Carrick M. Eggleston

The structure of hematite (α-Fe2O3) (001) surfaces in aqueous media: scanning tunneling microscopy and resonant tunneling calculations of coexisting O and Fe terminations

The iron oxide–water interface is of interest not only in geochemical and environmental processes, but also in fields ranging from corrosion to photocatalysis. The structure of α-Fe2O3 (001) surfaces is not fully understood, and questions have arisen recently concerning different terminations of (001) terraces; a so-called Fe-termination is expected, but under some conditions an O-termination may also be possible. Ultra-high vacuum (UHV) scanning tunneling microscope (STM) studies report evidence for an O-termination in coexistence with an Fe-termination, but other studies find no evidence for an O-termination. Molecular mechanics studies suggest that an O-termination should be possible in an aqueous environment. An O-termination could result from dissolution; if Fe atoms were to dissolve from an Fe-termination, an O-termination would presumably result (and vice-versa). We imaged hematite (001) surfaces in air and aqueous solution using STM. To aid interpretation of the images, we use a resonant tunneling model (RTM) parameterized using ab initio calculations. Our STM and RTM results are consistent with mixed O- and Fe-terminated (001) surfaces. For acid-etched surfaces we find evidence for a periodic (with wavelength of 2.2 ± 0.2 nm) arrangement of nominal O- and Fe-terminated domains. Two different borders between domains should occur, one in which the Fe-termination is high relative to the O-termination, and the reverse. The different domain borders have significantly different heights, consistent with RTM calculations. This agreement allows us to conclude that the Fe-termination is topographically high on most terraces. Surface domains are observed in aqueous solutions at the atomic scale, and appear to be very unreactive on tens-of-seconds time scales at pH 1.

Surface Complexation of Sulfate by Hematite Surfaces: FTIR and STM Observations

Abstract Sulfate adsorbed to hematite surfaces from aqueous solution is examined using Fourier transform infrared (FTIR) spectroscopy and scanning tunneling microscopy (STM). FTIR was carried out using an Attenuated Total Reflectance (ATR) element coated with a fixed layer of hematite particles; this configuration allowed in situ variation of pH and sulfate concentration. The FTIR results are consistent with an inner-sphere monodentate surface complex. On dried samples, sulfate may form bidentate or possibly monodentate bisulfate complexes. STM was applied to samples that were removed from solution and imaged in air, conditions corresponding to those of the dried samples in FTIR. The images show mobile adsorbates whose lifetimes were greater than 5 ms and less than 240 ms, times that bracket the average lifetimes of aqueous FeSO4+ complexes (∼50 ms). In addition, the images show pairs of bumps, in agreement with STM images of bisulfate adsorbed on Pt(111) electrode surfaces ( Funtikov et al 1995 ). Although the STM images do not provide chemical identification, they are consistent with imaging of adsorbed inner-sphere sulfate (STM is incapable of seeing outer-sphere adsorbates). Our results suggest that categorization of adsorbates into inner-sphere and outer-sphere on the basis of macroscopic adsorption information is perhaps oversimplistic. Instead, a spectrum of intermediate behaviors is likely. Adsorbates classed (macroscopically) as outer-sphere may be those for which a relatively small proportion of adsorbates are in inner-sphere complexes at any given time.

Dissolution kinetics of magnesite in acidic aqueous solution, a hydrothermal atomic force microscopy (HAFM) study: Step orientation and kink dynamics

The dissolution kinetics of features on the magnesite (104) surface were studied in aqueous solutions from pH 4.2 to 2 and at temperatures between 60 and 90°C by hydrothermal atomic force microscopy (HAFM). At pH = 4.2, HAFM images showed magnesite step orientations that are comparable to the step orientations on calcite. Similar to calcite (104), there is anisotropy in the step velocity, but the magnitude of the anisotropy is much greater for magnesite. Furthermore, below pH = 4.2, changes in the dominant step orientation were observed. These results are discussed in terms of a nearest neighbor kink dynamic model, and the associated kink dynamics were tested with kinetic Monte Carlo (KMC) simulations. The KMC results suggest that the kink dynamic model does not account for the experimental observations and that further details such as second-nearest neighbor interactions or surface/edge diffusion cannot be excluded from the model. The dominant step orientations at low pH also point toward mechanisms stabilizing steps along periodic bond chain directions.

