (>90%) silver film at a low temperature (80°C) compatible with conventional additive manufacturing processes. We show that by reducing porosity and transitioning from interface-rich sintered nanoparticles to dense solid metals, we achieve DC conductivities up to ~75% of bulk silver for aerosol jet printed conductive traces, which aligns with our predictive microstructural transport model. Furthermore, we show through an initial demonstration that the RF performance of these devices far exceeds that of standard printed transmission lines and competes with conventional electroplating techniques. This work demonstrates a path toward high performance printed RF passive components that are robust and reliable for disruptive opportunities in aerospace materials and manufacturing processes.

[1] Eva S. Rosker, Rajinder Sandhu, Jimmy Hester, Mark S. Goorsky, and Jesse Tice, “Printable Materials for the Realization of High Performance RF Components: Challenges and Opportunities,” International Journal of Antennas and Propagation, vol. 2018, Article ID 9359528, 19 pages, 2018.

Ionic liquids (ILs) have gained considerable attention in the last two decades owing to their unique and tunable physico-chemical properties (e.g., low vapor pressure, and high thermal stability), which have made them potentially useful for a range of applications, including batteries, fuel cells, catalysis, and lubrication. Ionic liquids are particularly attractive in lubrication, since their properties make them suitable for components working under extreme conditions (e.g., high temperatures, low pressures), such as those found in engines, spacecraft, and micro-electromechanical systems. When ILs are used as lubricants, the interface between the IL and the surfaces of the components in relative motion plays a pivotal role in controlling the friction and wear response. Despite the scientific weight of published studies on the tribology of ILs, remarkably little is still known about the underpinning lubrication mechanism of ILs. The development of a fundamental understanding of the mechanism by which ILs reduce friction and/or wear requires shedding light on the processes occurring at nanoscale asperities within macroscale contacts. This constitute a significant challenge since observing and understanding the nanoscale mechanisms at play is inhibited by the hidden nature of the buried interface and the challenge of performing observations at the nanometer scale.
Here, we used atomic force microscopy (AFM) to visualize and quantify the processes occurring at sliding interfaces in situ, in single-asperity nanocontacts¹. In contrast to macroscale tribological experiments, where the multi-asperity nature of the contact does not allow for the accurate control of local contact stresses and geometries, in situ AFM studies, where the AFM tip mimics a single asperity in a multi-asperity macroscale contact, enable the precise control of critical parameters (e.g., contact stress) affecting mechano-chemical reactions, the visualization and quantification of phenomena occurring at tip/substrate interfaces (e.g., mechanically-induced nucleation and growth of films), and the local characterization of nanoscale properties (e.g., friction and adhesion). The AFM experiments, in which a diamond tip was slid on steel in phosphonium phosphate IL (PP-IL) at high contact pressure (>5 GPa), indicated a significant friction reduction only after the removal of the native surface oxide from steel. Even though the AFM experiments allowed for the identification of changes in topography and friction in situ while sliding in PP-IL, they could not provide any information about the composition and structure of the regions scanned by AFM. The analysis of these regions is a challenging surface science problem owing to their limited dimensions and the small thickness of the surface material modified by the mechanical action of AFM tips. To address this challenge and elucidate the origin of the friction reduction observed during AFM experiments, laterally-resolved ex situ analyses of the surface chemistry of steel were performed by synchrotron-based X-ray photoemission electron microscopy (X-PEEM), low energy electron microscopy (LEEM), and time-of-flight secondary ion mass spectrometry (ToF-SIMS). The analytical results indicated that the mechanically-induced exposure of metallic iron during AFM tests carried out in PP-IL leads to an increase in surface coverage of adsorbed phosphate anions together with a change in surface potential. Based on these results and atomistic simulations of the configuration geometry of phosphate ions on metallic and oxidized iron², a simple phenomenological model is proposed to account for the observed lubrication behavior.

(1) Gosvami, N. N.; Bares, J. A.; Mangolini, F.; Konicek, A. R.; Yablon, D. G.; Carpick, R. W. Science 2015, 348, 102-106.
(2) Ta, H. T. T.; Tieu, A. K.; Zhu, H.; Yu, H.; Ta, T. D.; Wan, S.; Tran, N. V.; Le, H. M. The Journal of Physical Chemistry C 2017, 122, 635-647.

Conducting polymers are an important class of Faradaic materials for energy storage, as they present opportunities to create energy dense supercapacitors. The vast majority of conducting polymers are p-type cathodes and already operating at or close to the upper voltage limit of non-aqueous electrolytes. On the other hand, it is generally safe for electrolytes to operate down to -2 V vs Standard Hydrogen Electrode. Therefore, extending the operational voltage in the negative range is an attractive route to increase energy density. Here we present promising n-type polymers that retained 90% of initial capacitance after 5000 charge-discharge cycles. The polymers are processed with solution based methods to ensure that the devices are scalable to large area. The electronic and ionic transport in the conducting polymers are monitored in operando by current-voltage and spectroscopic measurements, where spatial variations of film absorption would be mapped to discern electronic versus ionic degradation. We infer the mechanisms that lead to capacitance fade and suggests structural and electrochemical strategies to realize high-endurance energy storage devices for flexible printed systems.

In addition to energy storage, the narrow bandgap polymers are used to realize photodetectors that is relevant to a variety of applications including environmental monitoring and medical diagnosis. Here we show how the infrared polymers are processed into bulk heterojunction photodiodes with photoresponse up to wavelength of 1.7 micron. The performances of devices with different polymer structures were compared through metrics including detectivity, quantum efficiency, response time and rectification ratio. Example applications including hemodynamics and tissue composition measurements will be demonstrated.

Semiconducting metal oxides, particularly transparent conducting oxides, have attracted a great deal of attention for a variety of applications. Nevertheless, the solution chemistry and solution functionalization of these materials has been much less explored than that of traditional semiconductors, such as silicon.
One challenge in the stable functionalization of these materials is that their semiconducting nature enables efficient generation of long-lived photocarriers that can initiate chemical reactions that degrade the functionalization. This reactivity has typically been attributed to the undercoordinated metal cation sites at the surface, although direct evidence has been lacking. We have developed a new photochemical reaction that leads to the fluorination of all exposed metal cation sites on TiO₂. In contrast to approaches that use fluoride or HF-containing solutions, our method produces covalently bonded fluorine which is stable in aqueous solutions. To our surprise, these fluorine-terminated TiO₂ surfaces display higher photoreactivity than comparable “clean” surfaces. Possible origins of this behavior will be discussed.
We have also investigated the structure of solution-deposited monolayers on TiO₂ and demonstrated a rational approach to tuning intermolecular interactions and enabling long-range ordering. We show that simple electrostatic insights can be used to engineer away unfavorable intermolecular interactions, producing monolayers with exceptional long-range ordering. Quantitative measurements of the intermolecular interaction energies from molecularly resolved STM images are a factor of ~7 larger than those predicted by dispersion-corrected density functional theory (DFT). This finding suggests a new path to the production of highly ordered monolayers and superstructures of large molecules.

An interdigitated geometry of the interface between semiconductors remains a Graal in the photovoltaic field. Nanocylindrical solar cells based on a coaxial p-n or p-i-n junction have been generated in a small number of materials systems only so far, mostly among the classical semiconductors. The generalization of this interface geometry to alternative semiconductors, which are expected to profit of it most, has been hampered by the lack of appropriate deposition methods suited to coating complex three-dimensional structures in a conformal manner. Atomic layer deposition (ALD) provides this capability, but is limited by the necessity to use gaseous precursors in vacuum conditions. Modern classes of semiconductors such as the hybrid perovskites are not accessible by ALD due to their ionic nature.
In this perspective, we have transferred the principles of ALD to precursors dissolved in liquid solvents. The ‘solution ALD’ (sALD) method can be implemented in a variety of microfluidic or slot-die processing devices. We have demonstrated that the principles of classical, gas-based ALD —self-limiting surface reactions— are reproduced in sALD. Furthermore, sALD allows the experimentalist to design new, advantageous, reactions of known semiconductors (oxides, heavier chalcogenides), or to deposit materials otherwise inaccessible to ALD altogether, such as polymers, ionic solids, and metal-organic frameworks. They may be obtained in highly pure, crystalline form at room temperature, either as planar films or as conformal coatings of porous substrates. Based on these advances, we have been able to make the first sALD-derived hybrid perovskite cells.

Metallic inks with superior conductivity and printability are necessary for high-throughput manufacturing of printed electronics. In particular, gallium-based liquid metal inks have shown great potential in creating soft, flexible and stretchable electronics. Despite their metallic composition, as-printed liquid metal nanoparticle films are non-conductive due to the surrounding metal oxide shells which are primarily Ga2O3, a wide-bandgap semiconductor. Hence, these films require a sintering process to recover their conductivity. For conventional solid metallic nanoparticles, thermal and laser processing are two commonly used sintering methods, and the sintering mechanism is well understood. Nevertheless, laser sintering of liquid metal nanoparticles was only recently demonstrated, and to date, the effect of thermal sintering has rarely been investigated. In this talk, eutectic gallium–indium nanoparticle films processed separately by laser or thermal sintering in an ambient environment will be presented. Laser and thermally sintered films will be compared with respect to electrical conductivity, elemental composition, crystallinity and surface morphology. Both methods impart thermal energy onto the films and generate thermal stress in the particles, resulting in rupture of the gallium oxide shells and achieving electrical conductivity across the film. For laser sintering, extensive oxide rupture allows liquid metal cores to flow out and coalesce into conductive pathways. For thermal sintering, due to less thermal stress and more oxidation, the oxide shells only rupture locally and extensive phase segregation occurs, leading to non-liquid particle films at room temperature. Electrical conductivity is instead attributed to segregated metal layers and gallium oxide which becomes crystalline and conductive at high temperatures. This comprehensive comparison confirms the necessity of oxidation suppression and significant thermal stress via instantaneous laser irradiation to achieve conductive liquid metal patterns.

The rapid growth of technologies such as machine learning and autonomous vehicles is fueling a need for more powerful electronic devices requiring complex nanoscale structures. With the growing demands on nanostructure fabrication, selective deposition is gaining attention as a process to enable pattern features at the sub-10 nm length scale. This talk will focus on area selective atomic layer deposition (AS-ALD), a bottom-up growth strategy in which material is added only where desired, without the need for subsequent lithography and etch steps. ALD is a good choice for selective deposition because its chemical specificity provides a means to achieve selectivity on a spatially patterned substrate. One popular approach to AS-ALD uses inhibitory layers such as self-assembled monolayers (SAMs) to alter the native surface reactivity, allowing ALD selectivity to be achieved. There have been a number of studies demonstrating that different types of SAMs, including organosilanes, phosphonic acids and thiols, can successfully block ALD processes. However, the efficacy of the blocking varies strongly across different ALD processes. For example, for the same SAM system, the blocking selectivity of Al₂O₃ is limited to only ~6 nm whereas ZnO can be blocked for over ~30 nm. The difference in blocking highlights the importance of precursor chemistry for AS-ALD. In this work, we describe our studies to elucidate the mechanism of AS-ALD based on a comparative study of ALD precursors. Several precursors are studied, including trichloroaluminum (TCA), dimethylaluminum chloride (DMACl), trimethylaluminum (TMA), and triethylaluminum (TEA), each used for Al₂O₃ ALD, allowing comparison between polarity and steric effects of the precursors. Results of quantum chemical calculations of the reaction pathways show product energetics that are correlated with experimental observations, and Monte Carlo simulations address the role of sterics on surface coverage. AS-ALD of Al₂O₃ using octadecyltrichlorosilane SAMs as an inhibitor is also investigated; the blocking properties of the ALD precursors will be compared and the results discussed based on the growth mechanism. By pursuing first principles design of selective ALD processes, this work may enable new methods for additive nanoscale patterning.

