United States

Abstract Body: For the last few decades, most of the advancement in electronic and optoelectronic devices have been relying on our ability to integrate dissimilar materials having different lattice constants. By combining multiple different materials to form a heterostructure, we could engineer a particular band structure for a specific purpose. Nowadays the advancement on the integration technology is also motivated by the rapidly-growing field of Si photonics which requires integration of advanced non-Si infrared (IR) devices into the Si platform. However, to grow such devices on Si, one needs to address crystallographic challenges associated with high threading dislocations density caused by lattice mismatch. We have developed a novel materials integration technique, called metal-catalyzed lateral epitaxy, that has showed promising results in the Ge on Si system which has a 4% lattice mismatch. This technique uses a metal catalyst to control the growth location and a PECVD-grown capping layer to ensure growth laterally. The lateral growth is dictated by the vapor-liquid-solid mechanism which has shown a promising potential in suppressing the nucleation of threading dislocations by the extension of misfit dislocations. The main advantages of this technique over others are that it is wafer-scale, grown directly on the Si substrate, and allows direct electrical connection of the grown films to the substrate. In this talk, recent results of our systematic studies on the kinetics and the mechanism of this unique integration technique will be presented based on the heteroepitaxy of Ge-Si and InP-Si systems. Both are crucial heterostructures in Si photonics for high-performance IR photodetector and light source, respectively.

EMC 2021

Kinetics and Mechanism of Metal-Catalyzed Lateral Epitaxy for Growing Lattice-Mismatched Heterostructures via Vapor-Liquid-Solid

kinetics

vapor-liquid-solid

heteroepitaxy

The Electronic Materials Conference (EMC) is the premier annual forum on the preparation, characterization and use of electronic materials. This year’s event will feature a plenary session, topical sessions, a poster session and more.

EMC is coordinated with the Device Research Conference (DRC), which will be held during the same week, from June 20-23. This recognizes the strong interaction between device and electronic materials research and provides fruitful exchange of information between attendees of both Conferences.

![](https://assets.underline.io/uploads/markdown_image/1/image/eb26c24a18d5d1030616afa4f993cf89.png)
	
<a href="https://www.mrs.org/docs/default-source/meetings-events/mrs-conference-services/2021/emc-2021/2021-emc-program-at-a-glance-6-16-21.pdf?sfvrsn=e39f950c_4">Latest Program-at-a-Glance</a>

<a href="https://drive.google.com/file/d/1eiS1ao4p83VNy8xgHs3Rls7gTQjx7Rc4/view?usp=sharing">Full Program with Abstracts</a>

