R. Alan May

Reactive Ballistic Deposition of Nanostructured Model Materials for Electrochemical Energy Conversion and Storage

Porous, high surface area materials have critical roles in applications including catalysis, photochemistry, and energy storage. In these fields, researchers have demonstrated that the nanometer-scale structure modifies mechanical, optical, and electrical properties of the material, greatly influencing its behavior and performance. Such complex chemical systems can involve several distinct processes occurring in series or parallel. Understanding the influence of size and structure on the properties of these materials requires techniques for producing clean, simple model systems. In the fields of photoelectrochemistry and lithium storage, for example, researchers need to evaluate the effects of changing the electrode structure of a single material or producing electrodes of many different candidate materials while maintaining a distinctly favorable morphology. In this Account, we introduce our studies of the formation and characterization of high surface area, porous thin films synthesized by a process called reactive ballistic deposition (RBD). RBD is a simple method that provides control of the morphology, porosity, and surface area of thin films by manipulating the angle at which a metal-vapor flux impinges on the substrate during deposition. This approach is largely independent of the identity of the deposited material and relies upon limited surface diffusion during synthesis, which enables the formation of kinetically trapped structures. Here, we review our results for the deposition of films from a number of semiconductive materials that are important for applications such as photoelectrochemical water oxidation and lithium ion storage. The use of RBD has enabled us to systematically control individual aspects of both the structure and composition of thin film electrodes in order to probe the effects of each on the performance of the material. We have evaluated the performance of several materials for potential use in these applications and have identified processes that limit their performance. Use of model systems, such as these, for fundamental studies or materials screening processes likely will prove useful in developing new high-performance electrodes.

Desorption Kinetics of Ar, Kr, Xe, N2, O2, CO, Methane, Ethane, and Propane from Graphene and Amorphous Solid Water Surfaces

The desorption kinetics for Ar, Kr, Xe, N2, O2, CO, methane, ethane, and propane from graphene-covered Pt(111) and amorphous solid water (ASW) surfaces are investigated using temperature-programmed desorption (TPD). The TPD spectra for all of the adsorbates from graphene have well-resolved first, second, third, and multilayer desorption peaks. The alignment of the leading edges is consistent the zero-order desorption for all of the adsorbates. An Arrhenius analysis is used to obtain desorption energies and prefactors for desorption from graphene for all of the adsorbates. In contrast, the leading desorption edges for the adsorbates from ASW do not align (for coverages < 2 ML). The nonalignment of TPD leading edges suggests that there are multiple desorption binding sites on the ASW surface. Inversion analysis is used to obtain the coverage dependent desorption energies and prefactors for desorption from ASW for all of the adsorbates.

Flow-Based Multiadsorbate Ellipsometric Porosimetry for the Characterization of Mesoporous Pt−TiO2 and Au−TiO2 Nanocomposites

Au and Pt nanoparticle distributions within hierarchically ordered mesoporous TiO2 were explored using a combination of techniques including ellipsometric porosimetry (EP) and X-ray photoelectron spectroscopy (XPS). EP studies were used to examine adsorbate-TiO2 interactions and the influence of adsorbate polarity upon adsorption isotherms for mesoporous TiO2 films with and without Pt and Au nanoparticles. In particular, methods are described for modeling EP data to estimate the surface area and porosity of mesoporous TiO2 films and for estimating the pore size distribution (PSD) directly from the ellipsometry parameters Psi and Delta when fitting parameters alone are unable to extract reliable optical constants from the ellipsometry data. This approach reveals that mesoporous TiO2 films of approximately 200 nm thickness and approximately 10 nm pore diameter can be loaded with 1.7 nm diameter Pt and 3.9 nm diameter Au nanoparticles up to 26 and 21 wt %, respectively. The BET surface area of a representative mesoporous TiO2 sample using toluene as the adsorbate was found to be 44 m2/g with a mean pore diameter of 8.8 nm. EP and XPS depth profiling experiments indicate that 1.7 nm diameter Pt nanoparticles are well dispersed through the mesoporous TiO2 film, while 3.9 nm diameter Au nanoparticles are concentrated at the top of the film, blocking a significant portion of the available TiO2 pore volume. UV irradiation of the TiO2 films indicates that adsorbate-TiO2 interactions and surface wetting effects can play a critical role in the resulting isotherm and in evaluation of PSD.