The atomic layer deposition of cobalt using Co(DAD)₂ and tertiary-butyl amine (TBA) has nearly infinite selectivity (>1000 cycles) on metal vs. insulator (SiO₂ or low-k SiCOH) planar samples. However, on patterned samples, selectivity under identical ALD conditions is limited, due to the diffusion of molecularly-adsorbed metal precursor from reactive to non-reactive surfaces. Three strategies have been found to improve Co ALD selectivity: adding a passivant to remove insulator defect sites, increasing the purge time, decreasing the precursor dose, and periodic annealing. The periodic annealing technique allows reabsorption of the Co nuclei from the insulator surface to the growth surface and is consistent with a low temperature reflow process. This technique has the added advantage of allowing a lower temperature for reflow, potentially allowing a scaling of the diffusion barrier between the Co and the SiCOH which is normally employed. Co ALD was performed using Co(DAD)₂ + TBA at 180°C on 85 nm wide Cu stripes on SiO₂. The planar structure of these stripes is used to demonstrate the effectiveness of passivation, as top-down SEM imagery and XPS quantification can be used to monitor growth and presence of unwanted Co nuclei on insulator. To control precursor dose, multiple precursor pulses were employed in each cycle to limit the maximum pressure. XPS is performed without breaking vacuum to prevent oxidation of Co. This work is supported by the Applications and Systems Driven Center for Energy-Efficient Integrated Nanotechnologies (ASCENT), a Semiconductor Research Corporation program sponsored by JUMP and DARPA.

From thin film solar cells to metal-oxide-semiconductor devices in leading edge integrated circuits, the electronic structure at and near the interfaces between component materials determines the most important fundamental operating characteristics of those devices such as turn-on voltage, power dissipation and off-state current leakage. Fermi level location at buried interfaces, semiconductor band-bending, charge transfer, oxide defects and work functions of the constituent materials all contribute to device performance. In this talk I will describe how these important parameters can be determined by employing femtosecond photovoltage spectroscopy (FPS)[1], an extension of ultraviolet photoelectron spectroscopy (UPS) using ultrafast lasers. While UPS is fundamentally a surface sensitive spectroscopy, I will describe how pump/probe techniques add a new dimension to this venerable spectroscopy, permitting the accurate extraction of the underlying band bending in Si. When combined with valence band edge location of the semiconductor and oxide, and determination of the system Fermi level, full characterization of the electronic structure of an MOS stack can be obtained providing key insights on device operating properties. I will describe the use of FPS to investigate device stacks ranging from Si and III-V based MOS structures to thin film solar cells formed from earth-abundant elements. In each case surprising new details were uncovered that led to performance optimization of these technologically important devices.
[1] Richard Haight and Adra V. Carr, "Industrial Applications of Ultrafast Lasers", World Scientific Press, 2018, ISBN:97898145690002

For a long time, the etching technique played an important role in the electronics industry, and had a critical influence on the development of modern manufacturing technologies. Regarding to the operation of electronics devices, surfaces and interfaces significantly affect the functionality and performance of the devices. To investigate the formation mechanisms for characteristic surface morphologies during chemical etching, in this study an anisotropic chemical etching model was established and the formation and evolution of surface structures with resist-masks were simulated numerically. In this theoretical model, two major crystallographic orientations in a cubic system, such as <100> and <111> directions, were taken into consideration and included in the reactions of surface diffusion and anisotropic etching. By adjusting the numerical parameters corresponded to the different experimental conditions, various surface profiles could be created and the influence of the experimental parameters on the surface morphologies would be revealed. From the calculations of anisotropic etching along <100> preferred-directions, a V-shape profile with faceted surface morphologies appeared. A larger loss of linewidth or a higher etch bias was displayed with the increase of the etching rate. While an anisotropic etching along <111> preferred-direction was examined, etch pits were firstly produced in the early stage of etching process. A different surface evolution during a chemical etching was distinctly demonstrated due to the crystallographic structure. In addition, since the resist-mask was included in the numerical model, the phenomenon of undercutting in an anisotropic chemical etching could be also discussed in this study. These numerical simulations could provide a fundamental information to understand the formation mechanism of etched surface, and broaden the applications of chemical etching in the advanced manufacturing technologies.

Metal halide perovskite nanocrystals (NCs) are attractive materials for optoelectronics because of their optical and electronic properties, such as tunable direct band gaps, high absorption coefficients, low exciton binding energies, relatively high electron and hole mobilities, narrow emission line-widths, and high photoluminescence quantum yields (Φ_PL). However, further improvements in stability and reproducibility in Φ_PL of NCs are essential for enabling future optoelectronic applications. Inadequate surface passivation due to labile parent ligands, such as oleate and oleylammonium ions is a major cause of instability and less than unity Φ_PL. Herein, we probe the influence of multiple molecular and ionic ligand binding groups including thiols, thioethers, phosphine, phosphine oxide, sulfobetaine, thiocyanate, and sulfate salts on the colloidal stability and Φ_PL of CsPbBr₃ NCs. Among all the binding ligands, post-synthetic treatment with dodecanethiol reproducibly yields highly stable NCs with near unity Φ_PL for a range of synthetic conditions and initial Φ_PL of the as-synthesized NCs. A few of the other ligands can also increase the Φ_PL, but in ambient atmosphere they are unable to maintain stability more than 5 days. A mechanistic investigation shows that thiol addition to the as-synthesized NC dispersion leads to the thiol-ene reaction with octadecene, oleic acid, & oleylamine and formation of thiolate ions and thioethers. Both thiolates and thioethers are suspected to bind to undercoordinated Pb atoms and consequently eliminate metallic Pb particles commonly observed with transmission electron microscopy on the NC surfaces. The process of enhancing PL intensity and stability can be rapidly accelerated through exposure to blue or UV light.

The use of halogen chemistries on silicon surfaces offers a promising path towards atomic-scale manufacturing of acceptor-based devices. Atomically-precise fabrication of electronic devices in silicon has been widely demonstrated using a scanning tunneling microscope (STM) and hydrogen-based surface chemistries to precisely place phosphorous donor atoms on the surface. A lithographically patterned hydrogen resist is used to selectively incorporate PH₃ to produce metallic wires, quantum dots defined by electrostatic gates, and single donor atom qubits for quantum information (QI) research. However, interest in acceptor dopants and hole-based devices necessitates the development of alternate precursors and/or resist chemistries for device fabrication. To incorporate acceptors into silicon with atomic placement, halogen chemistry is potentially more favorable than hydrogen. To that end, we explore halogen-based resists including chlorine, bromine, and iodine in conjunction with metal-halide precursors to achieve area-selective deposition of acceptor dopants in silicon. Here, we present results on the complete surface passivation of Si(100) using Cl, Br, and I. We study the stability of halogen-terminated Si(100) surfaces in ambient environments using STM and x-ray photoelectron spectroscopy (XPS) to facilitate transport of samples outside of ultra-high vacuum environments. Finally, we present STM and secondary ion mass spectroscopy (SIMS) results on the adsorption and incorporation of metal-halide precursors on Si(100) as a viable path towards acceptor doping.

Polymer-dispersed liquid crystals (PDLCs), a significant class of materials have diverse applications in optics and electronics. Highly efficient and systematic buildings can be made utilizing the PDLC based switchable glazing as they exhibit switching behavior from opaque to transparent mode in presence of power supply, thus saving the energy and elucidating the privacy issues. In this work we prepared typical PDLC mixture using commercially prepared polymer NOA 65 and liquid crystals (like C7 and E7). The mixture was transferred on ITO glass using the wire bar coater and the cell was fabricated by coupling the glass in vacuum. No film was used between the glass as in typical roll-to-roll process, thus enhancing the transmittance. Via vacuum coupling, PDLC based switchable glazing can be fabricated on glass substrates instead of plastic polymers which degrade with the passage of time reducing the efficiency. The cells were then cured by irradiating UV light (intensity ~ 95.4 mWcm^-2) leading to polymerization and polymer induced phase separation process. As a result, the liquid crystal forms droplets segregating from polymer matrix. The cells exhibited a switching behavior characterized using oscilloscope and AC voltage driver at different voltages. The transmittance change of ~ 60% - 70% was measured with a driving voltage of 60 ~ 70 V. Thus, by using the vacuum coupling process we can fabricate PDLC based switchable glazing on large scale with maximum transmittance and enhanced stability as it is expected a green refurbishing material of the next generation.

Tin sulfide (SnS) is a desirable absorber layer for thin film solar cells. Its optical bandgap of 1.1 eV, absorption coefficient of 10⁵ cm^-1 in the visible region, inherent p-type electrical conductivity, non-toxicity, earth abundance of the constituent elements, and stability in humid and oxygen ambient are all attractive toward developing alternative solar cell technology. Nevertheless, solar cells of SnS have reached a photovoltaic conversion efficiency of only 4.8 %. The presence of secondary phases (SnS₂, Sn₂S₃) and small grains is among the efficiency limiting factors. Here, we present methods to enhance the quality of chemically deposited SnS films through using thin layers of Sn and SnCl₂ as flux materials produced by thermal evaporation and/or sulfur vapor to play a key role in grain growth during heating at 450 °C, while conserving its p-type electrical conductivity.
We prepared SnS thin films of 300 nm in thickness in an autoclave at 2.5 atmospheres and 55 °C with duration of deposition of 6 h. The deposition solution contained tin chloride and sodium thiosulfate. The as-deposited SnS thin films were of orthorhombic crystal structure, with an optical bandgap of 1.33 eV (indirect) and electrical conductivity of 1×10^-4 Ω^-1 cm^-1 in the dark and 1×10^-3 Ω^-1 cm^-1 under illumination. The films presented compact surface and dense cross-sectional morphology. We heated this film for 30 min, in a tubular oven in nitrogen atmosphere at 450 °C. Different flux materials were added during the heating to promote grain growth: sulfur, evaporated thin films of Sn metal, or SnCl₂. The film heated with sulfur powder was p-type, with electrical conductivity, 10 ^{– 4} Ω^{- 1} cm ^{– 1}. SnS films heated under the same conditions using Sn as well as SnCl₂ as flux materials retained the single phase of SnS with the formation of well-defined columnar grains of size equivalent to film thickness. They have a bandgap of 1.1 eV (indirect) and p-type electrical conductivity, 10^-3 Ω^-1 cm^-1 in the dark, which increases to 1.5 times, under illumination. Thus, heating SnS absorber films deposited at a low temperature using chemical methods, with Sn or SnCl₂ as flux materials is a promising methodology to achieve grain growth. We shall illustrate the effect of this in solar cells using SnS as absorbers.

Color image sensing by a smartphone or digital camera is usually performed by an array of sensor elements covered by color filters. The optical color filters cause color errors like the color aliasing or Color Moiré effect. Furthermore, the quantum efficiency of typical conventional optical color sensor consisted of side-by-side arranged color filters is limited to 1/N, where N is the number of color channels per pixels. Hence, the efficiency of the color detection process is limited to max. 33% by the side-by-side arrangement of sensor channels. In this study, color sensors consist of vertically stacked diodes using perovskite alloys is described with maximum quantum efficiency approaching 100%. The described sensor structure is color aliasing or Color Moiré error free, while conventional sensors using optical filters are limited by this error. The optical constants of the required perovskite alloys were determined by applying the energy shift model to experimentally determined optical constants of pure perovskites. The spectral sensitivities of the color sensors were determined by electromagnetically modelling. The spectral sensitivities of the sensors were colorimetrically characterized and compared to sensors in literature including conventional sensors using optical filters. Up to our knowledge for the first time it could be shown that a vertically stacked three color sensor exhibits a color error equal to, or smaller than errors of conventional sensors using optical filters as well as silicon based Foveon stacked color sensor. Details on the used materials, the device design, and the colorimetric analysis are provided.