technical paper

Abstract Body: A molecular beam epitaxy (MBE) approach to selective area epitaxy (SAE) of III-V semiconductors could enable the seamless integration of metals, dielectrics, and high-quality crystalline semiconductors. This technique could advance novel optoelectronic structures, such as high-contrast photonics, site-controlled quantum emitters, stacked pixel detectors, and photonic integrated circuits. Using patterned dielectrics and metals to define crystal seeding regions, SAE can be performed to embed microstructures into crystalline semiconductors1. While SAE by metal organic chemical vapor deposition has been widely successful due to its high material deposition selectivity, a SAE MBE method could enable further advances through its high layer precision and access to non-equilibrium growth conditions1,2. SAE is difficult to achieve with conventional (continuous) MBE growth due to III-V polycrystalline nucleation on the mask, even at growth temperatures as high as 700°C. As a result, Allegretti et al. developed periodic supply epitaxy (PSE), a method to inhibit polycrystalline deposition by cycling group III deposition under a constant group V flux2,3. SAE with PSE was achieved using high growth temperatures and low growth rates3-4. This mitigates nucleation of poly-GaAs through a combination of increased Ga adatom diffusion to seeding windows and desorption of remaining Ga adatoms3-4, enabling high-quality planar encapsulation of the mask1. While an all-MBE approach has been demonstrated to achieve selective growth of features ~2µm in width and ~300nm in height, applications requiring larger features are limited by nucleation of polycrystalline semiconductor on the amorphous surface due to poor deposition selectivity and low adatom surface diffusion1,3. We present experimental results of GaAs growth on SiO2 films that demonstrate decreased poly-GaAs formation at higher temperatures and lower As4/Ga flux ratios, enabling SAE of larger patterned features. However, this improvement in selectivity has yet to be explained quantitatively. In order to expand the applicability of all-MBE SAE, we developed a PSE model that (1) identifies selective growth regimes to clarify their underlying mechanisms and (2) allows optimized SAE growth conditions to be determined for a given SiO2 pattern. A numerical 1D model was developed to describe PSE selectivity by fitting adsorption, desorption, and diffusion constants to GaAs growth on SiO2 films at 600°C with a growth rate of 0.7 Å/s and 65x As4/Ga ratio5-8. Two growth regimes have been identified: desorption- and diffusion-limited. The desorption-limited regime depends only on thermal desorption from the mask surface to achieve selectivity and as a result, is feature-independent. This model has identified the desorption-based selectivity limit to be an 18.7% PSE duty cycle and was demonstrated by observing no polycrystal formation after growth of 100nm GaAs with 10% PSE duty cycle on an SiO2 film. The remainder of the selective growth space is diffusion-limited and sufficient III adatom diffusion off the mask is required to achieve selectivity. Experiments to fully investigate the influence of growth temperature and V/III ratio on selectivity are underway and will be reported at the conference. [1] D. J. Ironside et al., ACS J. Crystal Growth and Design (2019). [2] A. M. Skipper et al., 2019 MRS Electronic Materials Conf. [3] F. E. Allegretti et al., Journal of Crystal Growth (1995). [4] S.C. Lee et al. Journal of Applied Physics (2002). [5] S. Shankar. Diffusion in 1D and 2D, MATLAB. Retrieved April 11, 2020. [6] Aseev et al. Nano. Lett. (2019). [7] S. C. Lee et al. Cryst. Growth Des. 2016. [8] E.M. Gibson et al. Appl. Phys. Lett. (1990). This research was partly done at the Texas Nanofabrication Facility supported by NSF grant NNCI-2025227 and was supported by Lockheed Martin and NSF through the UT CDCM: an NSF MRSEC under Cooperative Agreement No. DMR-1720595, as well as CCF-1838435 and DMR-1839175.

Experimentally-Calibrated Modeling of Molecular Beam Epitaxy Selective Area Regrowth

Abstract Body: The (In,Ga)N material system is attractive for photovoltaic applications as its optical absorption covers the solar spectrum. Thick InGaN films (hundreds of nm thick) with InN mole fraction larger than 10% are required for this purpose. However, the growth of high In-content InGaN films remains challenging due to 10% lattice mismatch between InN and GaN and the large difference between the thermal stability of GaN and InN. While the optimum growth temperature of GaN by plasma-assisted molecular beam epitaxy (PAMBE) is ~700 °C-730 °C, the growth temperature of InGaN is typically below 600 °C. Therefore, lower growth temperatures are required to grow InGaN films within the miscibility gap. InGaN films can be grown pseudo-morphically on GaN below a critical thickness. As the thickness increases beyond the critical thickness, the InGaN film relaxes. This relaxation may occur through different mechanisms depending on the growth technique and growth conditions (e.g., growth temperature). V-defect formation from threading dislocations is a common way of strain relaxation, especially in InGaN films grown via metal organic chemical vapor deposition, and has been observed in molecular beam epitaxy (MBE) grown InGaN films as well [1]. Pyramidal slip through the Matthews-Blakeslee mechanism and basal plane slip of misfit dislocation are other mechanisms through which strain relaxation occurs in InGaN films. Critical thickness reduces as InN mole fraction in InGaN increases. Bazioti et al [2] studied relaxation mechanism in InGaN films grown by PAMBE. They showed that at lower indium contents, strain relaxation dominantly occurred via formation of V-defects, with concurrent formation of an indium-rich interfacial zone. As indium content increased, this mechanism was replaced by the introduction of a self-formed strained interfacial InGaN layer of lower indium content, as well as multiple intrinsic basal plane stacking faults and threading dislocations in the rest of the film. In this talk, we will present growth of 700 nm-thick InGaN with an average InN mole fraction of ~8% coherently strained to GaN confirmed by XRD reciprocal space map scans. This is several times thicker than the critical thickness of In0.08Ga0.92N. Surprisingly, we did not observe generation of new threading dislocations in the InGaN films or misfit dislocations at the interface. Instead, we show that the strain is managed by self-formation of a superlattice structure. The superlattice was confirmed via scanning transmission electron microscopy and atom probe tomography. Photoluminescence (PL) measurements at room temperature revealed peak wavelength of ~450 nm, corresponding to In0.18Ga0.82N with a full width half max (FWHM) of ~36 nm. Reference: 1. C. Bazioti et al. Structure and strain variation in InGaN interlayers grown by PAMBE at low substrate temperatures Phys. status solidi 252 1155–62 (2015). 2. C. Bazioti et al. Defects, strain relaxation, and composition grading in high indium content InGaN epilayers grown by molecular beam epitaxy J. Appl. Phy. 118,155301 (2015).

