Symposium CP04—Interfacial Science and Engineering—Mechanics, Thermodynamics, Kinetics and Chemistry

Solute segregation in nanocrystalline metals has been widely explored as a route to improve the thermal stability of fine grain microstructures. Efforts have revealed that segregation may not always be uniform or subtle, thereby shifting the microstructural stabilization mechanism [1,2]. Delineating the subtle differences between these two mechanisms maybe be possible through improved understanding of mass transport along solute decorated grain boundaries. In this work, we present very recent observations of grain boundary diffusion with concomitant microstructural evolution by employing a thin film diffusion triple configuration. A 10nm layer of pure Al then a 500nm layer of a nanocrystalline, highly textured Ni-Cr alloy are deposited via DC magnetron sputtering on to a coarse-grained Ni-Cr alloy substrate of the same nominal composition. These diffusion triples are annealed under flowing Ar at 600C and the asymmetric diffusion profile is extracted via STEM EDX and APT characterization. Comparison of these diffusion data are presented against similar thin film diffusion triples containing <1at% Y in the nanocrystalline films. Despite nearly identical initial microstructures, there are significant differences between these diffusion profiles, with implications for microstructural stability as well as environmental resistance.

[1] Darling, K. A., M. Rajagopalan, M. Komarasamy, M. A. Bhatia, B. C. Hornbuckle, R. S. Mishra, and K. N. Solanki. "Extreme creep resistance in a microstructurally stable nanocrystalline alloy." Nature 537, no. 7620 (2016): 378.
[2] Abdeljawad, Fadi, Ping Lu, Nicolas Argibay, Blythe G. Clark, Brad L. Boyce, and Stephen M. Foiles. "Grain boundary segregation in immiscible nanocrystalline alloys." Acta Materialia 126 (2017): 528-539.

The coffee-ring effect refers to the phenomena in which a drop of a colloidal suspension evaporates on a substrate and deposits an ordered ring-structure due to the outward radial evaporative flow. This phenomena has been well-investigated due to numerous practical applications including ink-jet printing, microscale separations, and sensing/diagnostics. However, the phenomena has been less quantified for a crystallizing solution. When a drop of salt solution is evaporated, the salt becomes supersaturated and crystals begin to emerge at the substrate. The emerging crystals alter the local wettability of the substrate and fundamentally alter fluid dynamics of evaporation, which in turn alters the form of deposit left behind. Here, we investigate the role of energetic interactions between the substrate, crystals, and solution. We systematically vary the substrate interfacial chemistry by deposition of silane groups of various surface energies, and use four different crystal systems to probe this effect. We find that crystals with favorable interactions with the underlying substrate chemistries alter deposit morphology, while non-interacting crystals behave similarly to colloidal particles, and develop phase diagrams predicting this behavior. These results can be used for fine-tuned control and understanding of crystalline evaporative deposits.

Magnesium metal is potential candidate for high strength to weight ratio alloys with wide application in aerospace and automotive industries. However, poor corrosion resistance under ambient environmental conditions is the bottleneck for industrial deployment. Designing passivation layers and/or corrosion resistance alloys require fundamental understanding of the corrosion process. The traditional ex-situ spectroscopic measurements of polycrystalline metal surface with ubiquitous surface impurities provided indistinct view of the corrosion process. To clearly distinguish the mechanism and sequence of corrosion process, we employed in-situ cryo-based x-ray photoelectron spectroscopy (XPS) measurements on well-defined pure and aqueous solution exposed Mg-single crystal surfaces in combination with ab initio atomistic modelling studies. Clean Mg (0001) surfaces were carefully exposed to pure and sodium chloride mixed water and the subsequent interfacial reactions were studies through integrated experimental and theoretical approach. This study provides atomistic view of magnesium hydroxide nucleation as main product of the corrosion process. Under salt conditions, the competitive nucleation process between magnesium hydroxide and magnesium chloride were observed along with the formation of magnesium chloride hydrate and magnesium hydroxide hydrate phases. By combining the energy requirements from computational modelling and the electronic states of corrosion products, the mechanism and sequence of corrosion process on Mg metal will be discussed.

The increasing research of nanowires (NWs) as building blocks for electronic, optoelectronic, energy devices and quantum computing is driven by the unique physical, and chemical properties originate from their nanoscale dimensionality. Catalyzed surface-guided NWs offer the possibility to control their position, direction and crystallographic orientation, which eventually leads to high-performance devices. To adequately control these features and gain predictive abilities, a deeper understanding of the growth mechanism of surface-guided NWs is required. Here, we experimentally and theoretically study the kinetics of planar catalyzed NWs. We present a model that considers two main regimes of the growth rate of NWs depending on their thickness: the Gibbs–Thompson regime, which dominates the growth of thinner NWs, and the surface-diffusion-induced regime, which dominates the growth of thicker ones. By developing this kinetic model and fitting it to the experimental kinetic data, we determine the dimensionality of the surface diffusion. We observe a good correlation between the model and the results for surface-guided ZnSe and ZnS NWs grown on sapphire. The dimensionality of the surface diffusion value was determined to be ~1.5 and ~1.8 for ZnSe and ZnS NWs, respectively, in contrast to the value of 1 for vertical NWs growth models, supporting the difference between the two growth morphologies. The newly developed model distinguishes between the growth mechanisms of horizontal and vertical NWs, underscores the important role of the substrate in the horizontal growth, and provides new insights into the mechanism of surface-guided NWs growth. Understanding the growth mechanism introduces the possibility of controlling the growth directions and crystallographic orientations of the aligned NWs accurately and helps gain some prediction abilities of the NWs properties, leading to better nanowire-based electronic and optoelectronic devices.

Organic-inorganic hybrid perovskites, such as the prototypical methylammonium lead iodide (CH₃NH₃PbI₃ or MAPbI₃), have emerged as promising light absorbers in photovoltaic (PV) cells or as emitters in light-emitting diodes (LEDs). The strikingly high energy conversion efficiency up to ~ 20% ensures hybrid perovskites as the key component of highly efficient solar cells. However, they generally suffer from moisture instability, which limits the long-term use of perovskite-based devices in ambient environment. In this work, in order to improve the surface moisture stability of MAPbI₃ in common humidity atmosphere, more specifically, to discover better ligands under a specific coverage on the [MAI]⁰ surface to greatly improve the moisture stability and decrease the ion dissociation rate, we have applied molecular dynamics (MD) simulations and reaction kinetics theory to model the ion dissociation process and estimate the associated free energy barrier with and without ligand passivation. We have developed a new MD force field for MAPbI₃, which can match the density functional theory (DFT)-predicted elastic properties, experimental water contact angle on MAPbI₃, and DFT-predicted water infiltration and adhesion energies. We design MAPbI₃ with ligand-passivated surfaces by replacing MA⁺ with ligands composed of long-chain alkyl-ammoniums. Ligands with different chain lengths, such as CH₃(CH₂)_nNH₃⁺ (n = 3, 5, 7), and under different surface coverages (σ = 25%, 50%, 75%, 100%), are considered here. We discover that ligand passivation can greatly help protect MA⁺ on the surface due to the much higher dissociation free energy barriers of these ligands compared to that of MA⁺. For iodine ions, ligand passivation can also shield them from water contacts, except for long-chain ligands, such as CH₃(CH₂)_nNH₃⁺ (n = 5, 7) under full surface coverage (σ = 100%), due to the reduced dissociation free energy barriers of long-chain ligands. As an interesting finding, the reduced dissociation free energy barriers for long-chain ligands under high surface coverages could be explained by their larger tendencies to micellize, which serves as additional driving force for their dissociation. This work significantly motivates future experimental efforts in designing new surface ligands to improve the moisture stability of hybrid perovskites.

We show how surfaces can be designed to redirect the momentum of the spreading lamella causing them to lift-off into 3-dimensional shapes thereby dramatically reducing the contact area. We design superhydrophobic surfaces with an in-plane discontinuity leading to the accumulation of vertical momentum resulting in the out-of-plane ejection of the lamella into water bowls. We demonstrate a two-fold reduction in the heat transfer between a cold rain and a warm surface. These insights can be broadly applied to other transport phenomena involving mass and energy exchange to limit heat loss under precipitation, icing of surfaces, reduce salt deposition on a surface exposed to ocean spray, or inhibit the formation of a water film on wings or wind turbine blades.

Reactivity and Ion transfer processes across heterogeneous interfaces is the backbone of electrochemical devices such as batteries and dye sensitized solar cells. The complexity of reactivity and charge transfer across interfaces often arise from convoluted relations between surface chemistry (such as defect density, crystallographic orientation and functional groups) and exogenous drivers such as temperature and electric field. For example, technologically significant nanocrystalline TiO₂ with diverse surface chemistry renders a complex interfacial processes. To correlate the surface chemistry of TiO₂ with interfacial processes (reactivity and ion transfer), we performed a multimodal spectroscopic analysis in combination with ab-initio molecular dynamics (AIMD) calculations. The interfacial processes of anatase and rutile phase with varying surface chemistry against lithium bis(trifluoromethylsulfone)imide (Li-TfSI) salt were analyzed. We observed that, ubiquitous oxygen defects at TiO₂ surface regime enhance the molecular decomposition of TFSI anions. The thermally driven splitting of –CF3 groups resulting in titanium oxyfluoride type bonding (O-Ti-F) at the near surface regime were observed in AIMD calculations and supported by multimodal spectroscopic measurements. The presence of free electrons due to fluoride ion transfer to TiO₂ surface were detected using low temperature electron paramagnetic resonance (EPR) analysis. In addition, separated Li⁺ from such a weakly coordinating anion, pass in through preferential transfer pathways across nanoparticle surfaces. The thermodynamics and kinetics of the molecular decomposition and associated Li⁺ ion transfer processes were derived from AIMD and high resolution nuclear magnetic resonance (NMR) analysis.

Complex oxide heterostructures possess a wide range of functional electronic and magnetic properties including metal-insulator transitions, superconductivity, ferroelectricity, and colossal magnetoresistance effects. At epitaxial interfaces formed between atomically thin complex oxide films, electronic, chemical and structural interactions can be used to effectively tune the physical properties of these materials. Using a combination of atomic-scale controlled thin film synthesis and high resolution synchrotron diffraction based imaging, we show that structural distortions the interfaces between polar La_1-xSr_xMnO₃ films and non-polar substrates can be effectively tuned by chemical modifications at these interfaces to control ferromagnetism in [001]-oriented
La_1-xSr_xMnO₃ films with thickness less than 1 nm. We show that atomic-scale chemical control at polar/non polar oxide interfaces provides a powerful route to engineer novel electronic and magnetic phenomena at complex oxide interfaces.

Adhesion and interface formation are essential processes in friction, particle aggregation, and grain boundary formation. It is generally accepted that adhesion is anisotropic and the adhesion force depends on the mutual orientation of attracting interfaces. However, direct experimental measurement of the adhesion force and its orientation dependence has never been realized. Here, by combining in situ environmental transmission electron microscopy and atomic force microscopy, we directly measured the adhesion between two rutile TiO₂ (001) surfaces with controlled mutual orientation. The results show that the adhesion force is primarily van de Waals attraction between the two surfaces and is highest when the lattices of the two surfaces are aligned, demonstrating strong orientation-dependence. Moreover, the orientation dependence is not influenced by surface hydration; though, the adhesion force drops considerably with increasing water vapor pressure. Our work demonstrates a powerful method for quantitative study of nanoscale interface formation and strength, providing new insights to the driving force of orientated attachment of particles and friction anisotropy.

Early grain boundary engineering work has demonstrated how significantly material properties can be improved with tailored microstructures. Unfortunately, continued improvement in this area will require complete structure-property relationships of grain boundaries, which are currently lacking. We describe tools and efforts to obtain both crystallographic structure-property relationships and atomic structure-property relationships. And since the two are must be combined to facilitate future design efforts, we detail efforts to find correlations between them. The properties of interest in this work are thermodynamic properties, like grain boundary energy, as well as kinetic properties, like grain boundary mobility.

The nonequilibrium nature of amorphous solids such as metallic glasses (MGs) shares intriguing commonalities with grain boundaries in nanocrystalline (NC) materials, owing largely to the multiplicity of energy states inherent to disorder. Here, we report on experimental studies of MG and NC materials and novel synthesis and processing routes for controlling the structural state – and as a consequence, the mechanical properties. Relaxation processes in disordered materials that facilitate atomic reconfigurations toward a lower energy state, such as low temperature annealing, have been shown to enhance mechanical strength while promoting shear localization. A particular focus in this talk will be on strategies for rejuvenation of disorder with the goal of suppressing shear localization and endowing damage tolerance. Parallels between our results and rejuvenation processes in glassy systems will be discussed in the context of controlling metastable structural configurations through novel processing routes.

Polymorphism in crystallography means the ability of a solid material to exist in more than one crystalline form, and the forms differ in physical and chemical properties even though their chemical components are identical. Recently, polymorphism in two-dimensional (2D) materials has been attracting interest in obtaining diverse properties that cannot be obtained in 3D forms [1, 2]. In this study, we have found 2D polymorphic nanostructures self-organized inside silicon ingots grown by Czchralski (CZ)-method; they are composed of different 2D periodic units just along Σ9{111}/{115} grain boundaries (GBs) with a 〈110〉 tilt axis. The atomistic structure of the 2D polymorphic GBs, as well as their segregation ability, are determined by a 3D chemical nanoanalysis technique using atom probe tomography (APT) combined with scanning transmission electron microscopy (STEM) and ab-initio calculations [3].

STEM revealed that, Σ9{111}/{115} GBs have a periodic interface structure with a period of 2 nm along the GBs, and the GB unit is composed of Σ9{221}-like and Σ9{114}-like nanofacets involving 2 〈110〉 reconstructed bonds (named type-I GBs in this paper) [4]. Ab initio calculations reveal the GB energy of 0.6 J/m², that is higher than the GB energy of Σ9{221} (0.18 J/m²) and Σ9{114} (0.36 J/m²) GBs. Also, the 〈110〉 bonds are rather longer in comparison with the 〈111〉 bonds in bulk Si. Thus, those long bonds would induce an energy in addition to the energy of Σ9{221}-like and Σ9{114}-like nanofacets.

The GB unit in type-I GBs is changed by annealing at 1100^oC. The GB unit in the annealed GBs, that was observed in a cast-grown Si ingot [5], is composed of Σ3{111}-like nanofacets involving 4 〈110〉 bonds, with the period of 2 nm (type-II GBs). Type-II GBs can be introduced by forming stacking faults of low GB energy (~0 J/m²) on the Σ9{221}-like and Σ9{114}-like nanofacets in type-I GBs. The GB energy of type-II GBs is estimated to be 0.45 J/m². Therefore, type-I GBs change into type-II GBs so as to remove Σ9{221}-like and Σ9{114}-like nanofacets of high GB energy, via the formation of Σ3{111}-like nanofacets of low GB energy, even though 2 〈110〉 bonds of high-energy are added in each GB unit.

APT reveals the oxygen segregation ability of type-I GBs (0.04-0.06 atoms/nm²) and that of type-II GBs (0.03 atoms/nm²). The ratio of the abilities is the same as the ratio of the GB energy of type-I and type-II GBs; i.e., the ability is proportional to the GB energy, as expected [3, 4]. We will discuss potential controls of the 2D polymorphic nanostructures, as well as of their segregation ability, by changing the growth condition.

1) M. Wu, et al., Phys. Rev. B 96 (2017) 205411.
2) M. Yoshida, et al., Nano Lett. 17 (2017) 5567.
3) Y. Ohno, et al., J. Microsc. 268 (2017) 230.
4) Y. Ohno, et al., Appl. Phys. Lett. 110 (2017) 062105.
5) A. Stoffers, et al., Phys. Rev. Lett. 115, 235502 (2015).
Acknowledgments: This work is supported by JST/CREST (Grant No. JPMJCR17J1, 2017-2023)

The scaling of behavior in ferroelectrics remains a topic of substantial interest given the need for device miniaturization and fundamental research. In recent years, a suite of reports has shown that the bulk piezoelectric functionality of thin ferroelectric films (approximately 30 nm or less) have a strong dependence on the surface chemical and structural state. Furthermore, these works suggest surface screening is a critical factor in understanding the bulk properties and switching behavior in thin film form, but precise mechanisms of this coupling are lacking. A key question that remains is the role of intrinsic as opposed to extrinsic screening on the properties of the thin ferroelectric films. In this work, we use BaTiO₃ (BTO) grown via pulsed laser deposition (PLD) as a model material system to explore the correlation between surface state and bulk properties. To carefully probe the surface and bulk properties of BTO thin films, we use a combination of atomically resolved scanning tunneling microscopy (STM), non-contact atomic force microscopy (AFM), piezoresponse force microscopy (PFM), X-ray diffraction (XRD), and in-situ X-ray photoelectron spectroscopy (XPS). The aforementioned characterization techniques provide an experimental pathway to evaluate the correlation between the surface state and bulk piezoelectric properties.

An assembly of hemispherical Ag nanoparticles is prepared by solid-state dewetting of thin Ag film deposited on the sapphire substrate. The in-situ nanomechanical compression testing of the particles with a flat diamond punch inside the scanning electron microscope demonstrates the deformation behavior typical for the nucleation-controlled plasticity: high elastic deformation followed by an abrupt particles collapse. The latter is associated with the dislocations nucleation in otherwise pristine particle. The average contact pressure in the contact zone at the onset of dislocation-controlled plasticity is about 8 GPa, and does not depend on particle size. This observation supports the hypothesis that the pseudoelasticity of much smaller Ag nanoparticles observed by Sun et al. [Nat. Mater. 2014, 13, 1007] is intrinsically related to their ultrahigh strength. A stress-induced diffusion along the particle-substrate and particle-punch interfaces is identified as a factor controlling the pseudoelastic deformation. The corresponding diffusion model allows estimating the room-temperature self-diffusion coefficient of Ag along the Ag-W and Ag-Zirconia interfaces, which was quite close to the estimated value of the grain boundary self-diffusion coefficient in Ag. Based on this finding, the map of pseudoelastic deformation of several metals is proposed.