For high switching speed HV Schottky diodes, with very high work function metal and extremely lightly doped epi, the built-in potential may be too high for thermionic emission to occur, when the applied external voltage is quite low (near the forward voltage: V_F = 0.07V). If the epi is lightly doped p type, the built-in potential (V_Built-in: potential difference between the metal and silicon Fermi levels) is 1.0V (measured with CV). If the external bias is 0.1V, near the measured V_F, it is not enough to overcome the built-in potential for thermionic emission as illustrated in the Figures. It is likely that in addition to thermionic emission, tunneling and diffusion currents also contribute to the total HV Schottky diode forward current. TCAD simulation of HV Schottky diodes with N⁺ guard bands suggests the potential barrier and electric fields at the Schottky junction are relatively high for thermionic emission to occur, when external bias V » V_F.

In the Advanced Technology Laboratory (ATL) at Northrop Grumman Corporation, high performance RF ASICs and power devices in silicon, GaAs, GaN and 4H SiC are simulated, designed, fabricated, and packaged. Extensive DC and RF tests are conducted. The performance of HV Schottky diode is superior to conventional p-n junction diode due to higher BV, lower reverse leakage, lower V_F and higher output current. We have fabricated HV Schottky Diodes. BV > 150V. V_F = 0.07V. High breakdown voltage, low forward voltage drop, and low reverse leakage are achieved with very high work function metal alloys and low doping epi films.

Lightly doped p (100) or n (100) type silicon films are deposited with epitaxy. Metal or metal alloys are deposited with sputtering. Figure 1 shows the measured I-V’s for the Schottky Diodes. Avalanche breakdown voltage > 150V is recorded with low reverse leakage current. The voltage required to turn the diode on is defined as V_F – when the current reaches 0.1uA.

Forward bias characteristics of Ti, Au, and NiCr are measured. These data indicate that in order to achieve low V_F and high current, very high work function metals are needed.

High frequency CV tests are conducted using the HV Schottky Diodes. These CV data are used to extract the doping profiles .

Metal work functons are extracted with the 1/C² vs. V plots. Interception with the x axis is DV, which is the potential difference between metal and silicon Fermi levels, or built-in potential V_Built-in. Metal work function is extracted with DV. One of the Figures shows the measured doping profiles and 1/C² vs. V plots. From the measured I-V and C-V data we may calcuate the avalanche breakdown critical electric fields and metal work functions. A summary of HV Schottky diodes fabricated with NiCr, Ti, Au in Si n and p epi films will be presented.

Energy band diagram is analyzed for HV Schottky diodes: NiCr in p epi (external bias = 0V) V_Built-in = 1.00V.

Another energy band is analyzed for HV Schottky diodes: NiCr in n epi (external bias = +0.50V). V_Built-in = 1.56V.

In conclusion: we have fabricated HV Schottky Diodes. BV > 150V. V_F = 0.07V. High breakdown voltage, low forward voltage drop, and low reverse leakage are achieved with very high work function metal alloys and low doping epi films. Diffusion and tunneling currents likely contribute to the total Schottky junction current.

Amorphous oxide semiconductor (AOS) based thin-film transistors (TFTs) have been rapidly evolved from academic curiosity to the display industry because of their many advantages such as low off-current, high uniformity over large area fabrication, and high visible light transparency compared to conventional amorphous silicon (Si) TFTs. Despite these advantages, the poor electrical performances such as low field-effect mobility compared to low-temperature polycrystalline Si TFTs have been an issue for high resolution and high frame rate displays. Thus, a wide range of research has been conducted to enhance the field-effect mobility in AOS TFTs: doping a nano-conductor in the back channel, adoption of high-k gate insulators, and capping reduction layer. Among such researches, one of the most prominent methods was applying heterostructure with different oxide semiconductors to the channel layers of TFTs; due to large on-current and high field-effect mobility. The large difference in the work function of oxide semiconductor layers results in bandgap mismatches, forming potential wells. Therefore, electrons are confined in the potential well forming a 2-dimensional electron gas (2DEG). However, to apply 2DEG-based TFTs to the actual display industry, accurate 2DEG control and formation mechanisms need to be further studied.
In this paper, high performance oxide TFTs were fabricated by sodium hydrosulfite (Na₂S₂O₄) reducing agent treated indium-zinc-oxide (r-IZO) as the front channel of zinc-tin-oxide (ZTO)/r-IZO heterostructure. Front channel IZO film was deposited by RF sputtering at room temperature. For reducing the IZO film, the IZO film was dipped into Na₂S₂O₄ solution. Applying the Na₂S₂O₄ reducing agent treatment on the front channel allows controlling the amount of confined electron at the interface between the front and the back channels for heterostructure oxide TFTs. For the back channel, ZTO film was deposited by RF sputtering at room temperature. ZTO/r-IZO heterostructure is suitable for the formation of 2DEG due to the large difference in work function, so it could produce high field-effect mobility even when fabricated at room temperature. Also, the large on-current could be achieved by precisely controlling the formation of 2DEG through Na₂S₂O₄ reducing agent treatment on the IZO front channel. To confirm the formation of 2DEG at the interface between the ZTO film and the r-IZO film, we performed a UV-vis spectrophotometer and ultraviolet photoelectron spectroscopy (UPS) measurements. The 2DEG was successfully formed by showing the difference of 170 meV work function at the heterointerface. However, conventional 2DEG-based oxide TFTs have high off-current because 2DEG free electrons remained at the heterointerface regions even when the TFT is in the off-state. To solve this problem, the ZTO back channel was applied to different oxygen partial pressures to the lower ZTO layer (20%) and the upper ZTO layer (30%) when deposited via sputtering. As a result, the high off-current could be adjusted to near the 10^-12 A range and ZTO/r-IZO TFTs exhibit field-effect mobility of 29.55 cm²/Vs, a subthreshold swing of 0.48 V/decade, and an on/off current ratio of 5.02×10⁸.
The results show that room temperature fabricated ZTO/r-IZO heterostructure TFTs can be used as a general route for high field-effect mobility TFTs and controllable oxide semiconductor devices for offering compatibility of the flexible and high-end display technology.

Electron paramagnetic resonance (EPR) techniques provide unrivaled analytical power for identifying the physical and chemical nature of defects in semiconductors. In conventional EPR, a sample is placed within a large slowly varying magnetic field while simultaneously exposed to an oscillating magnetic field. For paramagnetic defects, the large magnetic field splits the energy of the electron spin quantum numbers. When the energy of the oscillating magnetic field is equal to the difference of the electron energy levels, electron spins are “flipped”. For an unpaired electron unaffected by its surroundings, this resonance condition is [1] hv=g_eμ_BB where h is Planck’s constant, v is the frequency of the oscillating field, g_e is the free electron g, μ_B is the Bohr magneton, and B is the magnitude of the magnetic field at resonance. The resonance condition is primarily modified by spin orbit coupling and the effects of nearby magnetic nuclei. Electrically detected magnetic resonance (EDMR) is a detection technique which improves upon the sensitivity of conventional EPR. In EDMR, device current is measured as a function of magnetic field. When the resonance condition is met, the current changes due to either spin dependent recombination (SDR) or spin dependent trap assisted tunneling. EDMR can be applied to fully processed nanoscale devices.
Electron-nuclear double resonance (ENDOR) is a double resonance technique which combines EPR with nuclear magnetic resonance (NMR) at nearby magnetic nuclei. For a conventional ENDOR measurement, the EPR response of a sample is first measured. Following this measurement, the magnetic field is set and held constant at various points across the EPR spectrum, generally at the field of maximum microwave absorption. A third radio frequency field is then swept. At frequencies where magnetic nuclei experience NMR transitions the EPR amplitude changes. Every magnetic nucleus has a unique nuclear frequency, which allows for the direct identification of these nearby magnetic nuclei.
In this study, we demonstrate (to the best of our knowledge) the first measurements of electrically detected ENDOR (EDENDOR) on fully processed devices. The main results are presented on 4H-SiC BJTs. These EDENDOR measurements are made with the base-emitter junction forward biased while the base-collector junction is grounded with current measured through the base. For a range of forward biases below the built-in voltage for the base-emitter junction, an EDMR response is observed. This response is an isotropic central line with linewidth of 3.4 G and a zero-crossing g=2.003. The EDMR response is very similar to that previously been linked to silicon vacancies in 4H-SiC MOSFETs. The response is confirmed to be the result of SDR within the depletion region of the junction through a measurement of amplitude versus forward bias.
Following the EDMR measurements on the BJT, we hold the magnetic field constant and sweep another frequency from nearly DC up to 18 MHz. A large change in device current is measured when the magnetic field is held at the maximum EDMR response, centered at a frequency of 1.04 MHz. No response is measured when the magnetic field is shifted away from the EDMR response. This nuclear frequency corresponds to ¹⁴N, which is the n-type dopant in these 4H-SiC BJTs. A study of EDENDOR amplitude versus magnetic field confirms that we are measuring the electrical detection of ENDOR, as the response is a strong function of the constant magnetic field, falling off rapidly as the magnetic field is shifted away from the EDMR response location.
We believe that this new technique, EDENDOR, can be applied to the identification of paramagnetic defects within a wide variety of solid-state electronic devices.
[1] J. A. Weil, J. R. Bolton, and J. E. Wertz, Electron Paramagnetic Resonance: Elementary Theory and Practical Applications (John Wiley & Sons, Inc., New York, 1994).

Owing to the property of low cost and easy-to-fabricate property, organic/inorganic composite devices have attracted more and more attentions in OLED, OFET, organic solar cell and so on. However, the wettability of pristine silicon is poor and liquid cannot spread on these surfaces which leads to the growth of heterogeneous crystal film with defects and results in low efficiency. In this case, this work proposed the under-liquid film growth method on superlyophilic silicon surfaces which takes the advantage of the ultra-thin film formed on the superlyophilic silicon surface as a confined template. Compared with the epitaxial growth method in vacuum, this method can be carried out in atmospheric pressure by controlling the precipitation process between the solvent and the surrounding liquid to grown large-area crystal film which will greatly reduce the cost and is significant for the industry of organic/inorganic composite devices.

Infrared detector technology has shown remarkable progress towards the selection of detector materials and device architectures. Infrared detectors are an integral part of various surveillance devices, remote sensing system, night vision, all weather surveillance, missile tracking, thermography, enhanced vision systems for aeroplanes, automobiles and gas sensing.
Current InSb and HgCdTe based IR technologies have shown high performance because of excellent material properties, covering near infrared (NIR 0.75-1.1mm), short-wavelength infrared (SWIR 1-3mm), mid-wavelength IR (MWIR 3-5mm), long-wavelength IR (LWIR 8-14mm) and very long-wavelength IR (VLWIR 14-30 mm) range. However, these need to be cooled to cryogenic temperatures making them bulky and power hungry. Thus future infrared photodetector systems must possess reduced cooling requirements, consume less power, exhibit longer lifetimes, and have improved manufacturability. These targets have been partly addressed by the development of nBn detector architecture raising the operation temperature , reliability, higher sensitivity and low SWaP (Small Weight and Power ) .
Type2Superlattice (T2SL) detectors combine key advantages of InSb and Mercury-CadmiumTelluride (MCT)-based technologies namely scalability of focal plane arrays (FPAs), high sensitivity, lower tunneling currents and suppressed Auger recombination rates nad also facilitates a higher operating temperature. A super lattice is made up of a repeated series of very thin epilayers of different materials that leads to creation of sub-bands. The corresponding bandgap energies of these subbands determine the cut-off wavelength of the sensor. The III-V InAs –GaSb materials, based on nBn unipolar photodetector architecture, with an n-type absorption layer, a barrier layer, and an n-type contact layer have been fabricated and engineered for bandgap tunability. The barrier layer is constrained to have both a negligible valence band energy offset with the absorption layer, that suppresses the Shockley-Read-Hall (SRH) generation current thereby decreasing the primary noise In addition, detector is passivated to minimise the surface leakage current The nBn photodetector, through the near elimination of SRH and surface leakage currents, requires less cooling to operate optimally than the p-n junction photodiodes. Bandgap engineering and optimization of the material composition to achieve maximum quantum efficiency, low dark current, maximum absorption strength and higher operating temperature.
Appropriate tools suitable for the design of T2SL material systems are not yet fully developed due to the intricacies of T2SL material system and paucity of experimental data for input and for the validation of mathematical formulations. A holistic approach has been followed involving design of graded energy-gap T2SL structure, growth of material system, characterization of material, fabrication and characterization. Our design is aimed to grow T2SL epilayers with optimized thickness, material composition, energy gap grading and varied doping concentrations by Molecular Beam Epitaxy and detector fabrication with a combination of wet and dry etching. A favorable impact ionization coefficient leads to higher bandwidth range with low excess noise.to achieve maximum quantum efficiency, low dark current, high avalanche gain, low excess noise, maximum absorption strength and higher operating temperature. Design optimisation has resulted in high quantum efficiency of 75% validated with FTIR measurements and low NETD of 18mK at an elevated operation temperature of 140K. Current-Voltage measurements, quantum efficiency, noise equivalent temperature difference (NETD) and issues related with growth of graded energy-gap T2SL material system will form part of the full paper.