Demonstration of Self-Assembled InGaN/GaN Superlattice on GaN Template Grown by Plasma-Assisted Molecular Beam Epitaxy

Abstract Body: The (In,Ga)N alloy system is attractive for optoelectronic applications as it enables direct bandgap tuning within a large range (0.7 eV–3.4 eV). While highly efficient GaN-based blue light emitting diodes (LEDs) have been commercially available for a decade, GaN red LEDs suffer from low efficiency. Red LEDs require high In content InGaN (>35%) QWs. Growth of high In-content InGaN on the GaN substrate is challenging due to high vapor pressure of InN. InN starts decomposing at 500 °C which is well below optimum GaN growth temperature of ~700 °C in molecular beam epitaxy (MBE). There is a large lattice mismatch between InN and GaN (∼10%). Therefore, there is a great deal of interest in developing relaxed InGaN buffer as pseudo-substrate for optoelectronics applications. Various techniques have been employed in the past to obtain relaxed InGaN buffer layers including graded InGaN transition film and growth of InGaN on micrometer-size tiles of porousified GaN. Growth on oxide substrates, such as ScAlMgO4 and ZnO have also been explored as an alternative. ZnO has the same crystal structure as (In,Ga)N (wurtzite) and is in-plane lattice-matched with In0.2Ga0.8N [1]. Recently, we demonstrated growth of high-quality ∼300 nm-thick (In,Ga)N films with the In content varying from 11% to 23% on ZnO [2]. For this purpose, we developed low-temperature atomically smooth GaN thin films (~2 monolayers (ML)) to suppress the interfacial reaction between nitrides and the ZnO substrate at elevated temperatures using metal-enhanced epitaxy (MEE). In this talk, we will present successful growth of high-quality 1µm-thick InGaN with In content ranging from 19.5% to 30.5% on ZnO substrate by plasma-assisted molecular beam epitaxy. PL analysis showed a relatively uniform InGaN composition on samples grown at 600 °C and 580 °C. Lowering the growth temperature to 560 °C led to clusters with higher In content. The uniformity of low temperature InGaN growth was improved by periodic desorption of In from the surface. Although, STEM image showed a high threading dislocation density generated at the substrate interface, TDD reduced significantly through annihilation of dislocations as InGaN thickness was increased. A self-assembled superlattice structure of the InGaN on ZnO was confirmed by scanning transmission electron microscopy and atom probe tomography. We will show the relaxation of InGaN on ZnO by XRD reciprocal space map measurement. Reference: 1. A. Kobayashi et al. Pulsed sputtering epitaxial growth on m-plane InGaN lattice matched to ZnO, Sci. Rep. 7 1-6 (2017). 2. K Khan, Growth of high quality (In,Ga)N films on O-face ZnO substrates by plasma-assisted molecular beam epitaxy, AIP Adv. 10 075120 (2020)

