Symposium MD1 : Materials, Interfaces and Devices by Design

In this talk, we stress that big data of materials are structured in a way that is typically not visible by standard tools. Furthermore, with respect to a certain (desired) property, the practically infinite-dimensional space of different materials is very sparsely populated. Indeed, the key issue in data-driven materials science is to find the proper descriptive parameters (descriptors) that characterize the materials and their property.
We will show that and how compressed sensing, originally designed for representing a complex signal in the lowest possible dimensionality, can select, out of a huge-dimensional space of potential descriptors (features), a low dimensional descriptor. Examples are crystal-structure and stability prediction and the prediction of the band gap of binary and ternary compounds.
By applying stability analysis, supervised pattern discovery, and causal inference techniques, we discuss the causal relationship between the selected descriptors and the predicted physical properties.
(*) in collaboration with Mario Boley (FHI and MPII Saarbrücken), Emre Ahmetcik (FHI), Christopher Sutton (FHI), Runhai Ouyang (FHI), Jan Vybiral (Charles University, Prague, Czech Republic), Benedikt Hoock (Humboldt University, Berlin), Karsten Hannewald (Humboldt University, Berlin), Claudia Draxl (Humboldt University, Berlin)

We are actively developing various data driven techniques to facilitate and improve the way we implement experimental combinatorial materials science in technologically relevant topics. The major component of this overall effort is the integrated materials discovery engine, where we apply a dual track high-throughput experiment-and-theory approach to exploring and discovering new materials with frequent feedback between the two tracks. A cornerstone of our approach is the automated analysis of x-ray diffraction patterns taken from hundreds of spots across composition spreads. We have developed the metric approach where the similarity matrix is calculated using a variety of machine learning metrics in identifying the optimum algorithm for performing clustering. This effort is then combined with automated peak indexing of known phases. Finally, the experimental patterns are rapidly cross-referenced against peaks from predicted structures from AFLOWLIB. We will also discuss our latest effort to mine a large amount of experimental data taken from the literature. This work is carried out in collaboration with A. G. Kusne, S. Curtarolo, Y. Iwasaki, and A. Mehta, and it is funded by ONR, NSF, NIST and DOE.

This presentation presents and overview of how the use of manifold learning methods can be used not only to search for targeted properties among an existing database but to also discover new structure-chemistry-property relationships. The value of using these methods is to identify pathways for discovery. The focus of this talk is to advance a new paradigm in which one explores the structure of data to ascertain and uncover relationships that would not be easily identified via computational and/or experimental methods. Examples are given in crystal chemistry, alloy design and chemical imaging.

Accelerating the discovery and deployment cycle of new advanced materials is essential to meeting our energy and technology needs. Advancements in materials science has led to high-throughput materials synthesis and characterization, able of outputting a large number of materials a day, each of them associated with complex characterization information. As the data are being collected at faster and faster rates, the success of this approach relies on efficient, robust and scalable automated analysis techniques in order to interpret all the collected data and identify the next-generation materials. In collaboration with materials scientists at Cornell and Caltech, we introduce a novel approach that analyzes composition and structure data from a combinatorial library, and identifies the constituent phases and the corresponding phase diagram. The novelty lies in the integration of complex a priori scientific domain knowledge into state-of-the-art combinatorial optimization techniques and how experts in materials science as well as complete novices can provide valuable input to this solution approach.

Building block self-assembly has enabled the fabrication of a wealth of functional materials with applications as diverse as photonics and “smart” drug delivery systems. The inherent many-body nature of self-assembly has inhibited the construction of good order parameters to describe the configurational phase space of aggregate structures, making it difficult to probe the underlying mechanisms and thermodynamics driving assembly and rationally engineer materials and conditions to assemble desired structures. Utilizing our prior work applying nonlinear machine learning to computer simulations of self-assembly, we have applied machine learning directly to experimental particle tracking data for the non-equilibrium self-assembly of metallodielectric Janus particles under an applied oscillating electric field. Representing the first application of machine learning directly to experimental self-assembly data, we recover the set of collective order parameters quantifying aggregate morphology and structural transitions to infer the underlying low-dimensional assembly landscape. By characterizing changes in this landscape, we quantitatively measure the impact of field strength, oscillation frequency, and salt concentration on assembly pathways and the stability of different self-assembled aggregates. By recovering low-dimensional assembly “roadmaps” directly from experimental data, our approach enables the rational design of experimental conditions to direct the assembly of desired terminal aggregates.

*Long et al. Soft Matter. 2015, 11, 8141-8153

Many technologically relevant materials are in face kinetically stabilized and not at their true thermodynamic free energy minimun, i.e., they are metastable. These metastable materials can show improved functionality over their thermodynacally stable counterparts. However, it is difficult at best to predict the formation pathways for non-thermodynamically stable materials. In this work, we aim to understand how the initial state of amorphous thin films of VO_x influences the final crystal structure of the film. We employ a range of synchrotron based techniques: grazing incidence pair distribution function analysis (GIPDF), x-ray absorption spectroscopy (XAS), and x-ray diffraction (XRD) during in-situ crystallization under different atmospheric conditions. We focus on amorphous VO_x films deposited by pulsed laser deposition (PLD) with varying laser repetition rates and oxygen partial pressures. These different synthesis conditions create several different amorphous structures, as identified from GIPDF and XAS analysis. In-situ 2D-XRD during film crystallization shows that the different starting amorphous structures result in the formation of different crystal structures or polymorphs. We use these results to relate the local structure of the inititial amorphous structures to the VO_x polymorph structure. The work is part of a larger effort to help better understand the pathways needed to synthesize different polymorphs.

Currently available techniques used to synthesize nanostructured materials do not provide a high degree of control over the final microstructure. The ability to manipulate the size and aspect ratio of grains in nanostructured films/coatings would give us unprecedented control over the mechanical behavior of such materials. Here, we present a bottom-up synthesis process to systematically alter the microstructure of TiAl (45 at.% Ti) films by controlling its crystallization from the amorphous phase using thin crystalline seed layers.

Amorphous TiAl thin films were deposited on Si/SiN_x substrates using DC magnetron sputtering at room temperature. Non-contiguous, crystalline layers of Ti and Al (thickness <2 nm) were deposited in between the amorphous layers of TiAl to serve as preferential grain nucleation sites (seeds). The films were subsequently heat treated in vacuum to obtain a crystalline microstructure. Transmission electron microscopy and x-ray diffraction were used to study the microstructure evolution and crystallization kinetics of these films.

We found that the seeds aided crystallization, with seeded films crystallizing faster than non-seeded films for the same annealing temperature. In addition, the final microstructure of the seeded films was in the nanocrystalline regime up to a temperature of 700^oC, whereas the unseeded films underwent uncontrolled recrystallization to form large microcrystalline grains. In situ TEM heating experiments showed that the seeds significantly increased the number of nucleation sites in the amorphous matrix, which led to a refinement in the final grain size. In contrast, unseeded films crystallized by the formation and coalescence of large single crystal islands. Plan view and cross-sectional TEM analysis revealed that, as anticipated, the final grain diameter scaled with the in-plane distance between the crystalline seeds, while the grain height scaled with the seed spacing along the thickness.

Semiconducting nanoparticles (quantum dots) offer a wide range of potential applications, due to their unique and size dependent physical and chemical properties. A major issue today concerns the make of such particles on an industrial scale with a sufficient control of the particle size, shape and polydispersity, which needs a good understanding of the formation mechanisms involved. To get such a fundamental insight into the undisturbed formation of CdS in aqueous solution, we have developed a free liquid jet setup which accesses a so far unexplored time regime from 20 µs up to 10 ms. The key advantages of this setup compared to capillary based setups are: 1) the access of very early stages (1000 times faster than in stopped-flow experiments), 2) the high time resolution (down to 10 µs), 3) no radiation damage in the sample, and 4) high quality data evaluation because of missing container scattering. Via SAXS experiments the morphology of the early particle states are accessible while simultaneously acquired WAXS patterns give insights into the evolution of the crystalline structure.
The results of both SAXS and WAXS studies show, that CdS quantum dots formation is along a non-classical two-step nucleation pathway starting with the formation of primary clusters driven by the fast diffusion of cadmium and sulfur ions in water. Further growth is by cluster attachment where the diffusion of the primary clusters appears as the growth-limiting factor. Temperature dependent data yield an Arrhenius-like diffusion with an activation energy of E_g=0.6 eV. During the entire observation period the growing particles are not jet fully crystalline.

Flexible electronics have generated tremendous interest among scientific and engineering society. Due to their light-weight, flexibility, large-area printability, and potential biodegradability, flexible electronics hold great potential in applications like wearable markets and medical devices. The ability to self-heal upon rupture or scratch can significantly improve the devices’ lifetime. Here we describe a new class of elastomer material with transition metal ions serving as the cross-linkers, which have showed excellent stretchability and self-healing ability. Furthermore, when integrated into organic field-effect transistors as a gate insulator, this new class of material exhibits stable capacitance in a wide range of frequencies up to 10⁵ Hz. The superior self-healing ability along with no mobile ion effects of this new dielectric material demonstrate a significant contribution to the area of flexible electronics.

