The arsenal of synthetic methods for solid-state chemistry includes the direct combination high-temperature approaches, synthesis from fluxes and melts, hydrothermal synthesis, and synthesis from solutions.

Because of the very low diffusion coefficients in solids (in the order of 10^-12 cm²sec^-1), the traditional solid-state synthesis requires high thermal activation. In this method, reactants are measured out in a specific ratio, ground together, pressed into a pellet, and heated to high temperatures.

Synthesis from fluxes can be viewed as slow cooling of a melt that can have a composition very different from the resulting crystalline phase. This technique is popular for crystal growth of oxides. Selection of a suitable metal as a flux medium is based on a number of criteria such as nonzero solubility of the reagents, low melting point, and inertness. The reagent solubility provides high reactivity while the low melting point allows for kinetic control over the reaction and eventual isolation of new phases. Removal of the flux is done in a number of ways such as chemical etching with various solutions or mechanical decanting/centrifuging at elevated temperature while it is still molten.

Hydrothermal syntheses involve chemical reactions in water above ambient temperature and pressure in a sealed or closed system and are a special type of chemical transport reaction that relies on liquid phase transport of reactants to nucleate formation of the desired product. Under autogenous conditions, water functions both as a pressure transmitting medium and as a solvent. In a sealed vessel, the vapor pressure of water increases as the temperature is raised above its normal boiling temperature, but below its critical temperature. The selected reaction temperature and the degree of fill, or the percent of the reaction vessel free volume that is filled with water at room temperature, determine the prevailing experimental pressure. When using water as a solvent the dielectric constant and viscosity are also important. These decrease with rising temperature and increase with rising pressure, the temperature effect predominating. Owing to the changes in the dielectric constant and viscosity of water, the increased temperature within a hydrothermal medium has a significant effect on the speciation, solubility and transport of solids.

If elevated temperatures alone are insufficient to dissolve the solid reagents, then a mineralizer can be added, whose ions form complexes with the solids and render them more soluble. These solubilizing reagents include acids, bases and alkali salts. We use one of the aforementioned types of mineralizers in almost every hydrothermal reaction we study. Our transparent conducting oxides (TCOs) are synthesized under basic conditions and our reactions yielding oxide fluorides use either hydrofluoric acid or an alkali fluoride as the mineralizer.

With standard PTFE Teflon-lined stainless steel pressure vessels, one experiment per vessel can be performed, and exploratory synthesis would require numerous vessels and consume a significant amount of time. To overcome this obstacle, we use semi-combinatorial exploration to rapidly gain understanding of a chemical reaction under hydrothermal conditions. Multiple FEP Teflon pouches are employed as individual reaction vessels placed within a single pressure vessel of significantly larger size. The sealed pouches are semipermeable to water and air under reaction conditions, but not to the solid materials. Thus, a specific reaction can easily be studied by changing one variable across all pouches without being concerned about possible differences in reaction conditions.

Noncentrosymmetric solids, or materials without any centers of symmetry, exhibit important dielectric and elastic properties, including ferro-, pyro- or piezoelectricity, ferroelasticity and optical activity. An example of the practical application of noncentrosymmetric materials can be found in the area of photonic technologies, which require high-performance nonlinear optical (NLO) materials with enhanced optical properties at the microscopic and macroscopic level. Polar distortions in metal centered octahedra are believed to be the origin of the nonlinear optical response in metal oxides. Octahedrally coordinated d⁰ transition metal cations in Groups 4 and 5 such as Ti⁴⁺ and Nb⁵⁺ are unstable in mixed metal oxides, and tend to undergo a distortion in which the transition metal cation moves away from the center of its octahedron. While the noncentrosymmetric crystal classes required for solids to exhibit these properties have been mathematically derived and are well known, the structural design principles that facilitate the supramolecular assembly of materials with specific chemical properties are less well defined.

In our work, first we are carrying out a series of syntheses aimed specifically at elucidating the fundamental behavior of [MO_xF_6-x]^n- anions in purely inorganic, solid-state environments. Second, we are also studying the effects of crystallizing the oxide fluoride anions with multiple, different alkali cations. Finally, we are exploring the use of HF(aq) as a fluoride source and mineralizer.