Investigation of 1µm-Thick InGaN Films Grown on O-face ZnO by Plasma-Assisted Molecular Beam Epitaxy

Abstract Body: Transition-metal dichalcogenides (TMDs) have attracted a lot of interest owing to their direct bandgap as they are reduced down to thickness of a monolayer (ML) [1]. Thus, they have potential for numerous applications in electronic and optoelectronic devices such as transistors, photodetection and sensing [2]. MoS2 is probably one of most investigated TMD material due to the ability to tune its band structure with applied strain or electric field [3, 4]. As a naturally occurring crystalline mineral, tuning the properties of MoS2 as a representative TMD material provides further insights into the effects of strain and doping on its properties before moving on to investigate rare and more complex TMD materials. The strain and doping in MoS2 can be modulated in various ways such as mechanical bending [5], use of different substrates [6] or by introducing a metal- MoS2 interface [7]. A metal-semiconductor interface provides an interesting platform to investigate the plasmon-exciton interactions and the charge transfer mechanism between the metal and the semiconductor. In this work, we create plasmonic hot spots between gold nanoparticles to modify the Raman and luminescence response of ML MoS2. Nominal 2, 4 and 7 nm of Au and 2nm of Al were evaporated on SiO2/Si (oxidized silicon). Au and Al do not wet well SiO2, resulting in the formation of nanoparticles. Au and Al nanoparticle films exhibit a plasmonic band centered around 550 nm and 300 nm respectively. Plasmonic hot spots are created upon excitation within the band of plasmonic response of the nanoparticles. Monolayers of MoS2 were mechanically exfoliated directly on the nanostructured metallic layers and on SiO2/Si as a reference sample. Raman spectroscopy and photoluminescence were performed to characterize the optical response of the MoS2 to the plasmonic hot spots and compared to the reference sample. By varying the thickness of the Au film from 2nm to 7nm, we observe a shift and splitting of the E12g Raman modes into E12g and E12g’. The A1g mode also red shifts. The shift and splitting are only observed when the plasmonic resonance is excited during the optical measurements, indicating a plasmonic hot spot effect. We also investigate the effect on the photoluminescence (PL). In comparison to the reference ML MoS2, the PL intensity from the ML MoS2 on Au is drastically reduced. As the thickness of the Au film increases, the PL emission from A and B excitonic peaks is quenched until the B-exciton is no longer seen for ML MoS2 on a 7nm Au film. This can be explained by the electron transfer from MoS2 to Au, which hinders the recombination of electron−hole pairs created by the photoexcitation. As the work function of MoS2 is lower than that of Au, the electrons in the excited state of MoS2 transfer to Au, leaving a hole behind, which causes p-doping in MoS2. The charge transfer phenomena, along with the strain effects on optoelectronic properties of MoS2 MLs will be presented and dicussed. This work shows the potential of combining plasmonics with 2D materials to modify their optical and phonon response. [1] Choi, W. et al. Mater. Today 20, 116–130 (2017). [2] Lorenz, T., Joswig, J.-O. & Seifert, G. 2D Mater. 1, 011007 (2014). [3] Castellanos-Gomez, A. et. Al. Nano Lett. 13, 5361–5366 (2013). [4] Chakraborty, B. et al. Phys. Rev. B 85, 161403 (2012). [5] Conley, H. J. et al. Nano Lett. 13, 3626–3630 (2013). [6] Sun, Y., Wang, R. & Liu, K. Appl. Phys. Rev. 4, 011301 (2017). [7] Moe, Y. A., Sun, Y., Ye, H., Liu, K. & Wang, R. ACS Appl. Mater. Interfaces 10, 40246–40254 (2018). Acknowledgement: This work is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 765075 (Project LIMQUET). Funding from Swiss National Science Foundation (project 196948) is also acknowledged.