Three-dimensional (3D) ZnO porous films composed of interconnected skeleton were fabricated successfully through atomic layer deposition method using carbon nanoparticles as template. After surface modification, they showed an excellent superhydrophobic property with a contact angle larger than 160° and a sliding angle less than 1°. Based on the superhydrophobic 3D ZnO porous films, self-cleaning ultraviolet photodetector devices were fabricated. The devices exhibited a rise time of 42.03 s, a recovery time of 5.84 s and a responsivity of 9 mA/W at a 5 V bias under low light illumination of 14.38 μW/cm². The mechanism for the enhanced ultraviolet photoresponse from ZnO porous films is discussed.

We developed and demonstrated a ZnO/ZnS/Au composite photoanode with significantly enhanced photoelectrochemical water-splitting performance, containing a ZnS interlayer and Au nanoparticles. The solar-to-hydrogen conversion efficiency of this ZnO/ZnS/Au eterostructure reached 0.21%, 3.5 times that of pristine ZnO. The comparison of the incident photon-to-current efficiency
(IPCE) and the photoresponse in the white and visible light regions further verified that the enhancement resulted from contributions of both UV and visible light. The modification of the Au NPs was shown to improve the photoelectrochemical (PEC) performance to both UV and visible light, as modification encouraged effective surface passivation and surface-plasmonresonance effects. The ZnS interlayer favored the movement of photogenerated electrons under UV light and hot electrons under visible light, causing their injection into ZnO; this simultaneously suppressed the electron-hole recombination at the photoanode-electrolyte interface. The optimized design of the interlayer
within plasmonic metal/semiconductor composite systems, as reported here, provided a facile and compatible photoelectrode configuration, enhancing the utilization efficiency of incident light for photoelectrochemical applications.

Digital colloids, a cluster of lock-and-key “halo” particles bound to the surface of a central particle, were recently proposed as a novel soft materials system for high-density information storage*. To deploy these colloids in real memory applications requires in depth understanding of the conformational stability and transition kinetics governing changes in configurational bit state. We apply nonlinear machine learning techniques to Brownian dynamics simulations of these digital colloids to extract the low-dimensional intrinsic manifold governing digital colloid morphology, thermodynamics, and kinetics. Parameterized by the key collective variables governing the colloidal dynamics, these low-dimensional maps provide a quantitative framework in which to understand configurational transition mechanisms and rationally engineer digital colloid structures. By focusing on the relative size ratio between halo particles and central particles, we highlight the size-dependent stability and transition mechanisms for the 2-state tetrahedral (N=4) and 30-state octahedral (N=6) digital colloids. Finally we extend our analysis to the larger icosahedral (N=12) digital colloid capable of storing 2.86 bytes per cluster, or approximately 1.4 terabytes in a tablespoon of material.

*Phillips et al. Soft Matter. 2014, 10, 7468-7479

Microwave annealing has been reported to be an efficient, rapid process that could improve the electrical performance of alloy thin films and could possibly be an ideal process to maximize conduction in AgTi thin film. The microwave annealing process was done on 110 nm thick films with composition of Ag-93% and Ti-7%. The metallic layer was prepared by co-sputtering onto a silicon surface and annealed at temperatures varying between 65-160 degrees Celsius in air. The four-point-probe analysis of the thin film confirmed an inverse relationship between sheet resistance and annealing time. Prior to annealing, the sheet resistance was measured to be approximately 4 - 5 Ω/square. After annealing, a decrease in sheet resistance was observed and measured to be approximately 0. 94 - 3 Ω/square, depending on annealing time. The properties had been studied further through Rutherford Backscattering Spectrometry (RBS) and X-ray Diffraction (XRD). The RBS spectra imply the formation of titanium oxide on the surface and XRD analysis suggests the change in the metal’s lattice constant which correlates to the amount of Ti in solution. Our results confirm a change in resistivity and implies microwave annealing is a rapid, low-temperature self-encapsulation process that has the ability to form titanium oxide on the AgTi surface, induce grain growth and reduced resistivity in the AgTi alloy. For these reasons, AgTi demonstrates a promising future in contact metallization for silicon based solar cells.

Complex doping schemes in RE₃Al₅O₁₂ (RE=rare earth element) garnet compounds have recently led to pronounced improvements in scintillator performance. Specifically, by admixing lutetium and yttrium aluminate garnets with gallium and gadolinium, the band-gap was altered in a manner that facilitated the removal of deleterious electron trapping associated with cation antisite defects. Here, we expand upon this initial work to systematically investigate the effect of substitutional admixing on the energy levels of band edges. Density functional theory (DFT) and hybrid density functional theory (HDFT) was used to survey potential admixing candidates that modify either the conduction band minimum (CBM) or valence band maximum (VBM). We considered two sets of compositions based on Lu₃B₅O₁₂ where B = Al, Ga, In, As, and Sb; and RE₃Al₅O₁₂, where RE = Lu, Gd, Dy, and Er. We found that admixing with various RE cations does not appreciably effect the band gap or band edges. In contrast, substituting Al with cations of dissimilar ionic radii has a profound impact on the band structure. We further show that certain dopants can be used to selectively modify only the CBM or the VBM. Specifically, Ga and In decrease the band-gap by lowering the CBM, while As and Sb decrease the band gap by raising the VBM, the relative change in band-gap is quantitatively validated by HDFT. These results demonstrate a powerful approach to quickly screen the impact of dopants on the electronic structure of scintillator compounds, identifying those dopants which alter the band edges in very specific ways to eliminate both electron and hole traps responsible for performance limitations. This approach should be broadly applicable for the optimization of electronic and optical performance for a wide range of compounds by tuning the VBM and CBM.

Generally, initial stress inside of material and processing stress during milling are the main causes to make 7075 components to deformed. VSR( vibratory stress relief) can plays an unexpected role in the residual stresses relaxation, specially is suitable to aluminum alloy components．The experimental method is that firstly putting the component on a thick and big platform made of cast iron, and then starting a polarization apparatus to find the sub-resonance region. Finally process of vibratory stress relief is completed under different frequency and time. By stress measurement of X-ray diffraction and 3D shape scanning, the experiments results show that vibratory energy can be added to high energy area of processing layer in materials, which perceives the materials of surface may be yielded, so VSR techenique shouled make surface high stress of component reduced effectively, after that the bending moment on surface materials is decreased, which means deformation of component will be controlled by VSR.

The potential of carbon nanotube (CNT) networks for high performance structures can be undercut by non-ideal network structures, including short CNT lengths, disorganized network architectures, and weak van der Waals bonds among CNTs. Converting many CNTs into larger, covalently-bonded graphitic structures offers a path to increased CNT-based material performance. In [H.Y. Jung et al., 2014], networks of single-walled CNTs were fused into multi-walled CNTs and graphitic nano-ribbons by applying moderate AC voltage pulses, greatly improving their thermal and electrical properties. Fusion into larger graphitic structures also offers potential for improving mechanical properties by replacing van der Waals connections with covalent bonds. However, optimizing the fusion process remains an open challenge, with both the ideal pre-fusion starting material and the ideal fusion process parameters still to be determined.
The present research focuses on how the design of the material’s pre-fusion structure and its preparation process affect mechanical properties after fusion; data mining will extract additional patterns from these and related simulation results. The CNTs’ proximity to and alignment with each other may affect the ease with which neighboring CNTs fuse, and CNT alignment relative to the network’s axis may affect how current interacts with the CNTs. CNT yarns are wet-twisted from 14 tex CNT tapes with a 1.2 N applied load and low, medium, and high process twists (0.5, 1.0, and 1.5 rot/mm) that result in three final yarn structures (12°, 18° and 23° average angles of twist). The twisted specimens are then fused by applying 500 cycles of 9 V AC voltage across a length of 7 mm at a frequency of 2.5 Hz in 10 mTorr of Ar gas. For each angle of twist, 7 mm-long as-spun unfused yarns, unfused yarns that were heated at 100°C for 24 hours to dry them, and fused yarns undergo tension testing.
The results show that fusion affects both stiffness and strength, and the change upon fusion depends on the initial angle of twist. Fusion increased the stiffness of the as-spun low and medium twist yarns by 22% and 23% respectively; the stiffness of the as-spun high twist yarns increased by only 7%. The improvement in stiffness is accompanied by a decrease in ultimate strength, likely reflecting reduced slip among fused CNTs. Whereas the strength of the low twist yarns decreased by only 27%, the strengths of the medium and high twist yarns declined by 53% and 58%, respectively. In contrast, heat-treated low, medium, and high twist yarns respectively have stiffness values that are only 11%, 5%, and -13% higher than those of the as-spun yarns described above. Since heating to fully dry the spun yarn does not replicate the results of fusion, the improvement upon fusion is attributed to electrical rather than drying effects of the fusion process. The results indicate that fusion offers an optimal improvement in structural performance for lower values of twist angle.

Transition metal dichalcogenides (TMDCs) are emerging two-dimensional materials for next generation field-effect transistors, light-emitting diodes, and biosensors. Molybdenum disulfide (MoS₂) differentiates itself from other 2-D materials, such as graphene, since it possesses an inherent bandgap and intrinsic photoluminescence. Adaptation of the solution and electronic properties of these 2-D materials hinges on functionalization at the surface to realize the broad applicability and smooth integration into devices. While reports of graphene functionalization are numerous, examples of the modulation of MoS₂ properties are relatively scarce and primarily focus on covalent modification or ion implantation methods which distort lattice symmetry. Non-covalent modification using polymers facilitates the fabrication of processible, electronically relevant nanomaterials not yet reported.

