Symposium EL03—Emerging Ionic Semiconductors—Research and Applications

Most conventional semiconductors are covalent materials with four-fold coordination. The recent emergence of semiconducting perovskites, with six-fold coordination metal atoms in the skeleton framework and an exceptional success in photovoltaics, however, represent a different type of semiconductors distinctly different from any existing ones. For example, BaZrS₃ (a chalcogenide perovskite) has a direct band gap in the visible close to that of GaAs, but its ionicity, on average, is close to that of AlN. In other words, conventional semiconductors and perovskites represent two extremes in the ionicity spectrum. In this talk, I will discuss our recent theoretical efforts in advancing the knowledge on various forms of chalcogenide perovskites as a new class of semiconductors for electronic and optoelectronic applications [1-4]. Noticeably, a chalcogenide perovskite often exhibits a significantly higher density of states near the band gap than a conventional semiconductor due to the presence of transition-metal d states [1]. While the d bands are usually flatter than the (s, p) bands which is undesirable for carrier transport, their hybridizations exhibit noticeable anisotropy to resulting in favorable conduction channels with reasonably small effective masses. Also, indirect gap materials are often perceived as poor optoelectronic materials. However, due to these d bands, high optical absorption in indirect chalcogenide perovskites can take place at energies only a few tenth of an eV higher than the fundamental band gap [3], which could be beneficial for photovoltaic applications requiring both high absorption in the visible and long carrier lifetimes.
[1] Y. Y. Sun, et al., Nano Lett. 15, 581 (2015).
[2] Y.-Y. Sun, et al., Nanoscale 8, 6284 (2016).
[3] M. L. Agiorgousis, et al., Adv. Theory Simul. 2, 1800173 (2019).
[4] H Zhang, et al, Chin. Phys. Lett. 37, 097201 (2020).

The electron count in complex oxide compounds can be controlled during growth via compositional substitution or formation of defects such as oxygen vacancies. Reversible control can be achieved through gating in heterostructures or through introduction of H or alkali metal interstitials. In this talk, I will present first-principles investigations of the effects of added charge on the crystal structure and bandstructure of illustrative systems, including BaTiO3, ZnSnO3, SmNiO3, La1/3Sr2/3FeO3 and perovskite titanate superlattices. In some systems, the added electrons or holes become free carriers, their mobility and concentration determining the resulting conductivity. When the system is ferroelectric, for example BaTiO3 and ZnSnO3, key questions include the effect of the free carrier concentration on the polar distortion, and the switchablity of ferroelectrics with free carriers. Alternatively, the added charges can localize on transition metal ions and order, breaking translational and/or point symmetries. Combining the symmetry-breaking of the charge order with other symmetry-breaking factors, such as cation layering, epitaxial strain or the arrangement of the added interstitial ions, can produce charge-order-driven ferroelectricity. In these systems, coupling of charge order to the lattice is a key consideration in determining switchablity between states of different polarization. Connections to experimental observations in the literature and to ongoing experiments will be discussed.

N-type transparent conductors (TCs) are key materials in the modern optoelectronics industry. Despite years of research, the development of a high-performance p-type TC has lagged far behind that of its n-type counterparts, delaying the advent of “transparent electronics” based on transparent p-n junctions.[1] Here, we computationally investigate three layered oxychalcogenide structural motifs to try to predict new p-type TCs, namely the [Cu₂Ch₂][A₃B₂O₅] (325), [Cu₂Ch₂][A₄B₂O₆] (426) and [Cu₂Ch₂][A₂BO₂] (212) structural motifs. Specifically, we have used a materials informatics approach (SMACT) to screen through the search space using low-cost heuristic tools.[2] This reduces the potential combinations from 1800 to a more computationally tractable 228, which then undergo DFT calculations to assess thermodynamic and dynamic stability. In this talk, I will present an update on how our search has predicted >50 novel semiconductors with potential applications ranging from TCs,[3] to photocatalysts,[4] solar absorbers and thermoelectrics.

[1] A. Walsh and J.-S. Park, The Holey Grail of Transparent Electroncs, Matter, 3, 604 (2020)
[2] D. W. Davies, K. T. Butler, A. J. Jackson, A. Morris, J. M. Frost, J. M. Skelton and A. Walsh, Computational Screening of All Stoichiometric Inorganic Materials, Chem, 1 617 (2016)
[3] B.A.D. Williamson, G.J. Limburn, G.W. Watson, G. Hyett, and D.O. Scanlon. Computationally Driven Discovery of Layered Quinary Oxychalcogenides: Potential p-Type Transparent Conductors?, Matter, 3, 759 (2020)
[4] G. J. Limburn, M. J.P. Stephens, B. A. D. Williamson, A. Iborra-Torres, D. O. Scanlon and G. Hyett. Photocatalytic, structural and optical properties of mixed anion solid solutions Ba₃Sc_2−xIn_xO₅Cu₂S₂ and Ba₃In₂O₅Cu₂S_2−ySe_y, Journal of Materials Chemistry A, 8, 19887 (2020).

The interplay of first-principles density functional theory and experiments has led to the optimization of well-known families of oxides and chalcogenides whose members display a range of tunable band gaps, polarizations, and compositions. Here we combine crystallographic database mining and high-performance computing to discover and design new semiconductive functional materials that could be readily synthesized and potentially operate in different regimes of temperature, pressure, humidity, etc. Ultimately, we would like to examine a family of materials whose compositions span a wide set of cations and anions, specifically oxide and chalcogenides, to complement well-established materials like the ABX₃ perovskites and Ruddlesden-Popper phases. We focus on finding new examples of functional materials called ferroelectrics and antiferroelectrics since semiconductive ferroelectrics can find use as photovoltaics and antiferroelectrics are proposed to find use as materials capable of energy storage. Our work explores a candidate family of materials with a general formula of A₂BX₃, where members have been assigned nonpolar, antipolar, and polar structure types. We use density functional theory to map out the potential energy landscape of this family by including both known and as-yet to be identified structure types. The results of our analysis show several known members to be either ferroelectric or antiferroelectric. We present a targeted set of oxide and chalcogenide semiconductors, and their compositionally tuned variants, that warrant further investigation as new functional materials.

Materials design from first principles enables exploration of uncharted chemical spaces. Broad computational searches have been reported for ternary mixed-cation oxides and nitrides. More recently, mixed-anion systems are gaining interest for computational discovery studies. Within this class, oxide-nitrides are particularly intriguing, because the two constituent systems exhibit remarkable commonalities (e.g., bond strength, elastic properties) and differences (e.g., formation enthalpies, electronic structure). A common limitation of computational discovery approaches is the reliance on prototype structures for total energy calculation which can miss the lowest energy structure. In addition, some known ternary oxide-nitrides form an ordered or disordered configuration derived from the crystal structure of a binary oxide, which may not be captured by prototypes selected from ternaries. We approach this challenge by letting two complementary structure sampling approaches compete. We use the Kinetically Limited Minimization (KLM) approach for high-throughput unconstrained crystal structure prediction in smaller cells up to 20 atoms, and, on the other hand, a configurational sampling on larger binary prototype structures. Using this approach, we searched 65 different charge-balanced oxide-nitride stoichiometries, where 6 known systems were included as control sample. Within this set, we predict 8 new stable ternary oxynitrides, and an additional 5 new oxynitrides that are expected to be stable under activated Nitrogen conditions. The control sample was correctly recovered, providing confidence in the approach and clearing the path for future studies in a more extensive search space.

Scientists and practitioners have long dreamt of designing materials with novel properties. Yet, a hundred years after quantum mechanics lay the foundations for a systematic description of the properties of solids, it is still not possible to predict the best material in applications such as photovoltaics, superconductivity or thermoelectric energy conversion. This is a sign of the complexity of the problem, which is often exacerbated by the need to optimize conflicting material properties. Hence, one can ponder if design routes for materials can be devised.
In recent years, the focus of our work has been on designing advanced functional materials with attractive opto-electronic properties, including phase change materials, thermoelectrics, photonic switches and materials for photovoltaics. These materials are typically discussed as unconventional semiconductors, often but not always, with appreciable charge transfer. Phase Change Materials have provided a special challenge for materials optimization. They possess a remarkable property portfolio, which includes the ability to rapidly switch between the amorphous and crystalline state. Surprisingly, in PCMs both states differ significantly in their properties. This material combination makes them very attractive for applications in rewriteable optical and electronic data storage, as well as photonic switches. In this talk, the unconventional material properties will be attributed to a unique bonding mechanism (metavalent bonding). Further evidence for this bonding mechanism comes from a quantum-chemical map, which separates the known strong bonding mechanisms of metallic, ionic and covalent bonding. The map reveals that metavalent bonding is a new, fundamental bonding mechanism, which differs substantially from metallic, covalent and ionic bonding. This insight is subsequently employed to design phase change as well as thermoelectric materials. Yet, the discoveries presented here also force us to revisit the concept of chemical bonds and bring back a history of vivid scientific disputes about ‘the nature of the chemical bond’.