Organic thin-film transistors (TFTs) are being developed for portable and wearable electronic applications to be powered by small batteries or solar cells, making low-voltage operation an important prerequisite. Hybrid gate dielectrics composed of a thin metal oxide layer and an organic self-assembled monolayer, are useful for low-voltage organic TFTs by providing a large dielectric capacitance while minimizing undesirable gate leakage and improving overall TFT performance. We have fabricated organic TFTs in which the gate oxide is an ultrathin aluminum oxide (AlO_x) layer prepared by oxidizing the surface of an aluminum gate electrode (deposited by thermal evaporation in vacuum and patterned using a shadow mask) in an RF-generated oxygen plasma. One advantage of forming the gate oxide by plasma oxidation, rather than by physical or chemical vapor deposition, is that the oxide forms a natural, high-quality interface with the gate metal, reducing the possibility of defects at the interface. Another advantage is that the oxide is formed only where it is required for TFT operation, thus eliminating the need for subtractive patterning. When prepared with a typical set of plasma parameters (oxygen flow rate = 30 sccm, oxygen partial pressure = 0.01 mbar, RF power = 200 W, duration = 30 s), the AlO_x layer has a capacitance of 1 µF/cm² and when used as a dielectric in organic TFTs, a gate leakage current of the order of only 10 pA. However, the exact properties of our oxygen-plasma-grown AlO_x, such as the thickness and stoichiometry, are currently unknown. We have therefore investigated our plasma-grown AlO_x by transmission electron microscopy (TEM), and these measurements reproducibly indicate the AlO_x to have a thickness of 5.8 nm, which (taking into account the measured capacitance) yields a permittivity of 9, similar to aluminum oxide deposited by physical vapor deposition. Furthermore, we have used atom probe tomography (APT) to investigate the stoichiometry and composition profile of our plasma-grown AlO_x layers. APT is a combination of pulsed field evaporation of surface atoms and their chemical identification using time-of-flight mass spectrometry. Specimens were prepared in the form of sharply pointed needles with a tip radius below 100 nm using a focused ion beam process. In an APT experiment, layers of atoms are sequentially removed from the sharpest point of the tip by thermally activated and time-controlled field evaporation events initiated by irradiation with an ultrafast pulsed laser and collected at a position-sensitive multichannel plate detector. From the data collected by the detector, a three-dimensional representation of the original atomic positions in the sample was digitally rebuilt, providing information about the stoichiometry of the material with atomic-scale resolution. In this work, we present results about the composition of the plasma-grown AlO_x layers and correlate it to its electronic properties as a gate oxide in organic TFTs.

Since the discovery of graphene, fabrication of atomically thin two dimensional (2D) materials with nontrivial band topology has attracted enormous attention primarily because of their dissipationless conduction. This led to the discovery of a family of 2D quantum materials with exotic properties. In particular, the prediction of stanene, a buckled tin honeycomb layer, to be a quantum spin Hall insulator with Dirac cone-like linear energy dispersion and a large gap of 0.3 eV has stirred up efforts to realize it on substrates. Here, we have studied the growth of Sn on Au(111) under different conditions and demonstrate presence of linear Dirac-like bands crossing the Fermi level at the zone center with Fermi velocity comparable to graphene from angle resolved photoemission spectroscopy. The Dirac-like bands occur only in a specific Au-Sn surface phase with unit cell of oblique symmetry (γ =70.9° and b:a= √7/√3) that is formed at high temperature, but is stable when cooled to RT. Form X-ray photoelectron spectroscopy (XPS), we have determined the composition of this phase to be Au₂Sn. We have also studied the evolution of the Au-Sn composition with thickness and obtained other phases and surface structures. The layer, the interface, and the substrate related components in the Au 4f and Sn 4d core-level spectra obtained using XPS vary with deposition parameters. AuSn is found to form by deposition at room temperature. On the other hand, Au rich compounds form as a result of heat treatment through interdiffusion of Au and Sn. Deposition at high temperature forms more Au rich compositions compared to postannealing at the same temperature due to the kinetic energy of the impinging Sn atoms in the former case. The XPS valence-band spectra of AuSn and Au₅Sn layers show good agreement with the density functional theory calculation. However, the influence of antisite defects is observed in Au₅Sn. Low-energy electron-diffraction study reveals that although the AuSn layer is ordered its top surface is disordered at room temperature. Surface order is obtained by annealing or deposition at elevated temperatures and dispersing bands are observed by angle-resolved photoemission spectroscopy. Both electronlike and holelike bands are evident for the (√3 x √3)R30° phase, while a nearly free electronlike parabolic surface state is observed for the p(3 x 3)R15° phase.

In Biomedical and Scientific industries scientist have been perfecting the application of bone regeneration by using biodegradable implants. Surgeons all over the world want to ensure bone fusion can occur as painless as possible for pre and post surgery. Bones absorb Magnesium (Mg) elements naturally and its alloys. Magnesium (Mg) elements are of interest due to easier processing methods with conventional machining , elements are low density, have high specific strength, the modulus of elasticity com-parable to that of bone, and Magnesium ions stimulate bone regeneration the biodegradable implant material as the bone heals.The objectives of our research is to study the influence of acid treatments on the surface morphology of material, and to analyze if the electrochemical corrosion of Magnesium (Mg) alloys are potentially biocompatible.Also, specific structures of the material could potentially be used to help bones regenerate faster. Electrical Discharge Drilling (EDD) is a thermal erosion process used to produce through holes in Mg Alloys to fabricate perforated structure similar to open cell porous structure. Biodegradable materials will eventually replace current methods of implantation such as plates, screws, rod, nail, wires, or any external fixator that has been used to heal the bones. Fabricating meta-materials from biodegradable metallic implants are significant due to materials degradation after they have fulfilled their function of bone fusion. The process for removing metal implants requires pre and post surgery. Overall, the fabrication of biodegradable implants are of interests for future applications in scientific and biomedical research.

Metal Oxide UV photodetectors have a large photoresponsivity making them useful in applications such as detection of bacterial fluorescence. Their photoresponse is controlled by the adsorption and desorption of oxygen molecules on the surface, so a model describing the adsorption kinetics would aid in controlling the gain and response time of metal oxide photodetectors. However, current metal oxide photoconduction models describe the adsorption process with Elovich kinetics which are only an approximation to the actual kinetics and are not applicable to photodetectors with large on-current to off-current ratios due to the significant change in the surface potential. In this work, a new model is developed which overcomes the limitations of the Elovich kinetics by using an analytical solution to the adsorption rate equation. The adsorbed concentration is related to the sample conductance through Poisson’s equation. The model provides the transient photocurrent given the device geometry, material parameters, atmospheric environment, voltage bias, and illumination intensity and wavelength, and closed-form expressions are derived for the response time and responsivity. These expressions are compared to experimental results from ZnO thin films and nanowires. Future work will use this model to develop improved heterostructure devices such as ZnO quantum dot/graphene channel UV photodetectors.

Attempts to achieve a wide-bandgap bipolar transistor have so far been promising. However, the limiting factor for improving performance has been the use of p-type gallium nitride (GaN), which suffers from poor hole mobility that further degrades for high-doping levels. This reduces the base transit time and ultimately leads to gain degradation in a bipolar transistor. In this study, we focus on the electrical properties of an n-GaN/p-Si junction, where p-Si was chosen as an alternative to p-GaN for its high hole mobility even at high-doping levels, despite the lattice mismatch between n-GaN and p-Si. We propose three separate experiments utilizing a N₂, O₂, and Ar plasma as surface treatments, then using direct wafer bonding as an alternative to epitaxial growth to form n-GaN/p-Si junctions. This method dually addresses the low mobility of p-GaN and the lattice mismatch of the two materials. Additionally, in a separate experiment, Al₂O₃ was used as an adhesion layer at the n-GaN/p-Si interface. I-V characteristics will be presented and C-V measurements will be shown and compared to the theoretical V_bi of the junction, demonstrating the potential for better performance in future HBTs.

The replacement of traditional CdS buffer layer with magnesium doped zinc oxide (ZMO) has been demonstrated helpful to boost higher CdTe solar cell performance exceeding 18% due to the reduced interface recombination and parasitic light absorption by buffer layer .¹ However, due to the atmosphere sensitivity of ZMO film, The post-treatments of ZMO/CdTe stacks, including CdCl2 treatment, back contact deposition et. al, which are critical for high performance CdTe solar cells, became crucial challenge. In order to maximize the expected benefit of ZMO cooperation, plenty of investigation need to be accomplished in the future. Here, copper thiocyanate (CuSCN) is demonstrated to be a powerful back-contact material with multi-advantages without degrading ZMO layer in CdTe solar cells, even with ammonium hydroxide as the solvent. The thickness of CuSCN film as well as the subsequent annealing temperature are systematically investigated and the highest efficiency 16.5% has been achieved with open-circuited voltage 850 mV, short-circuited current 26.2 mA/cm² and fill factor 74.0%.


1 Munshi, A. H. et al. Polycrystalline CdSeTe/CdTe Absorber Cells With 28 mA/cm² Short-Circuit Current. IEEE J. Photovolt. 8, 310-314, DOI:10.1109/JPHOTOV.2017.2775139 (2018).