Semiconductor heterostructures enable band-gap engineering by combining two or more dissimilar materials, and are essential for many optoelectronic applications such as multi-junction solar cells, lasers and light-emitting diodes. However, differences in lattice parameters between film and substrate give rise to interfacial misfit strain. Misfit dislocations are introduced beyond a certain critical thickness, which will likely severely limit device performance by acting as sites for electron-hole pair recombination. Identifying the sources of these defects and understanding their evolution as a function of film thickness would possibly lead to identifying a path for minimizing defects in such heterostructures by optimizing growth conditions, film thickness, etc. In this study, the creation and evolution of structural defects in low-mismatched (misfit strain ∼0.6%) GaAs (cap)/GaAs_0.92Sb_0.08/GaAs(001) heterostructures grown by molecular beam epitaxy have been investigated using transmission electron microscopy as well as high-resolution x-ray diffraction and atomic force microscopy (AFM). Three distinct stages of strain relaxation were identified in GaAs_0.92Sb_0.08 films with thicknesses in the range of 50-4000 nm, whereas 50-nm-thick GaAs capping layers exhibited only initial Stage-I relaxation. Aberration-corrected electron microscopy revealed that glide-set dissociated 60° dislocations were primarily formed during Stage-I relaxation. Lomer edge dislocations with compact core structure and curved dislocations prevalently with edge character formed in Stage-II and Stage-III of relaxation. Evolution of dislocation density at cap/film interfaces showed a strong correlation with the evolution of surface morphology at the growth front. Even though misfit strain was highest at the cap/film interface of the heterostructure with 4000-nm-thick film, dislocation density was lower in comparison with heterostructures with 1000-nm-thick and 2000-nm-thick films. AFM results indicated that smoothening of the growth front in the heterostructure with 4000-nm-thick film caused a decrease in the areal density of surface troughs which may act as preferred sites for heterogeneous nucleation of dislocation half-loops. Detailed consideration of different sources for threading dislocations and the experimental evidence of correlation between decreased dislocation density and decreased density of surface troughs led to the conclusion that heterogeneously nucleated surface half-loops are the major source of threading dislocations in these low-mismatched heterostructures [1].
Reference
[1] A. Gangopadhyay, A. Maros, N. Faleev, D.J. Smith, Strain relaxation in low-mismatched GaAs/GaAs_1-xSb_x/GaAs heterostructures, Acta Mater. 162 (2019) 103-115.

III-V-Bismides are attractive materials for near- and mid-IR device applications due to their dramatic decrease in band gap with small incorporations of bismuth. The growth space of GaAs_1-xBi_x has been explored by molecular beam epitaxy, with the highest obtained bismuth incorporation being x=0.22 [1]. However, resulting film quality has been a barrier to incorporating high bismuth content materials into optoelectronic devices. GaAsBi films are typically grown thin (<50 nm) on GaAs under-layers to prevent strain relaxation from the epitaxial mismatch. The resulting compressively strained films have exhibited a variety of phase separation characteristics, including lateral composition modulations [2], droplet induced composition inhomogeneity [3], and CuPt_B-type atomic ordering [4]. We have explored growing on InGaAs buffer layers as a method for reducing epilayer compressive strain, suppressing surface droplet formation, and eliminating vertical phase separation through the bismide film layers.

GaAsBi epilayers were grown by solid-source molecular beam epitaxy on a Veeco GENxplor MBE using a valved As₄ source and an effusion cell for bismuth. Samples were characterized by HRXRD to determine the average bismuth content and the strain state of the layers. AFM was used to analyze the droplet formation on the surface of the films. TEM and HAADF STEM were used to qualitatively characterize the bulk phase separation. By comparing samples grown on GaAs to samples grown on InGaAs buffers, we will draw connections between film strain, surface droplet formation, and phase separation. We are able to reduce the degree of phase separation that occurs in GaAsBi films, by growing on InGaAs buffer layers; thereby increasing film homogeneity and general suitability for optoelectronic applications.

[1] R.B. Lewis, M. Masnadi-Shirazi, and T. Tiedje Appl. Phys. Lett., 101 082112 (2012)
[2] E. Luna, M. Wu, J. Puustinen, M. Guina, and A. Trampert, J. Appl. Phys., 117, 185302 (2015)
[3] C. R. Tait, L. Yan, and J. M. Millunchick, Appl. Phys. Lett., 111, 042105, (2017)
[4] A. G. Norman, R. France, and A. J. Ptak, J. Vac. Sci. Technol. B, 29, 03C121, (2011)

C54 titanium disilicide (TiSi₂) has attracted great attention in Fin Field-effect transistor (FinFET) application, since it has accompanied with promising advantages, such as low resistivity, outstanding reliability and low mobility in silicon substrate. In order to further investigating the structural properties, we observed the formation process of C54-TiSi₂ by in-situ TEM in this study. During the heating process, Ti atoms diffused into Si_xGe_1-x substrate and nucleated afterward to form C54-TiSi₂ . It is remarkable that we also observed the occurance of a novel type of texture-axiotaxy between C54-TiSi₂ (110) plane and Si (001) plane at around 800 celcius degree. Axiotaxy is convinced to be produced by the strain implemented on the low index planes. Additionally, the elemental distribution information was provided by Energy Dispersive Spectrometer (EDS) analysis with scanning transmission electron microscopy (STEM) image. With this atomic scale analysis, we are able to establish the connection between microstructure and properties, further enhance the performance of Complementary Metal-Oxide-Semiconductor (CMOS) technology.

Heteroepitaxial thin film growth on substrates with large lattice mismatch is an important area of crystal growth research to overcome applications-based restrictions on substrate choice. For some material systems, one crystalline phase may exhibit growth of domains with different epitaxial relationships to the substrate or two phases may grow simultaneously with different orientations. This process, known as double epitaxy, creates thin film materials with many grain interfaces at fixed angles with respect to the substrate. This work demonstrates double epitaxy in FeSe, an important superconducting compound with a complex phase diagram which is highly sensitive to stoichiometry. Conditions exist during pulsed laser deposition (PLD) that lead to the simultaneous epitaxial growth of the tetragonal (β-FeSe) and hexagonal Fe₇Se₈ (γ-FeSe) phases of FeSe in intermingled domains throughout the film, creating many phase boundary interfaces. We investigate the impact of PLD parameters on the double epitaxy of this interface-dominated material and identify the conditions under which double epitaxy is suppressed.. The lattice parameters of β-FeSe and γ-FeSe are unsuitable for lattice-matching epitaxial growth on (001) MgO. In the case of β-FeSe, the lattice parameter (a = 3.672-3.769 Å, depending on stoichiometry) differs from that of MgO (a = 4.211 Å) by more than the 9% normally tolerated for conventional epitaxy. The cubic structure of MgO, however, exhibits various domain matching epitaxial relationships with the various FeSe crystalline phases. This opens opportunities for engineering the internal interfacial boundaries of different FeSe phases by controlling the rotation of the epitaxial domains with processing parameters. A KrF excimer laser (1.4-3.4 J/cm²) was used to ablate a pressed, sintered target synthesized from FeSe powder. Targets were sintered in sealed quartz ampules at 700°C for 12 hours. Targets were ablated in high vacuum (10^-6 torr) with substrate temperatures in the 350-550°C range. FeSe films grown at temperatures between 350°C and 450°C with laser fluence fixed at 3.4 J/cm² showed the simultaneous epitaxial growth of both (001)-oriented β-FeSe and (101)-oriented γ-FeSe. The relative fraction of these two phases changes smoothly from majority β-FeSe at 350°C to majority γ-FeSe at 450°C. The β-FeSe phase grows aligned with the substrate and the γ-FeSe phase grows in two rotated domains, one dominant domain aligned with the substrate and a minor domain rotated by 45°. Atomic force microscopy (AFM) reveals the granular surface morphology with apparent grain sizes of 220-300 nm in films grown at 350°C and 450°C. The film grown at 400°C has a nearly equal fraction of each phase and the apparent grain size is found in the 120-200 nm range. The smaller grain size points to a larger number of phase interfaces, corresponding to an estimated number density of β-FeSe/γ-FeSe interfaces of 3.5x10⁹ cm^-2. Substrate temperature of 550°C results in a change in epitaxial orientation of β-FeSe from (001) to (101)-oriented. The (101) oriented β-FeSe grows in three domains, one aligned with the substrate and two others rotated by ±30°. AFM of the (101)-oriented FeSe shows large, micron scale grains leading to a much smaller grain interface number density than the biphasic films on the order of 10³ cm^-2. This growth technique shows the orders of magnitude scalability of in-plane interfaces of these epitaxially-oriented grains. At 550°C, reducing the laser fluence to 1.4 J/cm² increased the relative fraction of the γ-FeSe phase. The parameters of the PLD plasma plume are expected to influence the simultaneous epitaxial growth of the two phases by delivering a locally non-stoichiometric flux. Measurements of space and time resolved ion densities of the PLD plume will be presented and correlated with the emergence of γ-FeSe to determine the impact of the relative density of arriving ions on phase crystallization.

Grain boundary complexion engineering was recently introduced as a framework to be used to design grain boundaries and govern bulk material properties. The main conceptual tools of grain boundary complexion engineering are grain boundary complexion equilibrium diagrams and grain boundary complexion time-temperature-transformation (TTT) diagrams. Importantly, when the thermodynamic and kinetic diagrams are used in tandem, scientists and engineers have the ability to tailor classical processing conditions (e.g. chemistry, pressure, temperature, and time) to intelligently control grain boundary complexions, and specifically, complexion transitions from one complexion type to another. The ability to control complexion transitions is particularly important because a complexion transition could theoretically convert a tough material to a brittle material or an insulator to a conductor. Overall, the first segment of this talk will review the concepts of grain boundary complexion engineering and outline various approaches to develop the necessary complexion equilibrium and time-temperature-transformation diagrams. The second segment of this talk will summarize a grain boundary complexion engineering case study of silica and rare-earth oxide doped boron suboxide armor ceramics. Boron suboxide is being investigated as a promising armor ceramic because of its low density and high hardness, however, boron suboxide exhibits poor sinterability and fracture toughness. This case study outlines how grain boundary complexion engineering is used to control the thermodynamics and kinetics of grain boundary complexion transformations to improve sinterability and enhance fracture toughness. Boron suboxide powders were co-doped with silica and rare-earth oxide additives and the resultant microstructures were analyzed using advanced electron microscopy techniques to characterize grain boundary structures and quantify grain boundary segregation. The main observations include clear differences in microstructural evolution between the silica and rare-earth dopants, modified atomic structures with different excess grain boundary segregation concentrations, and improvements in bulk hardness.

Ni/Al reactive nanolaminates are systems of current scientific interest due to their highly customizable exothermic combustion characteristics. The high combustion temperatures and wave speeds attainable by these systems have found industrial application in soldering microelectronic components and have been proposed as microinitiators in military ordnance. Ni/Al nanolaminates are typically synthesized through a physical vapor deposition (PVD) technique, allowing precise control of nanostructure which has been shown to have significant impacts on combustion characteristics. The most fundamental structural attribute is the bilayer thickness, which is defined as the combined thickness of one Ni and one Al layer. This thickness has been shown to affect both the combustion temperature and wave speed of the reaction. Correlated with the bilayer thickness, the thickness of an individual material layer has been shown to affect grain size. The PVD process forms columnar grains perpendicular to nanolaminate layers, which have average grain diameters on the order of the individual layer thickness. The current work investigates the role of grain size on reaction rates and peak temperatures. Decreasing grain dimeter provides increased rates of solid-state diffusion and subsequent solid/liquid dissolution, increasing overall reaction rates. Two bilayer thicknesses are evaluated in this work, on twice the thickness of the other, revealing stronger grain size dependence in nanolaminates with larger bilayer thicknesses.

Nickel silicide (NiSi) thin films have various advantages for microelectronics processes, such us low resistivity and low formation temperature. Therefore, it has been widely used in semiconductor devices. The formation of NiSi thin film mainly depends on the mechanism of diffusion between Ni and Si. In order to understand its formation kinetics, in-situ transmission electron microscopy (TEM) is a powerful tool to study the diffusion behavior during the process. In this work, we successfully observed the process of diffusion by in-situ TEM. We chose silicon-germanium (Si_0.7Ge_0.3) as strain layer to improve the electronic performance. During the heating process, Ni atoms would diffuse into Si_0.7Ge_0.3 layer at 400°C in the duration of 10 minutes heating. The HRTEM results and the corresponding FFT-DP demonstrated that the product was NiSi. In addition, we found that there is a specific orientation relationship of [120]_NiSi//[01-1]_Si0.7Ge0.3 and (002)_NiSi//(1-1-1)_Si0.7Ge0.3. Furthermore, the Energy Dispersive Spectrometer (EDS) analysis established the elemental information for both as-grown and after-annealed samples, giving strong evidence for the formation of the NiSi. These results provide us useful information to grow low resistivity NiSi thin film.

High-carbon steels (pearlitic steel) are widely used in the fields, including bridge cable, automobile tire cord and high-rise building, owing to the outstanding high strength and low cost. Micro-alloying is the common method to increase the strength of pearlite steels. In the simplest Fe-C-M (M: Mn, Si, Cr, Ni) ternary system, the effects of alloying elements have been extensively studied. For example, Mn, as the austenite stabilized element, delays the pearlitic transformation, improves the stability of cementite and retards the spheroidization of cementite lamellar; Si, as the non-carbide forming element, refines the lamellar structure of pearlite, increases tensile strength, improve softening resistance during tempering, and etc. However, in the complicated multicomponent system the partition and interactive effects among alloying elements needs to be attention, especially the interaction of alloying elements during the eutectoid transformation need to be further studied.
Therefore, in the present work the atomic interaction among Si, C and Mn in Fe-C-Mn-Si quaternary system during the eutectoid transformation is investigated using atom-probe tomography combining first-principle calculations. Two types of pearlitic carbon steels with different Si contents are prepared. At the initial stage of eutectoid transformation, atom probe tomography observations indicate that Si is enriched at the ferrite side of ferrite-cementite interface, while Mn is enriched at the cementite side. This interfacial segregation phenomenon gradually diminishes or disappears as the transformation proceed. Based on the results of first-principle calculations, Si moves from the core of cementite layer to the ferrite side of ferrite-cementite interface can obviously enhance the chemical bonding and stability of pearlite system. In the same manner, Mn atom moves from the core of ferrite layer to the ferrite-cementite interface, which improves the thermal stability of pearlite system. Therefore, Si/Mn segregates at each side of ferrite-cementite interface at the initial stage of pearlitic transformation. Furthermore, the partitioning ratio of Mn in high Si steel is higher than that in low Si steel, leading to more Mn partition into cementite. Based on the results of first-principles calculations, due to the strong repulsive force between Si and Mn at the interface of pearlite, the enrichment of Si in ferrite promotes the partitioning of Mn into cementite.

We have performed an ab initio study of tensorial elastic properties of the interface states in Ni₃Si associated with the Σ5(210) grain boundary (GB). Ni₃Si is well known structural material with a large potential for high temperature applications in corrosive atmospheres, which however suffers by low ductility with tendency to intergranular fracture. Rather complex tensorial elasto-chemical aspects of the studied periodic GB approximants were found in the case of states with different atoms at the GB plane (either only Ni atoms (Σ5(210)^Ni,Ni) or both Si and Ni atoms (Σ5(210)^Si,Ni)). The Σ5(210)^Ni,Ni GB state is predicted to have only a slightly lower interface energy of the two studied variants. The elastic constants are found to depend very sensitively on the GB plane chemical composition. In particular, the GB variant containing both Ni and Si atoms at the interface is shown to be unstable with respect to a shear deformation (one of the elastic constants, C₅₅, is negative). This instability is occurrs for a rectangular-parallelepiped supercell obtained when applying a standard coincidence-lattice construction. Our elastic-constant analysis allowed us to identify a shear-deformation mode reducing the energy and eventually to obtain a mechanically stable ground-state characterized by a shear-deformed parallelepiped supercell. Nevertheless, a three-/eight-fold reduction of the elastic constant C₅₅ (when compared with the bulk value) is identified as the crucial weakest link of the mechanical stability for the studied GB states. We have also partly stabilized this GB interface state by Al substituent replacing Si atoms at the GB.
Next, an origin of this elastic softening and instability of the rectangular-parallelepiped Σ5(210)^Si,Ni GB variant is discussed in terms of chemical inter-atomic interactions described by the crystal orbital Hamilton population (COHP). Lattice-dynamics properties represented by projected force-constant matrices on the unit vector along each bonding direction were considered as well. Such complex analysis reveals a weak interaction far from the GB interface between the Ni atoms in the 3^rd plane and the Si atoms in the 5^th plane. However, this bond weakening is a consequence of a very strong interaction between the Si atoms in the GB plane and Ni atoms in the 3rd plane off the GB interface. The same strong interaction was not observed when Si atom at the GB is replaced by Al. Thus the strong interaction near the GB plane makes this GB variant mechanically unstable.
Our study clearly shows the importance of anisotropic elastic-constant treatment as well as interatomic-interaction analysis for next studies of interface states close to GBs when determining origin of their mechanical (in-)stability. The sensitivity of tensorial elasto-chemical properties, which was additionally illustrated by studying the impact caused by Al atoms substituting Si atoms at the GB interface plane, paves a new way towards a solute-controlled design of GB-associated interface states with controlled tensorial elastic properties and stability.

For decades, diamond has shown superior properties that make it desirable for novel electronic materials. However, fabricating large size high quality single crystal diamond wafers faces multiple challenges. Lateral outgrowth and mosaic growth of CVD diamond make it highly possible for large size diamond wafers to be realized. Mesoscale characterization on the interface and surface of the grown CVD diamond is key to understanding how to eventually fabricate large size high quality single crystal diamond wafers.

In this work, mosaic and lateral single crystal diamonds were grown by Microwave Plasma Assisted Chemical Vapor Deposition (MPACVD) on High Pressure High Temperature (HPHT) seeds and characterized by a micro x-ray diffraction (µ-XRD) mapping technique with a 300 µm slit to show the evolution of diamond crystal structure information during growth and provide feedback on growth strategies. By understanding how crystal structure evolves during growth we can understand the mechanisms linking processing conditions the structure and properties of grown material.

Our results demonstrate after growth and regrowth by MPACVD, variation in the structure and distributions of these variations evolves towards homogeneity. Measurements of the (400) phase for mosaic and lateral growth samples by high-resolution x-ray rocking curve (HRXRC) and 2θ-ω µ-mapping technique. The overall misorientation of the merging CVD layers is shown to decrease with the regrowth from 0.225° (bottom layer plate) to 0.092° (uppermost layer plate). µ-HRXRC mapping results show a low mosaicity and μ-HR 2θ-ω scans show a low average Full Width at Half Maximum (FWHM), 0.032° and high crystallinity. Typically lateral outgrowth shows a uniform misorientation at the growth front, ranging from 0.092° to 0.035°. In contrast, our mosaically joined samples show a non-uniform distribution in misorientation at the merging interface, displaying a biaxial deviation parallel and perpendicular to the interface. The preliminary data shows that after regrowth, this biaxial distribution tends toward uniformity, with regions of independent preferred orientation overtaken by a single, new orientation. We will discuss this phenomenon in terms of step-flow overgrowth and energetically favorable states of the diamond crystalline phase during epitaxial lateral overgrowth in the extreme plasma environment.

Continuum models will be critical towards the computational characterization of the diffuse layer at electrochemical interfaces. Recently, a number of different approaches have been proposed. Here we will present a new class of methods based on a planar averaged approximation and the Gouy-Chapman theory for electrochemical interfaces [1]. These models potentially constitute a powerful set of tools for investigating one-dimensional surfaces, with several advantages over their more generalized counterparts. They exploit the analytic solution for the one-dimensional complex electrostatic problem, are computationally inexpensive, and are not affected by any imposed periodicity. Charged metal solvent interfaces and their experimentally recorded differential capacitance values present a benchmark to compare candidate models. We will present the theory, implementation, testing, and applications.