In the talk I shall discuss recent efforts to computationally identify light-absorbing materials for use in solar cells and water splitting devices in particular with a tandem configuration. A range of materials are considered including sulfide perovskites, quartenary chalcogenides, and more broadly experimentally synthesized compounds, which have been structurally characterized. Key descriptors such as structure, stability, band gap, band structure, and defect states will be addressed. The possibilities of using machine learning to improve the efficiency of computational screening and structure determination will also be discussed.

Nonlinear optical (NLO) materials are used to convert the specific coherent wavelengths produced by lasers into wavelengths in other spectral regions such as UV-visible (0.2–2 µm) and infrared (IR) (3–20 µm), where lasers have poor efficiency. Metal chalcogenides with their smaller band gaps are appropriate for IR lasers, which are useful for visualization of tissue, environmental monitoring, and security applications. There are plenty of commercially available UV-visible NLO materials, but there only a few commercially available IR NLO materials. Even these materials suffer from problems such low laser damage threshold and two photon absorption, thus there is a need to search for new NLO materials in the IR-region. Alkali-metal chalcoarsenates (AAsQ₂) one such system that were discovered a decade ago which showed impressive NLO properties.
AAsQ₂ (A = Li, Na and Q = S, Se) compounds are made up of 1-dimensional (1D) (1/∞) [AQ₂]^— chains which are connected by pyramidal AQ₃ units as observed in β-LiAsQ₂ (Cc) and γ-NaAsSe₂ (Pc) structures (Q = S and Se). γ-NaAsSe₂ have one of the highest second harmonic generation (SHG) intensities observed (~ 75 x AgGaSe₂) while, adding Na to the β-LiAsQ₂ structures resulted in increased SHG response as observed in γ-Li_0.2Na_0.8AsSe₂ (~ 65 x AgGaSe₂) and β-Li_0.6Na_0.4AsS₂ (~ 30 x AgGaSe₂). Structurally, the substitution of the smaller Li ions with larger Na ions weakens the interactions between adjacent (1D) (1/∞) [AQ₂]^— chains. A problem with γ-NaAsSe₂ (Pc) is a phase transition to the centrosymmetric δ-NaAsSe₂ (Pbca) upon melting, making it a challenge to grow large single crystals for NLO applications.
Here we have studied the effects of substituting γ-NaAsSe₂ (Pc) with the larger cation K to see if the interactions between the adjacent (1/∞) [AQ₂]^— chains are further weakened, thereby, its effect on their NLO properties. Additionally, K containing chalcoarsenates are usually low melting making them attractive for crystal growth which are essential for their application as NLO materials.

The smaller band gap compared to pure oxides yields a superior light-absorption efficiency of perovskite oxynitride photocatalysts. While the bulk structure and properties of these materials have been extensively studied by computational methods, significantly less is known about their surfaces and consequently their photocatalytic reaction mechanisms. In this talk, I will present results of our density functional theory (DFT) calculations, highlighting anion-ordering phenomena at perovskite oxynitride surfaces, the effect of defects on their oxygen-evolution reaction (OER) activity as well as the potential of strain engineering these materials to become ferroelectric photocatalysts.

Thermal energy is an unavoidable by product of myriad processes, spanning the industrial, domestic and transportation sectors, and presents an abundantly available and largely untapped source of clean energy. Thermoelectric devices, which are able to convert thermal energy into electricity, provide a route to significantly increase the overall energy efficiency of existing process across a range of sectors via waste-heat harvesting and yield a supply of clean energy.^[1] The effectiveness of a thermoelectric material is measured using the dimensionless figure of merit ZT, with the world record set at 2.6 for single-crystal SnSe along the through-plane direction.^[2] Realising reasonable conversion efficiencies generally requires high electrical conductivity and low thermal conductivity, with the maximum ZT of a material often limited by the strong correlation between these properties.

Despite efforts in the past few decades to identify promising novel thermoelectric materials, the champion thermoelectric materials tend to rely on non earth-abundant and toxic elements, such as bismuth telluride (Bi₂Te₃) and the lead chalcogenides. Oxides generally present properties valuable for thermoelectric applications, including low cost, thermal and chemical stability and environmental benignity, but broadly thermoelectric performance has been hindered by the inherent high lattice thermal conductivities.^[4] The performance of oxide thermoelectrics generally lagging behind the efficiencies achieved by chalcogenide-based materials, with poor carrier mobilities and high intrinsic thermal conductivities hampering progress.^[3]

In this study, we predict the maximum theoretical ZTs of a range of novel mixed-anion quaternary systems using state-of-the-art methods based on density functional theory. We have identified three materials with complex crystal structures, Bi₂YO₄Cl, Bi₂YO₄Br and Bi₂YO₄I, which possess lower lattice thermal conductivities than usually seen in oxides, and excellent charge transport properties, calculated using methods superseding the traditionally used constant relaxation time approximation.⁵ We conclude that these materials have the potential to achieve ZTs greater than many existing oxide thermoelectric materials, and have the potential to perform as high-performance thermoelectric components. Additionally, the effects of nanostructuring these materials is discussed, to guide potential synthesis methods.

1. L. E. Bell, Cooling, heating, generating power, and recovering waste heat with thermoelectric systems. Science 321, 1457–1461 (2008)
2. L-D, Zhao, S-H. Lo, Y. Zhang, H. Sun, G. Tan, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals. Nature 508, 373-377 (2014)
3. G. Tan, L-D. Zhao, and M. G. Kanatzidis, Rationally designing high-performance bulk thermoelectric materials, Chemical Reviews 116, 12123–12149 (2016)
4. J. He, Y. Liu and R. Funahashi, Oxide thermoelectrics: The challenges, progress and outlook, J. Mater. Res. 26, 1762-1772 (2011)
5. Ganose, A. M.; Park, J.; Faghaninia, A.; Woods-Robinson, R.; Persson, K. A.; Jain, A. Efficient Calculation of Carrier Scattering Rates from First Principles. arXiv:2008.09734 [cond-mat, physics:physics] (2020)
6. Einhorn, M.; Williamson, B. A. D.; Scanlon, D. O. Computational prediction of the thermoelectric performance of LaZnOPn (Pn = P, As) J. Mater. Chem. A, 8, 7914–7924 (2020)

Chalcogenide Perovskites are a new class of semiconductors, which have tunable band gap in the visible to infrared part of the electromagnetic spectrum, large density of states with potentially high carrier mobility, and emergent photonic properties with anisotropy and non-linearity. In this talk, I will review the developments in thin film growth of perovskite chalcogenides and discuss some of the advances made in my research group on developing pulsed laser deposition as a growth method for chalcogenide perovskite thin films. We have successfully achieved textured growth of chalcogenide perovskite thin films on single crystalline oxide substrates. This advance presents new opportunities to probe the physical and chemical properties of this family of novel electronic materials, their device implementations and applications, and also study the interfacial properties of oxides and chalcogenides. Specifically, I will outline the growth of model chalcogenide perovskite, BaZrS₃ and the quasi-1D hexagonal Perovskite-related sulfide, BaTiS₃ on compatible perovskite and other oxide single crystal substrates. Preliminary investigations on the physical properties accompanied by detailed structural and chemical characterizations and future outlook for the thin film growth of these materials will also be discussed.

Despite long-standing research efforts to develop p-type transparent conductive materials (TCMs), the current generation of optoelectronic devices still relies exclusively on n-type TCM contacts due to their much better trade-off between conductivity and transparency. An important issue in oxide-based p-type TCMs is the deep energy and localized nature of the 2p oxygen states in the valence band, which has negative consequences on both hole dopability and hole mobility in most oxides.