CdSe_xTe_1-x alloys now used in state-of-the-art CdTe solar cell technology can be further characterized and optimized. Se alloying can significantly reduce the bandgap of the CdTe absorber due to the strong bandgap bowing between CdTe and CdSe. This has been demonstrated to the enhance quantum efficiency in the infrared and boost the device short-circuit current (J_SC). Interestingly, despite having a lower bandgap, devices using Se have achieved open-circuit voltages (V_OC) on par with (unalloyed) CdTe-based devices. Recent time-resolved photoluminescence and cathodoluminescence data suggest that the CdSe_xTe_1-x materials have lower recombination rates and improved grain-boundary passivation relative to CdTe, and this appears to be making up for bandgap-related V_OC losses[1]. Further characterization studies should reveal additional relationships between absorber processing and the CdSe_xTe_1-x composition, current collection, recombination, etc., that can be used to optimize device performance.
There are two methods of CdSe_xTe_1-x growth. The first method creates a CdSe window layer for a CdTe device then allows diffusion to create a dopant gradient. The second processing method deposits a CdSe_xTe_1-x alloy layer and then a CdTe layer. Previous characterization studies have focused on the CdSe/CdTe device structure and the effects of changing window layer thickness and composition. A study by Poplawsky, et al. [2], indicated that, as CdSe window thickness increased, device performance increased then quickly decreased. Quantum efficiencies and correlative SEM/TEM analysis indicated that if the Se concentration was above 40 at%, grains could change phases from zinc-blende to wurtzite and stop contributing to the device photocurrent. In the CdSe_xTe_1-x /CdTe structure, the Se concentration can be kept below 40 at% to prevent the formation of the wurtzite phase.
To better understand the impact of Selenium concentration on device photo-collection, an Electron Beam Induced Current (EBIC) study is presented which studies the impact of Selenium on device collection for the CdSe_xTe_1-x /CdTe device structure. EBIC is an SEM-based characterization technique which uses an electron beam to locally excite carriers, which are subsequently collected and analyzed to understand microstructural effects on recombination. A current collection map can thus be created with sub-micron resolution. EBIC has been previously used to identify local variation in collection current at microstructural features such as heterojunction interfaces and grain boundaries[2]. In this study, selenium concentration was varied in the alloy layer from 0 to 40 atomic percent while maintaining constant deposited layer thickness.
EBIC analyses of these alloy junctions indicate that the highest carrier collection was near the CdSe_xTe_1-x /CdTe interfaces with a second collection peak near the TCO/CdSe_xTe_1-x interface. This finding is indicative of a buried junction in the CdSe_xTe_1-x /CdTe device. Previous studies on the CdSe/CdTe structure have shown highest collection near the TCO contact with no buried junctions. The position and strength of this buried junction could decrease the effect of interface recombination at the TCO interface, but it could also decrease drift velocity, negatively impacting performance. These competing effects need to be better understood for the optimization of CdSe_xTe_1-x /CdTe device performance.
The presence of a buried junction in the CdSe_xTe_1-x /CdTe device implies a complex change in carrier concentrations and band structure within these devices. In order to better understand the implications of this observation, these results are correlated with a finite element device model. Using this model, it is possible to calculate the equilibrium band structure of the device and estimate the built-in potential between at each junction.

1.Zheng, X., et al. APL Materials 7(7).
2.Poplawsky, J. D., et al. Nat Commun 7: 12537.

Fast synthesis of copper sulfide nanoparticles through two different methods is described in the present investigation. The first reaction was carried out by applying of heating in constant temperature water bath while in the second case the mixture was sonicated. As copper and sulfur sources, copper nitrate and thiourea solutions were used respectively. Main physicochemical properties of the synthesized samples were studied using several characterization techniques. Formation of copper sulfide with hexagonal crystalline structure was corroborated, coinciding with previously reported values for covellite. The most important results corresponded to the un-sonicated sample, which exhibited spiky-like nanoballs morphology and better size distribution. Optical tests revealed band gaps of 1.69 and 2.05 eV, while photoluminescence emissions at 421 and 430 nm, showing promising possibilities for applications in solar cells, catalysis and optoelectronic devices.

Thin crystalline silicon (less than 50 µm thickness) with outstanding flexibility has been considered as a means for developing flexible crystalline silicon (c-Si) solar cells. However, the optical management of the thin c-Si is required for developing the highly efficient c-Si flexible solar cells due to the high surface reflection (more than 30%) and absorption loss over 900 nm wavelength region. Here, we introduce novel high efficiency c-Si flexible solar cells which include random inverted pyramidal polydimethylsiloxane (RIP-PDMS) optical sticker on the 20-µm-thick thin c-Si solar cells. Due to the enhanced light absorption at the entire wavelength region, the thin c-Si with RIP-PDMS optical stickers showed the improved integrated photon flux absorption up to 85.4% (75.0% to 85.4%). The RIP-PDMS sticker enhances the light absorption capability of the thin c-Si by increasing the optical path length in the thin c-Si up to 10%. Further, we analyzed the optical enhancement of the thin c-Si with the RIP-PDMS optical sticker through the finite-difference time-domain (FDTD) simulation. From the optical simulation, we confirmed that the optical path length in the thin c-Si increased because the angles of incident light at the front and rear surface of the thin c-Si tilted up to 10° and 15°, respectively, by attaching the RIP-PDMS optical stickers. Notably, the thin c-Si solar cells with the RIP-PDMS optical stickers showed 18.4% power conversion efficiency while maintaining superior flexibility (bending radius of 10 mm). In addition, our flexible thin c-Si solar cells with the RIP-PDMS optical stickers showed high bending stability without any change in efficiency, even with 1000 bending cycles with a bending radius of 10 mm. Thus, we believe that the thin c-Si solar cells with RIP-PDMS optical stickers can be a promising candidate for the highly efficient flexible solar cells.

Multifunctional textiles have gained special attention due to their great possibility for healthcare, wearable sensors, flame resistance, thermal regulation, UV protection, self-cleaning, energy harvesting, electrical conduction and toxic chemical protection. Among them, electro-conductive fabric is most widely investigated for various wearable applications. Conventional conductive textiles
based on metal coated fiber have some problems, such as high fabrication cost, poor scalability and environmental issues. Graphene based textile, however are considered as one of the most promising materials in eco-friendly fabrication of scalable e-textile. In this work, we introduce scalable fabrication of graphene based fabrics and fibers using dip coating/padding. The electrical conductivity of graphene based e-textile undergoes radical growth at a certain surface deposition of reduced graphene oxide, which is known as percolation threshold. The
electric conductivity of graphene e-textile increased by orders of magnitude at certain number of dip coating cycles. As fabricated graphene e-textiles and fibers have shown great possibilities in wearable energy applications such as energy harvesting. An impedance spectroscopy was used to characterize impedance responses of graphene e-textiles, and equivalent circuit fitting was used to analyze the measured impedance. The electrical properties were
fitted well with nested RL circuit which can provide some important information of charge transport mechanism of graphene e-textiles. Graphene coated fiber can also be used as chemical sensors because electrical properties result from
multiple microstructural processes such as the movement, concentration, and hopping of charge carriers like electrons and ions. These properties can change if additional features are present in the material, such as defects, interfaces and presence of other external chemical species.

GaN based high electron mobility transistors (HEMTs) are considered of great potential to be the building blocks in the next-generation microwave power amplifiers due to their excellent power handling capabilities Ohmic contact and passivation technology are the main factors of device power and efficiency. Improving device efficiency is the premise of improving application system efficiency. Nowadays, More and more efforts are made to improve the efficiency of GaN HEMTs. In this paper, we report the AlGaN/GaN HEMT with improved ohmic contact and double layer pssivasion. the device show an excellent power characteristics.
The epitaxial layers consists of a 1.5nm GaN cap layer,22nm Al0.25GaN, 2nm AlN and GaN buffer layer. The process for GaN HEMT devices is as followed the steps: a)maker, b) ohmic contact etching ion; c) isolation; d) gate; e)doubel passivation by Al/SiN;f) via and metal. The source-drain and gate length are 2.40µm and 0.25µm.the TEM result shows the cross section of recessed TiN/Ti/Al/Ni/Au. For comparison. Conventional structures of MOCVD HEMT is also fabricated simultaneously.
Measurements are performed in continuous wave (CW) mode, load-pull power measurements of this 4×50μm device at 10 GHz is carried out. The device is biased at a voltage of 35V.the output power reached 31.50dBm (7.0W/mm) with 70%
power-added efficiency (PAE) and 14dB gain. Such a good RF power performance comes from the improved ohmic contact and new double passivasion layer.
The interface of ohmic contact is analyzed by means of TEM. It is found that tin thin layer combined with recessed process is more conducive to the formation of low contact resistance.

In this work we report the development of CuS thin films coated with composites of bioglass 45S5/AgNps and urease functionalized as a potential biosensor of urea. The method of chemical bath deposition was used to obtain CuS thin films and through spin coating a thin layer of bioglass 45S5/AgNps was deposited as a substrate for urease immobilization on the surface of the films. The characterization of the composites (FTIR and UV-Vis), the thin films (UV-Vis, SEM/EDX) and urease-urea reaction essays (UV-Vis and 4W) are also reported. The FTIR characterization demonstrate the characteristic molecular vibrations of the bioglass 45S5 and the composites with the AgNps. SEM and EDS characterizations showed us the morphology and the homogeneity of the biomaterial layer deposited over CuS thin films. Through 4W characterization changes in the resistivity of the films were obtained after the urease-urea essays, these changes depend of the urea concentration, which make this thin films a good candidate for its application in a enzyme-based urea biosensor.

As devices continue to trend towards the micro- and nanoscale, the influence of individual crystal defects on their optical, electronic, and thermal properties becomes significant. As an example, the thermoelectric properties of these nanodevices are closely related to the presence of crystal defects, grain boundaries, and strain. While the effects of interfaces and grain boundaries on thermoelectric performance have been widely studied in semiconductor materials, there remains an incomplete understanding of the role that crystal defects play in metals. A major challenge in pinpointing the origins of thermoelectric variation in metallic nanodevices is the difficulty in preparing high quality crystalline thin-film metal microstructures on thermally and electrically insulating amorphous substrates.

Controlling the micro- and nanostructure of metals offers the opportunity to investigate and engineer the properties of metal nanotechnologies. Here, we report a high-throughput method for preparing patterned crystalline gold microstructures grown directly on amorphous insulating oxide via rapid melt growth. In this growth process, thin-film gold-platinum metal-seed microstructures are encapsulated in an insulating crucible and annealed at temperatures above gold’s melting point. During cooling, liquid epitaxial growth directs single crystal solidification [1]. We can extend this platform to controllably prepare single grain boundaries [2], enabling the systematic study of the role of defects in governing local thermoelectric behavior. A scanning photothermoelectric (PTE) setup to allows us to probe the microscopic variations in photovoltage across the gold single crystal and bicrystal wires. Surprisingly, we observe a photovoltage distribution where local changes in Seebeck coefficient are highly influenced by long-range strain fields and concentrated dislocations and uncorrelated with plane grain boundary defects. Our results showcase the link between the presence of elastic strain and dislocations to the PTE effect using a model material system.

[1] K. Zhang, X. B. Pitner, R. Yang, W. D. Nix, J. D. Plummer, J. A. Fan, Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 685.
[2] L. T. Gan, R. Yang, R. Traylor, W. Cai, W. D. Nix, J. A. Fan, Adv. Mater. 2019, 1902189.

The creation of durable liquid-repellent surfaces with optical functionality has been extremely challenging. Major challenges have included low optical transmission, low optical clarity, lack of scalable fabrication, condensation failure, and inability to self-heal. Eliminating light reflection from the top glass sheet in optoelectronic applications is often desirable across a broad range of wavelengths and a large variety of angles. In this paper, we use a multiobjective learning and Bayesian optimization approach to guide the experiments of glass substrate simulation and fabrication with multifunctionality. We demonstrate that nanocone structures have the best anti-reflection (AR) performance and additionally provide physical insight into the AR performance of different structures. Simulations suggest nanocone arrays are able to achieve a solar integrated normal and 65° incidence angle reflection of 0.15% and 1.25%, respectively. In addition, the fabricated glass exhibits resistance to condensation or antifogging properties with an antifogging efficiency of more than 90% and demonstrates the departure of water droplets smaller than 2 μm. The surface can restore liquid-repellency after physical damage through heating for 15 minutes. We envision that these surfaces will be useful in a variety of optical applications where self-cleaning, antifouling, and antifogging functionalities are important.

Superconducting resonators and qubits are essential for the development of quantum-computing architectures. These systems depend on high quality superconducting materials and interfaces, in which the two-level systems loss is minimal. Our previous work focused on understanding of the effects of plasma modification of vacuum/silicon and vacuum/superconductor interfaces on the performance of TiN/Si superconducting microwave resonators. In this paper, we discuss plasma-based modification of TiN/Si interface. We provide a systematic study of every step in the preparation of the TiN/Si heterostructures, namely the chemical treatment with hydrofluoric acid of SiO₂/Si surface, the plasma treatment of H-terminated Si and the TiN film deposition processes. We find that both the gas environment and the plasma processing conditions affect the modification of Si, which in turn affects the composition of the TiN/Si interface. The impact of this work lay in providing new approaches to TiN/Si interface preparation which may enable reduction of metal/dielectric interface losses thus improving the superconducting microwave resonators performance. This work was partially supported by the office of Naval Research through Naval Research Laboratory Base Program.