We present the synthesis of polymers bearing pendent tetrathiafulvalene (TTF) as physisorbing moieties to enhance the solution stability of chemically exfoliated MoS₂ nanosheets, and participate in ground state charge transfer and work function modulation of pristine MoS₂. Well-defined polymers with a methacrylate or polynorbornene backbone were synthesized with TTF incorporation ranging from 1-50 mole%. Notably, these TTF polymers display dimerization and a concerted two-electron transfer to MoS₂, while TTF itself (i.e., as a small molecule additive) undergoes two, one-electron transfer processes, supported by spectroelectrochemical experiments. These redox interactions lead to a change in the work function of pristine MoS₂, by 0.2-0.3 eV, as indicated by Kelvin-probe force microscopy. Experiments using chemically exfoliated MoS₂ nanosheets show that even low incorporation of TTF into a polymer scaffold extends the colloidal stability of the nanosheets by days, whereas unfunctionalized polymers and control suspensions undergo restacking within hours of resuspension with high incorporations of TTF extending the colloidal stability to months. Stabilization with polymers bearing a pi-electron rich pyrene functionality restacked within hours demonstrating that solution stability primarily depends upon sulfur-sulfur and sulfur-pi interactions. All experiments were performed in conjunction with density functional theory (DFT) calculations to probe TTF-polymer/MoS₂ surface interactions and allow insight into the mechanism of electronic modification.

Composites, which are made by combining two or more distinct materials and give superior characteristics that did not exhibited in individual components, have been utilized in diverse applications, covering nearly every area of industries. Composites can offer various advantages such as high strength, stiffness and lightness. The performance of composites is largely dependent on the structure and properties of the interface between different materials because the interface is the weakest point of composite. Among many joining technologies such as mechanical fastening, adhesive bonding and welding, the adhesives provide benefits over other joining technologies. In particular, epoxy resin provides high adhesion strength to weight ratio, chemical inertness, great flexibility and easy manufacturing. Despite the outstanding adhesion properties of epoxy resin, the understanding of adhesion mechanism of epoxy is still unclear. Thus, we investigated the mechanism of epoxy adhesive (DGEBA) on Fe(100) surface by calculating the adhesion energies and analyzing electronic structures using DFT calculation in this study.

Acknowledgement
This research was supported by Global Frontier Program through the Global Frontier Hybrid Interface Materials (GFHIM) of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (No. 2013M3A6B1078865 and No. 2013M3A6B1078869).

Alkali chalcometalates (with Cu, Ag, and Au) represent a large family of compounds with intriguing structural features and physical/chemical properties.^[1] The alkali chalcocuprates and -aurates are rather well investigated and a remarkable number of new representatives have been synthesized during the past two decades. In contrast, only a few alkali chalcoargentates have been reported up to date. All known, chalcoargentates are Ag^I and contain Q^2– (Q = S, Se, and Te) ligand, except AAgSe₄^[2] (A = Cs, Rb) with Se₄^2– ligand. The Q^2– containing alkali metal chalcoargentates were classified in terms of A₂Q-n(Ag₂Q) (A= alkali metal, Q = chalcogen atom).^[3] This series include, the KAgSe (n = 1),^[4] A₂Ag₄Q₃ (A = K, Rb; Q = S, Se and n = 2),^[5] the AAg₃Q₂ (A = Rb, Cs; Q = S, Se, Te; n = 3),^[6] the AAg₅Q₃ (A = Rb, Cs; Q = Se, Te; n = 5),^[3,7] the K₂Ag₁₂Q₇ (Q = Se, Te; n = 6),^[7b] and the CsAg₇Q₄ (Q = S, Se; n =7).^[8]
Here, we report synthesis, crystal structure, detailed chemical bonding analysis, and thermal stability of K₄[Ag₁₈Te₁₁], a novel potassium telluroargentate(I). K₄[Ag₁₈Te₁₁] crystallizes in the cubic space group Fm-3m (no. 225) with the cell parameter a = 18.6589(6) Å. The crystal structure contains a [Ag₁₈Te₁₁]^4– complex cubic 3D anionic framework characterized by polarcovalent Ag–Te bond and weak Ag^…Ag interactions supporting the complex anionic character of the title compound. The band structure suggests good thermoelectric properties (flat band below, steep band above the Fermi level, “rattling” of cations), possible superconductivity (flat and steep bands) and/or topological insulator properties. Electrical conductivity measurements are the subject of future studies in order to determine electric and thermal transport properties of K₄[Ag₁₈Te₁₁]. The title compound is thermally stable up to 450°C and chemically resistant in air, organic solvents and water, some inorganic acids and base, like HCl, H₂SO₄, and NH₄OH.

References
[1] a) O. Fuhr, S. Dehnen, D. Fenske, Chem. Soc. Rev. 2013, 42, 1871–1906; b) I. Chung, M. G. Kanatzidis, Chem. Mater. 2013, 26, 849–869.
[2] M. Emirdag, G. Schimek, J. Kolis, J. Chem. Crystallogr. 1998, 28, 705–711.
[3] J. Li, H.-Y. Guo, X. Zhang, M. G. Kanatzidis, J. Alloys Compd. 1995, 218, 1–4.
[4] G. Savelsberg, H. Schäfer, J. Less-Common Met. 1981, 80, 59–69.
[5] a) W. Bronger, H. Schils, J. Less-Common Met. 1982, 83, 287–291; b) W. Bronger, C. Burschka, Z. Anorg. Allg. Chem. 1976,425, 109–116.
[6] a) S. W. Klepp, W. Sparlinek, Z. Kristallogr. 1996, 211, 363; b) W. Bensch, P. Dürichen, Z. Kristallogr. New Cryst. Struct. 1997,212, 97; c) C. Burschka, W. Bronger, Z. Anorg. Allg. Chem. 1977,430, 61–65.
[7] a) M. Eanes, G. Schimek, J. Kolis, J. Chem. Crystallogr. 2000, 30, 223–226; b) M. Emirdag, G. L. Schimek, W. T. Pennington, J. W. Kolis, J. Alloys Compd. 1999, 282-295, 287–296.
[8] J. Huster, J. Alloys Compd. 1992, 183, 377–384.

This research explores the variation in metal-boride phase formation during microwave plasma enhanced chemical vapor deposition (PECVD) boriding on cemented tungsten carbide. An evaluation of the borides lends insight into various options for the boron strengthening of cemented tungsten carbide. Additionally, boriding as a pre-treatment by surface blocking of elemental cobalt for improved nanostructured diamond (NSD) growth and adhesion is examined. We show that surface temperatures from 800-1100 °C result in chemically discrete metal-boride surface layers consisting of WCoB, CoB and/or W₂CoB₂ with average hardness from 23-27 GPa and average elastic modulus of 600-730 GPa. A combination of glancing angle x-ray diffraction, x-ray photoelectron spectroscopy, nanoindentation and scratch testing was used to evaluate the surface composition and material properties. In all cases the surface was composed of hexagonal tungsten carbide, with a remainder of metal-borides; no elemental cobalt was observable on the surface at 800°C and was significantly reduced up to 1100°C. Distinct boride phases then underwent PECVD diamond deposition for nanodiamond film growth. Diamond growth was established on all boride surfaces prepared between 800-1100 °C. However, delamination occurred in all cases other than the W₂CoB₂ formed at 900 °C. Diamond surface coatings were evaluated with Raman spectroscopy, scanning electron microscopy and energy dispersive x-ray spectroscopy. Early results indicate that CVD boriding at 900 °C for W₂CoB₂ formation improves NSD adhesion without damaging the cemented carbide substrate.

Ultraviolet photodetectors have drawn extensive attentions due to their broad applications, including ozone hole sensing, flame detection, convert space-to-space communication, and water purification, etc [1]. In the past several decades, ZnO has been investigated for photodetectors in the UV range due to its wide direct band gap (3.37 eV), low defect density, and strong radiation hardness. And our group proposed some distinctive works, such as ZnO hollow sphere nanofilm and ZnS/ZnO biaxial nanobelt based ultraviolet photodetectors [2,3]. However, owing to the lack of high quality and stable ZnO p-n homojunction photodiodes, and high performance single Schottky barrier diode, the performance of ZnO-based photovoltaic UV photodetectors is still lower than expected and need much lower applying voltage and much higher responsivity. Up to now, approaches are available for the improving performance of existing photodetectors via modifying their active areas, especially surface plasmon [4,5].
To explore a simple and practicable approach to realizing high performance ZnO-based photovoltaic photodetectors, we proposed a new type ZnO photovoltaic photodetectors based on asymmetry Au#1-ZnO-Au#2 structure. Its responsivity at 0 V was significantly enhanced by increasing the asymmetric ratio between the two electrodes, and could reach as high as 20 mAW^-1with an asymmetric ratio of 20 : 1. This value is higher than that of other kinds of self-powered ZnO-based photodetectors such as p–n junctions, Schottky junctions, and heterojunctions [6]. The photoresponse of our self-powered device was very fast, highly stable and reproducible. The 10-90% response time and decay time were measured to be ~710 ns and ~4 μs, respectively. In addition, the mechanism of as fabricated devices is mainly derived from artifical modulation of Schottky barrier and surface charge effect, and we will discuss about it in detail in the following.



Reference

[1] H. Y. Chen et al. Mater. Today (2015), http://dx.doi.org/10.1016/j.mattod.2015.06.001.
[2] M. Chen et al. Small, 2011, 7(17): 2449-2453.
[3] L. F. Hu et al. Adv. Mater., 2012, 24(17): 2305-2309.
[4] H. Y. Chen et al. J Phys. Chem. C, 2013, 118(1): 679-684.
[5] H. Y. Chen et al. Appl. Phys. Lett., 2014, 104(9): 091119.
[6] H. Y. Chen et al. J Mater. Chem. C, 2014, 2(45): 9689-9694.