Recent computational work has identified the mixed ionic-covalent compound CaCuP as a promising material that could potentially outperform the existing p-type TCMs. On the one hand, the covalent bond between Cu and P produces a highly dispersive valence band, which favors high hole mobilities. On the other hand, the presence of an electropositive cation (Ca) implies that more ionic Ca-P bonds are also present in CaCuP. This plays an important role in widening the band gap (thus ensuring high optical transmission in the visible) and making CaCuP potentially suitable as a p-type TCM.

Using a unique combinatorial sputter chamber equipped with reactive PH₃ gas, we have grown CaCuP thin films for the first time. We have mapped the compositional and thermal phase space of CaCuP and evaluated its potential as a p-type TCM using high-throughput methods. Due to the high concentration of copper vacancies, the electrical conductivity of CaCuP under optimized growth conditions is almost on par with the conductivity of state-of-the-art n-type TCMs such as ITO and FTO, even in the absence of any extrinsic dopant. However, CaCuP films are only moderately transparent in the visible region due to unexpectedly high absorption strength above its indirect band gap. We will discuss possible solutions to this problem, together with the general prospects of CaCuP as a p-type TCM.

Recently published work reveals that the ionic chalcogenide perovskite BaZrS₃ possesses suitable properties for PV application, such as direct band gap of 1.7 to 1.8 eV, high stability against moisture and pressure, and strong light absorption. However, the lack of thin films hinders further exploration of this material’s fundamental properties. Here we report the first work on BaZrS₃ films, by sulfurization of precursor films deposited by pulsed laser deposition and magnetron sputtering. The films are n-type semiconductors with carrier densities in the range of 10¹⁹ to 10²⁰ cm^-3. The hall mobility ranges from 2.1 to 13.7 cm²/Vs depending on the sulfurization temperature. UV-Vis result shows an absorption coefficient of >10⁵ cm^-1 at a photon energy of >1.97eV. We further incorporate Ti to reduce the band gap of BaZrS₃ for better solar performance. Our thin film results pave the way to future device fabrication and validate BaZrS₃ as a viable candidate for optoelectronics.

Chemical intuition, first-principles calculations, and recent experimental results suggest that chalcogenide perovskites are an outstanding class of semiconductors [1]. Chalcogenide perovskites feature the large dielectric response familiar in oxide perovskites, but also have band gap in the VIS-IR and strong light absorption. Preliminary results suggest that chalcogenide perovskites feature excellent excited-state charge transport properties familiar in halide perovskites, while also being thermally-stable and comprised of abundant and non-toxic elements. Nearly all experimental results on chalcogenide perovskites to-date were obtained on powder samples and microscopic single crystals [2-7]. Realizing the full potential of chalcogenide perovskites for optoelectronic applications will require the availability of high-quality, single-crystal thin films.

Here we report synthesis of BaZrS₃ thin films by molecular beam epitaxy (MBE) on oxide substrates. The film composition is confirmed by X-ray fluorescence and X-ray photoelectron spectroscopy. The phase is confirmed by X-ray diffraction and Raman spectroscopy. Photoluminescence spectroscopy showed peaks at 1.8 and 1.95 eV, suggestive of pristine and partially-oxidized material. We use spectroscopic ellipsometry to measure optical properties, confirming strong band-edge light absorption. As time allows we will present results of scanning transmission electron microscopy, impedance spectroscopy, the growth of BaZr(S,Se)₃ alloys with tunable band gap.

[1] R. Jaramillo, J. Ravichandran, APL Materials 7(10) (2019) 100902.
[2] S. Niu, D. Sarkar, K. Williams, Y. Zhou, Y. Li, E. Bianco, H. Huyan, S.B. Cronin, M.E. McConney, R. Haiges, R. Jaramillo, D.J. Singh, W.A. Tisdale, R. Kapadia, J. Ravichandran, Chemistry of Materials 30(15) (2018) 4882-4886.
[3] K. Hanzawa, S. Iimura, H. Hiramatsu, H. Hosono, Journal of the American Chemical Society 141(13) (2019) 5343-5349.
[4] S. Niu, J. Milam-Guerrero, Y. Zhou, K. Ye, B. Zhao, B.C. Melot, J. Ravichandran, Journal of Materials Research 33(24) (2018) 4135-4143.
[5] S. Niu, H. Huyan, Y. Liu, M. Yeung, K. Ye, L. Blankemeier, T. Orvis, D. Sarkar, D.J. Singh, R. Kapadia, J. Ravichandran, Advanced Materials 29(9) (2017) 1604733.
[6] Y. Nishigaki, T. Nagai, M. Nishiwaki, T. Aizawa, M. Kozawa, K. Hanzawa, Y. Kato, H. Sai, H. Hiramatsu, H. Hosono, H. Fujiwara, Solar RRL 4(5) (2020) 1900555.
[7] S. Filippone, B. Zhao, S. Niu, N.Z. Koocher, D. Silevitch, I. Fina, J.M. Rondinelli, J. Ravichandran, R. Jaramillo, Physical Review Materials 4(9) (2020) 091601.

SrHfS₃ belongs to the family of chalcogenide perovskite semiconductors with a bandgap of 2.43eV. Recently SrHfS₃ bulk crystals have been shown to exhibit strong green photoluminescence. It was also shown that it can be heavily doped to be both p- and n-type with strong emission, suggesting its bipolar doping capability and defect tolerance. It is thus a promising candidate to fill the green gap for solid
state lighting applications. In this work, we show that SrHfS₃ thin films can be prepared by magnetron sputtering and CS₂ sulfurization. Raman spectrum matches closely with the theoretical calculations of a distorted perovskite structure. The film quality such as PL intensity and carrier concentration is highly dependent on sulfurization temperature.

PbSe has emerged as a prototype for the study of unusual bonding in semiconductors. Recent work reveals the nature of bonding in PbSe and related IV-VI narrow band gap semiconductors as being distinct from covalent, ionic, and metallic.¹ We investigate the impact of such bonding on thin film growth, dislocation generation and dynamics, band-to-band carrier recombination, and field evaporation in epitaxial films grown directly on III-V substrates. These high quality engineered substrates serve as an excellent platform to probe and manipulate the properties of PbSe and its alloys.

We first demonstrate a route to prepare single crystal rocksalt PbSe films both on nearly lattice-matched and highly mismatched zincblende III-V substrates like InAs and GaAs using molecular beam epitaxy.² Our synthesis technique yields sharp heterointerfaces and we see clear structural distortions in the first few monolayers of PbSe mediating the significant bonding mismatch. These bare thin films strongly luminescence in the mid-infrared at room temperature even in the presence of threading dislocation densities exceeding 10⁹/cm², in sharp contrast to what may be expected from covalently bonded narrow bandgap III-V materials. We find minority carrier recombination lifetimes in our unintentionally doped PbSe films under low excitation in the 20–30 ns range and suggests a three carrier Auger process, although some very thin films deviate from this trend towards even longer lifetimes. Overall, these results are promising for the development of heterogeneously integrated mid-infrared light emitters.

Exploring this theme of unusual bonding further, the occurrence of elevated multi-atom evaporation events in atom probe tomography (APT) was recently hypothesized as a means to evaluate the nature of bonding in materials.³ We indeed see high rate of events with multiple atoms evaporating at once in voltage-pulsed APT of PbSe, supporting this suggestion, and finding even more such events localized around dislocation cores. However, in the same experiment we also see a high rate of multi-atom evaporation events in epitaxial layered-structure SnSe, which should be a mix of conventional covalent intra-layer and Van der Waals inter-layer bonding, showing exceptions are possible.

¹ M. Wuttig, V.L. Deringer, X. Gonze, C. Bichara, and J.-Y. Raty, Advanced Materials 30, 1803777 (2018).
² B.B. Haidet, E.T. Hughes, and K. Mukherjee, Phys. Rev. Materials 4, 033402 (2020).
³ M. Zhu, O. Cojocaru-Mirédin, A.M. Mio, J. Keutgen, M. Küpers, Y. Yu, J.-Y. Cho, R. Dronskowski, and M. Wuttig, Advanced Materials 30, 1706735 (2018).