The process by which strongly-photoexcited semiconductors return to their ground state consists of a series of complex, strongly-correlated many-body interactions which overlap in space and time, resulting in an evolution of behaviors that is both difficult to study and valuable, both for fundamental insights into the quantum nature of matter and for control of device and material behaviors in electronic and optoelectronic applications. Extensive work has been done in this regard, illuminating the sequence of hot carrier excitation, carrier-carrier scattering, carrier-lattice scattering and thermalization, recombination, and lattice heat diffusion. After strong photoexcitation creates a high-energy bath of “hot” carriers, electron-phonon coupling results in the emission of optical phonons, which decay into lower-energy phonons, either optical or acoustic. The resulting nonequilibrium phonon populations can in turn influence hot carrier relaxation and transport, an inextricability which has been demonstrated through transient Raman scattering [1-3]. The inseparable nature of structural behavior from that of excited electron-hole plasmas is further demonstrated by the result that when not in equilibrium, acoustic mode phase velocities change as a function of time due to residual Coulomb interaction caused by imperfect screening by electrons [4]. Thus, a thorough understanding of ultrafast post-excitation structural dynamics is essential to a full comprehension of semiconductor behavior.
As electronic devices continue to shrink, the associated increased importance of device surfaces, interfaces, and defects add an additional layer of complexity to the structural behavior. Techniques typically employed to investigate post-excitation responses operate in reciprocal or momentum space and require spatial averaging, reducing real-space resolution. It is known that interfaces change the electronic structure in their surrounding regions, strongly influencing carrier dynamics, and it has been shown in our previous work that phonons are structure-directed, propagating specifically from interfaces and repeatably traveling perpendicularly to the launching interface [5]. In sum, electron-phonon coupling behaviors are known to be influenced by structure, making real-space information relevant to a complete picture of post-excitation behavior.
Here we use ultrafast electron microscopy (UEM) to image phonons in real space and thereby demonstrate the influence of both the meso-scale and atomic structure of GaAs on the post-excitation phonon behavior. Making use of combined picosecond-nanometer real-space resolutions, we have directly imaged excitation of GHz coherent strain waves propagating exclusively from specimen-vacuum interfaces with initial hypersonic velocities which decay over time to the speed of sound in GaAs. We also report the change in the properties of these waves upon laser annealing of the specimen, epitaxially crystallizing the focused-ion-beam-induced amorphous surface layer and resulting in measurable hardening of the specimen. This appears in the phonon behavior as an increase in the velocity of the phonons. By imaging the phonon interactions with both the specimen-vacuum interface and the amorphous-crystalline interface within the specimen, we demonstrate the importance of interfaces in fully understanding the phonon dynamics in photoexcited GaAs.

[1] A. Othonos, Journal of Applied Physics 83, 1789 (1998), p. 1789.
[2] D. Von Der Linde, et al., Physical Review Letters 44, 1505 (1980), p. 1505.
[3] J. A. Kash, et al., Physical Review Letters 54, 2151 (1985), p. 2151.
[4] B. Y.-K. Hu, et al., Physical Review B 44, 11067 (1991), p. 11067.
[5] D. R. Cremons, et al., Physical Review Materials 1, 073801 (2017), p. 073801.

The need to increase coherence times in quantum devices simulates the search for materials systems that can exhibit electronic states not affected by structural and chemical inhomogeneities. Owing to the advances in synthesis of III-V semiconductors, concentrations of point defects and impurities in these materials can be well controlled, while stacking faults (SFs) provide asymmetric quasi-2-dimensional potential wells that behave as internal interfaces capable of trapping excitons [1]. Here we use ab initio simulations to examine how the stability and electronic properties of SFs change as a function of the distance between them and how the details of their configuration affect spatial distribution of the electron and hole components of excitons and their effective masses. Our simulations suggest that a potential drop of the order of 10 meV is associated with an isolated SF, which is sufficient for the hole to localize predominantly at the SF and to induce an asymmetric distribution of the excited electron density in the vicinity of the SF. For interacting SFs, we find that the electron distribution is sensitive to both the type of the SFs and the distance between them. This coupling provides a pathway to the design of materials with spatially resolved excitonic properties and photo-emission energies by controlling the density and relative arrangements of the stacking faults.

[1] T. Karin, X. Linpeng, M. M. Glazov, M. V. Durnev, E. L. Ivchenko, S. Harvey, A. K. Rai, A. Ludwig, A. D. Wieck, K.-M. C. Fu, Phys. Rev. B 94, 041201(R) (2016)

Metal-assisted chemical imprinting (Mac-Imprint) is a contact-based wet etching process which combines metal-assisted chemical etching (MACE) and nanoimprint lithography (NIL) and can pattern 3D features in group IV and III-V semiconductors with the resolution down to 20 nm [1-3]. The scalability of this process however is restricted by the limited mass-transport of the reactants towards the catalytic metal coated stamp-substrate contact interface. Several approaches have been implemented to enable diffusion through interconnected porous networks embedded into either substrate or in the stamp’s catalyst thin layer [1, 2, 4]. However, these approaches are unable to sustain large area patterning and etching of deeper features (i.e. >1 µm) due to limited volume for etchant storage at the micrometric gaps between stamp and substrate. In this work, we introduce a new polymeric stamp materials for Mac-Imprint that are fully porous through its thickness supporting greater storage volume and a diffusion pathway to the bulk of the etchant solution. Due to the (i) interconnected pore structure, and (ii) precise control over the stamps’ pore size in the sub-100 nm range, Mac-Imprint of not only nano-, but also microscale features is demonstrated in silicon for the first time with high patterning fidelity and greater depths. Additionally, we will present a detail characterization of topography and surface defects of Mac-Imprinted silicon through correlative AFM-Profilometry-Raman imagining. Finally, our results demonstrate wafer-scale Mac-Imprinting of bare silicon wafers of functional optical elements such as antireflective coatings, diffraction gratings, parabolic concentrators.

References:
[1] B. Azeredo, Y. Lin, A. Avagyan, M. Sivaguru, K. Hsu and P. Ferreira, Adv. funct. mater., vol. 26, no. 17, p. 11, 2016.
[2] E. Torralba, M. Halbwax, T. Assimi, M. Fouchier, V. Magnin, J. Harari, J. Vilcot, S. Gall, R. Lachaume, C. Cachet-Vivier and S. Bastide, Electrochem. Commun., vol. 76, p. 4, 2017.
[3] J. Zhang, L. Zhang, L. Han, Z.-W. Tian, Z.-Q. Tian and D. Zhan, Nanoscale, vol. 9, pp. 7476-7482, 2017.
[4] A. Sharstniou, S. Niauzorau, P. M. Ferreira and B. P. Azeredo, PNAS, vol. 116, no. 21, pp. 10264-10269, 2019.

Through the implementation of an ultrafast pulsed UV thermoreflectance measurement system, we demonstrate that one can interrogate the surface temperature of non-metal materials, including wide band-gap semiconductors. This is in contrast to typical visible-wavelength thermoreflectance techniques, where the optical penetration depth in non-metals greatly complicates analysis. This experimental advance allows for the typical multi-layer thermal analysis of other thermoreflectance techniques, with the elimination of the metal transducer. This leads to a net reduction of interfacial thermal resistances in the system, thus greatly enhancing the sensitivity to additional interfaces of interest. This is demonstrated with measurement of a Galium Nitride - Aluminum Nitride (GaN-AlN) interface, where the inclusion of a metal transducer and corresponding additional metal-GaN interface would overshadow any sensitivity to this GaN-AlN interface. In addition to thermal measurements, acoustic measurements can also be done in a typical visible-wavelength thermoreflectance system. This continues to be the case with the UV implementation, however acoustics are also made easier with UV, as the elimination of the transducer layer can yield cleaner results for subsequent layers. Lastly, the presented UV pump-probe system provides a unique insight into the ultrafast carrier dynamics of both wide band-gap semiconductors and metals. In semiconductors, electron excitation and subsequent decay can be observed and measured. In the case of metals, the shorter wavelength regimes allow access to additional electron orbitals, including the direct measure of d-band electron populations in Au following ultrafast excitation.

Copper (Cu) doping is a crucial process to improve the hole concentration in CdTe and reduce the back-barrier height at the interface of CdTe/back electrode, which are critical for boosting the performance of CdTe-based solar cells. Here, we use a wet chemical method to incorporate Cu in CdTe using a CuCl solution prepared by dissolving CuCl powders in ethanol. Temperature-dependent current density-voltage and capacitance-voltage measurements show that the CuCl treatment reduces the back-barrier height and slightly improves the hole concentration. Time resolved photoluminescence measurement shows increased minority carrier lifetime for CuCl treated cells. These characterizations provide convincing evidence for the observed improvement in power conversion efficiency of CdTe solar cells from 13.5%, with an open-circuit voltage (V_oc) of 0.835 V, short circuit current (J_sc) of 24 mA/cm², a fill Factor (FF) of 67%, to 15.6%, with a V_oc of 855V, a J_sc of 24.6 mA/cm2 and a FF 74%.

4H-SiC metal-oxide-semiconductor field-effect transistors (MOSFETs) have great potential in high power and high temperature applications. However, near-interface defects degrade the effective channel mobility and thus limit the performance of 4H-SiC MOSFETs. In this work, we utilize electrically detected magnetic resonance (EDMR) to study hyperfine interactions with spin-dependent recombination (SDR) of silicon vacancy defects at the 4H-SiC interface in 4H-SiC MOSFETs. The physical foundation of EDMR is electron paramagnetic resonance (EPR). In EPR and EDMR, a sample is subjected to a large, slowly varying magnetic field and an RF or microwave oscillating field at a frequency ν. In the simplest case, EPR is achieved when hν=g_eβH.¹ Here, h is Planck’s constant, ν is the electromagnetic radiation frequency, g_e is the free electron g factor (g_e ≈ 2.0023), β is the Bohr magneton, and H is a large externally applied magnetic field. At resonance, the electron transitions from its +1/2 to -1/2 spin state (or vice versa). The analytical power of EPR and EDMR comes from deviations from this simple case caused by the paramagnetic site environment. These perturbations are, for the most part, caused by hyperfine interactions, interactions of the electron with magnetic nuclei, and spin-orbit coupling. EPR detected via SDR allows for measurements of small numbers of defects via spin-dependent changes in device current. The sensitivity of SDR/EDMR is very nearly independent of the field or frequency at which the measurement is made.²
Prior continuous wave EDMR (cwEDMR) studies have linked interface and near-interface defects to carbon dangling bond defects³ and silicon vacancy defects.⁴ The identification of silicon vacancy defects in 4H-SiC MOSFETs was based on measuring nearest neighbor carbon 13 hyperfine interactions.^{4 29}Si next-nearest neighbor (NNN) hyperfine interactions are too small to resolve in an EDMR measurement due to the width of the center part of the resonance signal. The interactions of NNN ²⁹Si are isotropic with a value 2.96 G determined from conventional EPR. Broadening in EDMR obscures these ²⁹Si NNN hyperfine interactions. We have overcome the EDMR broadening problem using the Breit-Rabi effect,⁵ a shift in the hyperfine side peaks towards lower fields and frequencies. This effect is sufficiently large so that it may be detected in ultra-low field and frequency measurements. We utilize frequency swept EDMR (fsEDMR). With fsEDMR, tuning the resonance condition is performed by varying a low static field. In fsEDMR, the magnetic field H is held constant as frequency is swept. Our experiments were carried out at fields less than 10 G and frequencies less than 30 MHz. We observe a centerline with a sharp peak on the low field side only at these low fields and frequencies. We attribute the peak to otherwise unresolved hyperfine interactions made observable by shifting due to Breit-Rabi effects. The peak has fairly large amplitude. EasySpin⁶ was utilized to simulate the results accounting for the presence of one ²⁹Si NNN. The results of the simulation are in good agreement with our experimental observation.
¹ W. Gordy, Theory and Applications of Electron Spin Resonance (John Wiley & Sons, 1980).
² D. Kaplan, I. Solomon, and N.F. Mott, J. Phys. Lettres 39, 51 (1978).
³ J. Cottom, G. Gruber, G. Pobegen, T. Aichinger, and A.L. Shluger, J. Appl. Phys. 124, (2018).
⁴ C.J. Cochrane, P.M. Lenahan, and A.J. Lelis, Appl. Phys. Lett. 100, 023509 (2012).
⁵ G. Breit and I.I. Rabi, Phys. Rev. 38, 2082 (1931).
⁶ S. Stoll and A. Schweiger, J. Magn. Reson. 178, 42 (2006).</p>

The continuous innovation in nano materials has allowed the development of new antireflective coats (ARCs) for optical lenses, solar cells, data storage, displays and others that require a reduced reflectance to increase efficiency. Multi-layered ARCs particularly have attracted attention lately for demonstrating low and tunable spectral reflectance.