Two-dimensional materials have become the focus of much attention in recent years - due to their fascinating physical properties and potential for novel technological applications. While graphene, boron nitride, and transition-metal dichaclogenides are widely studied, there is broad potential for more promising additions. Here, we start from the systematic exploration of experimental databases of inorganic materials to search for new two-dimensional candidates, using first simple chemical and geometric criteria that provide an initial pool of ~5000 different candidates. These are then further studied with first-principles calculations, focusing first on their ease of exfolation, thermodynamic stability, and electronic properties.

In this presentation we will discuss the development of entropy descriptors for the AFLOWLIB.org ab-initio repository and the path leading to synthesis of the new family of entropy stabilized oxides.
[Nat. Comm. 6:8485 (2015)].

The discovery of Gum Metal, a TiNb based super-elastic, high strength alloy has produced a great deal of interest within the metallurgy community over the past decade. These alloys display numerous "super'' properties. Most notably the bulk alloys appear to deform near their ideal strength. Previous studies have indicated that deformation of a bulk material near ideal strength is possible if (1) dislocation motion can be made so difficult that dislocations remain immobile at ideal strength and, (2) the material is intrinsically ductile in the sense that the ideal strength is limited by a shear instability rather than a tensile instability. In TiNb alloys, this mix of properties is obtained by tuning the composition to be very near to a body-centered-cubic (BCC) to hexagonal-close-packed (HCP) transition, and is marked by a spreading of the dislocation core structures. It is interesting to explore if other alloys engineered to be near the BCC to HCP transition will also display "super'' properties.

Accordingly, we have begun exploration of LiMg disordered alloys. These are particularly attractive candidate alloys, as they have the potential to show a very high specific strength. First principles calculations of the elastic moduli of the alloys as a function of composition shows that LiMg alloys behave elastically similarly to TiNb alloys in that their shear moduli are reduced to near zero as the BCC to HCP transition is approached from the BCC side. Further, direct computation of the screw dislocation core structures reveals significant spreading. Both of these characteristics suggest that synthesis of a LiMg Gum Metal should be possible, and that these alloys have the potential to display "super'' properties as well. However, analysis of the ideal strength of the LiMg alloys suggests that they are intrinsically brittle. Strategies for reducing the brittleness are explored.

The rapid rise in available computing power has made it possible to accelerate materials design by using ab-initio methods to calculate properties for tens of thousands crystalline materials. The calculation of many of these properties requires the evaluation of an integral over the Brillouin zone, which is commonly approximated by sampling a regular grid of points, known as k-points, in reciprocal space. We have developed an automated tool for generating k-point grids that significantly accelerates the calculation of material properties compared to commonly used methods. Our tool, which is being made freely available to the public, is capable of generating highly efficient k-point grids in a fraction of a second for any crystalline material. We present an overview of our method, benchmark results, and a discussion of how it can be integrated in a high-througput computing environment.

The Dow Chemical Company is a world leader in process and product technology development with more than 6000 product families manufactured at nearly 200 sites. Dow has been active in the development of high throughput experimental techniques since the early 1990s and has developed unique approaches to experimental data integration. In the current global market environment, advanced data modeling, experimental design, and a greater critical understanding of the relationships between laboratory data and commercial process implementation are a necessity to be successful. Rapid screening techniques must be adapted to the ever increasing demands of relevant process conditions and customer applications. Combination of advanced data mining and analysis to simplify the experimental domain will play a significant role in future of improvements in rapid experimentation within commercial applications. This talk will present a discussion of the requirements for success in bringing new materials from the rapid screening laboratory through to commercialization and how the data analysis techniques can be used across multiple process scales.

Understanding the interphase behavior with respect to surface chemistry and underlying mechanisms of polymer chain dynamics and the impact on thermomechanical, optical, and dielectric properties is the key to prediction of properties and design of novel nanocomposite materials with tailored property combinations.

In this work, we present a data-driven heuristic approach to investigate interphase mechanism in polymer nanocomposites and how it facilitates prediction of optimal macroscale properties. We make use of collected data from NanoMine, a living, open-source data resource for polymer nanocomposites that builds upon the rich tapestry of data of polymer nanocomposites from experimental and simulation efforts in the literature and individual labs’ generated data. NanoMine stores raw material data reported from literature and recorded in labs, including the source of data, constituent characteristics, processing conditions, characterization methods, microstructure and nanophase dispersion, and measured macroscale properties. A wide range of material components have been included to date in terms of the types of polymer matrices, nanoparticles, and surface chemistry. We have also been recording property data that are both single valued (e.g., glass transition temperature), condition dependent (e.g., storage modulus vs. frequency), and microstructural images.We analyze how constituents and processing interact to achieve interphase response as well as desired microstructure and macroscale response.

To characterize interphase behavior, we concentrate on the deviation of interphase from the neat polymer and how such deviation is first obtained from the dispersion of nanophase as well as the macroscale property, and later used to design nanocomposites with desired properties. As case studies, we focus on thermomechanical and dielectric properties in nanocomposites. The interphase property is represented by a set of shift parameters to account for the change of polymer chain dynamics in interphase. Values of such shift parameters are subject to the types of constituents in the nanocomposites, which are obtained from statistical analysis and machine learning techniques on bulk composite property across a wide range of constituent combinations. For example, shift and broadening factors in elastic modulus are correlated with types of polymer matrix, particle and surface treatment and validated by comparison between experimental data and results from finite element simulation. Once these factors are determined for a particular type of nanocomposite, effect of dispersion can be introduced by design of experiments on microstructure and additional simulations with interphase property as an input to the model until an optimal property is reached. We demonstrate that a desired property can be predicted by taking a collection of reported data from the data resource, analyzing the data to obtain interphase property, and performing assistive physics-based modeling.

A major deficit in suitable dyes is stifling progress in the dye-sensitized solar cell (DSC) industry. Materials discovery strategies have afforded numerous new dyes; yet, corresponding solution-based DSC device performance has little improved upon 11% efficiency, achieved using the N719 dye over two decades ago. Research on these dyes has nevertheless revealed relationships between the molecular structure of dyes and their associated DSC efficiency.
Two molecular engineering approaches are presented in this talk, which illustrate how one can exploit structure-property relationships to design new DSC dyes.
A ‘top down’ approach involves large-scale data-mining to search for appropriate dye candidates [1]. Here, structure-property relationships for DSC dyes have been codified in the form of molecular dye design rules, which have been judiciously sequenced in an algorithm to enable large-scale data mining of dye structures with optimal DSC performance. This affords, for the first time, a DSC-specific dye-discovery strategy that predicts new classes of dyes from surveying a representative set of chemical space. A lead material from these predictions is experimentally validated, showing DSC efficiency that is comparable to many well-known organic dyes. This demonstrates the power of this approach; and with further development of this approach, the materials discovery of higher-performing materials is anticipated.
A ‘bottom up’ approach concerns case studies on families of well-known laser dyes that are transformed into functional DSC dyes using molecular engineering [2,3]. The underlying conceptual idea is to implement certain electronic structure changes in laser dyes, using molecular engineering, to make DSC-active dyes; while maintaining key property attributes of the parent laser dyes that are equally attractive to DSC applications. This requires a concerted experimental and computational approach, interleaving results from single crystal X-ray diffraction, UV-vis absorption spectroscopy, cyclic voltammetry, density functional theory, and time-dependent density functional theory. A comparison of the frontier molecular orbital energy levels with the conduction-band edge of the classic TiO₂ DSC photoanode and the redox potential of a DSC electrolyte, allows the prediction of these re-functionalized parent laser dyes as dye co-sensitizers for DSC applications.
References
[1] J. M. Cole, K. S. Low, H. Ozoe, P. Stathi, C. Kitamura, H. Kurata, P. Rudolf, T. Kawase, “Data Mining with Molecular Design Rules Identifies New Class of Dyes for Dye-Sensitised Solar Cells” Phys. Chem. Chem. Phys. 48 (2014) 26684-90.
[2] S. L. Bayliss, J. M. Cole, P. G. Waddell, S. McKechnie, X. Liu, “Predicting solar-cell dyes for co-sensitization”, J. Phys. Chem. C 118 (2014) 14082–14090
[3] F. A. Y. N. Schroeder, J. M. Cole, P. G. Waddell, S. McKechnie, Advanced Energy Materials 5 (2015) 1401728 (1-12).

Developing proton conducting solid electrolyte would decrease the operating temperature of solid oxide fuel cells in turn improving reliability and operational efficiency. In spite of extensive literature on proton conducting solid electrolytes, there is very little insight into fundamental questions such as (a) How dopants spatially organize at various dopant concentrations? (b) How spatial organization of dopants influence proton transport? (c) How disorder and strain in a material influence its ionic transport. We will present ab-initio results of over 160 perovskite compounds via high-throughput computations[1], benchmarked by neutron scattering, transport and microscopy measurements to draw novel insights into the aforementioned questions.