As the device's feature sizes continue to shrink, traditional floating gate NAND flash memory faces reliability issues such as cell-to-cell interference, reduced charge loss tolerance, and susceptibility to stress-induced leakage currents. 3D NAND flash memory was developed to overcome the above-mentioned problem. In 3D NAND flash memory, the charge storage material is silicon nitride (SiN_x), and the main trend for increasing memory density is not scaling shrinking but stacking layers. Low-pressure chemical vapor deposition (LPCVD) is one of the most popular deposition methods to deposit SiN_x thin films on semiconductors due to its low hydrogen content and excellent step coverage and thermal stability. However, as more layers are stacked to increase memory density, the device's aspect ratio increases. The 1st generation 3D NAND flash memory has 24 layers and has an aspect ratio of 40:1. And currently, devices with more than 100 layers are required. As above, with the continuous aspect ratio increase, there is a demand deposition technology with accurate thickness control and high step coverage. Among various deposition methods to apply at high step coverage device, atomic layer deposition (ALD) is one of the best solutions to satisfy above-mentioned requirements. Due to self-limited ALD reaction, it enables to deposit thin film with high step coverage and good thickness control. Particularly, remote plasma ALD (RPALD) was utilized to enhance the reactivity between precursor and reactant gas for high film density with minimizing plasma damage. In RPALD, the plasma generation section is remotely outside of the reaction chamber and the radicals in plasma generation region enter into the reaction chamber for deposition.
In this study, we developed low-temperature SiN_x using bis(dimethylaminomethylsilyl)-trimethylsilyl amine (C₉H₂₉N₃Si₃, DTDN-2H2) and N₂ reactant plasma as the precursor and reactant plasma. We studied the effect of plasma treatments that are 2 methods such as post plasma treatment and during deposition process plasma treatment. Various plasma processing processes were used in deposition processes to satisfy the defect density levels (5x10¹⁹ /cm³). We investigated the effect of plasma treatment on the properties of SiNx thin film. Auger electron spectroscopy (AES) was utilized to measure the stoichiometry. Samples treated by plasma treatment were stoichiometry thin film by 1:1.33 Si:N ratio. Because hydrogen content affects fault density, the hydrogen content of SiNx film deposited was measured using secondary ion mass spectrophotometry (SIMS). X-ray photoelectron spectroscopy (XPS) was utilized for chemical binding state. N1s peak of H₂ plasma treatment SiN film had low-binding shift to397.5eV from 397.7 eV (N-Si bond). The density of plasma-treated thin films was compared by X-ray reflectometry (XRR). In the case of thin film with plasma treatment, film density was higher than as deposition sample. And thin film with post H2 plasma treatment was the highest density of 2.9g/cm³. In addition, studying defect properties of each thin film, we fabricated metal-Al₂O₃-silicon nitride-SiO₂-Si (MANOS) device. The trap density required by CTF device was satisfied by Ar plasma treatment during ALD process, post Ar plasma treatment, and post H₂ plasma treatment SiN thin film. Trap density of Post H₂ plasma treatment SiN films was increased to 7.65x10¹⁹ /cm³

Recently, the application of superionic conductors with liquid-like sublattice has been extended to the field of thermoelectrics. Extremely high thermoelectric figure of merit with the maximum zT > 2.0 have been observed and reported in a large family of novel Cu- and Ag-based liquid-like materials. However, despite the high performance, the stability and reliability of these liquid-like materials are key concerns for long-term service in the real applications. In this study, through systematically investigating the behavior of various liquid-like materials in an electric field or thermal gradient, we reveal the relations of atom migration, deposition, and material degradation. A general model is proposed to reveal the threshold for decomposition of liquid-like materials to prevent metal deposition upon atom migration. It is found that each liquid-like material has a threshold for metal deposition, named as the critical voltage. When the voltage stressed on the material is lower than the V_c, the mobile ions just form a steady concentration of gradient inside the material without any deposition. In this case, the liquid-like materials will demonstrate similar good stability with the traditional stable materials. With these understanding, a new strategy is proposed to design the stable and efficient TE modules based on liquid-like materials via geometry and interface optimization. By using this strategy, a TE module based on Cu₂Se with both good stability and high energy conversion efficiency up to 9.1% was successfully achieved.

Perovskite materials can have self-passivated surfaces without resorting to reconstructions or other passivating agents. This feature renders perovskite materials intrinsically suited for making two-dimensional (2D) materials. Previous studies often considered the structure of 2D perovskites to be directly terminated from the bulk. However, octahedron rotation (OR), being the distinctive structural feature of perovskite materials, appears to be an unexplored degree of freedom in 2D.
In this work, the OR patterns in 2D perovskites are systematically studied for the first time by employing an adapted Glazer’s notation. Taking 2D Ba-Zr-S system as an example, we establish the relation between the OR pattern and slab thickness. It is found that as the thickness decreases, the OR pattern undergoes a transition by suppressing out-of-plane rotations. The OR in 2D chalcogenide perovskites could result in an anti-confinement effect, i.e., reducing the band gap to even below that of the bulk by countering the quantum confinement effect. In addition, we show that the Ba-Zr-S 2D perovskites exhibit reasonable electron mobility of ~150 cm² V^-1 s^-1 and large exciton binding energy of ~0.9 eV. Our results suggest that the structure of OR is worth attention in future studies on 2D perovskites, and it provides a novel tuning knob for enhancing the properties of 2D materials, which is absent in existing 2D materials, such as graphene and transition metal dichalcogenides.

Transparent conducting materials (TCMs) have been widely used in optoelectronic applications such as touchscreens, flat panel displays and thin film solar cells. These applications of TCMs are currently dominated by n-type doped oxides. High-performance p-type TCMs are still lacking due to their low hole mobility or p-type doping bottleneck, which impedes efficient device design and novel applications such as transparent electronics. Here, based on first-principles calculations, we propose chalcogenide perovskite YScS₃ as a promising p-type TCM. According to our calculations, its optical absorption onset is above 3 eV, which allows transparency to visible light. Its hole conductivity effective mass is 0.48��₀, which is among the smallest in p-type TCMs, suggesting enhanced hole mobility. It could be doped to p-type by group-II elements on cation sites, all of which yield shallow acceptors. Combining these properties, YScS₃ holds great promise to enhancing the performance of p-type TCMs toward their n-type counterparts.

As one of low-cost and widely-utilized materials, pervoskite type metal oxides have been extensively investigated and applied in environmental remediation and protection, energy conversion and storage, etc.^[1-7] Most of these diverse applications are results of a large diversity of the electronic states of metal oxides. Noticeably, however, many metal oxides present obstacles for applications in catalysis, mainly due to the lack of efficient active sites with desired electronic states. Continuous efforts and strategies have been devoted to create new structural units and functional activities of metal oxides. Single-site catalysis is one of most attractive solutions to explore new activities and enhance catalytic efficiencies.^[8-10] Here, we demonstrate the fabrication of single tungsten atom oxide (STAO), in which the metal oxide’s active site unit reaches its minimum and a new electronic state is created. The catalytic mechanism in the STAO is determined by a new single-site physics mechanism, which is fundamentally distinct from the traditional size effect, and is also in contrast to the standard condensed matter physics. The photogenerated electron transfer process is enabled by an electron in the spin-up channel excited from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO)+1 state (both are largely tungsten atomic d-orbitals), which can only occur in STAO with W⁵⁺. STAO results in a record-high and stable sunlight photocatalytic degradation rate of 0.24 s^-1, which exceeds the rates of available photocatalysts by approximately two orders of magnitude. The fabrication of STAO and its unique single-site photocatalytic mechanism lays a new ground for achieving novel physical and chemical properties using various single metallic atom oxides.

Complex chalcogenides have demonstrated a variety of unique physical properties that are directly related to specific bonding schemes, and therefore continue to be investigated for different applications of interest. In addition, new chalcogenide compositions expand our library of materials and provide potential new avenues for discovery. Research that not only allows for a continuity of ongoing work into new and novel chalcogenide material systems but can also provide pathways for the design of materials or processing techniques with targeted properties for specific applications is paramount. The intellectual merit of these investigations is very closely tied to the particular bonding and structure types these materials possess. Developing a fundamental understanding of the underlying physical properties and their structure-property relationships is also of interest. I will present some of our recent progress on the structure-property relationships of specific materials, including layered materials, lanthanide metal polytellurides and new quaternary chalcogenides, provide insight into the search for new materials, and present results that demonstrate strong electron-phonon coupling and atypical transport of new chalcogenide materials.