The out-of-center distortions present in [MO_xF_6-x]^n- anions result from both the inherent differences in the nature of M-O and M-F bonding and, to a lesser degree, from electrostatic interactions between the anions and the extended structure of the material. Here, the 2p valence orbitals of the oxide ligands mix with the transition metal d orbitals more strongly than the fluoride orbitals because the energy of the fluoride valence orbitals is extremely low. The result is an out-of-center movement of the central cation towards the oxide ligand(s) that is inherent to each [MO_xF_6-x]^n- anion. This "primary" distortion can be directed toward a corner, edge, or face of the octahedron, depending on the number of oxide ligands (one, two, or three, respectively) that are coordinated to the transition metal cation.

The less obvious, and more difficult to quantify, perturbation of an out-of-center distortion is dependent on the three dimensional structure (crystal packing). When a ligand in a [MO_xF_6-x]^n- anion interacts with the surrounding bond network, either covalently or through a hydrogen bond, the M-O or M-F bond is weakened slightly and its length increases. However, the oxide or fluoride ligand remains strongly bonded to the positively charged d⁰ central cation. The central cation responds to this distortion and forms shorter, stronger bonds with the other ligands in order to maintain its atomic valence. As a result, this "secondary" type of distortion is smaller in magnitude than a primary distortion.

In our work, we have synthesized via high-throughput hydrothermal conditions, inorganic solid state compounds with the [NbOF₅]^2- anion. Instead of replacing cations of a single alkali metal with [NbOF₅]^2- anions, however, mixed-metal cation systems have been employed to create multiple coordination environments around the oxide and fluoride ligands to aid in the crystallographic ordering of the anion. Two such compounds include KNaNbOF₅ and CsNaNbOF₅. Isolation of the [NbOF₅]^2- anion in these two compounds has allowed us to characterize primary and secondary distortions in a purely inorganic, solid-state environment.

Recently, we have synthesized mixed alkali cation compounds that contain the [MoO₂F₄]^2- and [WO₂F₄]^2- anions. These mixed metal cation systems further elucidate the effect of the bond network on the crystallographic ordering of the anion. The isostructural compounds are Rb₃Na(NbOF₅)₂·H₂O, Rb₃Na(MoO₂F₄)₂·H₂O, Rb₃Na(WO₂F₄)₂·H₂O and K₃Na(WO₂F₄)₂H₂O.

The (Ag₃MoO₃F₃)(Ag₃MoO₄)Cl compound was synthesized using aqueous HF as a mineralizer instead of pyridinium poly(hydrogen fluoride), which has been used in the majority of our previous research. Although it is a seemingly small change, the (Ag₃MoO₃F₃)(Ag₃MoO₄)Cl phase does not form with the more concentrated HF source. Ag₄V₂O₆F₂ is another example of a purely inorganic compound synthesized with an HF_(aq) mineralizer. This material is a promising cathode material for primary Li batteries.

The growth of thin films of potential p- (delafossites, etc.) and n-type (ZITO, GITO, etc) transparent conductors (TCs) is a critical step in realizing our ultimate goal of constructing transparent electronics or what are commonly referred to as invisible circuits. Researchers have employed numerous techniques (PLD, sputtering, etc.) for TC thin film growth, making it clear that growth conditions and treatment of transparent conducting films affect the crystal structure and therefore the band gap and conductivity. More specifically, work by Clark and Keszler (Inorg. Chem. 2001, 40, 1724) has shown that hydrothermal treatment of an amorphous film produces better crystallinity than high temperature annealing.

With the work of Clark and Keszler in mind, we have a devised a low temperature (< 210 °C) and low pressure (< 20 atm) hydrothermal technique to synthesize all the known copper CuBO₂ (B = Al, Sc, Cr, Mn, Fe, Co, Ga and Rh) and silver AgBO₂ (B = Al, Sc, Fe, Ga, Co, Ni, Rh, In and Tl) delafossite oxides in moderate to high yields. In contrast to two-step ion-exchange syntheses, or very high pressure syntheses, all the known silver delafossites have been synthesized by a convenient, practical and direct reaction for the first time, including several key compositions AgAlO₂, AgGaO₂, and AgScO₂, which have never synthesized in any practical (good yield) reaction before.

