Matt Beekman

Preparation and Crystal Growth of Na24Si136

The synthesis and single crystal growth of clathrate-II Na(24)Si(136) is performed in one step applying the spark plasma treatment to the precursor Na(4)Si(4). The reported results demonstrate a new route to intermetallic compounds facilitated by the electric field and current. SPS is revealed to offer significant opportunities as a novel preparatory method for synthesis and crystal growth of solid state materials.

Ferecrystals: non-epitaxial layered intergrowths

The synthesis, structure, and selected physical properties of a new class of layered intergrowth materials, referred to as ferecrystals, are reviewed. In these layered intergrowths, size and composition can be controlled at atomic length scales by utilizing structural interfaces between component compounds that lack an epitaxial relationship. Opportunities to observe and control size-dependent phenomena in intergrowths of technologically important compound semiconductors are discussed.

Framework Contraction in Na-Stuffed Si(cF136)

Systematic crystal structure refinements from powder X-ray diffraction data as well as density functional theory calculations demonstrate that the silicon clathrate II Si(cF136) exhibits a lattice contraction as Na is introduced solely into the Si(28) cages. When the Si(20) cages, in addition, begin to be filled with Na, a contrasting lattice expansion results. The nonmonotonic structural response to filling is an indication of markedly dissimilar guest-framework interactions for Na@Si(20) and Na@Si(28).

Controlling Size-Induced Phase Transformations Using Chemically Designed Nanolaminates

Since the initial observation of melting-point reduction with size for gold nanoparticles, the relative stability of crystalline phases has also been demonstrated to be size-dependent in a number of technologically important materials, including CdSe and CdS, Al2O3, [4] and various other metal oxides and chalcogenides. As the surface free energy contribution to the total free energy becomes increasingly important as the size of a system is decreased, a crystalline phase with lower surface free energy may be favored with respect to the thermodynamically stable bulk phase when the crystallite size is smaller than a critical value in one or more dimensions. The role of size as an effective thermodynamic parameter is of fundamental importance, but also provides a mechanism for controlling the crystal structure of a material, and therefore its properties. Although precise control of surface chemistry and nanocrystal size in well-defined material systems is prerequisite to understanding and controlling size-induced phenomena, it nevertheless remains challenging to achieve. Although impressive progress has been made in the preparation of ensembles of inorganic nanocrystals with relatively narrow size distributions, the preparation of nanocrystal ensembles of completely uniform size that can be tuned with atomic precision would constitute a significant synthetic advance, in particular for the application of size-dependent structural and physical properties. Using a combination of experimental and computational techniques, we demonstrate herein that chemically designed nanolaminates, consisting of an intergrowth of chemically and structurally distinct components, comprise a class of materials in which this level of control can be achieved for one crystallographic direction. As a consequence of the ability to precisely control size, the crystal structure of the components can be tuned via a size-induced transformation. In contrast to epitaxial superlattices, which experience structural distortions due to strain induced by epitaxial interfaces, the intergrowths reported on in the present work lack an epitaxial relationship between the components. This structural independence of the constituents allows the effect of size on crystal structure to be delineated from strain effects. It was recently discovered that [(MSe)1+d]m[TSe2]n intergrowths can be prepared by the modulated elemental reactants (MER) synthetic route. Here M= {Pb, Sn, Bi, Ce}, T is an early transition metal, and the integers m and n denote the number of consecutive layers of the respective components in the repeating unit of the intergrowth (Figure 1). The value of d reflects the difference in the inplane atomic packing density for the independent component structures in the intergrowth (hereafter we use the designation [MSe]m[TSe2]n for convenience). We chose to explore the SnSe-MoSe2 system for several reasons. Along with the prospect of developing novel nanocrystalline SnSe materials for optoelectronic and photovoltaic applications, bulk crystalline SnSe undergoes a secondorder (continuous) structural phase transition from a-SnSe (GeS structure type) to b-SnSe (ThI structure type) as the temperature is increased. As the parent compounds SnSe and MoSe2 (Figure 1a and b) are both layered semiconductors, significant charge transfer between the components in the intergrowth is not expected. A stable interface exists

Telluride Misfit Layer Compounds: [(PbTe)1.17]m(TiTe2)n

Telluride misfit layer compounds are reported for the first time. These compounds were synthesized using a novel approach of structurally designing a precursor that would form the desired product upon low-temperature annealing, which allows the synthesis of kinetically stable products that do not appear on the equilibrium phase diagram. Four new compounds of the [(PbTe)(1.17)]m(TiTe2)n family are reported, and their structures were examined by a variety of X-ray diffraction techniques.

Synthesis and Thermal Properties of Solid-State Structural Isomers: Ordered Intergrowths of SnSe and MoSe2.