The dependence of electronic structure on cation disorder in II-IV-V₂ materials with lattice parameters matched to their analogue III-V semiconductors makes these systems promising for tuning band gaps for use in energy-relevant devices such as LEDs and PV. Experimentally, however, properties do not always correlate with order as expected. In ZnGeN₂ specifically, the non-isovalent character of the disordered species (Zn²⁺ and Ge⁴⁺) subjects the cation ordering to strong short-range order effects which influence band structure by decreasing the band gap relative to ordered ZnGeN₂. ZnGeN₂ exhibits pronounced discontinuities in enthalpy, entropy, and structural order parameters, which correspond to a first-order phase transition which is key to understanding the tunability of electronic properties. The steep nature of this transition suggests intermediate degrees of ordering are difficult or even impossible to obtain in single crystals or within crystalline grains. To model cation disorder, we use Monte Carlo (MC) simulations implementing a cluster expansion to approximate formation enthalpy. Representative configurations are relaxed in supercells containing 1,024 atoms using density functional theory calculations. From the Monte Carlo structures, we calculate the fractions of all possible nitrogen coordination environments as a short-range order parameter and the Bragg-Williams and stretching parameters as long-range order parameters. We compare correlations between these order parameters as well as the mixing entropy and free energy of the system determined from thermodynamic integration to experimental literature on order in II-IV-V₂ materials.

Fe₂O₃ thin-films have been extensively investigated as photoanodes, with a variety of different strategies being proposed to mitigate key performance limiting factors, namely short carrier lifetimes and surface recombination kinetics.¹ On the other hand, significantly less is known about the properties of perovskite ferrite absorbers, including LaFeO₃,^2-4 YFeO₃,⁵ PrFeO_3,⁶ and BiFeO₃.⁷ These materials exhibit a wide range of intrinsic defects, which often lead to p-type conductivity. Our recent studies have shown that highly crystalline LaFeO₃ nanoparticles can promote hydrogen evolution under illumination at potential as positive as 1.47 V vs RHE,⁸ one of the highest photovoltages reported for a single p-type absorber layer. However, the performance of these perovskites is limited to quantum yields below 1%. In this contribution, we will examine the nature of the states involved in the loss of photogenerated carriers and the effect of partial substitution by alkaline-earth metal cations (AMC).
LaFeO₃ thin-film with a thickness of 95 nm were prepared by thermolysis (600 °C) of sol-gel precursors incorporating citric acid as chelating agent.⁹ The ratio of AMC (Mg²⁺, Ca²⁺, Ba²⁺ and Sr²⁺) to La³⁺ were adjusted in the range of 0 to 10% in the precursor solution, while keeping the Fe³⁺ concentration constant. XRD confirms the formation of single-phase cubic LaFeO₃ thin films across the whole composition range. Interestingly, we observe subtle trends in lattice constant variations which are closely correlated to shifts in the binding energies of Fe 2p_3/2 and O 1s measured by XPS. These trends are the result of the complex interplay between differences in ionic radii of the cations and changes in the oxidation state of Fe sites. Indeed, we establish a scaling factor between these two photoemission peaks, revealing a direct correlation between Fe oxidation state and Fe–O covalency. Electrochemical impedance spectroscopy (EIS) confirms the p-type characteristic of pristine LaFeO₃ thin-films, as well as the presence of sub-bandgap electronic state (A-states) close to the valence band edge. Partial AMC replacement leads to: (i) a decrease in the density of A-states, (ii) an increase in density of majority carriers (shallow acceptor states), and (iii) a shift of the valence band edge toward more positive potentials. In addition, AMC-substituted films exhibit deeper states centered at 0.6 eV above the valence band edge (B-states). These sub-band gap states have contrasting effects on the photoelectrochemical responses towards the oxygen reduction and the hydrogen evolution reactions. These trends are rationalized in terms of the position of the sub-bandgap states, majority carrier mobility, charge transfer and recombination kinetics.
References:
1. W. Yang, R.R. Prabhakar, J. Tan, S.D. Tilley and J. Moon, Chem. Soc. Rev., 48, 4979 (2019)
2. V. Celorrio, K. Bradley, O.J. Weber, S.R. Hall, and D.J. Fermín, ChemElectroChem, 1, 1667 (2014)
3. G.P. Wheeler and K.S. Choi, ACS Energy Lett., 2, 2378 (2017).
4. G.P. Wheeler, V.U. Baltazar, T.J. Smart, A. Radmilovic, Y. Ping, and K.-S. Choi, Chem. Mater., 31, 5890 (2019)
5. M.I. Díez-García, V. Celorrio, L. Calvillo, D. Tiwari, R. Gómez, and D.J. Fermín, Electrochim. Acta, 246, 365 (2017).
6. E. Freeman, S. Kumar, S.R. Thomas, H. Pickering, D.J. Fermín, S. Eslava, ChemElectroChem, 7, 1365 (2020)
7. D. Tiwari, D.J. Fermín, T.K. Chaudhuri, A. Ray J. Phys. Chem C, 119, 5872 (2015)
8. X. Sun, D. Tiwari and D. J. Fermín, J. Electrochem. Soc., 166, H764 (2019)
9. X. Sun, D. Tiwari and D. J. Fermín, ACS Appl. Mater. Inter., 12, 31486 (2020)

Metal chalcogenides define a large and important field of chemistry, and these materials continue to present appealing intellectual and scientific challenges. They are a large class with a broad set of chemical and physical properties involving diverse scientific phenomena and impact an astonishing variety of applications. The fields of science and technology impacted by chalcogenides continue to broaden. Some examples include nonlinear optics, thermoelectric energy conversion<!--[endif]---->, radiation detectors,<!--[endif]----><!--![endif]----><!--[endif]----> catalysis, topological insulators, science in two-dimensions, and unconventional superconductivity. Recently, the chemistry of these complex chalcogenides has witnessed the largest growth. We are in the middle of an impressive expansion in solid state chalcogenide chemistry with emphasis on materials with new compositions and structure types. In this regard, the development of novel synthetic methodologies is playing a major role in producing new materials, and this is a main objective in our program. We will highlight the general themes of synthesis science and structure-composition-property relationships with the following question being central: How does one develop the tools and concepts, both intellectual and experimental, to prepare new functional materials. There are several science drivers behind this research. Since crystal structure defines physical properties, of particular interest is learning to control structure dimensionality in complex systems. Another is learning how to incorporate two different metals in a single chalcogenide structure while avoiding phase separation. We will present relevant example in new materials with nonlinear optical properties and hard radiation and neutron detection capabilities and how they evolve as a function of composition and structure. <!--![endif]----><!--![endif]---->

Compounds in the structure type CaAl₂Si₂ have attracted great attention in recent years for their thermoelectric properties. In particular, MgMg₂Sb₂ and MgMg₂Bi₂ exhibit an impressive thermoelectric figure of merit zT due, in part, to their anomalously low thermal conductivity. In the present study, in situ high-pressure synchrotron X-ray diffraction was used to investigate the structure and bonding in binary and ternary AM₂X₂ compounds at pressures up to 50 GPa. Our results confirm prior predictions of isotropic in-plane and out-of-plane compressibility but reveal large disparities between the bond strength of the two distinct Mg sites. Using single-crystal diffraction, we show that the octahedral Mg–Sb bonds are significantly more compressible than the tetrahedral Mg–Sb bonds in MgMg₂Sb₂, which lends support to prior arguments that the weaker octahedral Mg bonds are responsible for the anomalous thermal properties of MgMg₂Sb₂ and MgMg₂Bi₂. Further, we report the discovery of a displacive and reversible phase transition (C2/m) in both MgMg₂Sb₂ and MgMg₂Bi₂ above 7.8 and 4.0 GPa, respectively. The transition to the high-pressure structure in these binary compounds involves a highly anisotropic volume collapse, in which the out-of-plane axis compresses significantly more than the in-plane axes. Further, we investigate the bond compressibility in ternary AM₂X₂ compounds using CaMg₂Sb₂ and YbMg₂Bi₂ to develop trends between bond compressibility and cation size and to study the stability of the CaAl₂Si₂ structure type.

As batteries are employed in larger numbers and for increasingly diverse applications, there is interest in electrode materials with improved safety, availability, and cost relative to commercial electrodes. Early transition metal oxides are one alternative material family showing promise, especially for high rate applications. However, we do not yet have a strong understanding of the role of composition, structure, and structural evolution with cycling for these materials, which tend to have large unit cells and complex structures.