We discover from ab-initio modeling that dopants tend to cluster in Y doped BaZrO₃ at high dopant concentrations (>20%), in agreement with recent calorimetric experiments[2], and our inelastic neutron scattering experiments. The calculated proton transition energy for clustered dopants is higher (~ 0.6eV) compared to dispersed dopants (0.1- 0.35eV), and is in good agreement with our Kelvin probe force microscopy as well as previous conductivity measurements[3]. Computations further show that the proton trapping is due to the strong hydrogen bonding between proton and the oxygen sub-lattice. The generality of this phenomenon is established by performing data analytics on data obtained from over 160 perovskite compounds and over a dozen dopants. These insights in turn naturally lead to new design principles for improving proton conductivity in perovskites.

To explore the influence of disorder on proton transport, we performed high-throughput computations to generate realistic models of lanthanum tungstate – a disordered fluorite material in both wet and dry environments. The results obtained from these models matched well with the inelastic neutron scattering experiments. From these computational models we discover that the proton adsorption energy decreases as the distance between O and vacancy site increases. Similarly, the proton adsorption energy decreases when the O is coordinated with many W atoms. The high-throughput computations also enabled to identify novel routes of improving ion transport in these types of materials.

We will also present our ongoing developmental activities in scaling the high throughput methodology on OLCF supercomputing facility.

Acknowledgement
This work was sponsored by Laboratory Directed Research and Development Program of Oak Ridge National Laboratory (ORNL). Computations were performed at NERSC and OLCF supercomputing facilities. Neutron experiments were performed at Spallation Neutron Source (SNS) at ORNL. Transport measurements were performed at Center for Nanophase Materials Sciences (CNMS) at ORNL.

References
[1] S.P.Ong et.al Comp. Mat. Sci. 68, 314–319 (2013).
[2] M.D.Goncalves et.al J. Mater. Chem. A, 2, 17840 (2014).
[3] D.Pergolesi et.al Nat. Mat. 9, 846–852 (2010).

Three-dimensional (3-D) semiconductor nanoarchitectures, including nano- and micro- rods, pyramids, and disks, are emerging as one of the most promising elements for future optoelectronic devices. Since these 3-D semiconductor nanoarchitectures have many interesting unconventional properties, including the use of large light-emitting surface area and semipolar/nonpolar nano- or micro-facets, numerous studies reported on novel device applications of these 3-D nanoarchitectures. In particular, 3-D nanoarchitecture devices can have noticeably different current spreading characteristics compared with conventional thin film devices, due to their elaborate 3-D geometry. Utilizing this feature in a highly controlled manner, color-tunable light-emitting diodes (LEDs) were demonstrated by controlling the spatial distribution of current density over the multifaceted GaN LEDs. Meanwhile, for the fabrication of high brightness, single color emitting LEDs or laser diodes, uniform and high density of electrical current must be injected into the entire active layers of the nanoarchitecture devices. Here, we report on a new device structure to inject uniform and high density of electrical current through the 3-D semiconductor nanoarchitecture LEDs using metal core inside microtube LEDs.
In this work, we report the fabrications and characteristics of metal-cored coaxial GaN/In_xGa_1-xN microtube LEDs. For the fabrication of metal-cored microtube LEDs, GaN/In_xGa_1-xN/ZnO coaxial microtube LED arrays grown on an n-GaN/c-Al₂O₃ substrate were lifted-off from the substrate by wet chemical etching of sacrificial ZnO microtubes and SiO₂ layer. The chemically lifted-off layer of LEDs were then stamped upside down on another supporting substrates. Subsequently, Ti/Au and indium tin oxide were deposited on the inner shells of microtubes, forming n-type electrodes of the metal-cored LEDs. The device characteristics were investigated measuring electroluminescence and current–voltage characteristic curves and analyzed by computational modeling of current spreading characteristics.

The Office of Data and Informatics (ODI) is a recently established organization within the Material Measurement Laboratory at NIST. ODI is a premier, pioneering resource for researchers and institutions in the biological, chemical, and materials sciences who need to leverage both large and information-rich data sets now common in many disciplines; who are faced with challenges of handling, archiving, storing and analyzing such data; and who would transform such data into products that can be reliably and broadly shared and used for sophisticated scientific endeavors. The ODI supports National needs such as the Materials Genome Initiative (MGI) and biological and chemical data integration, as well as the modernization of current NIST reference data services for use in state-of-the-art computer paradigms (i.e., virtual computing, parallel analysis, interoperability, semantic web, etc.) and the development of next generation NIST reference data services. The ODI also facilitates MML's adherence to the government open-data policy by providing tools for developing data management plans and guidance and assistance in the best practices for archiving and annotating research and data outputs. We are also investigating software and hardware configurations to improve data collection and data processing for the diverse instrumentation operated at NIST.

In the area of materials science NIST supports a number of national and international initiatives:
o A shared data repository for collaborations related to the Materials Genome Initiative, deployed on the DSpace data management platform.
o The US National Data Services Consortium and the Materials Data Facility pilot project (I. Foster, U. Chicago, PI).
o The Research Data Alliance Materials Data, Infrastructure, and Interoperability Interest Group, from which a Working Group is being formed to develop an international materials science data resource registry. ODI is working with colleagues in the NIST Information Technology Laboratory (ITL) on the registry infrastructure.
o The NIST ITL is leading the development of the Materials Data Curation System, a tool that supports the detailed annotation of data sets.

The Materials Project (materialsproject.org) uses supercomputing and informatics to compute and disseminate properties of all known inorganic bulk solid materials. Through web pages and web-powered interfaces, the Materials Project currently makes available over 65,000 compounds and over 43,000 bandstructures as well as an unprecedented number of other data such as elastic tensors. With roughly 15,000 users, it is one of the success stories of the Materials Genome Initiative. To achieve this result, the team has put a great deal of work and thought into the underlying computing and informatics infrastructure. This talk will attempt to describe, in clear and straightforward language without the usual alphabet soup of technology acronyms, the essential aspects of this infrastructure and lessons learned in its implementation. The talk will conclude with a preview of some of the data mining efforts and the "user contributions" framework for integrating experimental and other types of external community data into the core database.

Many of the most relevant chemical properties of matter depend explicitly on atomistic details, rendering a first principles approach mandatory. Alas, even when using high-performance computers, brute force high-throughput screening of compounds with electronic structure theory is beyond any capacity for all but the simplest systems and properties due to the combinatorial nature of chemical space, i.e. all the compositional, constitutional, and conformational isomers.
Consequently, efficient exploration algorithms should exploit all implicit redundancies present in high-throughput approaches. In this talk, I will discuss recent contributions which show how to generate machine learning models of quantum mechanical observables, after training in chemical space. Results will be shown for models capable of predicting atomization enthalpies, excitation energies, molecular properties, cohesive energies of crystals, atomic forces, NMR parameters and others. All predictions are done for out-of-sample systems, with high accuracy, and negligible computational cost.

We will discuss some progress in predicting materials for solar energy conversion, and materials for quantum information technology, using ab initio calculations. In particular we will focus on heterogeneous interfaces between photo-electrodes and water and between nanocomposites, and on defects in semiconductors and insulators for qubit predictions. We will also address the problem of building much needed tighter connections between computational and laboratory experiments.

High throughput and combinatorial methods have been successfully applied to the discovery of functional materials, and we describe their application to the discovery of functional interfaces. In the expanding effort to bridge materials discovery to technology deployment, and under the recognition that many technologies rely on high performance material interfaces, the materials science community must apply their best experimental and theoretical tools to the challenging task of materials integration. The photoelectrochemical generation of solar fuel provides an important example of functional interfaces, as coatings are often applied to semiconductor light absorbers to enhance performance metrics ranging from light management to charge separation to electrocatalysis. In this presentation we describe a foundational effort in combinatorial materials integrations in which thousands of integrated catalyst-light absorber photoanodes are screened for a variety of optical and electrochemical performance metrics. In addition to identifying promising photoanodes for the oxygen evolution reaction (OER), the results clearly demonstrate that relating interface properties to the properties of the constituent materials is not straightforward, yielding substantial challenges and opportunities for the concept of interfaces by design.

Structure prediction of bulk materials is now routinely performed, however the field of predicting the atomic structure of interfaces and other defects is still in its infancy. A detailed understanding of and ability to predict the atomic structure of interfaces is however of crucial importance for many technologies. Interfaces are very hard to predict due to the complicated geometries, crystal orientations and possible non-stoichiometric conditions involved and provide a major challenge to structure prediction. We present here the ab initio random structure searching (AIRSS) method and how it can be used to predict the structure of interfaces. Our method relies on generating random structures in the vicinity of the interface and relaxing them within the framework of density functional theory. The method is simple, requiring only a small set of parameters that can be easily connected to the physics of the system of interest, and efficient, allowing for high-throughput first-principles calculations on modern parallel architectures. We focus here on the prediction of grain boundaries, but application to heterostructure interfaces is straightforward. Examples for several grain boundary defects in technologically important materials will be presented: In particular grain boundaries in graphene, the prototypical two-dimensional material will be discussed, alongside with examples of grain boundaries in transition metal oxides, such as SrTiO₃ and TiO₂. Direct comparison to experiments will be made.

The interface between functional nanostructures and host substrates is of pivotal importance in the design of their nanoelectronic applications as it conveys energy and information between the device and environment. We report here an interface- engineering approach to establish a seamless, electrically insulating, while thermally transparent interface between graphene and metal substrates by introducing water intercalation. Molecular dynamics simulations and first-principles calculations are performed to demonstrate this concept of design, showing that the presence of interfacial water layer helps to unfold wrinkles formed in the graphene membrane, insulate the electronic coupling between graphene and the substrate, and elevate the interfacial thermal conductance. The findings here lay the ground for a new class of nanoelectronics setups through interface engineering, which could lead to significant improvement in the performance of nanodevices such as the field-effect transistors.