A family of structural isomers [(SnSe)1.05]m(MoSe2)n were prepared using the modulated elemental reactant method by varying the layer sequence and layer thicknesses in the precursor. By varying the sequence of Sn-Se and Mo-Se layer pairs deposited and annealing the precursors to self-assemble the targeted compound, all six possible isomers [(SnSe)1.05]4(MoSe2)4, [(SnSe)1.05]3(MoSe2)3[(SnSe)1.05]1(MoSe2)1, [(SnSe)1.05]3(MoSe2)2[(SnSe)1.05]1(MoSe2)2, [(SnSe)1.05]2(MoSe2)3[(SnSe)1.05]2(MoSe2)1, [(SnSe)1.05]2(MoSe2)1[(SnSe)1.05]1(MoSe2)2[(SnSe)1.05]1(MoSe2)1, and [(SnSe)1.05]2(MoSe2)2[(SnSe)1.05]1(MoSe2)1[(SnSe)1.05]1(MoSe2)1 were prepared. The structures were characterized by X-ray diffraction and electron microscopy which showed that all of the compounds have very similar c-axis lattice parameters and in-plane constituent lattice parameters yet distinct isomeric structures. These studies confirm that the structure, order, and thickness of the constituent layers match that of the precursors. The cross-plane thermal conductivity is found to be very low (∼0.08 Wm(-1) K(-1)) and independent of the number of SnSe-MoSe2 interfaces within uncertainty. The poor thermal transport in these layered isomers is attributed to a large cross-plane thermal resistance created by SnSe-MoSe2 and MoSe2-MoSe2 turbostratically disordered van der Waals interfaces, the density of which has less variation among the different compounds than the SnSe-MoSe2 interface density alone.

Clathrates and beyond: Low-density allotropy in crystalline silicon

In its common, thermodynamically stable state, silicon adopts the same crystal structure as diamond. Although only a few alternative allotropic structures have been discovered and studied over the past six decades, advanced methods for structure prediction have recently suggested a remarkably rich low-density phase space that has only begun to be explored. The electronic properties of these low-density allotropes of silicon, predicted by first-principles calculations, indicate that these materials could offer a pathway to improving performance and reducing cost in a variety of electronic and energy-related applications. In this focus review, we provide an introduction and overview of recent theoretical and experimental results related to low-density allotropes of silicon, highlighting the significant potential these materials may have for technological applications, provided substantial challenges to their experimental preparation can be overcome.

Precursor Routes to Complex Ternary Intermetallics: Single-Crystal and Microcrystalline Preparation of Clathrate-I Na8Al8Si38 from NaSi + NaAlSi

Single crystals of the ternary clathrate-I Na8Al8Si38 were synthesized by kinetically controlled thermal decomposition (KCTD), and microcrystalline Na8Al8Si38 was synthesized by spark plasma sintering (SPS) using a NaSi + NaAlSi mixture as the precursor. Na8AlxSi46-x compositions with x ≤ 8 were also synthesized by SPS from precursor mixtures of different ratios. The crystal structure of Na8Al8Si38 was investigated using both Rietveld and single-crystal refinements. Temperature-dependent transport and UV/vis measurements were employed in the characterization of Na8Al8Si38, with diffuse-reflectance measurement indicating an indirect optical gap of 0.64 eV. Our results indicate that, when more than one precursor is used, both SPS and KCTD are effective methods for the synthesis of multinary inorganic phases that are not easily accessible by traditional solid-state synthesis or crystal growth techniques.

Synthesis, structure, and thermal conductivity of [(SnSe)1 + y]n[MoSe2]n compounds

A series of semiconducting [(SnSe)1.05]n[MoSe2]n compounds where n = 1–4 were prepared from thin film precursors with designed local compositions and nanoarchitectures to promote formation of the desired products during low temperature annealing. Specular diffraction patterns of annealed precursors contain only 00l reflections that yield c-lattice parameters indicating the formation of layered intergrowths. The in-plane diffraction patterns contain the independent reflections from the two constituents and the in-plane lattice parameters are independent of n. Electron microscopy images suggest significant turbostratic disorder between the constituent layers and between Se–Mo–Se trilayers within the dichalcogenide constituent. The room temperature cross-plane thermal conductivity was found to be low, between 0.08 and 0.22 W m−1 K−1 for the series of isocompositional compounds investigated, and is independent of the density of interfaces.

Estimating Energy Conversion Efficiency of Thermoelectric Materials: Constant Property Versus Average Property Models

Maximum thermoelectric energy conversion efficiencies are calculated using the conventional “constant property” model and the recently proposed “cumulative/average property” model (Kim et al. in Proc Natl Acad Sci USA 112:8205, 2015) for 18 high-performance thermoelectric materials. We find that the constant property model generally predicts higher energy conversion efficiency for nearly all materials and temperature differences studied. Although significant deviations are observed in some cases, on average the constant property model predicts an efficiency that is a factor of 1.16 larger than that predicted by the average property model, with even lower deviations for temperature differences typical of energy harvesting applications. Based on our analysis, we conclude that the conventional dimensionless figure of merit ZT obtained from the constant property model, while not applicable for some materials with strongly temperature-dependent thermoelectric properties, remains a simple yet useful metric for initial evaluation and/or comparison of thermoelectric materials, provided the ZT at the average temperature of projected operation, not the peak ZT, is used.