To contribute to this fundamental understanding, we have studied the energy storage abilities and mechanism of complex niobate electrode materials KNb₃O₈ and NaNb₃O₈. Using ex situ and operando X-ray diffraction, complemented by nuclear magnetic resonance spectroscopy and first principles calculations, we identify local and average structure changes and relate them to cycling performance and properties, including electronic conductivity over a charge induced metal-insulator transition. We also reflect on the role of these results in establishing a more general understanding of the structure-property relationships of other early transition metal oxides, for energy storage and beyond. This understanding is a necessary step toward the selection and design of electrode materials for the quickly-evolving energy landscape and our fundamental understanding of complex oxide semiconductors.

Nitride semiconductors is an interesting class of materials studied for diverse properties. Simple binary nitride semiconductors with wurtzite crystal structure such as GaN have been long used in many practical applications ranging from optoelectronic to telecommunication. The structurally related ZnGeN2 or MgSnN2 derived by cation mutation from the parent binary compounds have recently attracted attention for their disorder-tunable properties. However, these ternary nitride materials are mostly limited to II-IV-N2 composition closely related to parent III-N compounds.
This invited presentation will focus on chemistry-inspired material design of unconventional ternary nitride semiconductors. First, chemical trends in crystal structure and thermodynamic stability of ternary nitrides containing Mg or Zn, and Zr, Nb, Mo transition metals will be presented. Second, the results will be shown for thin film synthesis and optoelectronic properties for some of these new ternary nitrides semiconductors beyond II-IV-N2 family, such as Zn2NbN3 and Mg3MoN4. Finally, kinetically controlled synthesis of metastable ternary nitrides will be discussed as one of the important next steps in the field.

Oxide and halide perovskites of the form ABX₃ are a thoroughly studied class of material. They display a great breadth of interesting properties, which make them suitable for a variety of application areas from high-temperature superconductors to photovoltaic absorbers. More recently, the possibility of forming stable nitride perovskites have caught the interest of several research groups. While these are fairly illusive, which can be understood in part due to the high oxidation states required on the A and B cations, the few that have been identified computationally and experimentally so far have shown interesting electronic properties.^1,2

In this study, we apply a computational screening workflow to further probe the composition space of nitride perovskite materials. By combining filters based on chemical heuristics from data-mined oxidation states³ we reduce the ABN₃ search space, were A and B are technologically relevant metals, from 3,906 to just 279 potentially feasible materials. We then carry out automated first-principles calculations to systematically investigate the stability of different octahedral tilting motifs in these materials. By mapping out the thermodynamic stabilities of the various tilts for each composition, we are able to provide the next level of resolution to the composition vs stability picture of this chemical space. Furthermore, we carry out phonon calculations on relevant subsections of the phase space to gain insight into dynamic stability and apply hybrid density functional theory (DFT) methods to determine accurate optoelectronic properties.

[1] R. Sarmiento-Pérez et al., Prediction of Stable Nitride Perovskites, Chemistry of Materials, 27 (2015)
[2] K. R. Tally et al., Synthesis of Ferroelectric LaWN₃ – The First Nitride Perovskite, arXiv preprint, arXiv:2001.00633 (2020)
[3] D. W. Davies et al., Materials Discovery by Chemical Analogy: Role of Oxidation States in Structure Prediction, Faraday Discussions, 211 (2018)

As the search for new semiconductors expands, new paradigms for property control are emerging. Varying stoichiometry, polymorph, and alloying have, along with doping, enabled property control in binary semiconductors. Ternary and multinary compounds have additional degrees of freedom, such as lattice site disorder, which may enable unparalleled property control. In II-IV-N₂ compounds, which are analogs of the III-N semiconductors, cation site disorder can reduce the bandgap without substantially impacting lattice parameter,¹ but several members of this class are underexplored due to the historical difficulty in synthesizing nitrides. MgSnN₂ is a II-IV-N₂ with a predicted ~2.3 eV bandgap and wurtzite-derived ground state structure, making it of interest for optoelectronic properties and integration with existing III-N compounds. Here, we report on one of the first syntheses of MgSnN₂ and explore control of cation disorder and polymorphism in this material.

We first demonstrate combinatorial radio-frequency co-sputtering of MgSnN₂ across a range of cation compositions and up to 500 °C.² The predicted wurtzite-type polymorph forms across the explored temperature range, while a rocksalt-type polymorph not captured by previous predictions is found as a secondary phase at high Mg compositions below 200 °C. This phase is substantially metastable (>70 meV/atom) compared to the wurtzite-type ground state, and is predicted to be a wide bandgap, high dielectric constant material. Synchrotron x-ray diffraction confirms that both wurtzite- and rocksalt-type phases are cation-disordered, consistent with the reduced optical absorption onset of < 2 eV for the wurtzite-type phase via spectroscopic ellipsometry. We also demonstrate the heteroepitaxial growth of mixed wurtzite/rocksalt MgSnN₂ on GaN at 400 °C, where the close effective lattice match between GaN and the rocksalt-derived phase promotes its formation outside of the previously mapped temperature and cation composition range.

The existence of two accessible polymorphs of MgSnN₂, in addition to the possibility of cation ordering may enable additional control of materials properties. We exploit the ability to nucleate both polymorphs of MgSnN₂ on GaN in ongoing work investigating cation ordering in the rocksalt-derived MgSnN₂ using post-growth annealing, and further discuss related polymorph and cation order phenomena in the analogous compounds MgSiN₂ and MgGeN₂. This work forms the basis for integrated materials control through both polymorph formation and lattice site disorder, bridging existing methods of manipulating semiconductor properties and emerging tools, while opening the door for a new set of semiconductors which can be integrated with well-established III-N compounds.

(1) Schnepf, R. R.; Cordell, J. J.; Tellekamp, M. B.; Melamed, C. L.; Greenaway, A. L.; Mis, A.; Brennecka, G. L.; Christensen, S.; Tucker, G. J.; Toberer, E. S.; Lany, S.; Tamboli, A. C. Utilizing Site Disorder in the Development of New Energy-Relevant Semiconductors. ACS Energy Lett. 2020, 5 (6), 2027–2041.

(2) Greenaway, A. L.; Loutris, A. L.; Heinselman, K. N.; Melamed, C. L.; Schnepf, R. R.; Tellekamp, M. B.; Woods-Robinson, R.; Sherbondy, R.; Bardgett, D.; Bauers, S.; Zakutayev, A.; Christensen, S. T.; Lany, S.; Tamboli, A. C. Combinatorial Synthesis of Magnesium Tin Nitride Semiconductors. J. Am. Chem. Soc. 2020, 142 (18), 8421–8430.

The binary Ta-N chemical system includes several compounds with notable prospects in microelectronics, solar energy harvesting, and catalysis. Among these, metallic TaN and semiconducting Ta₃N₅ have garnered significant interest, in part due to their synthetic accessibility. However, tantalum sesquinitride (Ta₂N₃) possesses an intermediate composition and largely unknown physical properties owing to its metastable nature. Herein, Ta₂N₃ is directly deposited by reactive magnetron sputtering and its optoelectronic properties are characterized. Combining these results with density functional theory provides insights into the critical role of oxygen in both synthesis and electronic structure. While the inclusion of oxygen in the process gas is critical to Ta₂N₃ formation, the resulting oxygen incorporation in structural vacancies drastically modifies the free electron concentration in the as-grown material, thus leading to a semiconducting character with a 1.9 eV bandgap. Reducing the oxygen impurity concentration via post-synthetic ammonia annealing increases the conductivity by seven orders of magnitude and yields the metallic characteristics of a degenerate semiconductor, consistent with theoretical predictions. Thus, this inverse oxygen doping approach – by which the carrier concentration is reduced by the oxygen impurity – offers a unique opportunity to tailor the optoelectronic properties of Ta₂N₃ for applications ranging from photochemical energy conversion to advanced photonics.

Many of the materials considered for photovoltaic applications suffer from one or both of the barring arguments for truly sustainable materials: They contain elements that are either toxic (e.g. Pb, Cd) and/or scarce (e.g. Te, In, Ga). The use of more than 1000 ppm Pb in electrical devices is, for instance, banned in the European Union¹ and most other countries. Zinc-group IV-nitrides are being considered as promising candidates for photovoltaic absorber materials, containing uniquely elements of low toxicity and low resource criticality.² They can be formally related to binary III-V nitrides by replacing the trivalent cations of the latter with equimolar amounts of divalent and tetravalent ions. Further to band gap tuning by alloying group IV elements – in analogy to cation alloying in III-V’s – it has been postulated based on DFT calculations that Zn-IV-N₂ compounds possess a second mechanism for bandgap tuning through cation disorder of the divalent and tetravalent species.³

The latter is unique to these ternary materials and marks the foundation of complex structure-property relationships in this class of materials, which need to be properly elucidated. Further, cation disorder is also triggered by oxygen substiting nitrogen on the cation sites and the effects mimic each other in some way.⁴ The latter is particularly important, since a degree of oxygen presence is hardly avoidable in most nitride syntheses. Understanding the different effects on the structural features of these materials is key to their performance as photovoltaic materials.