Current collapse limits the output power of AlGaN/GaN HEMTs and is regarded as one of the most critical issues to be solved for high power switching applications. Two major sources of current collapse have been identified in surface states and traps in epitaxial layers. Surface states located near the gate edge can be charged while biased into Off-state. By switching on AlGaN/GaN HEMT, the charged surface states cannot be immediately released, and partial 2DEG in the channel is depleted, rendering the device with high on-resistance.

In this work, three types of AlGaN/GaN HEMTs on p-type silicon substrates with different cap layers including undoped GaN cap (5-nm), Mg-doped GaN cap (5-nm), and in-situ SiN cap (4.1-nm) grown by MOCVD were investigated in breakdown and dynamic on-resistance characteristics. The epitaxial layers include 300-nm GaN channel, 1-nm AlN, 20-nm AlGaN, and cap layers. Devices with Mg-doped GaN cap were activated at various conditions before fabrication using rapid thermal annealing. All devices were fabricated in Schottky gate with passivation of 200-nm SiN layer.

The measured threshold voltages and drain currents of devices with different cap layers showed the similar results, demonstrating the impact of cap layer on device on-state characteristics is limited. However, device with in-situ SiN cap showed the lowest gate leakage current, highest On/Off current ratio (1.7Χ10⁸), and highest breakdown voltage (1200V) with gate-drain distance of 20-µm. In terms of dynamic on-resistance, devices with Mg-doped GaN cap and in-situ SiN cap showed improved results than device with undoped GaN cap at high Off-state voltage (> 100 V). At low Off-state voltage, device with undoped GaN cap is better in dynamic on-resistance.

In Off-state to On-state time transition measurements using Off-state conditions with positive substrate bias were used to investigate the impact of cap layers in these structures (later trapping with negligible vertical trapping due to the drain-substrate bias being zero). The measured results showed longest relaxation time in device with undoped GaN cap but shortest relaxation time in device with in-situ SiN cap.

Natural gas which is mainly composed of methane (~95%) is an affordable and accessible vehicular energy source comparing conventional petroleum-based gasoline and diesel fuel. Up to present, two main technologies to store natural gas are either compressing it at very high pressures (CNG) or liquefying it at very low temperatures (LNG). To fulfill these extreme conditions, costly high-pressure facilities or cryogenic cylinders are needed which limit their use in practical application. Alternatively, adsorbed natural gas (ANG) can be implemented with a promising porous materials that can store and deliver large amount of methane at higher temperatures and lower pressures with respect to LNG and CNG methods. Metal-organic frameworks (MOFs) with high surface area and large pore volume have been widely studied as methane sorbent materials during last decade. However, they suffer from several drawbacks such as complicated synthesis process, moisture sensitivity and scalability. In the recent project we designed and synthesized a series of highly porous carbons by chemical activation of a single source precursor. We deliberately selected a porous organic polymer with a phosphazene core to generate more porosity by evolution of volatile species during activation. Four various samples synthesized at different temperatures which feature diverse textural properties such as surface area, micropore and total pore volume. Among all, one sample shows outstanding surface area and pore volume of 5000 m²/g and 2.6 cc/g, respectively. The effect of porous texture, packing density and pore size distribution of four representative carbon sorbents on their methane storage performance were thoroughly investigated. Our results showed that samples with higher level of narrow micropores show higher methane uptake at relatively low pressure (below 10 bar), while wider micropores or micro/meso hierarchy of pores lead to higher uptake at higher pressures. The highest surface excess uptake was found to be around 0.300 g/g at 298 K and 65 bar.

Sodium borosilicate glass composites with up to 40wt% of annealed pyrolitic graphite (APG) have been fabricated through a sol-gel process. The APG was first encapsulated in a layer of mesoporous silica with a tunable thickness, found to substantially aid in dispersion and stability of the APG in a variety of alcohols. The silica coating on the APG provides an ideal interface between APG and a silica-based glass, achievable via a sol-gel process, as the aqueous precursors afford the opportunity to establish a uniform dispersion of the silica-coated APG and subsequently form a network around the dispersion. Multi-component gels of SiO₂, B₂O₃, and Na₂O with varying weight percentages of APG were produced. Dried gels were hot pressed under vacuum to yield fully dense, homogeneous glass free of phase separation or devitrification. Both unloaded and loaded sol-gel derived glasses have shown similar values for density and CTE to that of a traditional melt-quenched glass of the same composition. Furthermore, both X-Ray diffraction patterns and FT-IR spectra have shown similarities of the unloaded sol-gel derived glass to its melt-quenched counterpart. Loading of the sodium borosilicate glasses with silica-coated APG holds promise for enhanced thermal and electrical conductivity of the bulk material, as the high aspect ratio of the APG encourages formation of percolating networks within the glass matrix. Encapsulation of nanoparticles with silica provides a route for facile surface manipulation and obtaining stable, homogeneous dispersions in solution, while a multi-component sol-gel is attractive for other nanocomposite systems in which particle settling or agglomeration is a concern.

Heterocomposites comprising a variety of micro- or nano-scaled functional components and interfaces have become a promising class of thermoelectric materials in these years. The abundantly formed interfaces involving various dimensions, topographies, structures and scales would not only significantly suppress the thermal conductivity (κ) via enhanced phonon scattering, but even selectively modulate charge carrier transporting in matters under specific conditions and thus finally achieve an applicable level of thermoelectric figure of merit ZT defined as S²σTκ^-1. In this study, a series of novel periodically-assembled Bi_xSb_2-xTe₃/Ga₂Te₃ and Bi_xSb_2-xTe₃/carbon hetero-nanocomposite films were successfully fabricated using a dual-beam pulsed laser deposition (DBPLD) system. The influence of the assembling architectures on the corresponding enhanced thermoelectric properties was discussed.

Composition near the surface of alloys is not often consistent with that of the bulk which can dominate material properties in nanoscale systems. Experimentally observing formation of solid solutions and surface segregation effects in nanoscale alloys is challenging. Using energy dispersive x-ray spectrometry (EDX) in a scanning transmission electron microscope (STEM) we have observed the formation of a solid solution within the miscibility gap of the SnSe-PbSe bulk phase diagram in bilayers interdigitated with TiSe₂. With increasing bilayers, we observe surface segregation at the interfaces. These systems were synthesized using a modulated elemental reactant (MER) method which provides a versatile diffusion limited synthesis approach for self-assembly of targeted kinetically stable products. It has been shown that the nanostructure of the deposited precursor is preserved in the final products. In this work we demonstrate that using MER synthesis does not significantly influence surface segregation effects in (Pb_xSn_1-xSe)_1+δTiSe₂ and is a useful method for studying nanoscale miscibility and surface segregation. Using atomically resolved EDX maps, we section and integrate data cubes and use statistical methods to demonstrate a formation of a solid solutions in bilayers and quantify surface segregation in atomic planes.

Acknowledgements:
The authors acknowledge support from the National Science Foundation under grant DMR-1266217, National Science Foundation through CCI grant number CHE-1102637 and Sandia National Laboratories, which is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. A portion of this work is part of the Chemical Imaging Initiative at Pacific Northwest National Laboratory (PNNL) under Contract DE-AC05-76RL01830 operated for DOE by Battelle. It was conducted under the Laboratory Directed Research and Development Program at PNNL. A portion of the research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy’s Office of Biological and Environmental Research and located at PNNL.

Many studies have been conducted to reduce the weight of products or structures while maintaining their performance. Due to global pressure to enhance energy efficiency, in particular automotive industries have tried to develop new composites made of steel and polymers, e.g., steel/polymer/steel or steel/fiber-reinforced plastics (FRP)/steel hybrid composite. By locating polymer or FRP core between steel layers, the weight of resulting structure has been demonstrated to be reduced maintaining its performance. In addition, the hybrid composites have showed improved damping and soundproof properties compared to pure steel structure. In this study, first tensile tests were carried out for evaluating the influences of manufacturing parameters of hybrid composites including pressure, temperature and curing time. Damping and soundproof tests were also carried out. Finally, the adhesion strength between steel and polymers or FRP was characterized to investigate its effect on the mechanical and acoustic properties of the hybrid composites.

We propose an new approach to the design of functional chalcogenide materials. Our approach synergistically exploits both Edisonian and bottom up design approaches. Our materials discovery program employs both compuational-based Genetic algorithm-led materials optimisation and combinatorial composition-spread materials screening methods. As a proof-of-concept our methods are applied to Sb2Te3-GeTe phase-change materials. We show through experiment and simulation that our approach to materials design can accelerate the optimisation and discovery of new superlattices of van der Waals bonded 2-D chalcogenide crystal superlattices.

