Basic Research Needs for Geosciences - Energetics Meetings and ...

GRAND CHALLENGE: COMPUTATIONAL THERMODYNAMICS OF COMPLEX FLUIDS AND SOLIDSclustering, microcrystallization, defect structure, and other non-ideal properties. Of course, solidsunder extreme conditions of radiative flux or high temperatures or pressures can developnumerous internal defects. These systems are at best only at local equilibrium. Without a morerefined understanding of these materials on the atomic scale, it is difficult to reduce this behaviorto an EOS representation as required for the analysis of subsurface problems.Local thermodynamic EOS for colloids and surface adsorption. Typically colloidal particlessuspended in aqueous solutions contain hundreds of thousands of atoms. For systems this largethe possibility of developing a finite system thermodynamic description should be investigated.As in infinite solid solution systems, the effects of clustering, local structure, and surfacecompositional variation are difficult to quantify. For these systems the large ratio of surface areato bulk volume should be included in a thermodynamic treatment. However, recent calculationsof the stability of atmospheric aerosols using models based on bulk thermodynamics have shownremarkable agreement with available measurements (Moller and Weare, unpublished results).The chemical interaction of the surface of aqueous colloidal particles with solutes and solventspecies (e.g., sorption, dehydration, surface reactions, etc.) may be simulated with the firstprinciplesmethods discussed above. Because of the complexity of the chemistry in the surfaceregion, many particles in the surface and interface must be included in the simulation. Whilesuch simulations are today limited to 500 or so particles, the processor scaling issues that arediscussed above as limitations are less of a problem if the system size is increased. This meansthat with the availability of many more processors with next generation computers, it will bepossible to treat on the order of 1000 particles using present software. Other ways to expand thesize of the system without loss of accuracy, such as the QM/MM approach discussed above,should be explored. In addition there are ways to upscale time to provide rate information usingfree energy simulations and transition state theory (Voter et al. 2002). Extensions of suchmethods should be explored to treat problems such as colloidal coarsening and surface structure(including local ordering) and chemical reactions at the interface.The state of species adsorbed on solid and colloidal surfaces and their chemistry is a problem ofgreat importance to solute transport, mineral formation and dissolution, oxidation and reductionof solute species, etc. To treat these systems, quasi-thermodynamic models of surfacecomplexation have been invoked on an ad hoc basis (Dzombak and Morel 1990). These modelswork remarkably well when there are sufficient data for parameterization. With the developmentof the new surface probes discussed above, a much more precise, atomic-level view of thesurface interface region is becoming available. This is again a fertile region for the application offirst-principles methods of simulation using the QM/MM methods described above. As in theother areas in which these methods might be applied, the objective of these calculations is thedevelopment of an atom-based structural and dynamic view of the interface region (effects oflocal structure and defect structure as adsorption sites, formation of surface islands; seeFigure 23). This level of understanding would be used to develop more accuratephenomenological models for the interface region.Data collection vs. molecular simulation for EOS development. It is important to emphasize thatfor reliable application to real problems, EOS must be parameterized from the largest data setspossible in intensive parameter regions close to the region of application. This is because highaccuracy, theoretically-based EOS with good extrapolation properties are not available for evenBasic Research Needs for Geosciences: Facilitating 21 st Century Energy Systems 77