[1] F. Nattino, M. Truscott, N. Marzari, and O. Andreussi "Continuum models of the electrochemical diffuse layer in electronic-structure calculations" under review by J. Chem. Phys. (2018), available at https://arxiv.org/abs/1810.09797

Ethylene is a high-value chemical that is an essential building block for polymers and other designer molecules. Producing ethylene from ethane (a more plentiful feedstock) is, therefore, a high-value activity, but one that requires a lot of energy. The use of catalysts to more efficiently perform chemical reactions, for example by lowering energy barriers for reaction pathways, therefore has the potential to save billions of dollars annually at production scales. The oxide Mo-V-Nb-Te-O (M1) is one such catalyst, but one whose efficacy degrades at elevated temperatures. In this work, we aim to understand the dynamics of ethylene on the surface of M1 to inform strategies for improving M1’s efficacy or lower its optimal operating temperature. Specifically, we develop tools that facilitate molecular modeling of organic molecules including ethane and ethylene on M1 surfaces and use them to study surface dynamics. We implement a combination of the OPLS-AA and UFF force fields and measure surface residency times of ethylene and ethane as a function of temperature. These measurements show ethane preferentially adsorb near Vanadium and Tellurium sites, informing strategies for increasing adsorption selectivity towards Vanadium. We investigate candidates for surface patterning and discuss anticipated changes in catalyst effectiveness.

Hybrid organic-inorganic perovskites (HOIPs) such as methylammonium lead iodide (MAPI) and methylammonium lead bromide (MAPBr) have attracted much attention as solution-processed semiconductors that could be used in high performance optoelectronic devices such as photovoltaics, light emitting diodes, lasers, and X-Ray detectors. These materials exhibit long carrier diffusion lengths, large absorption coefficients, and high photoluminescence quantum yields. For commercial applications, however, HOIPs suffer from significant humidity-induced degradation. Here, we report the stabilizing and destabilizing effect of B-site substitutional doping in HOIP materials, and discuss how surface doping at the B-site can induce preferential crystal orientation.
Substitutional doping of MAPI with Bi was found to lead to a stabilizing effect at high humidity, but a destabilizing effect at lower humidity. With 5 mol% Bi in MAPI, UV-Vis-NIR absorptance spectroscopy and X-ray diffraction (XRD) showed that at 90% relative humidity, the films took 72 h to fully degrade compared to 24 h for the undoped MAPI films. At the lower humidity of 60% RH, the undoped MAPI films were more stable than the Bi-doped MAPI films. The MAPI films were stable for 14 days; whereas, the Bi-doped MAPI films degraded after only 10 days.
While MAPI is a promising photovoltaic material, certain tandem cell and LED applications require wider band gap HOIPs, such as MAPBr. However, the random orientation of grains in MAPBr thin films made from commonly employed synthetic methods may hinder these devices’ effectiveness. Here we report the incorporation of Ag into MAPBr films induces a preferential (100) crystallographic orientation on any substrate and increases crystal grain size of thin film cubic MAPBr. X-ray photoelectron spectroscopy (XPS), energy-dispersive X-ray spectroscopy (EDS), and time-of-flight secondary ion mass spectrometry (TOF-SIMS) revealed that Ag accumulates at the interfaces of the MAPBr film, which appears to be the reason for the preferred orientation of the film on the substrate. Elemental maps and grazing incidence wide angle X-ray scattering (GIWAXS) of the Ag-doped films also showed the emergence of secondary methylammonium silver bromide phases. These regions in the film could create recombination centers in photovoltaic devices (PVs). PVs were fabricated with MAPBr and Ag-doped MAPBr films. Despite the differences in crystallographic orientation and crystal grain size, there were no observed differences in the performance of these perovskite PVs up to 2 mol% Ag incorporation, at which point PV performance became worse with increasing silver content.

Cholesterol is an amphiphilic molecule present in cellular membranes at different concentrations depending on the cell’s role and functionality. It is uniquely structured as its hydrophilic head is much smaller than its bulky hydrophobic tail. When present in a phospholipid leaflet cholesterol behaves according to the umbrella model, filling the gaps in between phospholipid molecules and reducing the area per lipid. This provides an ordering and condensing effect for the membrane. The effect of cholesterol on cell membranes has been investigated and it has been seen to minimize its permeability and move the membrane from a liquid disordered to a liquid ordered state. These mechanics may be measured through the measurement of the interfacial tensions of lipid monolayers and bilayers with and without cholesterol.
In this work, the droplet interface bilayer (DIB) technique is used to investigate cholesterol’s effect on surface tension and the membrane structure. Based on lipid-coated aqueous droplets adhering in an oil medium, the DIB method enables linking droplets mechanics and membrane mechanics for interpreting changes in the membrane structure through droplet-droplet adhesion and the angles of contact at the droplet intersections. It is a purely fluidic method where the droplets are minimally constrained and the membrane is free to expand dependent on the interfacial tensions. By observing the changes in the droplets’ adherence while measuring the membrane properties through electrophysiology, one can calculate membrane’s tension, specific capacitance, applied stress and resulting strain.
Pendant drop tensiometry was used to investigate the changes in the monolayer surface tension with increasing amount of cholesterol. Higher monolayer tensions were observed with increasing cholesterol concentrations, along with increasing bilayer tension and adhesion energy. Next, membrane properties with varying cholesterol were measured. The membranes exhibited a non-linear reduction in thickness with the increasing cholesterol mole fractions. To understand cholesterol’s impact on membrane’s structure and rigidity, an electric field was gradually increased across the membrane until failure providing membrane compression beyond the equilibrium dimensions. The membrane thickness along with the corresponding electrical stress were measured for each voltage. Plotting the pressure versus membrane thickness for different cholesterol concentrations shows that cholesterol does not affect the critical stress at failure. However, it significantly reduces membrane thinning. Adding 30% cholesterol reduces membrane thinning by more than half. This indicated that cholesterol enhances the stiffness of the membrane, making it more rigid and resistant to transverse compression.

Controllability of the shape, size, and internal structure of block copolymer (BCP) particles as well as their uniformity is crucial to determine their utility and functionality in the practical applications. Here, we demonstrate the particle restructuring by solvent engineering (PRSE) strategy that combines membrane emulsification and solvent annealing processes, to produce monodisperse particles using functional BCPs with controlled size, shape, and inner structure. Importantly, the advantage of our PRSE approach is the general applicability to various types of functional BCPs including polystyrene-block-poly(1,4-butadiene) (PS-b-PB), polystyrene-block-polydimethylsiloxane (PS-b-PDMS), and polystyrene-block-poly(4-vinylpyridine) (PS-b-P4VP). PRSE starts with producing monodisperse BCP spheres in a wide range of particle size (from hundreds of nanometers to tens of several microns) using membrane emulsification, followed by successful transformation to shape-anisotropic BCP particles by solvent annealing under neutral wetting condition. Monodispersity of particle size was maintained during the PRSE, and the shape transformations to both prolate and oblate shaped ellipsoids were successfully achieved. Our approach was effective in controlling the aspect ratio (AR) of particles over wide ranges, from 1.0 - 2.5 and 1.0 - 5.0 for prolate and oblate ellipsoids, respectively. Also, observed AR of the particles was well-supported by theoretical calculation based on the model describing the particle elongation. Further investigation on the shape transformation kinetics during the PRSE revealed that the morphology transformation was driven by reorientation of BCP microdomains, whose kinetics was strongly associated with the overall molecular weight of BCP and the annealing time.

Surface energy and hydro-affinity play a significant role in device manufacturing and semiconductor processing. Surface reactivity, bonding and passivation is as critical as structure and composition for gate oxides, ohmic contact formation, heterostructures, opto-electronic and piezo-electronic device integration.

In this work, the effect of doping species upon Si(100) total surface energy, γ^T, hydro-affinity, and reactivity, is investigated for three n-type dopants, phosphorus, arsenic or antimony, and two p-type dopants, boron or gallium, [2] by new quantitative measurements via Three Liquid Contact Angle Analysis (3LCAA) [1] correlated with Ion Beam Analysis (IBA), as a function of concentration.

3LCAA and the van Oss-Chaudhury-Good (vOCG) theory yield γ^T from three interactions, the Lifshitz-van der Waals interaction energy γ^LW, the interaction energy with electron donors γ^-, and acceptors γ⁺, measured via the contact angles of three liquids.

A new automated image analysis algorithm, DROP™, [3], yields contact angles within <1° and γ^T within 3% in minutes. DROP™’s accuracy makes possible quantitative analysis of γ^T, γ^LW, γ^-, and γ⁺ as a function of doping species and concentrations.

Ion Beam Analysis (IBA) can detect oxygen coverage and elemental composition. High resolution IBA combes ��(O16, O16)�� 3.039±0.01 MeV nuclear resonance with <111> channeling detects oxygen within 5%, or about 6x10¹⁴ at/cm². Rutherford Backscattering Spectrometry (RBS) combined with <111> axial channeling can detect dopants with a sensitivity that increases quadratically with their atomic number, thus ~4x10²⁰ boron/cm³, or 1 at.%; ~4x10¹⁹ phosphorus/cm³, or 0.1 at.%; ~1x10¹⁹ gallium/cm³, or 0.02 at.%; ~1x10¹⁹ arsenic/cm³, or 0.02 at.%; and ~4x10¹⁸ antimony/cm³, or 0.01 at.% [2]. Oxygen coverage measured by IBA, along with doping levels and dopant species, can be correlated with γ^T, γ^LW, γ^-, and γ⁺ measured by 3LCAA on both as-implanted and Rapid Thermal Annealed (RTA) Si(100). Si(100) amorphizes with increasing ion dose and defect concentrations, which catalyzes native oxide formation. During RTA, Si(100) recrystallizes and incorporates dopants into substitutional lattice sites. However, native oxide formation and dopant segregation modify final native oxide composition and thickness, as well as total surface energy and the surface interactions with donors and acceptors.

Higher dopant concentrations in both p-type and n-type doped Si(100) result in larger oxygen coverage and thus thicker native oxides, as measured by IBA, and lower surface energies as measured by 3LCAA. As expected, resulting surfaces are more hydrophobic and less reactive than as-implanted Si(100).

3LCAA also finds that as Si resistivity increases, γ^T also increases, showing that lower doping concentrations result in a more hydrophilic and more reactive surface [3]. Implanted Si is then etched, which removes surface oxides, and recrystallized by RTA to electrically activate dopants, regrowing a new native oxide. Longer RTA processes cause thicker oxides to form, along with higher segregation of n-type dopants at the Si-SiO₂ interface. Hence, electrically activated n-type dopants enhance oxidation. The opposite effect is observed for recrystallized p-type doped Si, which shows thinner native oxides than as-implanted Si. 3LCAA, combined with IBA, can accurately measure changes to native oxide surface energies, hydro-affinity and reactivity as a function of dopant species and dopant concentration. These insights from 3LCAA expand understanding of dopants’ influence upon native oxide formation and of surface engineering for heterostructure formation by heteroepitaxy, conventional wafer bonding and NanoBonding™ [3].

[1] US Patents #9,018,077; #9,589,801, Herbots N. et al (2015); (2017)
[2] P. A. Cullen, Ph.D, Massachusetts Institute of Technology PhD Thesis (1991)
[3] US Patents Pending, Nicole Herbots, Saaketh Narayan, Jack Day, et al. (2018)

Ultrahigh-pressure (HP) polymorphs such as diamond exhibit unique physical properties. There are many HP polymorphs that can be recovered to the ambient pressure to be utilized in industry. However, some of the HP polymorphs that are thermodynamically stable under HP (> 10 GPa) transforms into amorphous structure or different crystalline phases during decompression. If this back-transformation can be suppressed, we can obtain some useful HP polymorphs to be utilized in future. As a possible solution, we combined the epitaxial thin film growth technique with HP synthesis method using a multi-anvil apparatus.

We first investigated a HP-phase, α-PbO₂-type titanium dioxide (α-TiO₂, orthorhombic, a = 0.454 nm, b = 0.549 nm, c = 0.491 nm). Unfortunately, most of the reported data was about the product in form of powder, and only a few reports about the fabrication of single crystals are currently available. In particular, single-phase α-TiO₂ epitaxial thin films has not been reported. In this study, we report the fabrication of the high-quality single-phase α-TiO₂(100) epitaxial thin films.

Thin films of epitaxial rutile TiO₂(100) (thickness: ~100 nm) were deposited as precursors on Al₂O₃(001) (5 mm in diameter and 0.5 mm in height) using pulsed laser deposition. HP treatment for thin films was performed using a Kawai-type multi-anvil high-pressure apparatus. The precursor thin film was heated up to 1000°C under HP of 8 GPa, and then kept for 0.5 h. After the heating step, cooling was started down to room temperature (RT), followed by decompression.

As a result, a successful single-phase α-TiO₂(100) epitaxial thin film has been obtained. It should be stressed here that rocking-curve full width at half maximum of the 200 peak showed a quite small value of 0.113°, indicating very high crystallinity. Our present study indicates that HP treatment to thin film samples allows us to fabricate high-quality HP-phase epitaxial thin films.

Nanostructured carbons are common components of electrodes used for energy-storage and
-conversion applications to impart enhanced electronic conductivity that sustains high-rate operation, to support dispersed active materials (e.g., electrocatalytic metals and oxides), and to contribute to the physical structure and integrity of the electrode. Often overlooked in the design and characterization of such electrode structures is the critical nature of electronic/chemical/physical interactions at the junction of carbon with the nanoscale phases that impart desired storage or electrocatalytic functionality. Our own experience with such multifunctional compositions shows that we still “leave on the table” too much energy and power performance as a consequence of sub-optimal interfacial design. In order to explore these fundamental questions, we step back from the complexity of 3D electrode architectures to planar electrode configurations that otherwise mimic the material characteristics of practical electrode structures. Carbon films deposited onto planar substrates by vapor-phase pyrolysis of benzene and related monomers serve as model interfaces that we then modify with charge-storing metal oxides, electroactive polymers, and metal colloids [1]. The resulting 2D interfaces are characterized by classical electroanalytical methods to determine fundamental properties such as electron-transfer rate constants and impedance-derived response times. These substrates are also amenable to interrogation by scanning probe microscopy, including in-situ monitoring of conductivity and surface morphology under potential/current control. Lessons learned from these model 2D interfaces are readily applied to the redesign of practical 3D electrode structures in next-generation electrochemical capacitors and batteries.

[1] J. F. Parker, G. E. Kamm, A. D. McGovern, P. A. DeSario, D. R. Rolison, and J. W. Long, Langmuir, 33 (2017) 9415–9425.

Myriads of electrode materials with interesting nanostructures have been fabricated for a range of electrochemical applications. Beyond the increases in their surface area, unconventional properties have been observed along with the development of new approach and methodologies to analyze and utilize such properties. For instance, once a redox-active molecule diffuses into the nanoscale cavity of the electrode, it frequently collides with the inner electrode surface before it escapes the cavity, which is termed nano-confinement effect. This effect can specifically amplify the signals of slow electron transfer reactions (e.g., glucose oxidation) and has been established as a new mechanism for sensing the target materials that have slow electron transfer kinetics.

In this study, we fabricated a Au nanopillar electrode and devised a new electrochemical platform using redox cycling reaction to utilize the nano-confinement effect of the Au nanopillar electrode, which is applicable even to the fast electron transfer reactions as demonstrated by facilitating the electrochemical reaction of a specific biomarker (i.e., pyocyanin which has fast electron transfer kinetics) of bacterial infection (i.e., Pseudomonas aeruginosa). The Au nanopillar electrode was fabricated by first treating a polymer substrate with plasma to render an array of polymer nanopillars on the substrate and then depositing Au onto the plasma-treated substrate to the appropriate thickness to form nanoscale inter-pillar space. The Au nanopillar surface was modified with the redox cycling counterpart of pyocyanin so that the pyocyanin confined in the inter-pillar space underwent repetitive cycles of reduction (by the electrode) and oxidation (by the redox cycling counterpart tethered to electrode surface) reactions resulting in the amplification of the pyocyanin signal. This redox cycling-assisted signal amplification was clearly observed with the Au nanopillar electrode with higher amplification by taller Au nanopillars while the amplification was insignificant for the flat Au electrode, demonstrating the nano-confinement effect of Au nanopillar electrodes. As a result, the sensitivity in measuring pyocyanin was improved by the Au nanopillar electrode compared with the flat Au electrode. We believe that this new platform based on the surface-bound redox cycling reaction could overcome the limit in applicability of the nano-confinement effect.

In this paper we report high voltage MOS and Schottky Diode CV techniques for silicon and SiC power devices. 4H Silicon carbide is a wide bandgap semiconductor suitable for high voltage power electronics and RF applications due to high avalanche breakdown critical electric field, and thermal conductivity. The performance of various power devices, which may include MOSFET and Static Induction Transistor (SIT), can be affected by the deep level traps in the substrate and the oxide interfacial defects. We have characterized deep level trap (High Voltage Schottky Diode HF CV) and oxide interface trap densities (High Voltage HF MOS CV), measured the device channel doping profile for both 4H SiC and silicon, gate metal workfunction, and simulated the effects on DC/AC performance.

TCAD simulation is done for a high voltage RF SIT. Pinch voltage (V_Pinch, as defined by the gate voltage required to turn the device off for a certain applied drain voltage) may be affected by the interface charges. Generally speaking, the V_Pinch is very sensitive to the channel doping concentration.

If the oxide / SiC interface charges are positive charges, the V_Pinch is higher. Please notice the gain is also reduced – this may affect the AC performance.

The depletion width is derived from converting the measured Schottky Diode high frequency CV to Depletion Width vs. Voltage. The 1/C² vs. V plot may also be used to extract the gate metal work function.

SIT gate metal workfunction is extracted to be 5.42V.

SIT channel doping profile is generated by converting the high voltage HF Schottky Diode CV to 1/C² vs. V plot. When the frequency varies from 100 KHz to 1MHz, there is a small shift of the extracted doping profile. The difference between the measured doping concentrations at these voltages may be caused by the deep level traps in the 4H SiC epitaxial layer. When the test frequency is lower than 100 KHz, there is no significant shift. The SIT channel doping profile is extracted from the HF Schottky Diode CV. The deep level trap density is the difference between the two doping profiles.

The Surface Potential at the thermal oxide / 4H SiC interface vs. gate voltage is derived by the integration technique for the measured HF MOS CV.

Oxide / Silicon or 4H SiC interface trap density may be extracted by comparing the measured high-frequency MOS CV and a theoretical, defect-free high frequency MOS CV:

D_it (eV^-1×cm^-2) = 1/q ×C_ox × d(Delta_V_G)/ dØ_S (1)

Where q is the electron charge, C_ox is the oxide capacitance, D_it is the oxide interface defect density, Delta_V_G is the gate voltage difference between measured and theoretical HF MOS CVs, corresponding to the same capacitance in the CV plot, and Ø_S is the surface potential at the SiC / oxide interface. Surface Potential vs. Delta_V_G and Interface Trap Density vs. Surface Potential plots are presented in this paper.

