ACS PRF | ACS
All e-Annual Reports
44494-GB2
Mantle Generation of Heavy Hydrocarbons
Although it is well established that the vast majority of Earth's hydrocarbon reservoirs are the byproduct of decay from once-living organisms, recent work has demonstrated an abiogenic origin for hydrocarbon deposits scattered around the globe. For example, Sherwood Lollar et al (2002) used isotopic fractionation to determine that methane and heavier hydrocarbons in the Kidd Creek Mine deposits of the Canadian Shield were abiogenically synthesized. Several studies have experimentally demonstrated reaction pathways likely to operate in Earth's crust capable of generating methane and heavier hydrocarbons (e.g. Horita and Berndt, 1999). We have previously shown (Scott et al., 2004) that such abiogenic methane-forming reactions proceed at mantle pressures as well, further supporting the notion that abiogenic hydrocarbon production may be much more prevalent, when one considers the bulk Earth, than previously appreciated. Our Type G Start Grant was funded to continue this work in high-pressure hydrocarbon synthesis, and here we report progress during the first year of funding.
Our first priority has been to acquire, assemble and test laboratory equipment to reliably produce the high-pressure and temperature conditions of Earth's mantle. Due to the importance of making in situ observations, we utilize Diamond Anvil Cells (DACs). Briefly, DACs use two gem-quality diamonds as pressure-generating anvils, with an approximately 200 - 500 micron diameter by 75 micron thick sample chamber constructed between them. Due to transparency across much of the electromagnetic spectrum, observations (e.g. visual, X-ray or spectroscopic) are made directly through the anvils while samples are subjected to high pressure and temperatures. High temperatures are produced either by focussing infrared lasers directly on the sample or heating the anvils with resistive wires. For resistively heated experiments, temperature is measured by thermocouples which are in direct contact with the anvils. We have acquired DACs suitable for both types of measurements and a temperature controller to regulate the resistively heated DAC. The temperature controller regulates the amount of current flowing through the resistive wires by monitoring thermocouples which are attached directly to the anvils.
Pressure is measured indirectly in DAC experiments; a common technique utilizes pressure-induced fluorescence shifts in well-studied "calibrant" materials with strong fluorescence, such as Al2O3-ruby (e.g. Mao et al 1978). We have constructed a laser-induced fluorescence spectrometer specifically for this purpose with several criteria in mind: user safety, fine laser focus and positioning, ease of use, reliability and versatility. This instrument was completed at the beginning of August and is working very well. We have also acquired an Electric Discharge Machining (EDM) device to prepare sample chambers. As described above, DAC sample chambers are typically less than 500 microns in diameter. They are constructed by drilling a hole through a metal foil that has been pre-compressed between the anvils; the EDM enables us to accurately position holes in the center of the anvil indentation.
We received synchrotron X-ray beamtime at HPCAT (a dedicated high-pressure beamline of the Advanced Photon Source at Argonne National Lab) in August and used our recently completed in-house DAC facilities to load and prepare samples. For this round of experiments we were particularly interested in the potential production of hydrocarbons heavier than methane. Whereas our 2004 measurements used a starting material of FeO, CaCO3 and H2O, we decided to increase the relative amount of carbon in the system by using native iron rather than FeO and assumed a hydrocarbon-forming reaction similar to 3 Fe + CaCO3 + 2 H2O = CH4 + CaO + Fe3O4.
Using our fluorescence spectrometer and EDM, we loaded two DACs with an Fe:Calcite molar ratio of 3:1 and abundant water to pressures of 7 and 10 GPa. Subsequently, we traveled to HPCAT and laser heated both samples to 2100 - 2400 K. Diffraction was collected both during and after laser heating at many locations within the sample chamber. As expected, Fe3O4 is the dominant crystalline phase in the products, but we note the presence of many additional diffraction lines; the additional phase(s) has not yet been identified, but we are optimistic that these diffraction results will definitively determine the fate of calcium in this system (an unresolved issue in our 2004 study).
We have also made spectroscopic measurements using the Raman spectrometer at HPCAT. Although we have not detected hydrocarbons heavier than methane, the evidence for methane formation is extremely strong. Whereas methane was often difficult to find in our 2004 study, we readily found intense C-H stretching in both experiments. Indeed, the intensity of the C-H stretching vibration produced in the experiment at 10 GPa exceeds that of O-H from the H2O -- this greatly exceeds the typical intensities observed in our previous study. As expected, the greater concentration of C in the system enhanced methane production, but we are struck by the apparent efficiency of this reaction at pressures equivalent to several hundred kilometers of depth in Earth.