The theory of devices using SOI like a spin-filter are very important in the field of spintronics. The spin-filtering devices can be used to generate and detect spin polarized currents. Many researchers have reported about the spin-filters using linear Rashba SOI. However, the spin-filters using square and cubic Rashba SOIs are not yet reported. This is probably because the Aharonov-Casher (AC) phases acquired under square and cubic Rashba SOIs are ambiguous. In this study, we try to derive the AC phases acquired under their Rashba SOIs, which we call general Rashba SOIs, using non-Abelian SU (2) gauge theory. We have successfully derived these AC phases without the completing square methods [1]. Using the above AC phases under general Rashba SOIs, we investigate the spin filtering in double quantum dots (QDs) Aharonov-Bohm (AB) ring under general Rashba SOIs using the methods of the reference [2]. The double QDs-AB ring consists of elongated QDs and quasi-one dimensional quantum nanowires (QNs) under magnetic field. The spin transport is investigated from left nanowire to right nanowire in this structure within tight binding approximation. In particular, we focus on the difference of spin filtering among general Rashba SOIs. The calculation is performed for the spin transmission with changing the penetrating magnetic flux. As a result, we have obtained the penetrating magnetic flux dependence of spin polarization for the AB ring subject to general Rashba SOIs. It is found that the perfect spin filtering is achieved for all the Rashba SOIs. This result indicates that the double QDs-AB ring under general Rashba SOI can be a promising device for spin current generation. Moreover, they behave in totally different ways in response to penetrating magnetic flux, which is attributed to linear, square, and cubic behaviors in the in-plane momentum. This result enables us to make a distinction among linear, square, and cubic Rashba SOIs according to the peak position. We believe that this fact is very useful for many researchers.

Reference
1. N. Hatano, R. Shirasaki, and H. Nakamura, Phys. Rev. A 75 (3) 032107 (2007).
2. A. Aharony, Y. Tokura, Z. G. Cohen, O. Entin-Wohlman, and S. Katsumoto, Phys. Rev. B 84 035323 (2011)

Composite strengthening technique was applied on 15CrMo steel to receive good high-temperature mechanical properties with enough wear and fatigue resistance surface, which has better work performance than that of regular casted H13 steel. First, high-alloyed steel was paved on 15CrMo steel by welding method with addition of nitrogen to replace part of carbon to achieve a high strength and ductility and thermo-stability substrate. Microstructure and formation mechanisms of the martensite and nitrides/carbides inclusions were studied to understand the strengthening mechanism of the hard-facing alloy. Second, by using arc ion plating, a thin TiCN coating was deposited on the hard-facing alloy to make further enhancement on its high temperature hardness and wear resistance. The interface properties of coating and hard-facing alloy is emphasized in this paper, because the bonding strength plays a key role in the life time of coated die steel in service.

Over the last 60 years, implantable electronic systems and devices have undergone a significant transformation, becoming a valuable biomedical tool for monitoring, measuring and soliciting physiological responses in vivo. Increased in vivo stability, miniaturization and lower energy requirement of modern electronics led to a multitude of miniature wireless electronic devices, such as sensors, intelligent gastric and cardiac pacemakers, cochlear implant, implantable cardioverter defibrillators, and deep brain, nerve, and bone stimulators being implanted in patients worldwide.
Where in vivo sensing or stimulation is required for a short period of time, degradable implantable electronic devices can provide a solution to overcome inflammation and infections associated with long-term implant utilization. Moreover, subsequent surgical removal of these devices can be avoided which will diminish the pain and cost to the patient. Currently, fabricating a complex high-performing electronic system from entirely biodegradable, non-toxic set of electronic materials is of growing interests. Devices made of various biological or synthetic organic active and passive materials for in-vivo applications have been reported. An important recent advance in this field is the combination of silicon electronics with biodegradable polymer platform which offers both the flexibility of the device and sufficient bulk degradation. In spite of many works detailing the route to combine silicon with other organic and/or inorganic materials forming transient integrated circuits, sensors, communication devices and stimulators, a reliable electrical power source that suitable for those transient implantable electronic devices still has not be experimentally investigated previously.
In this paper, we report important breakthroughs on the power source for transient electronics. We first demonstrated a set of materials, manufacturing schemes, device components for a biodegradable triboelectric nanogenerator (BD-TENG) which can harvesting mechanical energy in vivo and physically disappeared at prescribed times without any adverse long-term effects. Tunable electrical output capabilities and degradation features were achieved by fabricated BD-TENG using different materials.
When applying BD-TENG to power two complementary micro-grating electrodes, DC pulsed electrical field was generated and the nerve cell growth was successfully orientated, showing its feasibility for in-vivo neuron repairing. Our work demonstrates a significant potential of BD-TENG as a power source for transient implantable medical devices.

The unusual artificial materials, known as metamaterials, have been developed in many ways to express interesting phenomena, such as invisible clocks, negative Poison’s ratio. Here, we present a new type of acoustic metamaterials consisting of an array of micro scale Helmholtz resonators. By using photolithography and thermal imprint process, the film type resonator array was fabricated, and the final acoustic metamaterials was produced by a laminating process of films above. The fabricated metamaterials have a high acoustic absorption characteristic near the resonance frequency which can be modulated by the design of micro scale Helmholtz resonators. Results demonstrate that the attenuation value is largely determined by the number of laminated layers and stiffness of micro structured surface with the maximum attenuation being ~ 90 dB/mm, which is comparable to that of widely used particle based polymer (~30 dB/mm) currently used in the medical applications. As continuous production and large area fabrication are realized, this film type acoustic metamaterials may be applied to various field including military equipment and medical diagnosis devices.

Ammonia (NH₃) has been extensively used in the refrigeration systems, explosive, chemical and fertilizer industries and used for material processing. Exposure and inhalation of NH₃ molecules into the respiratory organs cause irritation, lungs damage and premature death in extreme cases. Therefore, safety is a key issue to design a reliable and robust NH₃ gas sensor to counter the accidental leakage.
A number of techniques such as sputtering, thermal evaporation, chemical vapour deposition and sol-gel have been used with suitable catalyst incorporations to develop a highly sensitive and selective sensor for the detection of ammonia to meet the global need.
We report the gas-sensing properties of ammonia by exploiting a spin coated SnO₂ thin film (200 nm) via sol-gel route on a corning glass patterned with platinised interdigitated electrodes. The bare SnO₂ film showed little sensing response (S ~ 1.9). Platinum cluster of about 10 nm decorated SnO₂ sensor is found to be highly sensitive (S ~ 33.24) at a temperature of 240°C with reduced response and recovery times showing an excellent gas sensing response characteristics towards 500 ppm NH₃. The improved sensitivity is attributed to the enhanced role of spill over mechanism. The sensor also showed excellent selectivity under different interferrant gases. The optical, structural and microstructural properties have been obtained by UV-VIS spectroscopy, X-ray diffraction (XRD), Atomic force microscopy (AFM), Field Emission Scanning Electron Microscope (FESEM) and Transmission Electron Microscopy (TEM) images. The morphological properties of the sensor structure are correlated to its sensing characteristics.
Keyword: Ammonia, spin coating, sensitivity
References:
1. Ammonia sensors and their applications—a review, Sensors and Actuators B 107, 666–677, (2005).
2. Metal clusters activated SnO₂ thin film for low level detection of NH₃ gas, Sensors and Actuators B: Chemical, 194, 410-418 (2014).

Metastable layered perovskites containing interlayer transition metals can readily be obtained by simple ion exchange reactions on receptive hosts, such as those of the Dion-Jacobson and Ruddlesden-Popper structure types. The reaction with transition metal halides is particularly interesting since they often lead to novel architectures and magnetic behavior as seen in (CuCl)LaNb₂O₇ obtained by reaction between RbLaNb₂O₇ and CuCl₂. On subsequent heat treatment, these exchange products typically decompose to thermodynamically more stable phases. The newly synthesized the spin glass-like material, FeLa₂Ti₃O₁₀, obtained by ion exchange of Li₂La₂Ti₃O₁₀ with FeCl₂ at 350 ^oC, behaves differently. When heated to 700 °C under inert atmosphere, the compound undergoes a significant cell contraction (Δc ≈ -2.7 Å) with an increase in the oxidation state of iron. Details on the synthesis and characterization of this new phase, including Mössbauer studies, will be presented with an emphasis on the dramatic changes in the perovskite interlayer. Additionally, information will be given on the reactivity of the materials to reductive interaction with alkali metals and the effect on the magnetic behavior

The transportation industries require lightweight materials to achieve their goals of higher performance with increased efficiency. To this end, lightweight alloys such as AZ31 magnesium alloy have been implemented into current designs. However, certain aspects of these materials, such as stress triaxiality at high strain rates (10³ s^-1), have not been thoroughly investigated. This study analyzes the effect of stress triaxiality on the deformation mechanism of a hot rolled AZ31 Mg alloy. Room temperature tension tests at various strain rates (10^-3 to 10³ s^-1) were performed on specimens cut in the rolling, transverse, and normal directions with different notch geometries using a Split Hopkinson Pressure bar. Pre- and post-deformation microstructures were analyzed using electron microscopy and X-ray techniques. The results from mechanical testing indicate that the multi-axial stress state causes an increase in the flow stress required to deform the sample. Further, a shift in deformation mechanism is seen from dislocation based deformation to twin based deformation for the samples tested at high strain rates.