Summary:

Oxide / Si or 4H SiC interface trap density, deep level trap density, device channel doping profile, gate metal workfunction are characterized with high voltage (100V), high frequency (up to 1MHz) MOS and Schottky Diode CV techniques. TCAD simulation shows the effects of these defects on DC and AC performance. The deep level trap density of 4H SiC (average) = 1×10¹⁵ cm^-3 about 1 order of magnitude higher than lightly doped silicon (= 1×10¹⁴ cm^-3). The interface state density of 4H SiC is (average) = 4×10¹¹ cm^-2 eV^-1, also about 10 times higher than silicon.

Direct wafer bonding is replacing hetero-epitaxy in semiconductor-based hetero-structures such as tandem solar cells and sensors. Bonding of Li-based piezo-electronic perovskites, specifically LiTaO₃ (100) and LiNbO₃ (100) to Si (100) and a-quartz SiO₂ (100) is investigated because these perovskites exhibit unique acoustic and piezoelectric properties used in Surface Acoustic Wave (SAW) devices. A direct wafer bonding process called NanoBonding^TM [1] can bond LiTaO₃ (100) or LiNbO₃ (100) to Si using surface energy modification in ambient, low temperature (<220°C) conditions with neither plasma activation nor UHV.
The principle underlying NanoBonding^TM is to cross-bond at the molecular scale two surfaces over large interfacial domains, creating a 2-D bonding interphase. NanoBonding™ uses surface engineering to modify surface energy to promote electronic exchange during contact and bond activation.
Electron exchange is promoted between the surfaces by engineering them as hydrophilic-hydrophobic pairs, with each surface in a far-from-equilibrium state. The hydro-affinity of the initial surfaces is measured quantitatively using surface energy analysis via 3LCAA and the van Oss-Chaudhury-Good (vOCG) theory. Next, the hydro-affinity is modified to bring the surface far-from-equilibrium - from a hydrophobic to a hydrophilic state or vice-versa to bring the paired complementary far-from-equilibrium states.
NanoBonding™ includes planarization of both surfaces at three scales - nano-, micro-, and macro-scale. Creating extended atomic terraces by etching and reducing macroscopic wafer warp enable for direct mechanical contacting over large surface areas (Nano-contacting).
After Nano-contacting, cross-bonding is activated via electromagnetic radiation, such as heat, or ultra-violet illumination, or via isotropic pressurization using steam [1].
The surface energy ��^T is thus measured by 3LCAA on the “as received” LiTaO₃(100), LiNbO₃(100), Si(100), and a-quartz SiO₂(100) 4” wafers from TDC Corp. The uniformity of ��^T across the 4” LiTaO3 wafer is mapped in 10 locations in a Class 100 hood using 10 µL droplets. The three liquids used are 18 MΩ DI water, glycerin, and α-bromo-naphthalene. ��^T is then computed via the vOCG theory from the three interaction components that yield ��^T, the LifeChat-Van der Waals molecular interaction energy ��^LW, the interaction energy of the surface with donors, ��⁺, and with acceptors (��^-). Increasing the values of ��^- and ��⁺ promotes ionic bonding, electronic exchange and molecular cross-bonding.
��^T across the 4” LiTaO₃ wafer averages 43.3 ± 2 mJ/m², thus native LiTaO₃ surfaces are hydrophobic. Native Si(100) surfaces, by comparison, are always hydrophilic and have a typical surface energy of 53.0 ± 0.2 mJ/m², as measured over several 4” wafers. However, ��^T uniformity Si(100) 4” wafers is better by an order of magnitude than on LiTaO₃. Several NanoBonding™ experiments were then conducted with LiTaO₃ (100) on Si (100), and a-quartz SiO₂ (100) where the hydroaffinity of LiTaO₃ was reversed to hydrophilic by etching with aqueous HF and the hydroaffinity of native Si (100) reversed to hydrophobic.
Ion Beam Analysis and XPS is used to investigate the surface stoichiometry of LiTaO₃(100) before and after etching to correlate to ��^T and NanoBonding™ results.

[1] US Patents #9,018,077; #9,589,801, Herbots N. et al. (2015); (2017)

The deformation behavior of nanostructured metallic multilayers has been extensively investigated because of their superior mechanical properties, primarily via nanoindentation and micropillar compression experiments. Here, we report tensile measurements on a Cu/Co nanolaminate with 4 nm layer spacing. The nanolaminate was synthesized using magnetron sputtering at room temperature and tensile loading-unloading cycles were performed on freestanding samples parallel to the Cu/Co interface using MEMS devices. Cross-sectional TEM and XRD measurements showed the presence of a coherent Cu/Co interface. The nanolaminates exhibit a flow strength as high as 2 GPa and failure strains of 3-4%. Interestingly, significant inelastic strain recovery was observed both during and after unloading, with further recovery occurring upon heating. In addition, the stress-strain slope during the initial stages of unloading was considerably higher than the expected elastic modulus, suggesting the presence of a super modulus effect due to the large coherency strain.

Electric double layer capacitors, or supercapacitors, have high theoretical capacities and faster charge/discharge rates than Li-ion batteries. Because they store charge via electrostatic interactions between the electrode surface and ions in the electrolyte, increased specific surface area of electrode materials will increase the theoretical storage capacity of next-generation devices. Nanofiber-based electrodes have fewer issues with agglomeration and better long-term stability than nanoparticle-based electrodes, while maintaining extremely high surface area to volume ratios. As a potential electrode material, Mn₂O₃ is low-cost, naturally abundant, and environmentally benign. Electrospinning with subsequent thermal decomposition is a facile fabrication method for creating Mn₂O₃ nanofibers. Composite fibers consisting of a coordination polymer, polyvinylpyrrolidone, and oxide precursor, manganese acetate tetrahydrate, are electrospun. Calcination is used to burn out the polymer matrix and convert the salt to manganese oxide while retaining the high aspect ratio fiber morphology.

Although Mn₂O₃ nanofibers have been prepared by thermal decomposition of a coordinating polymer-oxide precursor composite fiber, the influence of calcination conditions on specific surface area have not been well-explored. The proposed work will examine the effects of calcination time, temperature, and environmental composition on fiber surface areas and corresponding electrochemical capacitance as compared to nanoparticle-based electrodes. Thermal analysis will be performed with thermogravimetry and differential scanning calorimetry to determine glass transition and decomposition temperatures of the as-spun composite fibers. The post-calcination fibers will be examined with X-ray diffraction to confirm phase and obtain information on crystallite size. Scanning electron microscopy will be used for morphological examination and qualitative examination of porosity resulting from changes in calcination parameters. To quantify changes in specific surface area, the Brunauer-Emmett-Teller (BET) method will be used on calcined fibers and compared to values measured for nanoparticles prepared from the oxide precursor. Increased surface area detected by BET may not necessarily be accessible to ions in the electrolyte, so to evaluate the accessible surface area, the fibers and particles will be processed into working electrodes. Electrochemical analysis including cyclic voltammetry and linear voltammetry will be utilized to assess the accessible surface area of fiber-based vs. particle-based electrodes based on measured capacitance. Preliminary results have shown that modifications to the gas composition and calcination time result in varied porosity, surface features, and cross-sectional morphologies. An understanding of the relationship between processing conditions and fiber morphology will allow for tunability of enhanced surface area, electrochemical capacitance, and supercapacitor performance.

HfSiO₄ (hafnon) can be used as a layer environmental barrier coating (EBC) materials placed between silicon dioxide and hafnia. Hafnon has a good lattice thermal expansion matching between silica and hafnia and can prevent oxygen diffusion in EBCs, thereby potentially lengthening the lifetime of EBCs. In this work, we conduct a comprehensive investigation of mechanical, thermodynamic and thermal transport properties of hafnon using first-principles density functional theory (DFT) calculations and compare our results to experiments performed on this material. The volumetric coefficients of thermal expansion (CTE) calculated by the quasi-harmonic approximation are in the range 9.18 - 19.07×10^-6 K^-1, when the temperature increases from 300 to 1500 K, in agreement with X-Ray Diffraction and dilatometer measurements. The thermal conductivity is nearly 10 Wm^-1K^-1 at room temperature (RT) using the Boltzmann transport theory. The time-domain thermoreflectance technique utilized to measure thermal conductivity yields 12.4 Wm^-1K^-1. The mechanical properties, such as bulk modulus and elastic constants are also compared between the first-principles calculations and the RT nanoindentation technique. This study validates our computational approach and can be used to further study thermomechanical properties of other high-temperature oxides.

Medical micropumps that utilize Magnetic Shape Memory (MSM) alloys are small, powerful alternatives to conventional pumps because of their unique pumping mechanism. This mechanism, the transfer of fluid through the emulation of esophageal contractions, is enabled by the magneto-mechanical properties of a shape memory alloy and a sealant material. Because the adhesion between the sealant and the alloy determines the performance of the pump and because the nature of this interface is not well characterized, an understanding of sealant-alloy interactions represents a fundamental component of engineering better solid state micropumps in particular, and metal-polymer interfaces in general. In this work we develop computational modeling techniques for investigating how the properties of sealant materials determine their adhesive properties with alloys. In particular, we develop a molecular model of the sealant material polydimethylsiloxane (PDMS) and parameterize its interactions with Ni-Mn-Ga alloy surfaces for use in molecular dynamics simulations. We perform equilibrium molecular dynamics simulations of the PDMS/Ni-Mn-Ga interface to iteratively improve the reliability, numerical stability, and accuracy of the models and associated data workflows. In sum, we develop a model combining OPLS-UA and UFF force fields for simulating PDMS/Ni-Mn-Ga interfaces and demonstrate its promise for informing the design of more reliable MSM micropumps.

A computational method has been developed to calculate the pressure required for grain boundaries to overcome the pinning effects of second phase particles. Multiple Pt bicrystals were constructed with small concentrations (<= 0.5 atom %) of Au segregated to the boundary using the Monte Carlo method. Molecular dynamics (MD) simulations employing the artificial driving force method were carried out on said boundaries to measure the relationship between boundary type, solute concentration, and boundary velocity. It was found that for a given solute concentration there exists a critical driving pressure necessary for grain boundary motion to occur. Sub-critical driving pressures resulted in flexing of the grain boundary about the pinning precipitates, but with no large scale grain boundary movement. Above critical driving pressures resulted in grain boundary motion although the amount of time required for grain boundaries to break loose from the pinning structures varied considerably. At solute concentrations of 0.1%, solute segregation to the boundary did not occur. The resulting lack of precipitates caused boundaries to behave no differently from pure boundaries, although at real timescales a reduction in mobility would be exhibited because of solute drag. Development of a method to quantify the effect of different solute concentrations on the stability of different grain boundaries will prove useful for understanding phenomena such as abnormal grain growth, as well as aid in the engineering of stabilized nanocrystalline alloys.

The ice accretion on metal surfaces in cold environments occurs through the process of water droplets freezing, which spans a wide range of length scales necessitates a multiscale computational analysis that can link the atomic-scale activities to its micrometer-level behavior.
Firstly, in this work, with aim of understanding the structure of grain boundary of polycrystal ice, we establish a computational protocol for measuring the energy on the grain boundary of bi-crystal ice by performing a coarse-grained atomistic simulation. Secondly, we apply this procedure to the bi-crystal ice in a complex chemical environment. Results show that the microstructure of the grain boundary of the ice is sensitive to the concentration of Na+ and Cl-. Then, to determine the parameter among several, droplet size, metal surface roughness, and chemistry, that controls the grain structure, product phases (cubic or hexagonal ice) in polycrystalline ice accreted on metal surfaces, coarse-grained atomistic simulations of water droplet freezing on an oxidized aluminum substrate are also conducted. By tracking the contact line position and the contact angle throughout the image sequences of the snapshots from the simulations, the contact angle change during a water droplet freezing can be monitored as a function of the real time. It is qualitatively shown that the contact angle decreases when the volume fraction of the ice crystals in a water droplet increase.

Ultrathin transition-metal oxide and dichalcogenide layers ranging from monolayer (2D) thickness to a few nanometers, exhibit a wide range of distinct physical properties when compared to their bulk counterparts. There is growing interest in understanding the crystallization behavior of these materials when approaching the 2D regime, in order to optimize the growth of already observed phases and realize the synthesis of predicted ones. Information on nucleation rates and critical nucleus size in homogeneous nucleation processes, the relative importance of various defects in heterogeneous nucleation, grain growth rates, and interface evolution in the 2D regime is highly sought to enable interface engineering in 2D materials. Tungsten trioxide (WO₃) is a particularly interesting system to study in ultrathin configurations because it exhibits a wealth of crystallization-dependent properties. In addition to its catalytic and ion-intercalation derived properties, such as charge storage and electrochromic behavior, it also shows evidence of a superconducting phase, an insulator-to-metal transition that may be achieved via electrostatic gating, and a transition from the conventional monoclinic, orthorhombic, and hexagonal polytypes observed in bulk and nanophases, to a stable free-standing, 2D single-layer honeycomb-like structure that is semiconducting and predicted to have a very high cohesive energy (10.5 eV) as well as one of the highest in-plane stiffness among metal oxide and dichalcogenide 2D materials (250 N/m). In this work we investigate the crystallization behavior of ultrathin WO₃ formed after oxidation of tungsten (W), using X-ray diffraction (XRD), atomic force microscopy (AFM), and impedance spectroscopy (IS). Polycrystalline W films of 10-nm thickness, deposited by atomic layer deposition (ALD) on n-type (001)-oriented silicon substrates, were annealed in oxidizing environment in a tube furnace at temperatures ranging from 275°C to 600°C. XRD on films annealed at 600°C shows the formation of the monoclinic phase of WO₃. The XRD patterns for annealed films are dominated by the sharp monoclinic (200) reflection at 2θ = 24.45°, indicating that the WO₃ crystallites exhibit preferential orientation with their [200] axis aligned perpendicular to the substrate. AFM shows surface morphology comprising distorted cubic-like features consistent with the underlying monoclinic structure observed by XRD. The AFM scans clearly reveal that the films are dominated by out-of-plane interfaces normal to the [200] direction of the WO₃ crystallites. At temperatures above 500°C, large crystallites are observed in the AFM scans, with heights in the 10-30 nm range, indicating significant 3D restructuring due to crystal growth after oxidation. The crystallite number density exhibits an Arrhenius relationship with the annealing temperature with an activation energy of 110 kJ/mol. Impedance spectroscopy measurements between 1 Hz and 7 MHz on thin film samples after the deposition of top nickel contacts, show Nyquist impedance arcs that depend on the annealing temperature and ambient. The impedance of samples with various WO₃ thicknesses, degrees of 3D restructuring, and crystallite number densities are analyzed in terms of equivalent circuits that yield the “grain interface” and “grain interior” contributions to the resistance and capacitance of the samples. The dependence of these electrical characteristics is compared with the structural features obtained by XRD and AFM to identify distinct kinetic regimes present during W oxidation, as well as crystal nucleation and growth of the WO₃ ultrathin layers.

The structural components of a very High Temperature Reactor (VHTR) are exposed to trace amounts of gaseous elements and moisture. These impurities can adversely affect the integrity of the reactor structural components. Among the impurity gases, carbon-bearing gases – methane, carbon monoxide and carbon dioxide – have significant detrimental effects on the protective oxide scales, which grow on nickel-based alloy surfaces, leading to carburization of the alloy.

In this work, we investigate the effect of carburizing gases on the stability of chromium oxide scales using electronic structure simulations. We have utilized Kohn-Sham density functional theory (DFT) as implemented in the Vienna ab initio simulation package (VASP) with the generalized gradient approximation; to describe the strong correlations between the d electrons of chromium, we have utilized the DFT+U approach. The molecular structure of the oxide surface without the gases is first optimized structurally. Fully relaxed carburizing gases are then allowed to be adsorbed on to the surface. We finally employ ab initio molecular dynamics (AIMD) simulations at high temperatures (up to 1200 K) to calculate the structural and dynamic properties of the interface, and to elucidate the mechanisms of thermal stability of the interface region at different temperatures and impurity coverage of the surface.

References
[1] Young, D. J. and J. Zhang (2018). "Alloy Corrosion by Hot CO₂ Gases." JOM 70 (8): 1493-1501.
[2] Funk, S., T. Nurkic, B. Hokkanen and U. Burghaus (2007). "CO2 adsorption on Cr(110) and Cr2O3(0001)/Cr(110)." Applied Surface Science 253 (17): 7108-7114.

Segregation of the alloying elements at the Grain boundaries (GBs) in a polycrystalline materials leads to a nanoscale chemical and structural variations along the GB region that can significantly alter the material’s performance. Hence, the experimental investigation of the atomic structure and chemistry of the grain boundaries by resolving the sites and chemical identities of the atoms comprising the interface is crucial to fully understand the implication of the atomic segregation phenomenon. Herein, we present the atomic scale analysis of ordered-structures induced by segregation of Cu atoms in Al 7075 alloy GBs along with the detailed examination of microstructural features (texture and grain boundary character distribution) by employing advanced microscopic characterization techniques: STEM-HAADF imaging, spectroscopic (EDS and EELS) analysis, and scanning precession electron diffraction (PED).
The <111> strong texture and dominant low coincidence site lattice (CSL) GBs were revealed in the films. Results demonstrated the segregation of Cu in all types of GBs, but with varying segregation patterns. Atomic-scale structure of each type of GBs showed two types of Cu GB segregation behaviors namely, point (highly segregated atomic column surrounded by low segregated columns, misorientation < 28^o) and parallel array (two highly segregated columns opposite to each other across the interface surrounded by low segregated columns, misorientation > 28^o). In addition to the single atom-interstitial hollow site (point) segregation behavior (predicted in theoretical studies and assumed to analyze the segregation effects), substitutional core site segregation of two columns of Cu atoms forming an ordered-structures with a parallel array of segregating atoms along the interface has been demonstrated. This is the first time that this type of parallel array segregation behavior has been reported. Furthermore, based on intensity profiles and scanning transmission electron microscopy Z-contrast principle, non-uniformly (high and low) segregated mixed atomic columns were predicted at the grain boundaries. We believe the knowledge of this type of experimental insight of the atomic scale arrangements of adsorbate on the GBs can potentially help to develop strategies for engineering the design of alloy compositions.

Zinc oxide (ZnO) is one of promising semiconductors for numerous (opto)electronic applications because of excellent properties such as a wide band gap, a high exciton binding energy, and large piezoelectric coupling coefficients. ZnO crystallizes in the wurtzite hexagonal structure, which forms the polar c-plane and non-polar a and m planes. ZnO thin films of different preferential crystal orientations exhibit different properties, which leads to various applications.
In ZnO films grown by chemical vapor deposition (CVD) techniques, the crystal orientation is mainly influenced by the orientations of the nucleation layer, which is, in turn, affected by diffusion of atoms on the substrate surface during film growth. Several parameters affect the diffusion process such as oxidation reactions, type of substrates, growth temperatures, and gas flow rates. In the growth of metal oxide films by CVD, oxidation rate can be controlled by varying the supply amounts of oxygen and metal sources.
Mist-CVD is one of the non-vacuum metal oxide film growth techniques¹ capable of controlling the oxygen and metal supply amounts precisely. The precise control of the oxygen/metal ratios was realized by using the mist-CVD system with two solution chambers² storing oxygen and metal sources (H₂O and Zn precursor in this case) separately. In this work, we found strong dependences of the crystal orientations, growth rates, and surface morphologies on the [H₂O]/[Zn] ratios in the ZnO films grown by the two chamber system. The equilibrium reaction of thermal decomposition of the Zn precursor was enhanced by addition of H₂O leading to an initial increase in the growth rate with the [H₂O]/[Zn] ratio. However, excessive addition of H₂O caused a decrease in the growth rate. Moreover, H₂O also influences the film crystal orientation. The diffusion of zinc and oxygen reactive atoms on the substrate surface during film formation was influenced by the [H₂O]/[Zn] ratios, which resulted in different ZnO preferential growth orientation and film surface morphology.
The details about fabrication of the ZnO films using the mist-CVD system with two solution chambers and a plausible ZnO film growth mechanism with different [H₂O]/[Zn] ratios will be presented in the conference.
References

[1] T. Kawaharamura, Jpn. J. Appl. Phys. 53, 05FF08 (2014).
[2] G.T. Dang, T. Yasuoka, Y. Tagashira, T. Tadokoro, W. Theiss, T. Kawaharamura, Appl.Phys. Lett. 113, 062102 (2018).