A hydrothermal atomic force microscope for imaging in aqueous solution up to 150 °C

We present the design of a contact atomic force microscope (AFM) that can be used to image solid surfaces in aqueous solution up to 150 °C and 6 atm. The main features of this unique AFM are: (1) an inert gas pressurized microscope base containing stepper motor for coarse advance and the piezoelectric tube scanner; (2) a chemically inert membrane separating these parts from the fluid cell; (3) a titanium fluid cell with fluid inlet–outlet ports, a thermocouple port, and a sapphire optical window; (4) a resistively heated ceramic booster heater for the fluid cell to maintain the temperature of solutions sourced from a hydrothermal bomb; and (5) mass flow control. The design overcomes current limitations on the temperature and pressure range accessible to AFM imaging in aqueous solutions. Images taken at temperature and pressure are presented, demonstrating the unit-cell scale (<1 nm) vertical resolution of the AFM under hydrothermal conditions.

Acidic dissolution of plagioclase: In-situ observations by hydrothermal atomic force microscopy

Hydrothermal atomic force microscopy (HAFM) provides in situ access to the surfaces of dissolving crystals at temperatures above the ambient boiling point of water. Here, we applied HAFM to the (001) surfaces of labradorite and anorthite at temperatures up to 125°C. In HCl solutions (pH 2) we observed the formation of a rough and soft surface layer on both minerals. By applying high loading forces to the scanning tip, the soft layer can be removed and the underlying interface (between the fresh solid and the altered layer) can be observed. In this way, in situ information about the thickness of the altered layer on plagioclase and the morphology of the underlying interface can be obtained. On labradorite, the thickness of this layer does not exceed about 30 nm within the first 5 hr of exposure to acidic solution at 125°C, but on anorthite thicknesses of up to about 300 nm were observed. The uncovered interface on anorthite shows a nonuniform morphology and either appears rough in AFM images or shows a step-like pattern. On anorthite, etch pits spread underneath the altered layer. This suggests that material must be released and transported through the layer without obvious changes in morphology of the layer’s surface. Based on the rate of spreading of etch pits, the dissolution rate was calculated to be about 2 × 10−6 mol m−2 s−1 at 125°C. This value agrees reasonably well with literature data and supports the suggestion that dissolution mainly takes place underneath the altered layer and not on its surface.

Point of zero charge of a corundum-water interface probed with optical second harmonic generation (SHG) and atomic force microscopy (AFM): New approaches to oxide surface charge

Abstract The p H and ionic strength dependence of light generated at a corundum-solution interface by the nonlinear optical process of second harmonic generation (SHG) is reported. A point of zero salt effect occurs in the p H range 5 to 6. The p H and ionic strength dependence of the SHG is qualitatively consistent with a model describing SHG from a charged mineral/water interface from Ong et al. (1992) and Zhao et al. (1993a, 1993b) , but certain aspects of the model appear inadequate to describe the full range of our data. Atomic force microscopy (AFM) force-distance measurements, though imprecise, were consistent with a point of zero charge (p.z.c.) for the interface also in the p H range 5 to 6. The SHG (and AFM) results are different from expectation; the observed p.z.s.e. (and presumably also the p.z.c.) is considerably lower than the accepted point of zero charge of clean alumina powders ( p H 8–9.4; Parks, 1965; Sverjenksy and Sahai, 1996) . Although the reasons for this are unclear, SHG holds promise as a probe of oxide-water interfaces that is independent of interpretation of acid-base titration stoichiometry.