Herein, we present a systematic study of Zn_1+xGe_1-x(N_1-xO_x)₂ as model system for oxygen containing Zn-IV-N₂ materials. While the oxygen rich oxide nitrides appear as completely cation disordered and crystallise in the wurtzite type,⁴ oxygen poor compounds crystallise in the β-NaFeO₂-type that is also adopted by stoichiometric ZnGeN₂.⁵ Both crystal structures are closely related to each other with the latter being a subgroup of the wurtzite-type. The β-NaFeO₂-type, however, consists of two crystallographically independent cation positions that allow ordered cations, as well as a quantifiable degree of disorder. Since Zn²⁺ and Ge⁴⁺ are isoelectronic, they are virtually indistinguishable using X-ray diffraction and hence neutron diffraction is used for the reliable determination of disorder in this class of materials.

Combining chemical analyses and both, X-ray and neutron powder diffraction, we endeavour to disentangle the different phenomena leading to cation disorder and find evidence for oxygen-content related disorder as well as for cation disorder that does not relate to oxygen. However, the effects of oxygen and sample composition are largely predominant. Finally, we relate these structural findings with the optical bandgaps of the materials obtained through diffuse reflectance UV-VIS spectroscopy.

References
1 Restriction of the use of certain hazardous substances Directive (RoHS) 2011/65/EU.
2 P. Narang, S. Chen, N. C. Coronel, S. Gul, J. Yano, L.-W. Wang, N. S. Lewis, H. A. Atwater, Adv. Mater. 2014, 26, 1235-1241
3 D. Skachov, P. C. Quayle, K. K. Kash, W. R. L. Lamprecht, Phys. Rev. B 2016, 94, 205201.
4 J. Breternitz, Z. Y. Wang, A. Glibo, A. Franz, M. Tovar, S. Berendts, M. Lerch, S. Schorr, Phys. Status Solidi A 2019, 216, 1800885.
5 Z. Y. Wang, D. Fritsch, S. Berendts, M. Lerch, J. Breternitz, S. Schorr, submitted.

In the past five years, Mg₃Sb₂ and Mg₃Bi₂ alloys have emerged as exceptional room-temperature thermoelectric materials, threatening to overthrow the decades-long reign of Bi₂Te₃. The success of these compounds is thanks in large part to their surprisingly low lattice thermal conductivity, which is rarely observed in simple, lightweight compounds. This talk will explore the chemical and thermodynamic origins of the anomalously low thermal conductivity in the Mg₃Pn₂ system (Pn = Sb, Bi). Temperature-dependent measurements of the elastic moduli and first principles phonon calculations were used to investigate the bond strength, rate of softening, and mode Grüneisen parameters. Compared with other isostructural compounds, we find that both Mg₃Sb₂ and Mg₃Bi₂ have anomalously soft shear moduli and large mode Grüneisen parameters. We attribute this behavior primarily to the small size of the Mg cations: Mg is undersized with respect to the 6-fold octahedral coordination environment, leading to weak anharmonic interlayer bonding and high rates of Umklapp phonon-phonon scattering. This is corroborated by the phonon spectra obtained from inelastic neutron scattering of Mg₃Bi₂ and YbMg₂Bi₂ single crystals, which show significant softening of the acoustic phonons when Yb is replaced by the smaller Mg cation. In addition, we used in-situ high-pressure synchrotron X-ray diffraction to investigate the structure and bonding in Mg₃Sb₂ and Mg₃Bi₂ at pressures up to 50 GPa. By extracting the pressure-dependent volume change of the polyhedra using Mg₃Sb₂ single crystal diffraction data, we show that the octahedral Mg-Sb bonds are significantly more compressible than the tetrahedral Mg-Sb bonds, lending further support to our argument that the octahedrally-coordinated Mg is responsible for the anomalous thermal properties of the Mg₃Pn₂ system. Further, we report the discovery of a reversible high-pressure phase transition in Mg₃Sb₂ and Mg₃Bi₂ to a monoclinic structure at 7.8 GPa and 4.0 GPa, respectively.

The lattice thermal conductivity of a material above its Debye temperature should either decrease with increasing temperatures, as in a crystal, or remain independent of temperature, as in a glass. However, multiple literature reports of the Yb₁₄MgSb₁₁ crystal indicate its lattice thermal conductivity increases with increasing temperature well above its Debye temperature. Herein we study the thermal conductivity and electrical transport of the Yb₁₄Mg_1-xAl_xSb₁₁ solid solution and show this increase in lattice thermal conductivity can be attributed to an electronic effect arising from multi-band transport. We go on to show the effect multi-band transport has on thermal conductivity is pervasive in literature but is often misattributed to the thermal conductivity arising from phonon transport.

The wide bandgap layered perovskites A₅M₄O₁₅ (A = Ba, Sr; M = Ta, Nb) are shown do be very active in photocatalytic water splitting under UV-light, resulting from the additional reaction sites in the crystal structure.[1] The activity can be improved either by the formation of heterojunctions such as Ba₅Ta₄O₁₅-Ba₃Ta₅O₁₅ or Ba₅Ta₄O₁₅-Ba₃Ta₅O₁₅-BaTa₂O_6.[2,3] Preparation of nanofibers can additionally solve the problem of the large mismatch between the small charge carrier diffusion length and the much larger light penetration depth, resulting in higher photocatalytic activity.[4,5] Visible light activity can be gained by the formation of heterojunctions such as Ba₅Ta₄O₁₅-AgVO₃[6] and Ba₅Ta₄O₁₅-C₃N₄[7] and via ammonolysis[8].
We are combining the positive effects of the nanofiber structure and the ammonolysis by preparing (111) layered perovskites nanofibers of Ba₅Ta₄O₁₅ and Ba₅Nb₄O₁₅ and converting them into perovskite oxynitrides. The samples were characterized with XRD and the conversion degree was determined. SEM characterization was performed, and Kr physisorption measurements were done to determine the BET surface area before and after ammonolysis. UV-Vis spectroscopy as well as hydrogen evolution measurements in water/methanol were performed.
In first experiments, conversion degrees of nearly 100% were obtained, yielding in a decrease of the band from 4.0 eV down to 2.0 eV for the niobium compound and from 4.6 eV down to 1.9 eV for the tantalum compound; giving visible light absorption abilities. The surface area is slightly increased. Morphology changes will be discussed in detail.

[1] H. Otsuka et al., Chem. Lett. 2005, 34, 822.
[2] R. Marschall, J. Soldat, and M. Wark, Photochem. Photobiol. Sci., 2013, 12, 671.
[3] J. Soldat, R. Marschall, and M. Wark, Chem. Sci., 2014, 5, 3746.
[4] N. C. Hildebrandt, J. Soldat, and R. Marschall, Small 2015, 17, 2051.
[5] A. Bloesser and R. Marschall, ACS Appl. Energy Mater. 2018, 1, 2520.
[6] K. Wang, X. Wu, G. Zhang, et al., ACS Sustain. Chem. Eng., 2018, 6, 6682.
[7] E. Hua, G. Liu, G. Zhang, et al., Dalt. Trans., 2018, 47, 4360.
[8] A. Mukherji, C. Sun, S. C. Smith, et al., J. Phys. Chem. C, 2011, 115, 15674.

Materials that conduct ions have long been studied for a variety of applications from solid state lighting to energy conversion technologies. Room temperature ionic conductivity is especially useful for energy storage applications and notable examples conduct monovalent Li⁺. Active materials in the cathode and anode must support facile ion diffusion and high electronic conductivity while solid-state electrolytes require high ionic conductivity but low electronic conductivity. Although Li-based systems work very well and have revolutionized energy storage, the desire to lower cost and increase availability motivates next-generation technologies based on new mobile ions including divalent Ca²⁺, Mg²⁺, and Zn²⁺. We will discuss divalent ion conductivity in this context with a focus on Zn²⁺ conductivity in ZnPS₃. ZnPS₃ supports Zn²⁺ conductivity with unexpectedly low activation energies (~350 meV) enabled by the flexible [P₂S₆]^4- polyanion that distorts into the Van der Waals gap at the transition state.