Electric double-layer capacitors (EDLCs), also known as supercapacitors, are energy storage devices that deliver high power densities and long cycle lives but exhibit lower energy densities than Li-ion batteries. Improving the energy density of EDLCs without sacrificing their power density or lifetime is desirable for numerous energy storage applications. In order to rationally design higher energy and power density devices, an improved understanding of how electrolyte and electrode properties affect the performance of EDLCs is critical. Recently the use of ionic liquid mixtures as electrolytes has been experimentally reported to improve EDLC performance under certain conditions. However, existing theoretical and computational work on ionic liquid mixtures in EDLCs remains limited, and there remains a need for detailed understanding of how ionic liquid mixture electrolytes affect EDLC performance.

Using all-atom molecular dynamics simulations, we study the nanoscale adsorption behavior of mixtures of ionic liquids comprised of 1-ethyl-3-methylimidazolium (EMIM⁺), bis(trifluoromethylsulfonyl)imide (TFSI^-), and tetrafluoroborate (BF₄^-) ions near planar carbon electrodes carrying various surface charges that mimic externally applied voltages. Near uncharged electrodes, we find that ion-electrode van der Waals interactions have a significant impact on the relative population of different ions in the first interfacial layer. By characterizing ion densities within a few nanometers of charged electrodes, we find that charging of the electrodes leads to greater overall changes in cumulative density of the smaller ion in each pure ionic liquid compared to the larger ion, as the larger ions tend to instead reorient and locally restructure close to the interface. Finally, in ionic liquid mixtures we identify an effective anion exchanging phenomenon near negatively charged electrodes that enhances counter-ion adsorption and provides a mechanism for capacitance enhancement. More broadly, our observation of modest capacitance differences between the ionic liquid compositions suggests that achieving substantially higher energy densities will ultimately also require tuning the structure and properties of the electrodes.

Grain boundaries (GBs) migrate via the motion of disconnections (line defects in the GB that have both step and dislocation character). While the motion of symmetric tilt boundaries via this mechanism is reasonably straightforward, application to more general GBs requires additional assumptions. In this poster, we report the results of molecular dynamics (MD) simulations of the motion of asymmetric GBs. In particular, we examine the faceting of migrating asymmetric GBs. In some cases, faceting may occur during migration even when no faceting is observed in the stationary case. We analyze the MD results in the context of a disconnection description of GB migration. Finally, we discuss implications of dynamic GB faceting on grain growth.

Thermotropic polymers own thermally tunable transparency which have shown applications in intelligent solar control coating, privacy windows, displays, and optical/thermal sensors. Usually, thermotropic polymer systems are constructed by physically dispersing phase transition materials in transparent hosting materials. However, they usually possess bad interfaces either between the dispersion and the matrix, or between the thermotropic system and the supporting substrates (e.g., glass), resulting in poor mechanical properties, weak interfaces to substrates, or bad long-term stability. Herein, we demonstrate a novel chemically interconnected thermotropic polymer, which is obtained by reacting dodecanedioic acid (DDA) with glycerol. In the system, some of DDA molecules were crosslinked to form a polyester network, poly(glycerol-dodecanoate) (PGD). Other grafted but non-crosslinked DDA molecules form semi-crystalline domains which possess a solid-liquid phase transition within the PGD matrix. The phase transition offers the resulting hybrid materials with tunable optical transparency. In addition, when applied for window coating, it results in tough interfacial bonding to glass substrates with toughness of > 6910 J m^-2 below its transition temperature and > 135 J m^-2 above its transition temperature. It increases the impact-resistance of the window by multiple times.

For the last several decades Time-of-flight Secondary Ion Mass Spectrometry (ToF-SIMS) became a prime tool for chemical characterization of wide range of materials and systems with sub-micrometer spatial and nanometer depth resolution. However, it lacks characterization of the morphological, physical and functional properties of the studied sample. This issue can be addressed by combination with Atomic Force Microscopy (AFM). Such combined multimodal AFM/ToF-SIMS platform enables a nanoscale characterization of both chemical and physical properties of the sample along with its surface morphology. In this case, chemical information acquired by the ToF-SIMS can be correlated with functional response measured by the AFM. Furthermore, it opens pathway for chemical characterization of the local materials behavior on the nanoscale, when local chemical phenomena induced by the physical fields of the AFM tip (e.g. mechanical or electric) are characterized by the ToF-SIMS.
Here we utilize commercial AFM/ToF-SIMS platform to explore interplay of chemical and physical properties in ferroelectrics during polarization switching in lead zirconate titanate (PZT, PbZr_0.2Ti_0.8O₃) thin films. Using this multimodal imaging platform, we demonstrated that chemical phenomena plays significant role in ferroelectric switching process. Specifically, we found that local ferroelectric switching by the AFM tip, significantly alters the chemical composition in the 3-nm-thick surface layer of the sample, forming reversible concentration wave, of Pb⁺ ions. Furthermore, investigations of the polarization cycling in the PZT sample with copper electrodes, showed penetration of the copper cations deep into the PZT structure. This explains ferroelectric fatigue phenomenon, leading to decrease in spontaneous polarization during spontaneous polarization cycling.
Altogether, explored chemical phenomena associated with ferroelectric switching will enhance fundamental understanding of ferroelectric phenomena and aid in the practical application of ferroelectrics in devices.
This research was conducted at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility and using instrumentation within ORNL's Materials Characterization Core provided by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy.

One of our goals in Bardeen’s group is to explore the possibility of using photochromic molecules to regulate photo-adhesion. We have used various molecules including spiropyran, DASA (donor–acceptor Stenhouse adducts) and anthracenes. These photochromic molecule shows new properties under light irradiation.
In one of our first projects we have used spiropyran (SP) molecules to increase adhesion of thin films of polystyrene (PS) to glass substrate. We have shown that photochromic reactions of SP doped in PS can significantly enhance the adhesion of PS to a glass surface. Shear adhesion test studies demonstrated that adhesion of polystyrene to glass slides increased by factor of 7 after irradiation. We hypothesize that the adhesion changes arise from localized polymer and molecular motions that eliminate void spaces and surface gaps at the polymer-glass interface. The results show that adhesive forces between a prototypical polymer and an inorganic substrate can be modulated by photochromic reactions of embedded molecules.
In the next studies we have used DASA molecules that has different properties compared to spiropyran. In Contrast with Spiropyran, DASA molecule is negative photochrome and its volume reduced after laser irradiation. This is in opposite direction of what we observed for spyropiran and we expected to see a decrease in adhesion of films to surface. We have used three different methods to measure adhesion before and after laser irradiation including water detachment, adhesion test and shear test and in all samples the adhesion decreased. Our experiments suggest that molecules’ photo induced shrinking may increase voids and pores and these changes will leads to a decrease in adhesion. Reduction of chemical interaction of molecules to surface may be another reason for this change.
As an application we tried to encapsulate organic dye -as a mode- between two glass surfaces which glued together by PS/DASA polymer. Then we measured time that the adhesion would break in water and irradiated samples released dyes 10 times faster than non-irradiated samples.
In conclusion we showed that we can remotely increase and decrease adhesion using various photochromics.

Grain boundaries (GBs) move by the formation and propagation of disconnections; line defects with both dislocation and step character. During the evolution of a polycrystalline microstructure, disconnections pile-up against triple junctions (TJs). This implies that TJs accumulate a net Burgers vector, B. As the GBs migrate, the net Burgers vector at the TJ grows, creating a back stress that effectively shuts off additional disconnection motion and stops GB migration. TJs have many options to relax the Burgers vector and back stress: (1) accumulation of cancelling disconnection from the other GBs meeting at the TJ, (2) emission of other defects (e.g., lattice dislocations, twins,...), and (3) formation of new types of disconnections along the GB with opposite sign Burgers vector. I will show molecular dynamics simulations, theoretical analysis, and continuum simulations that show what is going on at TJs during GB migration and examine its implications for microstructure evolution.

Twinning is a common deformation mode in hexagonal close packed metals such as magnesium. This atomistic simulation study characterizes the relationship between the kinetics associated with motion of the {10-12} twin boundary in pure Mg and the nucleation, growth and coalescence of disconnection terraces on the twin boundary. This problem is resolved adopting both 2D and 3D simulation geometries. This study shows that the kinetics of twin boundary migration can only be addressed in 3D, as 2D simulations predict twin boundary migration rates at least an order of magnitude higher than 3D simulations. A simple constitutive relationship is extracted from the atomistic simulations. Such a model, accompanied by observations of disconnection terrace nucleation, growth and coalescence, reveals the existence of an autocatalytic terrace nucleation regime for the shear induced motion of the {10-12} twin boundary in Mg.

The mobility of grain boundaries plays an important role in governing the kinetics of microstructural evolution in every class of polycrystalline materials. Of particular interest is the role of bicrystallography, characterized by the macroscopic crystallographic degrees of freedom, on the underlying atomistic mechanisms governing grain boundary mobility. In this talk, I will present an algorithm to automatically identify the atomistic motion mechanisms that give rise to mobility of an interface. We use machine-learning methods, inspired by recent work in disordered solids, to correlate local structure with the susceptibility for rearrangement of grain boundary atoms. We show that it is possible to automatically identify mobility mechanisms of grain boundaries with a diverse range of crystallographic character.

Semiconductor photoelectrodes for hydrogen generation in photoelectrochemical cells are a very good demonstration example for the role which defects can play in function of electrical devices.
We present an extensive study on metal oxide heterstructures synthesized by chemical deposition methods with soft x-ray spectroscopy and electroanalytical methods.
The matching of substrate current collector and photoabsorber layer forms an interface with a malign electronic defect state which can be "healed" by introduction of a buffer layer from a different metal oxide - to the extent that the performance of the photoelectrode increases considerably. The molecular origin of this effect is uncovered by using oxygen ligand spectroscopy with soft x-rays. Specifically, a malign state in the valence band at the interface is removed upon introdcution of the buffer layer. The charge transport increases accordingly.
Different roles of defects are necessary at the surface which is in contact with a liquid electroyte. Here the electrocatalytic properties are relevant, and correspondingly the absorber material is facing a challenge which its stoichiometry can not perform well. But a chemical after.processing step, specifically gas phase reduction alters the oxidation state of the surface of the absorber to an extent that it can increase the photocurrent and hydrogen production efficiency considerably. This is evidenced both with soft x-ray spectroscopy and electron spectroscopy.
With this study is shown how the proper processing of components in heterostructures can yield better component performance in a device and thus better device performance.

[1] Braun A, Aksoy Akgul F, Chen Q, Erat S, Huang T-W, Jabeen N, Liu Z, Mun BS, Mao SS, Zhang X: Observation of Substrate Orientation-Dependent Oxygen Defect Filling in Thin WO3−δ/TiO2Pulsed Laser-Deposited Films with in Situ XPS at High Oxygen Pressure and Temperature. Chemistry of Materials 2012, 24:3473-3480.
[2] Braun A, Hu Y, Boudoire F, Bora DK, Sarma DD, Grätzel M, Eggleston CM: The electronic, chemical and electrocatalytic processes and intermediates on iron oxide surfaces during photoelectrochemical water splitting. Catalysis Today 2016, 260:72-81.

Garnet Li₇La₃Zr₂O₁₂ (LLZO) has been considered as one of the most promising solid-state electrolytes that could enable solid-state-batteries employing metallic Li anodes. However, most Li₇A₃M₂O₁₂ garnets have been reported to react with air or moisture, which results in the loss of Li and also in the appearance of a main Li₂CO₃ secondary phase that is detrimental to the Li⁺ conductivity at the Li-LLZO interface. Interestingly, different to other garnets, cubic LLZO was recently reported to be more stable in humid atmosphere regarding no structure change and conductivity degradation upon exposure to humid atmospheres. Although it is known that generally Li⁺/H⁺ exchange took place when garnet expose to moisture, the mechanism of how Li₂CO₃ forms on the interface is still unclear, especially for cubic LLZO. There is also lack of study on the independent role which CO₂, H₂O and O₂ role plays in Li₂CO₃ formation. Here we performed time resolved Ambient-Pressure X-ray Photoelectron Spectroscopy to monitor Li₂CO₃ thickness growth kinetics on Al-doped cubic LLZO in the presence of dry CO₂, O₂ assisted CO₂, and H₂O_CO₂ moisture under different gas pressures (0.01 mTorr, 1 mTorr, 100 mTorr). We elucidated that small amount of O₂ significantly catalyzes the Li₂CO₃ formation on cubic LLZO. Moreover, the minimal Li₂CO₃ formed in humid CO₂ when expose to three atmospheres of 100 mTorr, probably went through COOH^- generation and is due to proton blocking effect. We hope this kinetic spectroscopic study would shed light on protecting carbonation of LLZO during processing and help to further understand the Li₂CO₃ formation mechanism in different gases.

ACKNOWLEDGEMENTS:
The authors thank funding from the Energy Biosciences Institute and Shell International Exploration and Production Company. This research used resources of the Advanced Light Source, which is a U.S. Department of Energy, Office of Science User Facility under contract no. DE-AC02-05CH11231. The synthesis used the facilities supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. Portions of the calculation work were supported by a User Project at The Molecular Foundry and its compute cluster (vulcan), managed by the High Performance Computing Services Group at Lawrence Berkeley National Laboratory (LBNL), and portions of this work used the computing resources of the National Energy Research Scientific Computing Center, LBNL, both of which are supported by the Office of Science of the United States Department of Energy under Contract DEAC02-05CH11231.

Many materials phenomena contain signatures of the coupling between chemistry (or electrochemistry) and mechanics. The coupling in the bulk of alloys is well acknowledged, for instance in studies of coherent phase transformations or of damping phenomena. Yet, strong coupling phenomena can also be found at interfaces. This may be exploited in novel materials concepts that afford unexpected functional behavior. Nanoporous-metal based hybrid materials that behave similar to piezoelectric ceramics exemplify this concept: Bodies of nanoporous gold impregnated with electrolyte emit exceptionally robust electric signals when subjected to external load. The metal-based material may thus be considered as piezoelectric, in a literal interpretation of the term. A predictive understanding of the interfacial stress-charge coupling, as the underlying microscopic phenomenon, remains elusive even after systematic experimental and numerical studies. On the other hand, the concepts which have been established in the discussion of bulk chemo mechanical coupling provide useful insights into a related phenomenon, namely the coupling between adsorption enthalpies and strain. This topic is of high interest to the field of strain-modulated catalysis. The talk will discuss relevant issues from the perspectives of experiment, phenomenological thermodynamics, and atomistics.

Sub-10nm metallic nanocrystals are expected to have unique elastic properties, strength and deformation due to the close proximity of surfaces, but are challenging to study due to the small forces and dimensions involved. Here, we use a diamond anvil cell to deform metallic nanocrystals under quasi-hydrostatic, and non-hydrostatic pressure up to 20 GPa in order to determine deformation mechanisms in this size regime and their dependence on microstructure. In particular, we are interested in the processes that lead to pseudoelasticity, or reversible shape recovery from large strain, that have have been observed previously in small noble metal nanocrystals. Au nanocrystals with sizes from 4-10 nm are synthesized using colloidal methods. Structural changes under pressure are measured using optical absorbance spectroscopy and high pressure X-ray diffraction. Optical absorbance spectroscopy is used to monitor plasmonic resonance, which is highly sensitive to nanocrystal size and shape. High pressure X-ray diffraction reveals changes in the position and breadth of diffraction peaks, which reveals elastic strain, microstrain and the formation of crystalline defects. These measurements are corroborated by transmission electron microscopy on nanocrystals after compression. We find that 4nm Au nanocrystals are able to reversibly recover their size and shape after load is removed, even after compressed of ~20% along one axis. Crystalline defects remain in the nanocrystals after recovery. Larger Au nanocrystals do not exhibit this pseudo-elastic shape recovery.

Surface topography strongly affects the functional performance of free surfaces. For example, in electromechanical switches and other applications that make and break contact, surface topography affects the mechanical and transport properties of the contact. Further, in coatings applications, the adhesion of thin films depends strongly on the surface topography of the coated surface. However, many experimental investigations to quantify the effect of surface topography on properties are inconclusive. This is because most surfaces have multi-scale fractal-like roughness, such that the values of measured topography parameters depend on the characterization size and method.

Here we use nanocrystalline diamond films of varying grain size as model systems to investigate the quantitative connection between surface structure and surface properties. In the first part of the talk, we demonstrate the quantitative characterization of surface topography using transmission electron microscopy and its combination with conventional techniques for multi-resolution characterization. Specifically, more than 100 individual measurements at scales from Ångströms to centimeters are combined using spectral analysis to yield scale-independent metrics for surface and interface topography. In the second part of the talk, we discuss experimental measurements of adhesion as a function of topography. Results are analyzed in the context of multi-scale continuum mechanics models. The investigation demonstrates that this multi-resolution approach to topography characterization is far more effective than any conventional (single-scale) roughness descriptors at predicting adhesion and other surface properties.

Understanding the sink efficiency of interfaces under irradiation is of paramount importance to tailoring materials for radiation tolerance. Using in situ irradiation coupled with precession electron diffraction analysis, defect absorption was tracked for increasing dose. Denuded zones were found to collapse, but in the absence of any detected changed in macroscopic degrees of freedom of the grain boundaries. Each grain boundary denuded zone experienced changes different doses, indicating a direct observation of the difference in grain boundary “immunity” to irradiation depending on character. Since a change in sink efficiency is likely due to a change in point defect absorption at the boundary, this indicates that something about the structure of these boundaries has changed. The grain boundary macroscopic character remains the same, however, leading to the conclusion that a change in microscopic character has occurred, possibly due to the formation of extrinsic defects in the boundary. To analyze this further, simulations were used to explore grain boundary microstates loaded with defects (mimicking a collision cascade) and assessed for vacancy formation energies. Overall, the results present a foundation for improving sink efficiency descriptions under irradiation, and take a step forward in understanding complex interfacial dependencies. Additionally, results will be shown for studies of grain boundary structure and stability, including faceting, in the context of grain boundaries acting as “phases.” These advances present the possibility that the extent to which a grain boundary’s microscopic degrees of freedom change controls sink efficiency, and thus presents an opportunity to tune such efficiency as the defect scale.