The Impact of Localized Hot Spots on the Raman Response of Monolayer MoS2

Abstract Body: INTRODUCTION: Graphene – an atomically thin layer of carbon atoms arranged in a hexagonal lattice – has gained interest as a bioscaffold for tissue engineering for its exceptional mechanical, electrical, and thermal properties1-5. Graphene’s structure and properties– tightly coupled to synthesis and processing conditions4– are thought to influence biomolecular interactions at the graphene-cell interfaces. In this study, C2C12 cells, a multipotent mouse myoblast cell line, were grown on graphene bioscaffolds with specific structure–property–processing–performance (SP3) correlations. We found that SP3 correlations significantly influenced C2C12 differentiation, myotube formation, and gene expression, suggesting that graphene–cell interfaces can be engineered to control biomolecule structure and function in adherent cells. The experiments performed are the first to make a direct comparison on scaffold SP3 correlations and their impact on genetic expression of multipotent myoblast cells. METHODS: Chemical vapor deposition (CVD) grown graphene film, transferred to quartz glass slides, resulted with surface roughness (Ra) of 2.37 nm and graphene ink– produced by solvent assisted exfoliation– was printed using a Dimatix inkjet printer onto glass coverslips (Ra of 10.0 nm). Epitaxial graphene on SiC wafers, produced by Si sublimation (Istituto Italiano de Tecnologia, Pisa, Italy) (Ra of 2.24 nm). Prior to cell integration, cell attachment was facilitated with agarose gel applied to the edges of all graphene bioscaffolds and corresponding control bioscaffolds (quartz glass, glass coverslips and SiC wafers). C2C12 cells, maintained in a humidified incubator at 37°C and 5% CO2, were seeded onto the scaffolds at 1.3 x 104 cells/cm2 in Dulbecco’s Modified Eagle Medium (DMEM) with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S). Cell growth and proliferation were monitored by transmitted light microscopy with images captured to study cell morphology. Gene expression patterns for each bioscaffold were analyzed for muscle myogenesis via qRT-PCR using a PCR array for muscle myogenesis and myopathy. RESULTS: Preliminary results show differences in cell morphology varied in a manner dependent upon the specific bioscaffold used. Myotube-like structures were evident in the control, CVD and epitaxial graphene (Fig.1A). Each of the three graphene bioscaffolds resulted in a unique overall genetic profile related to muscle myogenesis. However, four up-regulated genes were common to all biosaffolds when compared to their control. These genes are as follows: dystrophin (Dmd), myostatin (Mstn), leptin (Lep), pyruvate dehydrogenase kinase isoenzyme 4 (Pdk4) and Utrophin (Utm) (Fig. 1B). DISCUSSION: This preliminary work successfully shows that epitaxial, CVD, and printed graphene bioscaffold’s SP3 correlations impact the gene expression of C2C12 cells in culture. Each graphene bioscaffold presented with different and unique muscle gene expression profiles when compared to the control. The different morphologies of the cells as viewed in the transmitted light microscopy images confirms that the C2C12 cells grown on each bioscaffold, are uniquely different than the control. The following biomarkers: dystrophin, myostatin, leptin, pyruvate dehydrogenase kinase isoenzyme 4, and Utrophin were all upregulated in each of the graphene bioscaffolds. Epitaxial graphene shows the most genes upregulated that correlate to the muscle genotype followed by printed graphene and CVD graphene which showed the least number of upregulated genes relating to muscle (Fig. 1C). REFERENCES: 1. Nayak et al., ACS Nano., 2011. 2. Li et al., Sci Rep., 2013. 3. Krueger et al., ACS Biomaterials Science & Engineering, 2016. 4. Novoselov et al., Nature, (2012): 192-200. 5. Nieto et al., Advanced Functional Materials, 2015. ACKNOWLEDGEMENTS: Support kindly provided by NIGMS under P20GM103408 and P20GM109095 and NSF under 0619793 and 0923535