The Yb₁₄MgSb₁₁ structural system is a record-breaking high temperature p-type material with applications in radioisotope thermoelectric generators. However, the single parabolic band model that has guided previous optimization efforts, pointing toward decreasing carrier concentration, do not take into consideration a second valence band that alters effective mass and electronic transport in the Yb₁₄MgSb₁₁ system. Here we conduct the first investigation to increase carrier concentration by Na doping Yb₁₄MgSb₁₁ based on an improved multiband model. To understand defect-controlled carrier concentration, we apply density functional theory (DFT) to investigate equilibrium phases, defect formation enthalpies and band diagrams. We found that the enthalpy of formation was more favorable for Na as an interstitial rather than a substitutional atom. Furthermore, experimental transport data on Yb_14-xNa_xMgSb₁₁ (x =0, 0.05, 0.25 ,1, 2, 3) were prepared by mall milling and hot-pressing procedures. We show that Na increases Seebeck and resistivity leading to the conclusion that Na acts as a +1 electron donor interstitial atom decreasing carrier concentration.

Advances in CdO donor doping have realized carrier concentrations exceeding 10²⁰ cm^-3 while maintaining carrier mobilities of up to 500 cm²/Vs that together enable new mid-IR plasmonic applications. Despite this differentiating property combination, the underlying mechanisms dictating these properties remain incompletely understood. Unlike traditional semiconductors, mobility does not scale monotonically with carrier concentration in CdO as initial increases with dopant incorporation eventually level off and ultimately decrease. To explain these observations, we vibrationally resolve the lattice changes occurring upon doping and assess them in light with observed trends in mobility.
Raman spectroscopy, pre-resonant with the CdO band-gap, examined lattice strain, disorder, and exciton-impurity/phonon interactions occurring as a function of carrier concentration. Mobility is found to be inversely proportional to both the amount of strain and disorder that change with doping consistent with explanations for dopant incorporation based on defect equilibria. Additionally, longitudinal optical phonon-plasmon coupled modes are identified and found to scatter by an impurity induced Fröhlich mechanism. Taken together, these coupling strengths also correlate linearly with the mobility, with decreased coupling associating with increased mobility. When viewed as a whole, the effects of strain, disorder and exciton-impurity/phonon interactions together dictate the carrier mobility, with phonon properties affecting the electronic properties and vice versa. These results have the potential to inspire new plasmonic materials which are engineered at the phonon level, revealing a new axis of device tunability.

Acknowledgements: Sandia National Laboratories is a multi-mission laboratory managed and operated by the 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 No. DE-NA0003525.

Sunlight-driven water splitting has received much attention as a means of large-scale renewable hydrogen production [1]. Both efficiency and scalability of water-splitting systems are essential factors for practical utilization of renewable solar hydrogen. Particulate photocatalyst systems do not involve any secure electric circuit and can be spread over wide areas by inexpensive processes potentially. Therefore, it is highly impactful to develop particulate photocatalysts and their reaction systems that efficiently split water.

The author’s group has studied various semiconducting oxides, (oxy)nitrides, and (oxy)chalcogenides as photocatalysts for water splitting [2]. The water splitting activity of a SrTiO₃ photocatalyst can be boosted by two orders of magnitude by doping Al [3]. The apparent quantum yield of overall water splitting using a SrTiO₃ photocatalyst has been improved to 95% in the near UV region via refining the photocatalyst and cocatalyst preparation [4]. This quantum efficiency is the highest yet reported, and indicates that particulate photocatalysts can drive the uphill overall water splitting reaction as efficiently as the photon-to-chemical conversion process in photosynthesis.

The author's group has also been developing panel reactors in view of large-scale applications [3]. A prototype panel reactor containing Al-doped SrTiO₃ photocatalyst sheets splits water and releases product hydrogen and oxygen gas bubbles at a rate corresponding to a solar-to-hydrogen energy conversion efficiency (STH) of 10% under intense UV illumination. A 1-m²-sized photocatalyst panel reactor splits water under natural sunlight irradiation without a significant loss of the intrinsic activity of the photocatalyst sheets. A solar hydrogen production system with a greater size (100 m²) was recently built and its performance and system characteristics are under investigation. Panel reactors can accommodate various kinds of photocatalyst sheets and are expected to be built using light and inexpensive materials, thus being ideal for large-scale solar hydrogen production from water.

It is essential to develop photocatalysts active under visible light irradiation for practical solar energy harvesting. Ta₃N₅ and Y₂Ti₂O₅S₂ photocatalysts show activity in overall water splitting via one-step excitation under visible light irradiation [5,6]. Particulate photocatalyst sheets split water into hydrogen and oxygen via two-step excitation, referred to as Z-scheme, efficiently regardless of the size. In particular, a photocatalyst sheet consisting of La- and Rh-codoped SrTiO₃ and Mo-doped BiVO₄ split water into hydrogen and oxygen O₂ via two-step excitation, referred to as Z-scheme, and exhibit STH exceeding 1.0% [7,8]. Some other (oxy)chalcogenides and (oxy)nitrides with longer absorption edge wavelengths are also applicable to Z-schematic photocatalyst sheets.

In my talk, the latest progress in the development of photocatalytic materials and their reaction systems will be presented.

[1] Hisatomi et al. Nat. Catal. 2019, 2, 387.
[2] Chen et al. Nat. Rev. Mater. 2017, 2, 17050.
[3] Goto et al. Joule 2018, 2, 509.
[4] Takata et al. Nature 2020, 581, 411.
[5] Wang et al. Nat. Catal. 2018, 1, 756.
[6] Wang et al. Nat. Mater. 2019, 18, 827.
[7] Wang et al. Nat. Mater. 2016, 15, 611.
[8] Wang et al. J. Am. Chem. Soc. 2017, 139, 1675.

The nature of conduction band minimum is totally different from that of valence band maximum. Thus, the materials design concept different from the conventional covalent type semiconductors are required for these materials. In this talk we review our progress in materials and device application:
Material Design and Example of Transparent Bipolar Semiconductors
Ambipolar wide gap oxide semiconductors were restricted to CuInO2[1]and SnO2[2]. We reported ambipolar materials ZrOS based on new design concept[3] and Cu3N applying non-conventional doping approach[4].
Ultrawide gap amorphous oxide semiconductors and their TFTs
Ionic amorphous oxide semiconductors represented by IGZO is now widely used as the switching TFTs for high performance LCD and large-sized OLED TVs.[5] Here, ultra-wide gap( >3.5eV) amorphous oxide semiconductors[6] such as a-Ga2O3 [7] and their TFT characteristics are reported.
Crystal-amorphous nanocomposite semiconductors and their LED application
Nano-sized ZnO embedded in amorphous ZnO-SiO2 has low work function by ~1eV than bulk ZnO keeping mobility of ~1 cm2/Vsand can form ohmic contacts with a wide range of electrode materials [8]. These features make it possible to boost the performance halide perovskite LEDs [9]and OLEDs[10].

[1] Yanagi et al. APL 121, 15(2001),[2]Nomura et al. Adv.Mat.23,3431(2011), [3]Arai et al. JACS 139,17175(2017), [4]Matsuzaki et al.Adv.Mat.31, 1801968(2018),[5]Hosono, Nat.Elect.1,428(2018),[6] Kim et al. a, NPG Asia Mater. 9, e359 (2017),[7] Kim et al. APL Mat. 7, 022501(2019), [8] Nakamura et al. Adv. Electr.Mat.4,1700352 (2018), [9] Sim et al. Appl.Phys. Rev.6, 031402(2019), [10] Hosono et al. PNAS,114,233(2017).

Recent decades have seen exciting explosions of research into new and lesser-studied semiconductors, including such broad categories as complex halides, nitrides, and chalcogenides, and layered and two-dimensional materials. Even so, silicon has further consolidated its position as the leading material for computing, solar energy conversion, and even for some optoelectronics. In light of this friendly but often overmatched competition with silicon, we will ask four of the world’s leading researchers what motivates them to continue work on new semiconductor materials.

Panelists will be Hideo Hosono, George Nolas, Shengbai Zhang and Andriy Zakutayev. Moderated by Rafael Jaramillo.