Directed irradiation synthesis (DIS) and directed plasma nanosynthesis (DPNS) are robust, scalable approaches to design and fabricate surfaces and interfaces with unique, multi-functional properties at the nanoscale. In particular, DIS/DPNS can be used to rapidly fabricate hexagonal arrays of semiconductor quantum dots at III-V semiconductor surfaces in a single process step [1]. These are desirable for their excellent optoelectronic properties, including sharply-peaked density-of-states and sharp luminescence lines [2]. However, the ion irradiation process introduces significant damage and disorder to the surface, leading to the formation of a nanometers-thick, compositionally-complex amorphous layer at the interface [3], which significantly degrades the optoelectronic properties of the III-V quantum dots. Therefore, a fundamental atomistic understanding of the irradiation-driven defect dynamics and structural-compositional disorder is crucial for optimized fabrication of these nanostructures.

We have carried out large-scale molecular dynamics (MD) simulations of low-energy (500 eV) Kr+ ion irradiation of GaSb surfaces on the Blue Waters supercomputer at the University of Illinois [4]. These simulations have shown that the irradiation-induced altered compositional depth profile drives phase separation into clusters of the enriched component (Ga or Sb) at a given depth, with characteristic sizes of a few nm, surrounded by 50/50 amorphous GaSb. The nanoscale interface between these clusters and the “bulk-amorphous” 50/50 GaSb are a region of elevated potential energy, which can function as diffusion sinks or as pathways for accelerated transport at the atomic scale. From MD simulations of ion irradiation from the initial (pristine) surface to the structure formation threshold fluence, we observe the formation of small, unstable “protoclusters” of Sb throughout the surface from prompt ion effects. The Sb protoclusters are usually <1 nm in size and contain ~4% of all Sb atoms in the GaSb. However, no evolution of a global compositional gradient emerges from these simulations, indicating that long temporal scale effects such as defect-mediated diffusion are necessary to model the complex compositional evolution of the surface. Analysis of the atomic bonding in the irradiated surface indicates a strong preferential mobility of Sb atoms, suggesting accelerated Sb diffusion to the cluster/bulk interface as the protocluster growth mechanism.

To elucidate the defect dynamics and atomic diffusion mechanisms, we have developed a hybrid MD/KMC model to connect short and long temporal scales in a single simulation framework. The KMC component models atomic diffusion through the disordered layer via a lattice-free method which relies on “in-situ” characterization of the point defects generated by cumulative ion impacts into the GaSb surface. Using this hybrid model, we test the hypothesis that irradiation-enhanced diffusion of Sb atoms towards the Sb protoclusters interfaces provides the growth mechanism to stabilize Sb protoclusters at 2-3 nm diameters. This mechanism connects the prompt ion-induced structural changes and defect production to three-dimensional compositional evolution and surface morphology of III-V surfaces. We study the influence of the surface temperature on the defect dynamics and recovery rates, showing how manipulating the ratio between ion flux and surface temperature during DIS/DPNS processing offers a route towards minimizing damage and optimizing the properties of the resulting III-V quantum dots.

[1] S. Facsko, T. Dekorsy, C. Koerdt, C. Trappe, H. Kurz, A. Vogt, and H. L. Hartnagel, Science 285, 1551 (1999).
[2] P. M. Petroff and S. P. DenBaars, Superlattices Microstruct. 15, 15 (1994).
[3] O. El-Atwani, J. P. Allain, and S. Ortoleva, Nucl. Inst. Methods Phys. Res. B 272, 210 (2012).
[4] M. A. Lively, B. Holybee, M. Toriyama, and J. P. Allain, Nucl. Inst. Methods Phys. Res. B 409, 282 (2017).

Functional oxide interfaces exhibit useful emergent properties for electronics and sensors, but little work has been done to examine their degradation in high-radiation environments, such as in spacecraft and nuclear reactor cores. While grain boundaries in metals are known to respond much differently than the bulk, less attention has focused on the behavior of engineered oxide interfaces, which offer the potential to finely control structural, chemical, and electronic parameters. Here we examine the damage response of La₂Ti₂O₇ / SrTiO₃ and related heterojunctions using a combination of advanced scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS), and ab initio theory calculations. We observe that the interface can exhibit markedly improved damage resistance to 1 MeV Zr⁺ ion bombardment, in contrast to the bulk of the film or substrate. We discuss possible atomistic damage mechanisms and identify unique features of the interface that can influence the damage process. This work suggests that engineered oxide interfaces may offer a means to tune both functional properties and damage resistance, with potential implications for device design in extreme environments.

Laser-assisted probe tomography (LAPT) is a powerful tool for materials characterization due to its desirable combination of high spatial resolution and analytical sensitivity. In state of the art LAPT the thermal transient from a near-UV laser (E 3.5 eV to 3.6 eV; 343 nm to 355 nm) provides the energy to overcome the activation barrier for field ion evaporation. This technique is generally superior to voltage pulsing, which is limited strictly to conductors, and has allowed APT to expand its capability to effectively analyze a wide-range of materials including semiconductors and insulators. However, the thermal process is not without drawbacks. For example, LAPT data quality can be degraded due to thermal tails that limit sensitivity, formation of cluster ions that may have isobaric overlap with elemental species or dissociate into undetected neutral species which can adversely influence composition measurements. This is especially true for many ionic and covalent materials and can limit the recovery of bulk stoichiometry or composition to a narrow range of experimental conditions, if at all [1,2].

Ionizing radiation in the extreme ultraviolet (EUV) region of the electromagnetic spectrum
(10 eV to 100 eV; 124 nm to 12 nm) may offer a potential athermal field ionization pathway. Dependent on the particular photon energy used, EUV radiation is above the band gap, work function and ionization potential of any naturally occurring element. Photoionization cross-sections peak in the EUV band across the entire periodic table, and EUV is highly absorbed within the first few nm of the sample surface [3]. Coherent EUV radiation may also offer a potential in situ method for imaging the evolving specimen shape in real time through simultaneous coherent diffractive imaging or related methods [4].

I will present the first reported instrument design and results from a tunable EUV-APT that uses femtosecond-pulsed coherent radiation from phase-matched high harmonic generation in a noble gas cell. Initial experiments using Ar gas (EUV photon E = 41.85 eV, = 29.6 nm) demonstrate that pulsed EUV radiation triggers ion emission in e.g. fused quartz (amorphous SiO₂), BaTiO₃, and GaN. Thermal tails, peak widths, and the relative number of cluster ions are significantly decreased or eliminated entirely when compared with near-UV LAPT (E = 3.49 eV, = 355 nm) experiments on the same samples and specimens. Composition measured by EUV appears to be inherently stoichiometric; the bulk stoichiometry was recovered in the above three example materials under the same experimental conditions and without prior optimization of laser pulse frequency, laser pulse energy, or evaporation rate.

This finding has significant consequences. For example, EUV-APT may enable accurate nanoscale composition measurements in materials systems where the functionality depends critically on an unknown composition or stoichiometry at the local level, or where non-equilibrium phases play a role. This can be as simple as measuring and mapping the varying composition of an unknown reaction layer or compound formed during thin film deposition. Another example is in resistance switching oxides that have potential use in non-volatile memory and memristors in neuromorphic computing elements for beyond-CMOS quantum information systems. In these materials, oxygen deficiency (e.g. the oxygen vacancy concentration) plays a critical role in the nanoscale, conductive filaments responsible for the desired functionality. Other applications include hydrogen separation membranes, oxidation catalysts, ion conduction materials, and superconductors.

[1] Mancini, L. et al. Physical Chemistry C 118 (2014) 24136.
[2] Diercks, D.R. et al. J. Appl. Phys.114 (2013) 184903.
[3] Yeh, J.-J. and I. Landau. At. Data Nucl. Data Tables32 (1985) 1.
[4] Gardner, D.F. et al. Nature Photonics11 (2017) 259.

Interfaces are frequently the site of reactions and phenomena that dominate macroscopic properties, even though they are fractionally present relative to the bulk. Widely used vibrational spectroscopies like Raman and Fourier-transform infrared (FTIR) spectroscopy provide chemical fingerprints of bonding arrangements for materials characterization. However, only a few vibrational spectroscopies are effective to probe interfaces at the molecular level, viz. second-harmonic generation spectroscopy and sum-frequency generation spectroscopy. Advances in monochromation in the modern scanning transmission electron microscope (STEM) has made it possible to detect vibrational excitations using electron energy-loss spectroscopy (EELS) [1]. This new capability can be used to investigate the delocalized behavior of bulk as well as the localized behavior of surfaces and interfaces [2]. For a comprehensive understanding of the technique, experiments need to be performed on simple model systems. We explore the spatial variation in different vibrational modes when an electron beam is scanned across an abrupt SiO₂/Si interface. This investigation provides baseline data which can be used to further explore the influence of more complex interfaces on vibrational modes in materials.
A 3 μm layer of SiO₂ on a Si wafer was prepared for STEM EELS analysis by lifting out a focused ion beam (FIB) lamella using a Nova 200 NanoLab (FEI) FIB. A NION UltraSTEM 100 aberration-corrected electron microscope operated at 60 kV and equipped with a monochromator was used to perform EELS linescans across the SiO₂/Si interface. The electron interactions in vibrational EELS can be discussed in terms of dipole and impact scattering as done in surface vibrational spectroscopy using high-resolution EELS [3]. The energy-loss spectrum from amorphous SiO₂ shows the thin-film bond-stretch mode at 144 meV, a dipole scattering dominated signal, and the bond-stretch plus bend mode at 100 meV, an impact scattering dominated signal. We show that, in practice, the SiO₂ thin-film signal at 144 meV is delocalized to ~20 nm and that nanometer resolution is possible when selecting the SiO₂/Si interface signal at 136 meV. We also show that the 100 meV vibrational mode in SiO₂ was used to measure the interface abruptness to 0.5 nm. Finally, we demonstrate that with vibrational EELS in crystalline Si, which has only impact scattering dominant vibrational modes, a resolution of better than 0.2 nm was achieved with both optical and acoustic phonons [4]. Further experiments on the effect of interfaces such as grain boundaries and twin boundaries on the chemistry in Si will be presented.
References:
[1] O.L. Krivanek et al., Nature, 514 (2014), p. 209.
[2] K. Venkatraman et al., Microscopy 67 (suppl_1) (2018), p. i14.
[3] H. Ibach and D.L. Mills, Electron Energy Loss Spectroscopy and Surface Vibrations (Academic Press, 1982).
[4] K. Venkatraman et al., Science (under review).
[5] The support from National Science Foundation CHE-1508667 and the use of (S)TEM at John M. Cowley Center at Arizona State University is gratefully acknowledged.

Localization of strain into narrow bands of intense deformation plays a critical role in the mechanical response and performance of many structural materials. Such shear-bands can exhibit extremely complex and dense arrangements of nanoscale interfaces, especially in materials with low stacking fault energies and the possibility for shear-induced phase-transformations. In austenitic stainless steels, shear-bands present a range of phenomena including dislocation slip, mechanical twinning, and solid-state shear-transformations (e.g., forming hcp ε- and bct α’-martensite). Furthermore, the development of these features can be strongly affected by environmental effects, such as the presence of hydrogen.
In this presentation, we discuss scanning transmission electron microscopy (STEM) observations of the atomic- and nanoscale structure of shear-bands in a prototypical, low-stacking-fault-energy austenitic stainless steel (304L). We focus specifically on the effects of hydrogen on the shear-band development and the mechanisms of martensite nucleation and growth. We investigated both hydrogen-charged and non-charged specimens prepared from forged 304L material and subsequently deformed under tensile conditions to 5% and 20% strain. Using nanobeam scanning diffraction, with a 3nm near parallel-beam probe, we acquired crystallographic information and measured volume fractions of deformation-induced twins and phases within the observed shear bands. We complemented the nanobeam diffraction measurements with atomic-resolution imaging to determine the structure of the interfaces between the martensite phases and the FCC-austenite matrix and to establish the configuration of dislocations at these interfaces. At 5% strain, the non-charged specimens show ample dislocation content from the initial as-forged microstructure with twin-laden shear bands apparent throughout the specimen. In the hydrogen-charged material, dislocation activity is planar at 5% strain, with shear bands containing far less twinning and instead forming ε-martensite. This trend continues up to 20% strain, with non-charged samples only showing general dislocation plasticity and twinning. At 20% strain, hydrogen-charged samples showed large amount of ε-martensite shear bands with α’-martensite within the intersections of the shear bands.
Our observations support the long-standing notion that ε-martensite is a necessary intermediary in the formation of α’-martensite platelets, as has been described within the classical Olson-Cohen mechanism. However, detailed analysis of the observed structure of the shear-bands raises questions regarding the actual dislocation mechanisms underpinning this transformation. Atomic-resolution observations of an α’-martensite nucleus within an ε-martensite laden shear band shows low dislocation content surrounding the nucleus, indicating that the geometrically contrived distribution of shears of the conventionally invoked Olson-Cohen mechanism for α’-martensite nucleation are inconsistent with the experimentally observed shears. We will discuss alternative interfacial dislocation-based descriptions for the nucleation process and outline potential opportunities and needs for modeling to further clarify the transformation mechanism and its dependence on hydrogen.
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

In recent years, atom-resolved chemical mapping becomes possible by using aberration-corrected scanning transmission electron microscopy (STEM) -energy dispersive X-ray spectroscopy (EDS). STEM-EDS makes it possible to determine almost all the element atoms in periodic table at the same time, so it is advantageous to characterize material interfaces with local chemical inhomogenities. In this study, we applied this technique to understand complex local chemical structures at ceramic interfaces such as yttria stabilized zirconia grain boundaries. By using aberration-corrected STEM combined with high-sensitive silicon drift detectors (SDD), atomic-scale segregation profiles of solute and impurity atoms across ceramic interfaces can be clearly mapped. We demonstrate such chemical analysis elucidates very interesting atomic-scale segregation structures at ceramic interfaces. In this talk, the details of such ultrahigh-resolution chemical mapping and new knowledge gained by such observations on ceramic interface segregation will be reported.

In elemental metals and most binary alloys it is well known that the solid-liquid interfacial free energy decreases with decreasing temperature. In this work we use molecular dynamics simulations, the capillary fluctuation method and thermodynamic integration to show that the opposite trend is observed for Cu-Zr and Al-Sm alloys. This unusual temperature dependence is discussed in terms of the glass forming ability of these systems. In addition, we show that, in the case of Al-Sm, the anisotropy of the interfacial energy depends only on the structure of the liquid. Also, the anisotropy variation with composition is consistent with experiments and theoretical predictions of a dendrite orientation transition.

Gallium-based liquid metals are stable liquids at room temperature that exhibit many excellent thermo-physical properties. Liquid metals are useful in applications that require high electrical and thermal conductivity in addition to mechanical deformation. Combining liquid metals with other thermally conductive metallic particles could potentially result in enhancements of these transport properties [1]. However, liquid metals are notorious for alloying with and/or embrittling metals under prolonged exposure due strong intermetallic chemistries. Refractory metals such as tantalum and tungsten are generally resistant to these detrimental processes. However, these metals are unable to directly wet with liquid metals, which prevents them from forming composite fluids. Depositing a thin metallic layer onto these metals can potentially enable refractory metals to be mixed in liquid metal via reactive wetting processes [2]. Electroplating and electroless deposition techniques have long been used to fabricate bi-metallic powders in literature [4].
We show that electroless deposition of silver coatings on tungsten powder can facilitate reactive wetting of these powders in liquid metals. This work demonstrates a relatively simple procedure of forming tungsten-silver core-shell powders and their reactive wetting behavior in liquid GaInSn. In-situ focused ion beam (FIB) characterization is utilized for direct observation of this nanoscale wetting process. This work also investigates a time study of potential room temperature alloying processes of the silver layer in a gallium-rich environment under different volumetric fractions of liquid metal/powder. These studies can provide fundamental insights into liquid metal wetting phenomena and how to engineer material surfaces for thermal management, flexible electronics, and plasmonic applications.
References:
[1] Tang et al. ACS App. Mater Inter. 2017. 9(41), 35977-35987.
[2] Doudrick et al. Langmuir. 2014. 30(23), 6867-6877.
[3] Dezellis et al. J. Mat. Sci. 2010. 45(16), 4256-4264.
[4] Djokic et al. Journal of The Electrochemical Society. 2011. 158(4), D204-D209.

The properties of solid-liquid interfaces (SLIs) govern a wide variety of processes of technological import, e.g., wetting, heterogeneous nucleation, casting, and crystal morphology and growth. Here, we examine three chemically heterogeneous metal-metal SLIs, specifically Al(s)/Pb(l), Cu(s)/Pb(l) and Al(s)/Ga(l) to determine how interfacial structure and dynamics affect phenomena of experimental interest. For Al/Pb, transmission electron microscopy (TEM) experiments show that liquid Pb inclusions undergo Brownian motion within a solid Al matrix. Using molecular-dynamics (MD) simulations, we examine the role played by Al diffusion at the SLI interface in this phenomena and in the spreading of droplets. For Cu/Pb interfaces, earlier simulations predict a strong anisotropy in interfacial structure, with the (100) interface exhibiting surface alloying and the (111) showing a “pre-freezing” region, 2-3 lattice spacings thick, of crystalline Pb at the interface. Using an extensive series of MD simulations we show how this structural anisotropy contributes significantly to differences in the heterogeneous nucleation rates between these two interfaces. Finally, we examine the Al(s)/Ga(l) solid-liquid interface, an important material for the understanding of liquid-metal embrittlement, using both classical and ab-initio molecular-dynamics simulation.

Control of heterogeneous ice nucleation (HIN) is critical for applications that range from iceophobic surfaces to ice-templated materials. Multiple computational studies have been performed to study the effects of the substrate’s surface energy and topography and how their interaction determines phase transition temperature (PTT) and nucleation rates. Two particular well-studied substrates are graphene and silicon. In contrast to a considerable body of literature on computational approaches, few experimental studies have been reported concerning HIN on graphene and the extent of the effect of surface energy on the HIN on plasma-oxidized silicon. The study reported here focuses on the nucleation efficiency of ice on single layer graphene (SLG) deposited on three different substrates: single crystal silicon, fused quartz, and 300 nm SiO₂/Si, as well as plasma-treated silicon with multiple different surface energies. HIN was studied by placing a 10µL water drop on the substrate and observing the PTT using a freezing stage in a Raman spectroscopy system. After each freezing run, the WCA is measured, again. Findings are that the SLG displays surface transparency after only one experiment as indicated by the WCA changing from 95.4^o +/- 0.4 ^o (a value typical for SLG), to values ranging from 60.9+/- 0.3^o to 68.4^o +/- 0.3^o. Such a significant decrease in surface energy was not predicted by computational models suggesting surface topography effects. To investigate the latter, the substrate surfaces were characterize by scanning electron microscopy (SEM) and atomic force microscopy (AFM). The surprising findings were prominent wrinkles, where the SLG would peel away during the sequential HIN experiments, explaining the change in WCA which occurs due to the lower surface energy substrate being exposed. As a second material system, plasma-oxidized silicon was tested in the same way: as-received, and after 30 s, 120 s, and 300 s plasma exposure with resulting WCA ranging from 32.2^o +/- 0.3^o, to <2^o . In both systems, the range of observed PTTs was surprisingly small falling within a range from -18.6 +/- 0.6°C for the SiO2/Si system (contact angle from 2 to 33 degrees) and -22.5 +/-0.6°C for the graphene system (contact angle from 60 to 90 degrees). While the fact that the SLGs were compromised will have contributed to the results obtained for the SLG material systems, the independence from surface energy of the PTT in the case of the plasma-oxidized silicon is an important finding. Both outcomes are noteworthy: 1) The results from the SLG system illustrate the importance of a careful characterization of the substrate before and after nucleation experiments, as it may reveal, as in our case, the suboptimal quality of commercially available SLGs, which poses considerable experimental challenges for the experimental verification of computational results and puts into questions results obtained without such an analysis. 2) The results from the study of plasma-oxidized silicon, which suggest that surface energy or WCA alone are not good indicators of PTT, invite further exploration both experimental and computational.