Differential Gene Expression in C2C12 Cells Due to Scaffold Structure-Property-Processing-Performance Correlations

Abstract Body: Graphene plasmonic resonances have been broadly studied in the terahertz and mid-infrared (MIR) ranges because of their electrical tunability and large confinement factors, which can enable dramatic enhancement of light-matter interactions, among other optoelectrical applications. Plasmonic resonances occur in graphene when it is patterned into nanostructures (e.g. ribbons) which confine the graphene plasmons and enables coupling to free-space radiation. For nanostructures with dimensions greater than 15 nanometers, the characteristic scaling laws of the graphene plasmons are well approximated by the local model of graphene’s conductivity. However, we experimentally demonstrate that those scaling laws change for structures smaller than 15 nanometers due to non-local and quantum effects in the graphene. Those changes blue-shift the frequencies of the graphene plasmons, pushing their operational range from the MIR into the short-wave-infrared (SWIR). We realize these effects in centimeter-scale arrays of graphene nanoribbons as narrow as 12 nanometers that are created using a novel, bottom-up block copolymer lithography method. The devices we create exhibit tunable plasmonic resonances at wavelength as short as 2.2 μm, almost double the energy of the previous experimental works. Furthermore, the confinement factors of our resonators reach 137 ± 25, which is amongst the largest reported in literature for any type of 2D optical resonator. The combined SWIR response and large confinement factors coupled to a scalable and efficient fabrication method makes these graphene nanostructures an attractive platform to explore next-generation optoelectrical devices and strongly enhanced spontaneous emission in the SWIR.

Large-Area Arrays of Short-Wave-Infrared Graphene Plasmonic Resonators Exhibiting Strong Non-Local and Electron Quantization Effects

Abstract Body: Transition metal dichalcogenide (TMD) semiconductors are layered van der Walls materials that exhibit exceptional optoelectronic properties in monolayer form. Their atomically thin nature and reduced long-range dielectric screening make them ideal systems in which to study many-body electronic states. Notably, excitonic many-body phenomena govern the light-matter interactions in TMD semiconductors and open up new possibilities for novel optoelectronic devices, including new classes of quantum light sources. Thus, understanding the rich suite of many-body phenomena in these systems is motivated by potential applications in on-chip optoelectronics and by the insights that can be gained into the fundamental interactions that govern TMD material properties. Here, the dynamics of several higher-order excitons including trions, biexcitons, and charged biexcitons in monolayer-WSe2 are probed using temperature-, energy-, and power-dependent time-resolved optical spectroscopy. We find that the formation of multiexciton complexes is enhanced by increasing the optical excitation energy for a given density. This enhancement is attributed to formation from an electron-hole plasma which generates 200% more multiexcitons than a lower-energy exciton gas. We further find that the multiexciton enhancement is robust over many samples and diminished with increasing temperature. These studies reveal a complex interplay between the multiexcitons and single-excitons that depends on both the density and excitation energy of the initial exciton population and open new doors to a number of applications such as entangle photon pair production and lasing, highlighting the importance of understanding the formation and relaxation dynamics of the rich manifold of excitons in order to leverage 2D semiconductors for advanced technologies.