Toward New Uses for Hematite

rent). These back-of-the-envelope calculations suggest that receptor diffusion will have little effect at synapses with low to intermediate release probabilities and receptor occupancies. The larger effects of AMPA receptor diffusion on short-term depression observed by Heine et al. could arise from higher receptor occupancies due to glutamate release from multiple vesicles per contact, a higher fraction of mobile receptors, and/or slower recovery from desensitization at the synapse. In general, larger effects of receptor diffusion will occur at synapses where a large fraction of receptors desensitize, as expected for connections with high release probabilities and receptor occupancies (see the figure). Receptor diffusion could explain the lack of desensitization at such synapses (e.g., at the climbing fiber synapse) (6). Desensitization is prominent where the presence of glutamate in the synapse is prolonged (4,11) and where strongly desensitizing AMPA receptors are expressed (12). However, counteracting desensitization with diffusion also requires abundant extrasynaptic AMPA receptors, which are not present in all neurons (8,11). Although further studies are required to determine the contribution of rapid AMPA receptor diffusion to short-term depression at different synapses in the brain, the present study changes our concept of the postsynaptic density, from an array of receptors that is fixed on a time scale of minutes, to a highly dynamic structure with the potential to rapidly refresh itself from the surrounding sea of extrasynaptic receptors.

Dissolution kinetics of magnesite in acidic aqueous solution: a hydrothermal atomic force microscopy study assessing step kinetics and dissolution flux

Magnesite (104) dissolution kinetics were studied in acidic aqueous solutions (2.0 < pH < 4.2) at temperatures between 60 and 90°C by atomic force microscopy (AFM). Comparison of dissolution fluxes obtained by AFM and chemical methods revealed six to seven times larger dissolution fluxes obtained by chemical analysis. Corresponding empirical activation energies were found to be 74 ±22 kJ/mol and 41 ± 4 kJ/mol (at pH 4.2) for the AFM and chemical methods, respectively. The empirical reaction order with respect to proton concentration was 0.36 ± 0.13 and 0.47 ± 0.03 for AFM and chemical methods, respectively. These comparisons suggest that the two experimental measurement methods differ as a result of the different sampling length scales associated with the methods. Negligible changes in step dissolution velocity with changes in bulk pH were found, suggesting that the principal source of increasing dissolution flux with decreasing pH is an increase in step density. However, the observed stable step orientation, which is dependent on pH, suggests that more than one proton adsorption equilibrium should be used to describe the surface chemistry of magnesite in acidic solution.

Scanning tunneling microscopy of Cr(III) chemisorbed on α-Fe2O3 (001) surfaces from aqueous solution: Direct observation of surface mobility and clustering

We report scanning tunneling microscope (STM) observations of Cr(III) monomer and oligomer complexes chemisorbed to α-Fe2O3 (001) surfaces from aqueous solution. Monomers appear to prefer a particular surface site, one for each surface unit cell. This site is probably associated with an octahedral vacancy in the underlying oxide structure. Monomers were mobile; although “hops” from one site to a neighboring site were observed, most commonly adsorbates appeared upon or disappeared from particular surface sites. This probably represents adsorption-desorption coupled with outer-sphere surface diffusion. Activation energies for Cr(III) monomer motion estimated from the observed lifetimes of Cr complexes in surface sites are lower than those for ligand substitution in aqueous Cr(III) ions or its hydrolysis products. This may reflect an artifactual influence of the STM upon adsorbed Cr(III), but could also suggest that Cr(III) ligand exchange is not rate-limiting for Cr(III) adsorption-desorption or surface diffusion. A dimerization event was recorded, and larger surface clusters were observed. Clusters were immobile within the time-scale of our observations.

Photooxidation of Chloride By Oxide Minerals: Implications for Perchlorate On Mars

We show that highly oxidizing valence band holes, produced by ultraviolet (UV) illumination of naturally occurring semiconducting minerals, are capable of oxidizing chloride ion to perchlorate in aqueous solutions at higher rates than other known natural perchlorate production processes. Our results support an alternative to atmospheric reactions leading to the formation of high concentrations of perchlorate on Mars.