This paper presents a simple approach for forming anti-reflective coating stacks on bulk Ge and PECVD Ge grown on Si employing colloidal dispersions of Si nanoparticles (NPs).

The Si NPs have a diameter of 2.85 nm and are prepared by chemical etching of crystalline Si wafers in HF and H₂O₂ using electrical or hexachloroplatinic acid catalyst. The nanoparticles are recovered by sonicating the wafers in isopropyl alcohol, forming a stable colloid with concentration of 10 µM [1].

Each antireflective coat is applied by the drop casting of monodispersed Si NPs on a 3 by 3 cm pieces of both Ge bulk and PECVD Ge grown on Si. A total of 5 drops are applied with each drop consisting of 100 µL. After each drop, the samples are left to dry for an hour and a half.

For bulk Ge, (110) Sb doped (n-type) Ge with a resistivity of 0.1–0.5 Ωcm were used, while for the PECVD Ge growth, (100) P doped (n-type) Si with a resistivity of 0.01–0.02 Ωcm are used as starting substrates. An Oxford Instrument System 100 PECVD activated by 13.56 MHz radio frequency signal is used to grow Ge layers on Si substrates. High purity germane (GeH₄), H₂ and Ar gases are used as precursors for Ge growth. The undoped Ge film was grown using the two-step approach, which has been described in detail in our previous work [2].
The impact of multiple drop casting of Si NPs on the antireflective properties of the Ge surfaces were investigated by the analysis of their reflection spectra. The reflectance of the coating was measured at four locations on each sample by a UV–VIS spectrophotometer (Lambda 850, PerkinElmer). The samples were qualified by controlling the reflectance at the four points within an error of 1.2%. The corresponding reflection spectra were captured as a function of the incident wavelength, ranging from 250 nm to 1500 nm.

For PECVD Ge grown on Si substrates, the average reduction in reflectance has shown to increase as the incident wavelength decreases, with a decrease of ~2% after each 100 μL coating. After 5 drops (100 μL ×5), the overall reflectance across the whole spectra is on average 9.35% lower compared to the uncoated samples. The minimum decrease in reflectance was of 2.49% at 1290 nm and the maximum of 34% at 250 nm.

For Bulk Ge, the average reduction in reflectance has shown to increase as the incident wavelength decreases from 1500 nm to 320 nm. At 500 μL, the reduction in reflectance had an average of 6.16% (Min.: 2.73% at 1500 nm, Max.: 13.75% at 320 nm and Median: 5.1%). At lower wavelengths range from 250 nm to 320 nm), the amount of reflectance reduction is less compared to wavelengths above 320 nm.

After 5 drops (100 μL ×5) on PECVD Ge samples, the incident wavelength where the minimum reflectance occurred (λ_min), shifted from 1050 nm where reflectance is lowered by 34.5%, to 410 nm with 25.2% reduction in reflectance. While for bulk Ge sample, λ_min shift occurs from ~1480 nm to ~420 nm with a decrease in reflectance of 37.9% and 28%, respectively.

Results demonstrate that it is feasible to deposit a coating of Si NPs with tunable reflectivity over a wide range by drop casting multiple layers of the nanoparticles.

[1] Akcakir, O., Therrien, J., Belomoin, G., Barry, N., Muller, J.D., Gratton, E., Nayfeh, M., Appl. Phys. Lett. 76, 1857 (2000); https://doi.org/10.1063/1.126191
[2] Dushaq G., Rasras M. Nayfeh A., Thin Solid Films 636, 585–92 (2017); https://doi.org/10.1016/j.tsf.2017.07.009

Conventional thermal management solutions in microelectronics packaging suffer from an inability to effectively couple thermal conduction and mechanical compliance into a single material, routinely necessitating a sacrifice in thermal, and therefore device, performance in favor of mechanical reliability. For the past few decades, compliant thermal interface materials (TIMs) have been limited to polymers loaded with dispersions of unaligned or randomly distributed conductive elements which can only achieve thermal conductivities up to ~5 W m⁻¹K⁻¹. Here, we present the processing and characterization of a compliant, vertically-aligned metal nanowire composite ideal for achieving thin, flexible interfaces with low thermal resistance. The aligned nanowire composite is fabricated by electroplating copper into a 25-45% porous, sacrificial ceramic template, liberating the nanowires by chemically dissolving the template, and infusing the resulting nanowire array with a thermally stable polymer by capillary-driven wicking. An integrated bonding layer is added by sub-filling the template with copper and sequentially electroplating metallic solder to form localized solder tips on the nanowires. The entire fabrication process can be completed at wafer-scale with a lateral area up to 12.5 inches². The thermal performance of the composite TIM was evaluated with a modified thermal resistance measurement designed to mimic typical operating conditions. The double-sided construction of the composite and integrated bonding layers facilitate integration between any two arbitrarily-chosen surfaces by enabling the absorption of interfacial strain, mismatch, and roughness while achieving thermal conductivities up to 200 W m⁻¹K⁻¹, significantly advancing the state-of-the-art in TIMs.

In this talk, we report the use of surface-enhanced Raman scattering (SERS) to investigate the adsorption of thiol and isocyanide molecules on colloidal Ag nanocubes in situ. In a typical process, we mixed Ag nanocubes with ligand molecule solutions at different concentrations and then collected SERS spectra at different time points. We demonstrate that 4−nitrothiophenol (4-NTP) binds to the Ag surface rapidly because of the strong Ag−S chemical bonding. By decreasing the concentration of 4−NTP, we observe red shifts of the ν_NO2 band, which suggests the change to the orientation of ligands on the surface. Different from thiols, the 1,4−phenylene diisocyanide (1,4−PDI) binds to the Ag surface through σ−donation with weaker binding strength. By increasing the concentration of 1,4−PDI, our results indicate that the molecules tend to adopt an orientation that tilts away from the surface. More interestingly, we investigate the competitive binding behavior of these two ligands by introducing Ag nanocubes into a solution containing both 4−NTP and 1,4−PDI. Our results show that thiol is more preferential to bind onto the surface when the concentrations of the two ligands are comparable in the solution.

As environmentally benign and highly-efficient thin-film solar cells, low-cost Cu₂ZnSn(S,Se)₄ (or kesterite) solar cells, herein CZT(S,Se) have been steadily emerging as an alternative for the 2^nd generation Cu(In,Ga)Se₂ (CIGS) and CdTe solar cells in the past several years. CZT(S,Se) solar cells have several advantages such as high absorptions coefficient (>10⁴ cm^-1), tunable direct band gap energy (1 to 1.4 eV), and usage of abundant elements in earth's crust, becoming an ideal platform to harvest solar energy. However, the efficiency improvement and understanding of emerging CZT(S,Se) is still in an early stage compared to the 2^nd generation counterparts. In recent progress, our team reported that Germanium (Ge) incorporation into CZTSe solar cells has improved device performance, which contributed to deepening understanding of Ge-alloyed CZT(S,Se) solar cells, in particular, on the Schottky-type back-contact barrier.
We present the complete set of models about the Schottky barrier with the MoSe₂ interlayer between the absorber and Molybdenum (Mo) back-contact. Developed analytical models with in-house MATLAB simulators agree well with TCAD numerical simulation and experimental results. A new concept of an effective back-contact barrier is introduced, which provides the mechanism of the improved device performance with lower back-contact barrier for CZTSe Ge bi-layers devices. By incorporating nanoscale Ge bi-layers below and above the CZTSe absorber, the back-contact barrier is improved by 33 %, and power conversion efficiency is improved by 1.1 % than Ge single-layer devices. Based on the back-contact model, defect concentration of Ge reference devices is estimated to be 5.4 x 10¹⁶ cm^-3, whereas Ge bi-layers devices show an estimated <9.8 x 10¹⁵ cm^-3 at the back-contact interface. To compare models with experimental results, we fabricated three different types of CZTSe: nanoscale Ge devices. The property of the back-contact barrier is assessed by characterizing a temperature-dependent behavior of CZTSe alloyed devices.
Collectively, the reduced defect concentration by Ge incorporation in the MoSe₂ interlayer improves the barrier height of back-contact. As the defect concentration of the MoSe₂ interlayer increases, the benefit from the back-contact barrier lowering by the valence bands offset is reduced and essentially eliminated. Moreover, the increased effective back-contact barrier degrades the series resistance and the overall device performance. However, the barrier at the interface between MoSe₂ and Mo metal contact remains the same, even with MoSe₂ defect concentration increase. Incorporating thin Ge nanolayer at the interface between the absorber and MoSe₂ could positively influence and possibly reduce the defect states, which lowers the effective back-contact barrier from modeling and simulation results. Hence, the Ge layer at the back-contact interface successfully improves the conversion efficiency up to 8.3 % with the improved back-contact.

The metal-semiconductor contacts also known as Schottky diodes have their applications in solar cell, photovoltaics and other electronic devices. In order to understand the behavior of Schottky diode characteristics, it is essential to generate its current voltage characteristics using fundamental semiconductor device equations using various conditions and then analyze the derived barrier parameters in view of thermionic emission diffusion theory and the assumptions made.
The Schottky diodes are also fabricated by deposition of nickel on p-type silicon wafers. The current-voltage characteristics of the diodes are measured over wide temperature in low temperature regime. In this paper we discuss the phenomena of intersection in ln(I)-V curves observed in Ni/p-Si Schottky diode at low temperatures from 290K to 80K. The series resistance of the diodes was found to decrease with decrease in temperature which leads to saturation of the forward current of diode at higher bias, as a result intersection of ln(I)-V curves occurs. Moreover, the barrier height, ideality factor and series resistance has been evaluated from forward ln(I)-V curves using thermionic emission diffusion mechanism. Barrier height was found to decrease while the ideality factor increases with decreases the temperatures from 290K to 80K. The activation energy plot exhibits non-linearity in the measured temperature range which may be due to the presence of other current mechanisms such as tunneling, recombination along with thermionic emission through the barrier. The reasons for series resistance behavior causing intersection is analyzed.
The observed intersection behavior in ln(I)-V plots is compared with the intersection behavior observed in current-voltages plots of Schottky diodes simulated using semiconductor device equations. For this we have solved nonlinear Poisson’s equation along with the electron and hole continuity equations to simulate the current-voltage characteristics of Schottky diodes. The potential variation inside the bulk semiconductor near the metal-semiconductor contact was estimated first and then the current as a function of bias through the Schottky diode using silicon parameters were calculated for various conditions. From the simulated current-voltage characteristics the apparent diode parameters were extracted by fitting of current-voltage data into thermionic emission diffusion current equation. The derived barrier parameters are analyzed to study the effect of interface structure on the intersection behavior of current-voltage characteristics of Schottky diodes.

