Inside this Issue

• Global Tsunami Model
• Alaska Weather Modeling
• Geochemical Interfaces
• VRcussion
• Clusters at ARSC
• Intern Summer Research Program
• Visualizing Frost Boils
• Permafrost Visualization

Dr. Tom Trainor , University of Alaska Fairbanks
Dr. Anne Chaka , National Institute of Standards and Technology

Story by Lorien Nettleton

Whether it’s for tracing the environmental legacy of abandoned mines, examining why ten thousand arsenic poisoning cases have been caused from contaminated water in Bangladesh, or understanding the biogeochemical cycle of trace metals, studying the chemistry of metals in surface and groundwater as they make their way into the food chain requires an understanding of how the contaminants interact with surfaces of minerals. Naturally occurring mineral phases can act as both a sink (through precipitation and adsorption reactions) and a source (through de-sorption and dissolution reactions) of metals in natural waters. Developing a fundamental understanding of the interaction between aqueous metals and mineral surfaces starts with deciphering the molecular-scale details of what happens at the mineral-water interface.

A group of researchers representing several universities and national institutions are spearheading a cooperative effort to establish the groundwork for understanding mineral surface structures and their interactions with aqueous fluids. Dr. Tom Trainor, UAF Assistant Professor of Chemistry; UAF Chemistry students Kristen Williams-Jimenez, Vanessa Ritchie and Kunaljeet Tanwar; UAF postdoctoral associates Dr. Sarah Petitto and Dr. Cynthia Lo; along with Dr. Anne Chaka, Theoretical Chemist at the National Institute of Standards and Technology (NIST); Dr. Peter Eng and postdoctoral associate Sanjit Ghose, Physicists at The University of Chicago; and Dr. Gordon Brown Jr. and Juyoung Ha, Geochemists at Stanford University, are using collaborative modeling and experimental validation to obtain detailed molecular and nanoscale information that can help provide a structural and mechanistic understanding of how the mineral surface structures are modified by the interaction with water. The researchers are investigating how these hydrated mineral surfaces in turn bond with aqueous metals. While the ultimate goal of using the recent understanding of mineral surface structures to add mechanistic restraints to field-scale metal transport models is a long way off, the development of experimental and theoretical methods for studying mineral-fluid interface processes is advancing at a rapid pace.

The group’s current work is focused on examining iron oxide and iron hydroxide mineral surfaces. Iron oxides are ubiquitous, commonly found as fine-grain sediments or coatings on larger minerals in soils and sediments, and as suspended colloids in the water column. The typically small particle-size results in a large surface area of these materials being exposed to aqueous solutions. These phases also tend to be highly reactive with respect to various metals and organic compounds, and therefore are an important substrate for controlling the fate and mobility of contaminants in aquatic systems.

The study of iron oxides is being tackled from two parallel directions. Trainor, working at both the Natural Sciences building at University of Alaska Fairbanks’ West Ridge and at the Advanced Photon Source, Argonne National Laboratory, is executing an experimental approach to determine the molecular-scale structure and reactivity of the mineral surfaces. The second approach to the study, led by Anne Chaka and Cynthia Lo at the NIST Physical and Chemical Properties Division in Gaithersburg, Maryland, is based on high-level theoretical calculations to predict surface structure and the thermodynamics of mineral fluid interactions.

The experimental investigation of the structure at the mineral-fluid interface has only recently become feasible due to the advent of highly brilliant synchrotron x-ray sources, such as the Advanced Photon Source at Argonne National Laboratory.

“The main problem is the presence of liquid water,” says Trainor. “Most surface science techniques are based on electron or soft x-ray spectroscopic techniques that require ultra-high vacuum environments to probe surface structure and composition. To see through a layer of liquid water to probe the solid-fluid interface requires a technique that can penetrate through a thick water film, yet still be sensitive to the relatively small number of atoms present at the interface. Hard x-ray scattering methods are the ideal solution to this problem, but they require enormous x-ray flux, only available at synchrotron radiation laboratories.”

Using a surface-sensitive x-ray scattering technique called Crystal Truncation Rod Diffraction, the precise positions of atoms at the surface of a hematite crystal in contact with water can be identified. The experiments are carried out in an environmental cell in which the chemical conditions of the aqueous fluid can be controlled to simulate conditions in typical natural aqueous environments. This allows the researchers to study how changes in the water chemistry impact the surface structure, and how changes in surface structure impact important reactions such as the sorption or de-sorption of aqueous metals.

Making Modeling Available Worldwide

The theoretical component of the study consists of predicting the surface structure and energetics of interfacial reactions in systems identical to those that Trainor is experimenting on. Starting with the structure of the crystal, Chaka uses a method called density functional theory to predict ab initio surface structures and the thermodynamic properties of the mineral-water interface. These calculations are run on Cerebro, a Sun Microsystems cluster at ARSC, as well as clusters available at NIST. The use of multiple high performance computing clusters is critical for Chaka to be able to simulate the wide range of variables that can impact surface structure and reactivity in aqueous systems. The computational results are then compared to the results Trainor has arrived at through the experimental x-ray diffraction work.

Doing the experimental and computational work in parallel allows Chaka and Trainor to evaluate the accuracy of the model. The excellent results obtained thus far have given the researchers confidence that the theoretical models are producing realistic results. While the surface structure can be experimentally assessed, physical investigation of the energetics of interfacial reactions, especially across a broad range of soluble chemical variables, is a different challenge. If the theory is predicting the structure, the researchers are confident that it will also reveal the energetics. With confidence in the model, the team can go on to predict the structure and energetics of a wide range of environmentally relevant mineral-water interface systems at a much more rapid pace than could be done experimentally.

“Surface modeling theory has dramatically improved in the last few years, in part due to the increase in computing power and the available resources of high-performance computing systems,” says Chaka. “The recent advances in computational modeling theory wouldn’t be possible without the processing resources now available. With better machines, we are able to look at larger surfaces, examine the role of surface defects and predict how lead, chromium, arsenic, antimony, mercury, etc., bind to a surface under a variety of chemical conditions.”

The ability to make predictions such as these will ultimately allow for improved assessment of whether these metals will tend to associate with mineral surfaces, or remain in the solution phase where they can be easily transported and are more available for uptake by organisms.

The next step in the researchers study is to expand the investigation to a wider variety of systems and explore the chemical systematics of interfacial structure and thermodynamics. At the same time they plan to increase the level of complexity of their model systems to better reflect the complex milieu of chemical species typically present at the mineral-water interface in natural systems.

“The ultimate goal is to make predictions of chemical structure and thermodynamics that can be incorporated into field-scale reactive transport codes,” says Trainor. “But to develop robust predictions, especially for very complex interfacial systems, requires that we test the theory with solid experimental results.”

Sources of Support

This work is supported by the National Science Foundation Nanoscale Interdisciplinary Research Team Grant BES-0404400, the National Science Foundation Environmental Molecular Sciences Institute Grant CHE-0431425, The American Chemical Society Petroleum Research Fund, and the University of Alaska Fairbanks Arctic Region Supercomputer Center.