Expanding upon this work, we have started to explore the synthesis of other solid solutions containing different B-site cations. In the synthesis of delafossite-type oxides the solubility and amphoteric nature of the A- and B-site cations proved to be important factors in determining which delafossite-type oxides form. Looking at the Group 13 and Group 3 cations, their order of solubility (M) is as follows: Ga³⁺ (1), Al³⁺ (10^-1), Sc³⁺ (10^-4), and In³⁺ (10^-5). While we were unable to synthesize CuInO₂, the increased solubility of Ag⁺ over Cu⁺ (10^-2.5 M vs. 10^-4 M) allowed for the hydrothermal synthesis of AgInO₂, despite the low solubility of In³⁺. Similar to the synthesis of the simple delafossite-type oxides, initial research shows that the solubility of the A- and B-site cations play a critical role in product formation of solid solutions. At the current time, we have synthesized the following solid solutions: Ag(Ga,Al)O₂, Ag(Sc,In)O₂, and Cu(Ga,Al)O₂. It is apparent that a difference in B-site cation solubility ≥ 10^-3 renders the synthesis of that particular solid-solution inaccessible by our hydrothermal technique. These results are preliminary and the hope is that by altering the mineralizer concentration (pH), we will be able to expand upon this list.

Amorphous oxide semiconductors (AOS), a new class of TCOs, have recently gained attention as promising new channel materials in thin-film transistors (TFT). The amorphous structure allows the thin film deposition process to take place at room temperature and does not require a post-deposition annealing step. The transparency also affords the opportunity to fabricate transparent TFTs, which will lead the way to invisible electronics. Most importantly, AOSs demonstrate field-effect mobilities at least ten times higher than the currently used channel materials α-Si:H and pentacene. These high mobilities come from the unique band structure, and hence, atomic structure of the material.

We are studying the formation and structure of the TCO Zn-In-Sn-O (ZITO), which has shown high performance as a TFT channel as well as a highly conductive film. Amorphous ZITO is synthesized using solution chemistry and structural analysis is performed by using extended x-ray absorption fine structure spectroscopy (EXAFS), which can describe the local ordering around each metal center. Using EXAFS on top of other structural detection techniques, such as powder x-ray diffraction, electron diffraction and transmission electron microscopy, will give detailed insight into the structure of amorphous ZITO.

The focus of this research is to determine the surface structure of multi-component metal oxides and understand the reactions occurring on their surfaces. To reach this objective, single crystals with well-defined crystallographic orientations are needed. We study data from XPS, UV Raman, and TEM to determine connectivity and coordination environments. This information can be used to elucidate a reductive addition and oxidative elimination pathway.

Perovskite oxide surfaces have been used as heterogeneous catalysts and photocatalysts for reactions such as water splitting and photoreduction of carbon dioxide. Detailed knowledge of the surface structure at the atomic level is critical for a molecular level understanding of the reaction mechanisms involved, i.e. of the corresponding surface chemistry. This in turn allows for a rational design of a (photo)catalyst with an improved (photo)catalytic response. Much of the work on surface structure determination has been carried out on SrTiO₃ single crystals that are commercially available. In a parallel effort, the synthesis of SrTiO₃ powders consisting of nanoparticles of controlled size, shape and surface structure are being studied.

High quality single crystals with well-defined stoichiometries are needed to study the intrinsic properties of these catalysts. For fundamental surface studies the single crystals should be free of structural defects and impurity phases. To grow high quality single crystals, the traveling solvent zone method is used. This method requires precise compositions, high-density starting materials, and proper alignment, temperature, oxidizing/reducing atmosphere, and growth rate. Crystals of magnesium orthovanadate, several centimeters in length, were grown for the first time; these are significantly larger than crystals grown in our laboratory previously by conventional flux techniques. The single crystal growth of Mg₃(VO₄)₂ has been studied along with XPS, Raman, transmission electron microscopy, and adsorption studies of various gases (H₂, methyl radicals, and O₂). These studies probe the reductive addition (H₂ and methyl radicals) and the oxidative elimination (O₂) reactions occurring on the surface.

Our research on the systems MO-V₂O₅-MoO₃ (M = Mg²⁺, Zn²⁺ and Mn²⁺) systems revealed a new family of materials with the general formula M_2.5VMoO₈. X-ray single crystal diffraction determined that this series of compounds possesses a unique charge balance mechanism, in which variations in the V/Mo ratio are charge balanced by an appropriate increase or decrease in the concentration of the M²⁺ cation.