Thin films made of WO₃ possess a unique chromogenic behavior to be exploited in diverse sensing devices applications. The physical properties dependent main on microstructure and surface morphology and can be tuned by mixing with Pt. Our previous investigations indicate that pristine WO₃ holds a piezoelectric response for the monoclinic phase created during heat treatments at 400°C. In this study, we present platinum-loaded WO₃ thin films prepared by radio-frequency sputtering technique. The films were subjected to heat treatments at 300°C, 400°C, 500°C and 550°C to investigate crystal growth and piezoelectric response. The physical properties of the films were studied by grazing incidence X-ray diffraction (GXRD), scanning electron microscopy (SEM) and piezoresponse force microscopy (PFM), with complete atom probe tomography (APT) to understand Pt distribution in the atomic concentration profile, the polycrystalline nature of thermally treated films, the piezoelectric domains, as well as to track any segregation of tungsten and oxygen over silicon at Si/WO₃ interface which is correlated to material stability and performance for device fabrication and engineering applications.

Layered cobaltites Ca₃Co₄O_9-δ with incommensurate character are the most promising candidates for high-temperature Thermoelectric (TE) applications due to their high thermal stability, natural abundance, lightweight, and non-toxicity. The energy conversion efficiency, for calcium cobaltite, in a single crystal form shows excellent performance that approaches the well-developed conventional thermoelectric materials. However, the performance of polycrystalline ceramics remains low and only ~30-60 % of that found for the single crystals, largely because of low electrical Seebeck coefficient and the high electrical resistivity from the grain boundaries. Here we show that Bi as non-stoichiometric addition into polycrystalline ceramics renders the formation of intragranular dopants substitution and concurrent intergranular segregation at the grain boundaries. Such dopants grain boundary segregation resulted in the Bi substitution at both the Ca and Co sites at the Ca₃Co₄O_9-δ unit cell, companied by the dramatically decrease the electrical resistivity, simultaneously increase the Seebeck coefficient. Overall, Bi-non-stoichiometric addition in Ca₃Co₄O_9-δ Oxide Ceramics ultimately results in a remarkable high ZT of 0.67 at 800 °C.

Among various nanostructured metal oxides, molybdenum oxide (MoO₃), a wide band-gap n-type semiconductor, has attracted much interest as a promising gas sensor because of its excellent sensitivity, good reproducibility, low cost, high compatibility with microelectronic processing, low electron recombination rate, and high stability. The MoO₃ thin films have promising electrical properties for gas sensing applications.
To synthesize two-dimensional materials under ultra-high vacuum is a key advantage for enhancing the properties of nanostructured materials. In this presentation, we demonstrate the synthesis, characterization, and gas sensing properties of MoO₃ films, which were grown using Krypton-Fluoride (248 nm) excimer laser. Deposition was done under 200 mTorr Oxygen pressure at the substrate temperature of 600^oC. The targets used for deposition were synthesised using hydrothermal method. The deposition was done with the optimized parameters, to ensure good polycrystalline films on the Si (001) substrate. The optical, morphological and compositional properties of MoO₃ films were studied by using UV-Visible spectroscopy, Scanning Electron Microscopy and Raman Spectroscopy. The formation of MoO₃ was also confirmed using X-Ray Diffraction technique. The MoO₃ films showed excellent response to NO_X and NO_X gas moieties, at even low gas concentrations from 100 ppm onwards. The sensor not only exhibits high sensitivity but also good selectivity towards NO₂ gas.

Near-infrared filters, which selectively transmit a part of the optical spectrum and reject the rest of the infrared spectrum, have been developed for a number of applications including dense wavelength-division multiplexing networks, Raman spectroscopy, and spectral selectivity of visible lasers. However, a tunable filter with variable transmission that is desirable for dynamic applications has not been significantly studied.

Vanadium dioxide (VO₂) is a thermochromic insulator-to-metal phase transition material which exhibits a dramatic difference between the insulating and metallic phases’ optical properties. In this study a dichroic switchable optical transmission filter is made by placing a silicon spacer between a VO₂ thin film and a transparent quartz substrate. A Fabry-Perot resonance cavity is formed while the VO₂ is in the insulating phase, creating an optical filter with a peak wavelength around 2 µm and a full width at half maximum around 1.5 µm predicted by transfer matrix calculation. The transmission filter can be switched off by heating the VO₂ past the transition temperature; once VO₂ is in its metallic phase, the structure exhibits low transmittance across the visible and infrared spectra.

The structure was optimized so that it exhibits a large transmission difference between its on and off states. The structure is being fabricated via sputtering and electron beam evaporation deposition and will be measured using Fourier-transform infrared spectroscopy to confirm the theoretical calculations and successfully demonstrate the ability to switch transmittance in the infrared spectrum.

Copper is a commonly used interconnect in today’s semiconductor packaging industry due to its exceptional electrical and thermal properties. In applications of the 2.5 and 3-D microelectronics packaging technologies, Cu is utilized in through-silicon-vias (TSVs) and flip chip bonding to interconnect multiple layers of circuits [1]. These 2.5 and 3D packaging technologies can provide miniaturization, lower power and higher performance than current 2D technology.
Controlled collapse chip connection (C4), also known as flip chip bonding, is used in the microelectronics packaging industry to connect chips by means of electrically conductive solder bumps partially crushed between two chip surfaces, which allows for chip miniaturization and high performance. SnAg solder bumps are deposited on copper pillars by electroplating for better control of bump pitch and higher processing throughput and versatility [2]. Performance of the Cu surface is an important step in the process to ensure quality solder bumps. During electroplating, tight processing parameters must be met so that bump defects and poor height uniformity can be avoided. An understanding of the interactions at the solder bump/metal interface is needed, based on the electroplating parameters, to determine the best method for populating solder on the wafer surface.
In this study, Cu surfaces were treated by oxygen plasma etch to remove surface contamination and any potential surface oxide formation through a sulfuric acid dip prior to electroplating. Hull cell tests were performed prior to SnAg bump deposition to qualitatively determine the condition of the plating bath and to optimize current density plating parameters. SnAg solder bumps were then electroplated onto Cu pads at current densities varying from 200-30 A/cm². At higher current densities, greater than 60A/cm², bump height non-uniformity and dendritic growth were observed. Macro and nano-scale Cu surface examinations indicated the role of Cu surface nature on Cu/SnAg interface quality, which was evaluated based on surface topography and surface energy calculations. Electrochemical analyses were also performed to evaluate the rate of surface passivation on the interface and film growth quality [3].

[1] Song, C., Wang, Z., Chen, Q., Cai, J., Liu, L., “High aspect ratio copper through-silicon-vias for 3D integration”, Microelectronic Engineering, 85 (10), P1952-P1956, 2008.
[2] Woertink, J. et al., "From C4 to micro-bump: Adapting lead free solder electroplating processes to next-gen advanced packaging applications," 2014 IEEE 64th Electronic Components and Technology Conference (ECTC), Orlando, FL, P342-P347, 2014.
[3] Yagan, R., Basim, G. B. “A Fundamental Approach to Electrochemical Analyses on Chemically Modified Thin Films for Barrier CMP Optimization”, ECS Journal of Solid State Science and Technology, Special Issue on Chemical Mechanical Planarization for sub 10-nm Technologies, 8 (5), P3118-P3127, 2019.

This study describes the preparation of electrodes based on carbon aerogel with silver sulfide (Ag₂S) quantum dots (QDs) by Sputtering deposition method combined with solid vapor reactions. The chemicophysical properties and elemental composition of the samples were determined. Ag₂S-CA electrode designed showed external quantum efficiency (EQE) of 68 % in a wavelength of 525 nm, a band gap of E_g≈ 0.97 eV and conductivity of 1.72 S/cm ± 0.1 S/cm. Ag₂S QDs improved the absorption of visible and near infrared light and electric properties of the carbon aerogel electrodes providing good properties that can be used in future applications of photoelectrochemical biosensors.

Silver is the highest electrically conductive metal of all metals, therefore silver nanomaterials have been used in several applications regarding electrical conductivity. Even now, adequate research related to electrical conductivity of aqueous nanodispersions have not been done. Therefore, this study was conducted with the objective of assessing the influence of silver nanoparticles towards their electrical conductivity in nanofluids. According to literature, electrical conductivity of colloidal suspensions containing nanoscale conducting particles is related to the particle volume fraction, the electrical double layer (EDL) thickness and its motion. Particle shape can affect this EDL thickness which gives the conductive property for non-ionic particles in aqueous medium and increasing number of particles can affect the conductive pathway of nanoparticles by physical contact according to the literature. Thus, electrical conductivity of silver nanoparticles with different shapes and sizes were tested. Silver nanomaterials were synthesized using chemical reduction method and these were characterized using scanning electron microscopy, transmission electron microscopy, UV-visible spectroscopy and dynamic light scattering. Results obtained from characterization confirmed the successful formation of silver nanospheres (yellow), nanorods (red), nanowires (yellow) and nanoplates (blue). Electrical conductivity of each of these different nanomaterials was measured under different sizes and concentrations. The electrical conductivity measurements were taken after removing ions by centrifugation and redispersing silver nanoparticles in ultrapure water. For comparison of conductivity measurements, standard potassium chloride solution and ultrapure water were used as references. Generally, increasing the particle concentration increased electrical conductivity, regardless of shape or size on account of increasing number of electrical double layers around nanoparticle surface. The highest electrical conductivity reading was observed by silver nanowires and it range between 200 30 µS/cm for a very low initial concentrations of silver nitrate (1 x 10^-3 mol dm^-3). In silver nanowires, conductivity was shown to be increased with increasing the particle concentration more rapidly than other morphologies. Therefore, in conclusion, this study claims that there is an extraordinary electrical conductivity of silver nanowires due to the morphological deference. The increasement of the particle volume fraction, nanowires conductive pathway will be more favorable due to the physical contact of particles in the media and it can be taken as the key point of silver nanowires electrical conductivity. Further studies should be done to assess the effect of the type of metal and solvent towards electrical conductance in the range of nanoscale.

Keywords: silver nanowires, electrical conductivity, volume fraction, electrical double layer

Binary manganese oxide/cobalt oxide (MnO_2^/Co₃O₄) composites on the well-aligned electrospun carbon nanofibers (MnO₂/Co₃O₄@ECNFs) were synthesized via a simple and economical one-step template-free electrodeposition technique with a constant low-current (60μA) for 3h, and their electrocatalytic activity towards non-enzymatic glucose sensors was evaluated in detail. The MnO_2^/Co₃O₄ nanostructures were uniformly deposited on the well-aligned ECNFs with high porosity, effective interconnectivity and large amounts of extended number of conducting channels. The electrocatalytic performances of as-prepared MnO₂/Co₃O₄@ECNFs towards the oxidation of glucose were carried out by cyclic voltammetry (CV) and chronoamperometry (CA), and the experimental results show that the MnO₂/Co₃O₄@ECNFs exhibited a superior sensing performance towards glucose including a rapid glucose response within 5 seconds, a wide range (5 μM to 10.89 mM), excellent sensitivity of 973.91 μA mM^-1 cm^-2 and a low detection limit of 0.02 μM (S/N=3) with satisfactory selectivity, great reproducibility and stability. These excellent electrochemical performances were attributed to the synergistic effect of binary metal oxides and the unique porous architecture of the electrode. These features indicate that as-synthesized MnO₂/Co₃O₄ binary nanostructure decorated well-aligned ECNFs hold great potential in the development of a commercialized glucose sensor.