We utilize the spontaneous polarization in the ferroelectric polymer PVDF:TrFE (Poly vinylene fluoride trifluoroethylene) to modulate the surface charge density. This model ferroelectric polymer is used as the capacitive coupling layer to control the charge density in a polymer semiconductor (P3HT)¹ channel forming an all-polymer field effect transistors (OFET).
We demonstrate the ability to tune the charge transport behaviour at transport interface in OFET device using external applied electric field across the ferroelectric surface. The dielectric layer of the device is processed under electric field to orient the observed surface microstructure. Atomic force microscopy (AFM) and piezo force microscopy were deployed to probe the PVDF: TrFE surface evolution and polarization response. Detailed analysis of the correlation between OFET characteristics and the structural features are presented.
XRD measurements on PVDF:TrFE samples also indicate strong evidence of β- phase component in the system. The carrier mobility (µ_FET) increases by more than a factor of three upon introducing the electric-field polling treatment. It is also observed that the increase in µ_FET is more effective if the applied electric-field is restricted to ≈ 10⁵ V/m and is accompanied by formation of crystallite/domain of larger dimensions. For dielectric surfaces formed under large electric-field ( > 10⁵ V/m), the microstructure indicates ruptures and higher density of domain walls which consequently lowers µ_FET. The negative piezo electric response of PVDF:TrFE has been associated with lower crystalline sizes². Devices fabricated with n-type semiconducting polymers reveal similar trends. This controlled approach of tuning the interface using electric-field provides an elegant macroscopic approach to maximize the performance characteristics of OFETs.

Reference
1. Satyaprasad P Senanayak et al., Phys. Rev. B 85, (2012) 115311.
2. Katsouras, Kamal Asadi et al., Nature Materials (2015).

Reversible oxidation and reduction reactions in a confined space at the silica/Ru(0001) interfaces have been investigated using in-situ near-ambient pressure x-ray photoelectron spectroscopy (NAP-XPS). The weak interactions between the silica film and the ruthenium substrate allow oxygen and hydrogen molecules to intercalate the silica/Ru(0001) interface. In this paper, three types of silica films (bilayer silicate, bilayer aluminosilicate and MFI nanosheets) have been utilized to study the interfacial electronic structures upon oxidation and reduction. We found that the silica films stay essentially intact under these reactions, while the interface potential can be dramatically tuned up to 1.25 eV depending on the chemical states of the ruthenium substrate. These observations may help to understand variable catalytic performances of the silica by tuning the electronic states at the silica/Ru interfaces.

A micro-porous polymer film is utilized as a triboelectric layer of triboelectric energy harvester. The relationship between porosity of the triboelectric layer and output characteristics is analyzed for the first time. There are two key parameters found to influence the output performance of the triboelectric energy harvester: the surface charge density and effective capacitance of the triboelectric layer. Experiment, modeling, and simulation based on electrodynamics are performed to investigate how the two parameters affect the output performance. A triboelectric effect (triboelectricity) is a common phenomenon in our daily life. This often causes discomfort to
human beings. But, the triboelectricity can be a promising electric energy source due to its abundance in environmental vibration and human motion. A triboelectric energy harvester (TEH) generates electrical energy by means of two principles: triboelectric charging and electrostatic induction. Since the triboelectric layer (TL) is based on an organic dielectric, the triboelectric charge (σ) is localized on the surface of the TL when it is contacted. Despite the fact that the electrostatic induction is relevant to the capacitance of the TL, most of previous studies have mainly focused on σ enhancement through intentional patterning of surface.
In this work, we present a novel TL design that utilizes elastically porous dielectric film (polydimethylsiloxane, PDMS), resulting in improved triboelectric charge with increased effective capacitance. The correlation between the porosity and output power is analyzed through the experiment, modeling, and verified by simulation. Each contribution level of triboelectric charge and the effective capacitance to the output power are systematically evaluated. The triboelectricity is compared between the porous PDMS film and flat PDMS film. In addition, optimal resistance of TEH is changed according to the thickness of the TL. Long-term reliability enhanced by super-elasticity arisen from its inherent porous morphology is demonstrated. For more than a week, stable electrical outputs are achieved with no degradation. Moreover, the output performances are maintained under very humid environments.

The class of two-dimensional (2D) materials, which started with graphene, is rapidly growing and now includes dozens of metals, semiconductors, and insulators. These atomically thin materials exhibit unique optoelectronic properties with high technological potential. I the first part of this talk, I will give a general introduction to the electronic structure of 2D materials, including the characteristic features of the dielectric function and the collective excitations. The concepts will be illustrated by examples from the open Computational 2D Materials Repository (1) which presently comprises high accuracy first-principles data for a large range 2D materials. The 2D materials only form the basis of a new and much larger class of materials consisting of vertically stacked 2D crystals held together by weak van der Waals forces. In contrast to conventional heterostructures that require complex and expensive crystal-growth techniques to epitaxially grow the single-crystalline semiconductor layers, van der Waals hetero-structures (vdWHs) can be stacked in ambient conditions with no requirements of lattice matching. The latter implies a weaker constraint, if any, on the choice of materials that can be combined into vdWHs. However, at the same time the lattice mismatch poses a challenge to periodic first-principles calculations. I will show that the dielectric properties of realistic, incommensurable vdWHs comprising hundreds of layers can be efficiently calculated using a multiscale approach where the dielectric functions of the individual layers (the dielectric building blocks) are computed ab-initio and coupled together via the Coulomb interaction (2). The method is used to illustrate the 2D−3D transition of the dielectric function of multilayer crystals, the hybridization of quantum plasmons in thick graphene/hBN heterostructures, and the intricate effect of substrate screening on the non- Rydberg exciton series in supported WS2.

References:
(1) Computational Materials Repository, https://wiki.fysik.dtu.dk/cmr/(software), and https://cmr.fysik.dtu.dk/ (database).
(2) The Dielectric Genome of van der Waals Heterostructures, K. Andersen, S. Latini, and K. S. Thygesen, Nano Lett. 15, 4616 (2015)

The electronic structure and consequently, the charge transport properties of semiconducors containing transition metals (TMs) can be particularly sensitive to the state of magnetic order. The ground-state (T=0K) magnetic configuration is not always a good representation of the magnetic state at typical operational temperatures. For example, in thermoelectrics, the system is expected to operate at temperatures far above its Curie or Néel temperature in the para-magnetic state with possibly very different band edge characteristics and significantly different transport properties. Yet, computational identification of promising materials is routinely done by assuming either the ferro-magnetic order or by employing some approximation for the magnetic ground state, to calculate thermodynamic properties and electronic structures. In our work, we have assessed the effect of the magnetic order on predicted charge carrier transport properties and thermoelectric performance of ~1000 TM-containing chalcogenides by sampling a number of different magnetic configurations, including non-, ferro- and anti-ferro magnetic. The potential for thermoelectric performance is evaluated using our recently developed semi-empirical descriptor β_SE [1]. We find that for a non-negligible number of compounds the predicted transport properties vary significantly depending on the magnetic configuration. For these material systems, we will discuss the application of more appropriate representations of the spin disorder including magnetic special quasi-random structures, and their importance in achieving more reliable predictions of their electronic structure and related transport properties.

[1] J. Yan, P. Gorai, B. Ortiz, S. Miller, S. Barnett, T. Mason, V. Stevanović, and E. S. Toberer, "Material Descriptors For Predicting Thermoelectric Performance," Energy Environ. Sci. 8 (2015) 983-994.

There is a need for new dielectric materials with properties that satisfy diverse specification requirements of a multitude of applications, ranging from microelectronics to electric motor winding insulation. Currently, the number of compounds with known dielectric constant is in the order of a few hundred, which drastically limits the options available to the design engineer. Given the sheer size of the chemical compound space, attempting to experimentally search for new dielectrics is not practical considering the time required for synthesis and measurement of the dielectric response. In this work, we developed a high-throughput Density Functional Perturbation Theory (DFPT) screening methodology, specifically tailored to the discovery of new dielectric materials. We scanned more than 1,000 compounds and provided theoretical estimates of the dielectric constants.

In particular, our methodology used DFPT in two steps. We first defined and tested a workflow by benchmarking against experimentally reported values for the dielectric constant and refractive index. In doing so, we chose a set of 88 compounds consisting of 42 different elements and belonging to 14 different point groups, making it the largest comparison of the DFPT method against experiments to date for the aforementioned properties. We then applied our workflow to calculate the dielectric constant of more than 1,000 compound structures - for some of which, there is no prior evidence of ever having been synthesized in the lab. With this information in hand, we were able to highlight promising compounds and identify relevant applications. In order to help design engineers identify suitable dielectric materials, we integrated our results into the Materials Project [1] database, which lists multiple properties (e.g. band gap, elastic constant) and is a growing project.

References:
[1] A. Jain*, S.P. Ong* et. al. (*=equal contributions), APL Materials, 2013, 1(1), 011002.

The design and discovery of new materials have been pursued through a trial-and-error manner largely based on human intuition and serendipity. This traditional materials design process is time consuming and labor intensive, which have significantly delayed the research and development for novel materials that are critical to our societal needs. Computational techniques based on first principles are capable of predicting materials properties accurately with little experimental input. In this presentation, I will share our success stories of leveraging first principles computation techniques in resolving a number of key material challenges in all-solid-state Li-ion batteries, a new battery technology with potentially intrinsic safety, high energy density, and enhanced cyclability. I will first present how we use first principles computation methods to design new solid electrolyte materials. In addition, I will share our recent development in extending the computational techniques to investigate the interfacial phenomena. These techniques are applied to resolve the problems, such as interfacial degradation and high interfacial resistance, at the electrolyte-electrode interfaces in the all-solid-state batteries. For example, our first principles computation found that most solid electrolyte materials are not stable against Li metal or high voltages, and that the formation of solid-electrolyte-interphases in all-solid-state Li-ion batteries are ubiquitous. We revealed the mechanisms of artificial interfacial engineering to resolve these problems at interfaces. In the end, the computational approach to guide interfacial engineering will be discussed.