The effect of graphene conformal mapping on the wetting of a surface remains a controversial topic. Many groups have studied the wetting transparency of graphene coated surfaces by measuring the water contact angle on the surface, however, airborne contamination due to the exposure of the surface during contact angle experiment can alter the hydrophilicity/hydrophobicity of the surface. Here, we explore this controversy by comparing the interaction of water molecules with a surface through measurement of the adsorption energy of the water molecules on the surface with and without graphene coatings. Ambient-pressure X-Ray Photoelectron Spectroscopy and quartz crystal microbalance experiments will be reported.

Concentration-driven phase change, i.e., precipitation-dissolution equilibrium, is quite prevalent in many natural and artificial systems. With miniaturization, new technologies operate at small enough length scales (e.g., porous battery electrodes) where surface energies are expected to be a dominant factor. Even though the precipitation reaction is integral to functionality, its connection to interfacial effects has not been analyzed in the past. Here in we study the effect of surface energies on the nucleation dynamics of the precipitate phase. We also investigate the role of curvature on precipitate growth. This study proffers avenues for altering precipitation dynamics using interfacial characteristics.

A new density functional approach that employs both short and long range, rotationally invariant, multi-point particle interactions will be presented. This formalism will be shown to unify several closely connected phase field crystal (PFC) theories that have appeared in recent years, which are known for coupling important physics emergent at the atomic scale with diffusive kinetics governing microstructure evolution in most materials. Simulation results will be presented that demonstrate the interaction between nano--meso scales in PFC modelling of microstructure evolution. Specifically, recent studies of two-step nucleation in solidification, and cavitation in liquid pools during rapid cooling will be discussed. A formalism for coupling the PFC density order parameter to heat transfer will also be presented and shown to capture density rearrangement effects in latent heat release during solidification. We end by presenting a coupling of the PFC order parameter to electrical potential, and show results predicting void formation and time to failure in electromigration in polycrystalline materials.

Diffusion-induced grain boundary motion (DIGM) can be described as normal grain boundary migration phenomenon caused by lateral grain boundary diffusion. In this work, molecular dynamic simulations using Embedded Atom Method potentials and LAMMPS code are performed to investigate the physical origin of DIGM. A comprehensive approach constructed by combining various atomistic simulation techniques, i.e., the synthetic driving force method, the interface random walk method, and shear coupling, has been used to quantify the driving force that operates during DIGM. The driving force obtained from this method are then compared to those arising from coherency strain energy. The two values are consistent with each other, which shows that a key driving force for the phenomenon is that due to unbalanced coherency strain energy across the boundary. The simulation results also show that segregation plays an important role in promoting DIGM once GB is in close contact with the solute atoms. All observations made during the simulations are supported by the atomic configurations and graphical analysis at different stages of the process. This phenomenon is of immense technological importance because of its scientific and practical significance in materials.

Austenite (γ-Fe, face-centered cubic (FCC)) to ferrite (α-Fe, body-centered cubic (BCC)) phase transformation in steel is of great importance from the point of view of industrial applications. In this work, using classical molecular dynamics (MD) simulations, we study the atomistic mechanisms involved during the transformation of the ferrite phase from an austenite phase. The kinetics of the transformations, classified as martensitic and massive, is defined in terms of the interface mobility, which depends on the interface migration velocity and Gibbs free energy change. The simulations are performed by creating an FCC region, sandwiched between two BCC regions (interface formed
according to Nishiyama-Wasserman (NW) orientation relationship, with [110] of BCC oriented parallel to [111] of FCC) and studying how the former transforms into the later phase at elevated temperature, ranging from 1000 to 1400 K. Three configurations of BCC-FCC phase were created by tilting FCC phase with respect to z-axis with an angle of 3.11^o , 4.04^o and 5.77^o from the ideal NW, resulting into formation of steps or disconnections at the interface. Using MD simulations, the effect of the tilt on interface velocities, mobilities and activation energy followed by the interface migration mechanism in each of the orientation will be discussed.

In this talk, we will explore the nucleation and distribution of hydrides generated during electrochemical charging and solid phase redistribution in zircaloy-4. We will focus on the understanding of the distribution of hydrides and deuterides as a function of cooling rate in large grain “blocky alpha” and as received recrystallised Zircaloy-4 plate. At the macroscale, we use EBSD, optical microscopy, XRD and SEM and we observe that the location, phase, and clustering of hydrides is systematically dependent on cooling rate (through the precipitation transus temperature) and grain size; as well as how texture impacts hydride morphology. We follow these macroscale observations with nanoscale characterisation using atom probe tomography and high resolution TEM of deuterides, where we characterise the hexagonal close packed (HCP) α-Zr matrix and a face centred cubic (FCC) δ deuteride (ZrD1.5–1.65) and observe the presence of a nanoscale shell of a HCP ζ phase deuteride (ZrD0.25–0.5). In light of these observations, we have started to perform micromechanical testing using in-situ and elevated microcompression testing to understand how these hydrides impact mechanical performance.

The grand challenges in the hydrogen storage research are associated with the fact that the operating mechanisms often involve the complicated coupling of chemical/physical processes over vast ranges of multiple length and time scales. In particular, metal hydrides involve phase transformations coupled with surface and/or interfacial reactions during (de/re)hydrogenation. Essentially, these are collective dynamic processes of atomic/molecular species that determine the thermodynamics and kinetics of nano- or micro-level characteristics. However, the lack of clear understanding of associated mechanisms and their impacts on solid-state storage reactions has hampered the acceleration of materials discovery and development. Therefore, it is obvious that several modeling approaches that cover different length and time scales ought to be properly integrated in order to comprehensively explore the underlying mesoscale science during the co-evolution of relevant chemical and materials processes. The mesoscale continuum modeling framework can provide a unified platform for integrating the necessary atomistic approaches, which typically provide higher accuracy for nanoscale chemical processes, and continuum approaches, which offer far greater flexibility and scalability and can describe materials at the microstructural level. In this presentation, we demonstrate our integrated effort as part of the DOE Hydrogen Storage Materials—Advanced Research Consortium (HyMARC) under the Energy Materials Network (EMN) towards the multiscale modeling of phase transformations in metal hydrides for solid-state hydrogen storage. Specifically, we will discuss our recent efforts on the development of an integrated mesoscale modeling framework that can directly incorporate parameters derived from predictive atomistic modeling approaches such as ab initio calculations and molecular dynamics simulations. Representative case studies will be presented, including diffusional and/or structural phase transformations of simple or complex metal hydrides during the (re/de)hydrogenation.


This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

The graphitization of amorphous carbon (a-C) in thin film stacks with Ni was investigated as a function of initial stacking order, temperature, and time. Three different layer stacks, notably a-C/Ni, Ni/a-C, and Ni/a-C/Ni on fused silica, were exposed to heating ramps up to 700 °C in high vacuum. Thereafter, the bilayers had changed their stacking order, and in the triple layer the major fraction of carbon was diffused to the surface, and a minor part to the interface with SiO₂. Raman spectroscopy showed the formation of a two-dimensional layered graphitic structure for both layer exchange (LE) directions. While during outwards LE, where C is transported towards sample surface, a smooth layer with graphitic planes parallel to the interface with the substrate has formed through two-dimensional growth, a significant restructuring of the Ni layer appeared in the inwards direction of the LE, where C is transported towards the substrate. The fragmentation of the Ni layer at the interface with the crystallized carbon, as well as the regions with turbulence-like and folding defects indicated a three-dimensional growth in the inwards direction. The degree of LE (α_LE), quantified by Rutherford backscattering spectrometry and elastic recoil detection, is ≈ 95 % for the outwards direction, and of the order of 80 % in the inwards direction. Based on the calculation of surface and interface energies for the initial and final state, thermodynamic estimations were carried out. They point to the wetting of Ni grain boundaries by C atoms as the initial driving force for the LE and allow a consistent understanding of the α_LE directionality and of the final thin film microstructure characterized by a comprehensive structure analysis.
The mechanism of LE was studied in situ as a function of time and temperature by Rutherford backscattering spectrometry and Raman spectroscopy. The graphitization occurred simultaneously with the LE and is completed during the applied heating ramp to 700 °C. The temperature resolved measurements allowed the determination of the onset temperatures and transition rates as a function of the annealing temperature and the stacking order. Finally, the activation energy for both LE directions was estimated. In combination with the thermodynamic calculations, this in situ study allows to identify the metal-induced crystallization via wetting and diffusion along grain boundaries, instead of dissolution/precipitation as the responsible mechanism for graphitization of amorphous carbon with layer exchange in contact with Ni. The proposed model can potentially be used to estimate the catalytic transformation of group 14 elements in contact with transition metals.
Financial support by the Helmholtz Excellence Network DCM-MatDNA (ExNet 0028) is gratefully acknowledged.

Lifetime of structural alloys used in high temperature applications is often limited by their vulnerability to oxidation. To address this issue, different alloying and/or coating strategies have been developed depending on alloy system and application. Alloy chemistries may be tailored so that a slow-growing, passive oxide scale forms on the alloy surface and protects it from further attack by oxidation. The addition of small amounts of dopants can further reduce oxide growth kinetics and enhance scale adherence. While the principles leading to scale growth are generally understood, the mechanistic roles of alloying elements and synergy between alloy and oxide scale are often less clear. High resolution characterization techniques provide a unique opportunity to uncover new insights into the structure and evolution of oxide scales, and ultimately quantify mechanisms of alloy oxidation. Using our experimental observations, we will discuss the mechanisms of scale growth and the contributions of alloying elements during controlled oxidation exposure of model Ni-Cr-Al and Ti alloys.

Most models describing the formation of a passive oxide film on a metal are formulated in one-dimension where the principle assumption is that any interfaces in the system are planar. These models fail to describe the interfacial evolution of morphologically complex oxide films because they neglect interfacial energy and curvature, which are important for self-consistently describing morphological evolution for non-planar interfaces. We extend the Point Defect Model for growing oxide films to multiple dimensions and have modified the boundary conditions to include capillary contributions to the defect chemical potentials at the film interfaces. We perform a morphological stability analysis of the 1D steady-state as a first step toward describing the full morphology evolution of the oxide. Our analysis predicts that a localized thinning instability is present at concave regions oxide-solution interface due to anodic dissolution of the oxide. This instability could be the first step of the formation of pitting corrosion. Our results provide insights to the important role that interfacial energy plays in the evolution of the oxide film and suggests that many 1D models may be inapplicable in a broad class of corrosion processes.

Heterophase interfaces such as metal/ceramic and ceramic/ceramic interfaces are ubiquitous to a wide range of technological applications, such as the all-solid-state-batteries (ASSB), Cu(In,Ga)Se₂ (CIGS) based photovoltaic cells and numerous other solid-state devices. Transport of atoms along these heterophase interfaces may lead to metastable phase formation at the interface. In this work, we report the experimental observation of two such cases a) HCP type Ni(Fe) phase (ground state is fcc structure) and b) FCC Fe (ground state is bcc structure), after the heat treatment of (solid solution) Ni₈₀Fe₂₀/α-Al₂O₃ and (solid solution) Ni₇₃Fe₂₇/α-Al₂O₃ respectively. Using ab initio calculations in conjunction with cluster expansion techniques, within the recently formulated meta Generalized Gradient Approximation - SCAN functional, we develop a general model (based upon interface energies) that can be readily applied to establish the conditions under which metastable phases will exist at/near heterophase interfaces, and then predict their minimum and maximum equilibrium thickness under specified conditions. Our experimentally observed thickness of the hcp-Ni(Fe) metastable phase is within the bounds predicted by the model. This implies that the occurrence of stable phases at these heterophase interfaces may not always be as predicted by the equilibrium thermodynamics but in fact is dependent on the interface energies of these heterophase interfaces.

The crystallization of sputter-deposited amorphous strontium titanate (SrTiO₃, STO) films is affected by the proximity of free-surfaces and interfaces with other materials. We have crystallized STO films with a thickness of ~12 nm sandwiched between two Si and SiGe-alloy semiconductors in non-planar geometries. The amorphous STO films were obtained by sputtering onto a room-temperature Si-SiGe heterostructure with a built-in stress imbalance. The released STO/Si/SiGe heterostructure rolls into tubes with diameters of a few microns, placing the STO layers in the proximity of two non-planar (or curved) semiconductor surfaces. The effect of the oxide/semiconductor interfaces on the crystallization process is investigated by heating the rolled-up structures to induce crystallization. X-ray diffraction and Raman spectroscopy provide insight on how the proximity of interfaces may determine the balance between nucleation probability and growth of the crystalline phase. Additional effects associated with heating include a restructuring of the amorphous layer and a change in the tube diameter due to changes in the stress distribution within the rolls.

ACKNOWLEDGEMENT: This work was supported by NSF through the University of Wisconsin Materials Research Science and Engineering Center (award No. 1720415).

The unique properties displayed by canonical relaxor systems, such as the electric field induced transformation from a pseudo-cubic relaxor state to a polar ferroelectric state[1], the associated huge strain [2] and the phase transition mediated polarization reversal mechanism [3], make these materials highly interesting objects for fundamental studies as well as versatile components for industrial applications.
Especially, the achievement of high strains has drawn significant attention, the origin of which is based on the reversibility of the phase transformation between the ferroelectric and the relaxor state [4]. Many relaxor-ferroelectric systems exhibiting high actuating performance are located in the morphotropic phase region leading to the appearance of multiple crystallographic phases even within single grains. Their ability to take on different phases simultaneously and to change the phase ratio to adapt to external stimuli [5], is another important factor contributing to their exceptional electromechanical performance.

Here, we report on the thermal evolution of the different crystallographic phases in a multiphase (Bi_1/2Na_1/2)TiO₃-BaTiO₃ relaxor ceramic and the associated development of interfacial stresses. Upon electric field application, this system exhibits an “intrinsic core-shell structure” of a polar minority phase embedded into a polar majority phase. While the majority phase stays stable with increasing temperature up to the transition into the relaxor state, a gradual de-texturization of the minority phase is observed over the whole temperature range. The surface domain structure was found to decay already at significantly lower temperatures than expected from bulk observations. Development of interfacial stresses due to thermal expansion mismatch between majority and minority phases as well as differences in local stress state between surface and bulk are discussed as driving factors of the phase transition dynamics.
Tailoring of interfacial stresses through adaption of phase fractions opens up a pathway to optimize the strain performance of actuator materials and can become a useful tool for the stabilization of usually metastable crystallographic phases as well as for property tuning in piezotronics and multiferroics.

[1] Z.-G. Ye, Key Eng. Mater. 1998, 155–156, 81.
[2] S.-T. Zhang, A. B. Kounga, E. Aulbach, H. Ehrenberg, J. Rödel, Appl. Phys. Lett. 2007, 91, 112906.
[3] J. Glaum, H. Simons, J. Hudspeth, M. Acosta, J. E. Daniels, Appl. Phys. Lett. 2015, 107, 232906.
[4] M. Hinterstein, M. Knapp, M. Hölzel, W. Jo, A. Cervellino, H. Ehrenberg, H. Fuess, J. Appl. Crystallogr. 2010, 43, 1314.
[5] M. Hinterstein, L. A. Schmitt, M. Hölzel, W. Jo, J. Rödel, H.-J. Kleebe, M. Hoffman, Appl. Phys. Lett. 2015, 106, 222904.

Controlling the micro/nanostructure of thin films would enable us to explicitly tailor their mechanical behavior. Here, we describe a new process to synthesize thin films with precise microstructural control via systematic, in-situ seeding of nanocrystals, and subsequent crystallization of amorphous precursor films. It was found that varying the seed element affects the final phase of the films once annealed, such as NiTi forming an austenite or martensite phase when seeded with Cr or Ti, respectively. This gives the ability to control grain size, dispersion, and phase composition of thin films. Additionally, crystallization temperatures potentially decrease with the presence of these nanocrystals, reducing precipitate formation and enhancing mechanical properties.

Austenitic 316L stainless steel is presently used for orthopaedic implants due to its biocompatibility and high corrosion resistance. One issue, however, with the use of this material in such applications concerns its low wear resistance. Surface properties can be modified through case hardening treatments in order to improve wear resistance and fatigue life as well. In this study two such treatments were applied on stainless steel 316L. The first treatment was Surface Mechanical Attrition Treatment (SMAT) which is a severe plastic deformation (SPD) process that enhances the surface properties and the performances of a material without changing its chemical composition by creating a nanostructured layer on the top surface of the material. The hardness evolutions below the treated surfaces were measured using micro- and nano-indentation and it was shown that at the top surface an increase of 75% in the hardness can be obtained after SMAT. The effect of SMAT can extent deep beneath the treated surface with the creation of a high deformation zone following the nano-grained layer. This high deformation zone contains larger grains where local misorientations are present, and the material in that region is seeing strain hardening as well. Further to SMAT, some specimens underwent a second case hardening treatment, plasma nitriding, to further enhance their surface properties by the diffusion of nitrogen atoms in interstitial positions in the iron lattice creating additional residual strains at the top surface of the material. Using different microstructure characterization techniques, we were able to determine the microstructural evolution of the stainless steel after these different surface treatments and to determine the effects of creating new grain boundaries by SMAT on the nitriding process. The microstructure features investigated here include thickness of the nanocrystalline layer, size of the grains within the nanocrystalline layer and depth of diffusion of the nitrogen atoms within the material. The microstructure changes are here correlated to the improvement in hardness and wear resistance.

Significant interest has arisen in the use of solutes to provide thermodynamic and/or kinetic stabilization of nanocrystalline grains. Much of this work has primarily focused on their ability to mitigate grain growth under thermal annealing. In this work, we explore both the grain boundary specificity of such segregation under annealing as well as the mechanical responses of these stabilized nanocrystalline grains under mechanical loads. Using Ni(P) as a case study, we demonstrate a cross-correlative precession electron diffraction – atom probe tomography technique where the P solute segregation to specific grain boundaries was quantified. It was found to be relatively homogeneous on high angle grain boundaries, particularly with an increase in annealing temperature. Where variation was noted, the interfacial excess concentrations varied from zero to upwards of tens of atoms/nm^2. Subsequently, the alloy film was in situ mechanical loaded in a TEM where the granular evolution was characterized. The experimental results are coupled to atomistic simulations to predict the specific boundary segregation behavior as well as simulating how such partitioning is influencing the granular stability under loads in a thermally stabilized nanocrystalline alloy.