Enhanced Multiexciton Formation by an Electron-Hole Plasma in 2D Semiconductors

Abstract Body: Semiconducting two-dimensional (2D) materials possess intriguing optoelectronic properties owing to the unique layered confined structure, such as strong light-matter interaction and large exciton binding energies. Especially, their excitonic binding energies far exceeding thermal energy will allow for room-temperature optoelectronic applications, such as excitonic computing applications and optical interconnects. It is therefore of significant importance to engineer excitonic properties and dynamics. The most widely used approaches to modulate the semiconducting 2D materials' excitonic dynamics include modifying the dielectric environment,[1] applying strain,[2] and gate-tuning.[3] In this study, we study modulation of exciton dynamics in monolayer MoS2 via dielectric Coulomb engineering and local strain by correlating photoluminescence (PL) mapping and scanning probe microscopy characterization. A mixed-dimensional heterojunction (MDHJ) was fabricated consisting of monolayer MoS2, sub-μm 0-dimensional P(VDF-TrFE) copolymer islands, and hBN. An optical microscope image, AFM topography map, and side view schematic of the MDJS are showing in Figures 1a, 1b, and 1c, respectively. Photoluminescence (PL) maps were obtained; A and B exciton emissions were fitted to Gaussians for parameter extraction (Figure 2a and 2b) and correlated with the topography data (Figure 2c). A clear blueshift of A exciton emission energy was observed on 1L-MoS2 on P(VDF-TrFE) islands with respect to 1L-MoS2 on hBN (Figure 2b). MoS2 on the polymer islands (red scatter plot; draped over polymer islands) showed a noticeable blueshift of ~3 meV compared to that of planar MoS2 on hBN (blue scatter plot) (Figure 2c). The bandgap increase can be attributed to the band structure renormalization and binding energy decrease due to the strong dielectric screening of the P(VDF-TrFE) substrate (εr ≈ 13).[1] Effects of dielectric environment and local strain on excitonic dynamics were examined by analyzing the A and B exciton intensity ratio; a high A-to-B intensity ratio is a sign of low nonradiative recombination.[4] Regions of increased A/B ratio (up to 2x) correlate well with the P(VDF-TrFE) islands locations (Figure 3a). Suppressed B exciton from the MoS2/P(VDF-TrFE) region contributes to the majority of the A/B ratio enhancement. Moreover, the B intensity showed a negative correlation with height, suggesting that large substrate screening combined with higher strain from the morphology enhances the intravalley scattering from the B exciton state to the A exciton state (Figure 3b). The A and B exciton energy difference map (Figure 3c) demonstrates that in contact with P(VDF-TrFE), the spin-orbit coupling-induced valence band splitting reduces, which may lead to an increased intravalley scattering rate. The correlation of PL and topography map revealed that the substrate dielectric screening and local strain affect the exciton dynamics of monolayer MoS2. [1] A. Raja et al., "Coulomb engineering of the bandgap and excitons in two-dimensional materials," Nat. Commun., vol. 8, no. May, p. 15251, May 2017. [2] A. Castellanos-Gomez et al., "Local Strain Engineering in Atomically Thin MoS 2," Nano Lett., vol. 13, no. 11, pp. 5361–5366, Nov. 2013. [3] Z. Qiu et al., "Giant gate-tunable bandgap renormalization and excitonic effects in a 2D semiconductor," Sci. Adv., vol. 5, no. 7, p. eaaw2347, Jul. 2019. [4] K. M. McCreary, A. T. Hanbicki, S. V. Sivaram, and B. T. Jonker, "A- and B-exciton photoluminescence intensity ratio as a measure of sample quality for transition metal dichalcogenide monolayers," APL Mater., vol. 6, no. 11, p. 111106, Nov. 2018.

Dielectric Effects on Exciton Emission of MoS2/P(VDF-TrFE) Heterostructure—A Correlated Mapping Study

Abstract Body: Abstract The two-dimensional insulating material hexagonal boron nitride (h-BN) has attracted much attention for resistive switching applications, yet its performance needs to be improved. In this work we present our comprehensive study of the resistive switching behavior of monolayer h-BN films with different electrode materials. The h-BN monolayers and h-BN/graphene heterostructure are synthesized on Ni substrates using molecular beam epitaxy (MBE), then Ni and Al are deposited as top electrodes. Three types of resistive switching devices are fabricated, namely, the symmetric metal-insulator-metal (MIM) device with Ni/h-BN/Ni, asymmetric MIM devices with Al/h-BN/Ni, and metal-insulator-graphene devices with Ni/h-BN/graphene and Al/h-BN/graphene structures. The symmetric Ni/h-BN/Ni devices demonstrate a polarity independent bipolar resistive switching behavior, a good durance with up to 97 cycles at 100 mA compliance current, an average on/off window of 103, and a low set and reset voltage variability. The asymmetric Al/h-BN/Ni devices demonstrate polarity dependent set voltage, with the larger set voltage observed when positive bias is applied on Ni. The metal/h-BN/graphene devices demonstrate self-compliant current bipolar resistive switching. The formation of conductive filaments is attributed to the diffusion of metallic ions and trapped charges. Our work reveals the influence of electrodes on resistive switching devices and provide ways to control the resistive switching behavior of two-dimensional h-BN based devices.

