Why does "invisible gold" need arsenic?
Chris Palenik and coworkers (American Mineralogist, in press) found that
in Carlin-type and epithermal deposits, "invisible" gold is structurally
bound exclusively to arsenic-rich pyrite (arsenian pyrite). Recent advances
in electron microscopy techniques allow us to relate structurally bound gold
and gold nanoparticles to the arsenic-rich rim of arsenian pyrite particles.
However, the thermodynamic, electronic and structural interactions between Au
and As cannot be revealed from nanoscale experiments alone. To test the most
energetically favorable atomic configurations for gold and arsenic in pyrite,
M. Reich and colleagues (Goldschmidt Conference, 2004, abstract) performed
ab initio quantum mechanical calculations showing that a gold atom is stabilized
by about 1 electron volt when an arsenic atom is in its vicinity. Using quantum
mechanical calculations, we are beginning to understand why arsenic atoms are
necessary to incorporate one gold atom into the pyrite structure.
Why oxides and sulfides?
Some oxides and most sulfides are semiconducting and, therefore, the highest-occupied
(HOMO) and lowest-unoccupied (LUMO) orbitals are delocalized. Rosso and Becker
(Geochimica et Cosmochimica Acta, March 2003) demonstrated the extent of
such a LUMO orbital around a lead vacancy on a galena lead-sulfide surface.
Using quantum mechanical calculations, we are now able to show the electronic
relationship between two species that are some distance apart and explain the
thermodynamics of co-adsorption on mineral surfaces and/or co-incorporation
into their bulk. Surprisingly, the adsorbed oxygen atom is attracted by the
lead vacancy because of its peculiar electronic interaction through the near-surface
region of the semiconducting galena even though the laedvacancy (charge = -2)
and the negatively charged oxygen atom on the surface would repel each other
if just Coulomb interactions are considered.
Delocalization of orbitals also plays a role when iron as an oxidant can charge-polarize and spin-polarize corner sites somewhere in the vicinity of the adsorbed iron and subsequently mobilize hydroxyls to bond to such a corner site on a lead-sulfide surface. Another example of this effect is when a hydroxyl on a pyrite surface "pushes" electron density through a pyrite crystal toward the adsorption/oxidation site where an oxygen atom binds to a surface iron atom. This not only explains why pyrite oxidizes much faster in a hydrous environment but also shows that iron is the initial and sulfur the terminal electron donor. This type of co-reactivity also helps explain why pyrite oxidizes in patches.
Since these interactions work on the nanometer rather than the Ångstrom symbol scale, a "long-range" interaction influences the surface and bulk diffusion of atoms. Thus, diffusing particles "see" each other, greatly enhancing the formation of nanoparticles {emdash} for example, of noble metals in and on semiconducting minerals as shown by the formation of silver, gold and copper islands on molybdenite surfaces (Becker and colleagues, Geochimica et Cosmochimica Acta, special issue on sulfide and oxide surfaces, 2003).
Filling the proximity effect with life
These examples are just a few of the environmentally and geochemically important
reactions to which this scheme can be applied. Most of the evidence for these
co-reactivities over some distance in and on minerals is derived from quantum
mechanical calculations and atomic-scale observations. We are only beginning
to understand for what redox pairs and minerals this effect is important. What
reactions would not be possible without this effect and how it influences
rate laws that so far have only been determined for single adsorbed species
are still unknown. More fundamental quantum mechanical research will help explain
distance and directional dependencies, and experimental work at a larger scale
will be necessary to evaluate the importance of these long-range co-reactivities.
![]() |
Geotimes Home | AGI Home | Information Services | Geoscience Education | Public Policy | Programs | Publications | Careers ![]() |