Nanometallic multilayers (NMMs) are of considerable interest due to their high density of interfaces which lead to unique mechanical, chemical, and thermal properties. Additionally, NMM systems have been shown to undergo several microstructural transitions when annealed at elevated temperatures while still retaining nanoscale features. The microstructural evolution of these materials and the nanostructures that develop after annealing are not well understood but can provide insight of active thermal processes and stabilization mechanisms. To further understand the thermal evolution of these materials, several sputtered NMMs systems including Mo-Au, Hf-Ti, W-C and Ta-Hf were annealed to critical temperatures determined by DSC. A wide range of characterization techniques including TEM and APT were utilized to examine compositional changes and gain insight into thermally activated events such as layer roughening, solute grain boundary segregation and phase separation.

Al-Al₂Cu eutectic alloy with nano-laminated microstructure exhibit high strength and good plasticity. Due to the significant difference in elastic properties and plastic deformation capability between Al and Al₂Cu phases, structural instability occurs in Al₂Cu lamellae associated with elastic buckling when a compression stress is applied parallel to the lamellae. Al lamellae plastically deform via dislocations motion. Al-Al₂Cu interfaces play strong barriers for blocking dislocation motions in Al lamellae. Thus, elastic buckling in Al₂Cu lamellae is constrained by the elastic-plastic deformation of Al lamellae. To understand buckling behavior of nanoscale Al-Al₂Cu eutectic alloy, we developed a mesoscale crystal plasticity model and implemented it into finite element analysis. Plastic deformation in Al lamellae is described by the confined layer slip (CLS) mechanism. The critical resolved shear stress for dislocation slip varies with layer thickness, estimated by CLS mechanism. The strain hardening is described by a phenomenological power law. To capture the essential features of confined layer slip in Al lamellae (dislocation loops glide through the layer and deposite at interfaces), plastic deformation in Al lamellae should be the same through the thickness. However, interface constraints will cause different elastic deformation through the thickness of Al lamellae. We thus divided each Al lamella into three voxels that have the same plastic deformation but may develop different elastic deformation. In comparison, we also conducted a traditional CP analysis in which plastic deformation in three voxels through the thickness of the lamella is calculated individually according to phenomenological power law. In these simulations, plasticity in Al₂Cu lamellae is modeled with unusual slips on {011} planes which have been observed in our experiments. The first model is demonstrated to be effective in predicting mechanical properties of nano-scale lamellar materials. In contrast, the second model is more suitable to study deformation behavior of sub-micro scale lamellar materials. Simulation results reveal that critical compression strain corresponding to buckling increases as layer thickness decreases. Our predictions agree with micro-pillar compression tests. (This research is sponsored by DOE, Office of Science, Office of Basic Energy Sciences under Award No. DE-SC0016808. )

Grain boundaries tend to dominate mechanical response in oxides due to the oxides exhibiting limited dislocation and/or twinning plasticity at room temperature. Grain boundaries are often weaker than the bulk and may preferentially fracture or deform locally. The effects of interfaces and the bulk often convolute in bulk polycrystalline measurements making it difficult to isolate contributions from each. This talk emphasizes performing small scale and single interface measurements using in situ TEM in order to obtain fracture energies from real interfaces in polycrystals and describes efforts to characterize interface mechanics in nanocrystalline materials. Specifically, we will report on the roll of grain boundary sliding and grain rotation in inducing inverse Hall-Petch response in oxides.

Material interfaces such as grain boundaries, twin boundary, or phase boundaries are often introduced to resist dislocation motions in metallic materials. These interfaces are used to block dislocations and in turn significantly increase material strength. However, they may lead to a reduction in material ductility. By contrast, instead of blocking dislocations, the amorphous-crystalline interface (ACI) is recently introduced in amorphous/crystalline metallic composites (A/C-MCs) to absorb dislocations. This strategy has a potential to solve the “strength-ductility dilemma” from the bottom up. However, due to the lack of a commonly agreed mechanism for the co-deformation of crystalline and amorphous phases, the development of A/C-MCs is still at a “trial and error” stage update. In order to enable a rational design of high-performance A/C-MCs, this work aims to (i) predict the mechanical behavior in A/C-MCs under a plastic shear strain; (ii) determine the dislocation-ACI reactions; and (iii) quantify the coupled dynamics between dislocations and the activities of plasticity carriers, the shear transformation zones (STZs) in amorphous materials, through molecular dynamics (MD) simulations. The Cu/CuZr nanolaminate is chosen as the model material. The amorphous CuZr is formed under a wide range of cooling rates, ranging from as low as ~10⁴ K/s up to ~10¹³ K/s, and then integrated with Cu. Stress-controlled shear loading is then applied. When the Cu/CuZr is under shear, dislocations are found to nucleate in Cu first, migrate towards the ACI, and then get absorbed by the amorphous CuZr. The results show that: (a) the instability of amorphous CuZr can be induced by a dislocation absorption and follows a Mohr-Coulombic yielding criterion; (b) different icosahedral structures (ICOs, a dominant motif to resist the deformation in amorphous materials) from different cooling rates greatly affect the mechanisms of STZ nucleation and percolation in amorphous CuZr; and (c) the number of the absorbed dislocations can be directly correlated with the volume fraction of the STZs activated in amorphous CuZr. This work has a potential in providing researchers with opportunities to finely tune the multi-level microstructure of A/C-MCs for achieving a combined strength and ductility.

Nanoindentation is a promising method for deforming individual grain boundaries and determining their local yield behavior. However, it is difficult to de-couple the dislocation-grain boundary interactions, and understand the mechanisms that give rise to grain boundary deformation. In the present study molecular dynamics indentation simulations near grain boundaries of Fe-Si bicrystals provide information on dislocation absorption and transmission. It is shown that the occurrence of such events corresponds to ‘pop-ins’ in the load displacement plots. Analysis of the strain maps illustrates that grain boundaries begin deforming plastically at a critical applied stress, which can be termed grain boundary yield stress. Analytical expression for such a grain boundary yield stress can be obtained by gradient plasticity by introduction new types of interface energies that depend on the plastic strain at the interface, as well as, the specific grain boundary characteristics. The physical meaning of such a mechanically induced interface energy is explored in the context of multiple GB characteristics, including the slip compatibility of the two grain orientations and the level of solute material segregated at the boundary. Further, the inverse Hall-Petch effect is interpreted through indentation directly on grain boundaries, where softening can be directly tied to preferential nucleation of plastic deformation.

High-entropy alloys (HEAs) are an intriguing class of metallic materials due to their unique mechanical behavior. Achieving a detailed understanding of structure-property relationships in these materials has been challenged by the compositional disorder that underlies their unique mechanical behavior. For instance, extensive nanotwinning occurs at room temperature in the medium-entropy CrCoNi alloy, whose combination of strength, ductility and toughness properties approach the best on record. We present results of employ atomistic computer simulations to investigate twinning behavior in the medium-entropy CrCoNi alloys, considering model systems with random substitutional disorder as well as systems with a high degree of local chemical short-range order (SRO). The planar energetics associated with the twinning as well as parameter related to twinability are found to be strongly influenced local chemical SRO. Furthermore, the migration mechanism of twin boundary under deformation have been studied in the CrCoNi alloys with varying SRO.

Generalized stacking fault energy (GSFE) analysis is a valuable technique to provide some insights and information on dislocation core structures and mobilities which are controlled by crystal resistance to shearing along a specific crystallographic plane. This technique is composed of the following steps: 1) cutting a single crystal along a fault plane, 2) shifting rigidly the upper half of the newly created bi-crystal along an arbitrary fault vector which belongs to the fault plane, 3) releasing the shifted crystal on the top of the lower half crystal, 4) relaxing the bi-crystal atoms in the direction perpendicular to the newly generated fault plane. By generating GSFE profile, one should be able to recognize the local minima on the profile correspond to stable stacking faults (SFs).

However, the stability of the SFs should be checked once the atoms are allowed to relax in all directions. This issue could be heightened in metals with hexagonal close-packed (HCP), like magnesium or Mg, due to the unsymmetrical feature of atomic configuration on the two sides of the slip direction. In this work, two relaxing directions are considered, out-of-plane (N-direction) and in-plane (P-direction). N-direction is normal to the slip plane which is {10-11}, the first pyramidal or compression twinning plane in HCP metals. P-direction is parallel to the slip plane and perpendicular to the slip direction which is <-1012>. Accordingly, N-direction relaxation (one direction relaxation) and N-P-directions relaxation (two directions relaxation) schemes are used here.

To study the effect of different relaxation schemes on GSFE profile for {10-11}<-1012> slip system, a sample with 80,000 atoms is constructed in which the height of the sample is large enough to minimize the effect of free surface on the fault plane. Four different empirical potentials, three of which designed by embedded-atom method (EAM) and one with modified EAM (MEAM), available in the literature for Mg are employed in Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) to generate GSFE profile as described above with N-direction and N-P-directions relaxations.

We report here that results of N-direction and N-P-directions relaxations both coincide partly along direction, however, N-P-directions relaxation shows significant oscillations in the GSFE profile for all four potentials. This deficiency could be attributed to the following reasons. First, GSFE is a nonphysical projection of crystal slip. Dislocation emission and motion incorporate two energy components: (1) initial elastic deformation and (2) breaking of atomic bonds. While the second term can be assessed by the GSFE, the contribution of elastic deformation is absent from the rigid GSFE model. Second, the traditional way of cutting, shifting, releasing, and finally relaxing the atoms to generate GSFE profile due to the unsymmetrical feature of atomic configuration on the two sides of the slip direction causes a force along the P-direction, which influences not only the value but also the location of the calculated GSFE. Therefore, one may observe the fluctuations in the GSFE profile.

To resolve all the above mentioned issues related to the rigid GSFE profile, nudged elastic band method (NEBM) as a minimum path energy finder (MEP) is employed to find precise values of the saddle points in the GSFE profile. The SFE profile obtained by NEBM along direction shows a completely different energy profile in shape and magnitudes with respect to the GSFE one such that the magnitude of the energy barriers are reduced significantly thanks to the incorporation of elastic deformation energy in NEBM and non-straight shape of the slip trace obtained by NEBM. Therefore, NEBM can provide more accurate and robust results about SFEs compared to GSFE method.

Most polycrystalline Ni-based superalloys used in demanding environments contain a high concentration of annealing twins. In contrast to the concept of grain boundary engineering, where low Σ coincidence-site-lattice boundaries are essential to mitigate the intergranular cracking, coherent Σ3 twin boundaries (TBs) in Ni-based superalloys are found to be vulnerable for fatigue crack and hydrogen (H)-induced crack initiation. Despite recent extensive studies, the physical mechanism is still elusive, which hampers the alloy optimisation and performance prediction. Here we use multiscale microstructural characterisations and strain analysis, coupled with density functional theory calculation, to achieve a bottom-up understanding in the role of TBs on failure mechanism of γ"-strengthened Ni-based superalloys. We find that during mechanical loading much earlier dislocation plasticity and pronounced strain localisation happen at TBs, and H-induced cracks nucleate and extend along TBs. Multiscale characterisations, from centimetre scale down to sub-nanoscale, and DFT calculations demonstrate that the abnormal response from TBs originates from the distinctive precipitates formed along the TBs. Our work not only uncovers the mechanistic origin of TBs-related failure but also provides a practical method to alleviate the adverse effect of TBs on the performance of Ni-based superalloys in aggressive environments.

We describe an investigation of intergranular corrosion on the surface of pure Ni after cathodic charging. The intergranular corrosion takes the form of long trenches along coherent twin boundaries (CTBs). These trenches are formed by the growth and eventual overlap of conical cavities along <110>-type directions within the CTB planes. We propose that the corrosion cavities initiate at high energy, symmetric incoherent facets along CTBs and deepen due to their local anodic polarization relative to the sample surface. Our observations show that cathodic protection is not guaranteed to prevent corrosion. They also reveal that—far from being corrosion resistant—CTBs are especially susceptible to localized corrosion under some conditions.

Grain boundary mobility is affected by thermodynamic driving force and local velocity of atomic movement. It has been a long-time challenge to separate each individual contribution as they are intimately connect, but such understanding could help provide more reliable strategies for coarsening inhibition in nanocrystalline materials. Here we experimentally demonstrate the individual contribution of kinetics and thermodynamics in magnesium aluminate grain growth, in particular highlighting the nano-scale effects. Fully dense nanocrystalline MgAl2O4 grain growth was studied at different temperatures and a mobility shift - i.e. from a high mobility at small grains to lower mobility for larger grains - was observed below a critical nano-scaled grain size. By combining experimental grain boundary energies and kinetic analyses, we showed that the excess energies associated with grain boundaries at the nanoscale are increased due to the spinel inversion parameters, but there is also a change in grain boundary velocity, likely attributed to a complexion transition associated with the segregation of aluminum to grain boundaries.

High ionic conductivity is desired to optimize electrolyte performance, though it is significantly degraded by grain boundaries (GBs), which act as ionically-resistive blocking layers in polycrystalline electrolytes. Given the rich diversity in GB types, and the complex interplay between structure, composition, and chemistry at the atomic and nanoscale [1-3], there is considerable opportunity to elucidate fundamental science and performance optimization of GBs. Hence, studies should rely on GB datasets correlated across many length scales, with the aim of generalizing high spatial resolution observations to an entire GB population. This should facilitate bottom-up design of GBs with optimized properties, which remains a considerable challenge. By combining suitable modeling approaches with experimental measurements interrogating materials over different length scales, it becomes possible to estimate the electrical properties of individual GBs.

Here, we show that the enhancement of oxygen ionic conductivity by over two orders of magnitude in an electroceramic oxide is explicitly shown to result from nanoscale enrichment of a grain boundary layer or complexion with high solute concentration. A series of Ca_xCe_1−xO_2−δ polycrystalline oxides with fluorite structure and varying nominal Ca²⁺ solute concentration elucidates how local grain boundary composition, rather than structural grain boundary character, primarily regulates ionic conductivity. A correlation between high grain boundary solute concentration above ∼40 mole%, and four orders of magnitude increase in grain boundary conductivity is explicitly shown. A correlated experimental approach provides unique insights into fundamental grain boundary science, and highlights how novel aspects of nanoscale grain boundary design may be employed to control ion transport properties in electroceramics.

Additionally, we present a new data-driven interacting-defect model that has quantitatively described the nanoscopic composition of high solute concentrations at grain boundaries in ion-conducting ceramics. The successful model is a Cahn-Hilliard methodology for interfaces, introduced and demonstrated in this report. The model is applied to high spatial resolution composition data gathered at grain boundaries in calcium-doped ceria. The statistical methodology for the data-driven procedure shows definitively that gradient terms are required to model defect interactions to quantitatively describe the local grain boundary composition data. The model additionally predicts co-accumulation of negatively-charged acceptor dopants and positively-charged oxygen vacancies at the interface, which is in accordance with atom probe tomography in acceptor-doped ceria.

Acknowledgements
NSF Graduate Research Fellowship (DGE-1211230), NSF grant DMR-1308085, ASU’s John M. Cowley Center for High Resolution Electron Microscopy.

References
[1] W.J. Bowman, J. Zhu, R. Sharma, P.A. Crozier. Solid State Ionics (2014).
[2] W.J. Bowman, M.N. Kelly, G.S. Rohrer, C.A. Hernandez, P.A. Crozier. Nanoscale (2017).
[3] X. Tong, W.J. Bowman, P.A. Crozier, D. S. Mebane. (In Review).

Fe-based alloys are important structural materials for many engineering applications. Segregation of alloying elements and impurities to grain boundaries (GBs), either due to radiation induced segregation (RIS) or thermal segregation, can alter the GB cohesion strength and thus affect the fracture resistance of these alloys. Therefore, it is important to understand how the segregation of different alloying elements and impurities influences the GB cohesive strength. In this work, we use molecular dynamics simulations to study the changes of GB cohesive strength due to the substitutional segregation of P, Al, V, W, Cu, Ni and Cr at Σ3<111>, Σ3<112>, Σ5<012> and Σ5<013> GBs in body centered cubic (bcc) Fe. It is found that V, W increase the GB cohesive strength while P, Cu reduce it and Al, Cr, Ni have small effects. Additionally it is found that Σ3<111> GB is influenced the most by segregation and the alteration in GB cohesive strength is correlated to substitution energy in the bulk.

Grain boundary (GB) is one of the most important microstructures that govern the physical properties of materials. In many metallic systems such as nickel-based alloys, GBs are susceptible to intergranular corrosion, which can lead to intergranular fracture and stress corrosion cracking. Using first-principles calculations, we have investigated the strength of ∑5(012) and ∑3(111) GBs with different oxygen concentrations in both substitution and interstitial sites. As expected, our results show that oxygen is an embrittle element at both two sites. For each case, we compare the energy difference between mechanical distortion effect and chemical bond influence. Electronic analyses such as density of states (DOS) and charge density distribution are conducted to study the role of chemical bonds on GB strength.

Ceramic oxides are used for a wide variety of technologically relevant applications from electrochemical devices, novel resistive switching devices and oxygen sensors. Applications such as these typically rely upon the ability of oxides to conduct ions efficiently through the lattice. For oxides, such as CeO₂ (ceria), it is common to dope with aliovalent elements to increase the number of available charge carriers. While bulk-doping methods developed to optimize ionic conduction in the bulk have been successful, they have provided little insight when optimizing the grain boundary (GB) conductivity in these polycrystalline oxides. Unlike the bulk, GBs are often subject to complex segregation behavior, structural disorder, space charge effects as well as the standard defect-defect interactions typically present in the bulk. Recent nanoscale compositional characterization has shown different nominal concentrations could result in orders of magnitude increase in GB ionic conductivity relative to the undoped samples [1]. This effect has been attributed to increases in local dopant/solute concentrations at the GB. This study aims to understand how high concentrations of dopants, namely Ca, present at the GB increases the ionic conductivity.
To understand the key factors influencing the ionic conductivity, computational modeling is employed using molecular statics to optimize the GB interfacial structure for three symmetric tilt and one twist GB in CeO₂. The GB structures are doped according to three concentration profiles obtained in ref 1. Molecular dynamics simulations are performed to capture the changes in conductivity for various GB models. Migration energies, local strain and various oxygen migration pathways across each GB are compared to elucidate the role of local composition in modulating oxygen migration across GBs. Energetics and structural stability trends obtained using molecular statics/dynamics are validated using density functional theory calculations. This study further develops our understanding of the interplay between nanoscale composition and structure with high dopant concentrations enabling the development of methods such as selective doping to improve macroscopic ionic conductivity for both the grain and GB.

[1] Bowman, W. J. et al. (2017) Nanoscale, DOI: 10.1039/C7NR06941C
[2] We gratefully acknowledge ASU’s HPC staff for support and assistance with computing resources, NSF grant DMR-1308085, and ASU’s John M. Cowley Center for High Resolution Electron Microscopy.