The release of trapped gases from amorphous solid water films. II. “Bottom-up” induced desorption pathways

In this (Paper II) and the preceding companion paper (Paper I; R. May, R. Smith, and B. Kay, J. Chem. Phys. 138, 104501 (2013)), we investigate the mechanisms for the release of trapped gases from underneath amorphous solid water (ASW) films. In Paper I, we focused on the low coverage regime where the release mechanism is controlled by crystallization-induced cracks formed in the ASW overlayer. In that regime, the results were largely independent of the particular gas underlayer. Here in Paper II, we focus on the high coverage regime where new desorption pathways become accessible prior to ASW crystallization. In contrast to the results for the low coverage regime (Paper I), the release mechanism is a function of the multilayer thickness and composition, displaying dramatically different behavior between Ar, Kr, Xe, CH4, N2, O2, and CO. Two primary desorption pathways are observed. The first occurs between 100 and 150 K and manifests itself as sharp, extremely narrow desorption peaks. Temperature programmed desorption is utilized to show that these abrupt desorption bursts are due to pressure induced structural failure of the ASW overlayer. The second pathway occurs at low temperature (typically <100 K) where broad desorption peaks are observed. Desorption through this pathway is attributed to diffusion through pores formed during ASW deposition. The extent of desorption and the line shape of the low temperature desorption peak are dependent on the substrate on which the gas underlayer is deposited. Angle dependent ballistic deposition of ASW is used to vary the porosity of the overlayer and strongly supports the hypothesis that the low temperature desorption pathway is due to porosity that is templated into the ASW overlayer by the underlayer during deposition.

Evaluation of Lithium Ion Insertion Reactivity via Electrochromic Diffraction-Based Imaging

A microscope-CCD setup has been developed as an analytical tool for the detection of diffraction from one-dimensional redox-active transition-metal oxide gratings prepared with a combination of microtransfer molding (microTM) and cathodic electrodeposition. The diffraction efficiencies (DE) of tungsten trioxide, WO3, and binary molybdenum-tungsten trioxide, Mo0.6W0.4O3, gratings were measured during Li+ insertion/deinsertion experiments performed with both cyclic voltammetry and chronoamperometry in 1 M LiClO4/propylene carbonate. The DE was evaluated in terms of the optical constants of the grating materials determined by spectroscopic ellipsometry (SE) measurements of Li+ insertion/deinsertion into unpatterned thin films. The effect of grating thickness and the amount of inserted charge on DE has been analyzed. The diffraction method is used to quantitatively estimate lithium ion diffusion coefficients of electrochemically active metal oxide gratings.

Probing Toluene and Ethylbenzene Stable Glass Formation Using Inert Gas Permeation

Inert gas permeation is used to investigate the formation of stable glasses of toluene and ethylbenzene. The effect of deposition temperature (T(dep)) on the kinetic stability of the vapor deposited glasses is determined using Kr desorption spectra from within sandwich layers of either toluene or ethylbenzene. The results for toluene show that the most stable glass is formed at T(dep) = 0.92 T(g), although glasses with a kinetic stability within 50% of the most stable glass were found with deposition temperatures from 0.85 to 0.95 T(g). Similar results were found for ethylbenzene, which formed its most stable glass at 0.91 T(g) and formed stable glasses from 0.81 to 0.96 T(g). These results are consistent with recent calorimetric studies and demonstrate that the inert gas permeation technique provides a direct method to observe the onset of molecular translation motion that accompanies the glass to supercooled liquid transition.

Mobility of supercooled liquid toluene, ethylbenzene, and benzene near their glass transition temperatures investigated using inert gas permeation.

We investigate the mobility of supercooled liquid toluene, ethylbenzene, and benzene near their respective glass transition temperatures (Tg). The permeation rate of Ar, Kr, and Xe through the supercooled liquid created when initially amorphous overlayers are heated above their glass transition temperature is used to determine the diffusivity. Amorphous benzene crystallizes at temperatures well below its Tg, and as a result, the inert gas underlayer remains trapped until the onset of benzene desorption. In contrast, for toluene and ethylbenzene the onset of inert gas permeation is observed at temperatues near Tg. The inert gas desorption peak temperature as a function of the heating rate and overlayer thickness is used to quantify the diffusivity of supercooled liquid toluene and ethylbenzene from 115 to 135 K. In this temperature range, diffusivities are found to vary across 5 orders of magnitude (∼10(-14) to 10(-9) cm(2)/s). The diffusivity data are compared to viscosity measurements and reveal a breakdown in the Stokes-Einstein relationship at low temperatures. However, the data are well fit by the fractional Stokes-Einstein equation with an exponent of 0.66. Efforts to determine the diffusivity of a mixture of benzene and ethylbenzene are detailed, and the effect of mixing these materials on benzene crystallization is explored using infrared spectroscopy.