Resistive Switching Properties of Single Layer h-BN Film with Different Electrodes

Abstract Body: Graphene microstructures have been shown to exhibit controllable directional electron emission when carrying a current under an accelerating electric field. Through Phonon-Assisted Electron Emission (PAEE), graphene microstructures demonstrate electron emission at electric field intensities and lattice temperatures below what would be expected for carbon to demonstrate field emission or thermionic emission, respectively. Furthermore, such emissions tend to be out of the plane of the graphene microstructure, giving some control of the directionality of the emission current even before being directed with an applied electric field Thus far, such arrangements have tended to involve structural dimensions in the range of a handful of μm, commonly with CVD transferred graphene, flakes, or carbon nanotubes, with applied fields that can get up to the order of hundreds of kV/cm, and emission currents rarely exceeding a few nA. Herein are some examples of the characteristics and tunable variables demonstrated by multiple graphene structures with some dimensions up to cm, in fields below 4 kV/cm, and producing >1 μA emission currents. Quasi-freestanding epitaxial graphene on a silicon carbide substrate offers several advantages compared to transferred graphene or unzipped nanotubes, including lower defect density, greater structural integrity, ease of handling, and compatibility with simple photolithography fabrication techniques. The devices presented here were fabricated with a low-cost, wafer scalable, single-photomask process that does not require material transfer to a secondary substrate and is easily tailored to create variations in design parameters, which allowed multiple batches of devices to be rapidly manufactured with precise variations for the purpose of performance testing. The devices exhibited recognizable patterns of emission that seem to be influenced by several factors including, but not limited to shape and orientation, resistivity, input power, active time and on/off cycling frequency, device and substrate temperature, applied electric field, and both total device length and width dimensions (relative to current flow). Emission current seemed to show its most substantial dependence on input power, sometimes as much as doubling output with only a 25% increase in input; it is unknown if this relation holds for devices that vary other parameters alongside power input, necessitating further testing. The devices strongly imply a direct relationship between emission current density and device and substrate temperature, as some devices produced current outputs that exceeded measuring limits. Devices with larger dimensions also tended to produce more significant emission currents. However, the trends between increasing length and increasing width were not symmetrical: increasing device length (along the path of the input current) showed a stronger corresponding increase in output relative to % change in dimension than did changes in device width, and preliminary simulation results indicate both factors to have a roughly linear positive trend. While a general trend of larger dimensions and higher applied power levels tended to accompany higher emission currents, the output variations that may be attributed to other variables will allow for a broader range of device optimization parameters or tailoring of devices for specific purposes. Graphene microstructure electron sources may provide a means for device implementation in a 2D heterostructure environment or as a vehicle for further miniaturization of constructs requiring controllable electron emissions, such as electron microscopy or X-ray generation.

Downloads

Next from EMC 2021

Experimentally-Calibrated Modeling of Molecular Beam Epitaxy Selective Area Regrowth

Similar lecture

Si Dopants Enable Catalyst-Free n-Type GaAs Nanowire Growth by Modifying Facet Energies

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EMC 2021

Experimentally-Calibrated Modeling of Molecular Beam Epitaxy Selective Area Regrowth

Similar lecture

Si Dopants Enable Catalyst-Free n-Type GaAs Nanowire Growth by Modifying Facet Energies

Downloads