Our work has also focused on how trivalent metal oxides differ from divalent oxides in multicomponent vanadates and molybdates. Mg_2.5VMoO₈ shows good selectivity for the oxidative dehydrogenation of butane, Mg_2.5VMoO₈ and Zn_2.5VMoO₈ are likely to have similar properties. The M₂O₃-V₂O₅-MoO₃ (M=Fe³⁺, Cr³⁺) systems revealed new phases containing isolated [VMoO₇]^3- units (two corner-shared tetrahedra) in which one type of oxygen is bonded only to molybdenum forming a Mo=O double bond. These tetrahedral species are rare in the solid state. Raman Mo=O stretching frequencies were shown to be consistent with assignments made in previous surface studies of molybdate catalysts. The simple oxide α-MoO₃ also contains Mo-oxo double bonds, but in α-MoO₃ molybdeum adopts octahedral rather than tetrahedral coordination. The co-substitution of V⁵⁺ for Mo⁶⁺ and Fe³⁺ for Mg²⁺ in the quaternary vanadomolybdate results in no ion vacancy, which differs from the substitution of V⁵⁺ for Mo⁶⁺ in other ternary vanadomolybdates in which either cation or anion vacancy will be created. The present co-substitution mechanism gives rise to no ion vacancies, which differs from other substitution mechanisms observed in the ternary vanadomolybdates.

Implantable medical devices are used to address a variety of medical needs including nerve stimulation, drug delivery, bone growth, and cardiac management. One specific device, the internal cardioverter defibrillator (ICD), monitors the patient's heart in order to deliver therapy when the heart experiences tachycardia. For this, high power from a primary lithium battery needs to quickly charge two capacitors with as high as 30 - 40 J at 700 - 800 V which will be discharged into the heart. The present industry standard in ICDs is a primary lithium battery composed with a cathode of silver vanadium oxide Ag₂V₄O₁₁ (SVO) owing to its ability to provide a high current at high potential (above 3 V) associated to the reduction of silver. The synthesis of the new material Ag₄V₂O₆F₂ (SVOF) and the subsequent electrochemical characterization shows that the higher density of silver within the phase and the inclusion of the more electronegative fluoride provide about a 50% higher gravimetric capacity above 3 V and delivered at 300 mV greater than SVO. Owing to the higher potential (faster capacitor charge time) and higher silver density, there is significant commercial interest in using this material for lithium batteries in ICDs. The technical objective of this research is to improve upon the synthesis of SVOF toward smaller particles and scaled up quantities and to work with the interested industrial parties.

Since our discovery of the first silver vanadium oxide fluoride phase, our research has been focused on the synthesis of other silver transition metal oxide fluoride phases. Our reactions are usually in aqueous HF solution over a wide variety of temperatures in our hydrothermal pressure vessels. Upon discovery of new phases, we attempt to isolate single crystals good enough for single-crystal X-ray diffraction to solve the crystal structure. While we can experience the same O/F disorder as seen in our other work, completely ordered materials, like Ag₄V₂O₆F₂, are known. In addition, any new phase that is found will be tested as a cathode in a lithium battery.

Ni-YSZ cermets are commonly used in solid oxide fuel cell (SOFC) anodes because of their excellent electrochemical performance in hydrogen fuel. However, nickel is susceptible to sulfur poisoning and carbon coking, which are detrimental to anode performance. In order to avoid the challenges associated with Ni metal, several groups have studied conducting oxide materials for application as SOFC anodes. The most successful anodes, in terms of electrochemical performance, have been mixed oxygen-ion and electronically conducting oxides.

We are studying lanthanum strontium iron chromium oxide (LSFeCr) mixed with GDC as a potential anode and have evaluated it with two separate electrolytes, gallium doped ceria (GDC, Gd_0.1Ce_0.9O_2-δ) and La_0.9Sr_0.1Ga_0.8Mg_0.2O_3-δ (LSGM) with a La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ-GDC cathode. While LSFeCr with the GDC electrolyte showed promising behavior, results suggest that cation diffusiton occurs between the LSFeCr anode and LSGM electrolyte, which can be prevented by adding a GDC buffer layer.

Comparing the present anodes to La_0.8Sr_0.2CrO₃-GDC anodes, we show that the addition of iron to the structure greatly increased the power density. This arising from the much higher ionic conductivity of oxygen between the FeO₃ octahedra and FeO_2.5 pyramids, as well as the transport of the oxygen ions bridging FeO₃ and CrO₃ octahedra. Also attributed to better ionic conductivity is a decreased polarization.

Present and ongoing work in this area is addressing electrode poisoning, namely from an increase in sulfur owing to impurities in the fuel gas (e.g. H₂S). This is important owing to a possible wide variety of qualities of fuel and allowing fuel flexibility in real-world applications. In addition, we are analyzing how the anode ratio of LSFeCr/GDC affects SOFC performace.