LabNotes - April 2011
Unconventional Oil and Gas
Unconventional fossil energy resources consist of petroleum and natural gas that are either difficult to reach or challenging to extract. Examples include fine-grainedformations that contain oil and/or natural gas. These fine-grained source rocks generally have very low permeability, which for a long time made extraction uneconomical. Advances in directional drilling and hydraulic fracturing, to which NETL R&D contributed, were required before the resources could be recovered economically in commercial quantities (Figure 1).
Figure 1. Gas from a Marcellus Shale well in Greene County, PA is flared prior to a workover in the winter of 2010 (photo by Tom Mroz, DOE).
Current NETL research on unconventional fossil energy is focused in four areas: unconventional oil and gas resource characterization, environmental impact characterization, improved methods for enhanced oil recovery using carbon dioxide injection, and the physical behavior of crude oil at extreme depths and pressures, known as the equation of state. Each of these research areas is discussed below.
Although unconventional resources such as shale gas and shale oil are now being produced routinely using commercial techniques, many mysteries remain. The means by which hydrocarbons reside in and move through the pores of these fine-grained rocks is poorly understood, as is the response of the tiny pore structures and microscopic fractures to changes in pressure and fluids (Figure 2). For example, it is not known precisely how the gas is contained in an organic-rich, black shale. There is a free gas component in the natural fractures and pores, but there is also an adsorbed component of gas, which is chemically attached to the surfaces of organic materials and clay minerals, where it is held under a much higher density than free gas in the pores. How much gas is held by adsorption, and the conditions under which it might be produced are not well understood. Natural gas adsorbed in the shales is mostly methane. Other gases, such as carbon dioxide, adsorb more strongly than methane in other organic-rich materials like coal. It is possible that carbon dioxide may adsorb more strongly than methane in black shales as well, leading to the potential future use of these rock formations for carbon sequestration. Injection of carbon dioxide near the end of natural gas production could help maintain reservoir pressures to keep flowpaths open, and displace methane from adsorption sites, thereby enhancing the recovery of the remaining natural gas. If the injected carbon dioxide then remains adsorbed strongly onto organic and mineral matter in the shale, these formations could provide permanent sequestration sites to keep this potent greenhouse gas out of the atmosphere. A series of laboratory experiments with carbon dioxide and black shales is planned at NETL over the next few years to quantify and verify these behaviors.
Figure 2. Backscattered scanning electron microscope image of the Marcellus Shale, showing stacked orientation of clay flakes, flat structure of pores between the flakes, and a scattering of bright pyrite crystals near the bottom; the area photographed is about 40 microns wide (image by Dan Soeder, DOE).
Another poorly understood property of these fine-grained rocks is the interaction and movement of multiple fluid phases, such as gas, oil, and water in the pores. Shale pores are so small that liquids trapped by capillary forces can block the flow of gas. Most of the gas and oil-productive shales appear to be preferentially oil-wet, rather than water-wet, which means that water is repelled from the pores. This allows natural gas to flow fairly easily after hydraulic fracturing, because the water used in the fracturing process is easily displaced from the pores by gas pressure. Petroleum-based liquids do enter the pores quite readily, however, and the presence of these liquids in shale can inhibit gas production. In addition to methane, some shales produce light hydrocarbons such as propane, ethane, butane, and pentane that occur in a gaseous phase downhole, but condense into liquids after reaching the reduced pressures and temperatures on the surface. These light liquids are collectively known as condensate and are quite valuable. Once production has gone on for some time, however, the reduction in reservoir pressure may cause some of these hydrocarbons to become liquid downhole. This is called retrograde condensate, and these liquids can block shale pores to the point that virtually nothing is produced, either gas or liquid. Understanding the interaction of these phases at various levels of pressure, temperature, and relative saturation is important to engineering the efficient production and drawdown of these reservoirs.
Some Environmental Aspects of Unconventional Oil and Gas Extraction
A separate research program at NETL is investigating the environmental impacts of oil and gas shale drilling and production on air, water, landscapes, habitat and ecosystems; this research will be summarized in a future NETL LabNote.
However, in cooperation with that effort, the unconventional oil and gas program has investigated the potential for groundwater contamination from black shale drill cuttings, the movement of hydraulic fracture fluids underground, and the methods for forensic analysis of stray gas in water wells that may be related to nearby shale gas drilling.
One aspect of this research is a small pilot study being conducted in collaboration with Waynesburg University and a materials analysis lab called Separation Design Group to investigate the oxidation and leachate geochemistry of black shale drill cuttings. The issue is of concern because minerals in these cuttings are typically in a reduced or anoxic form, such as the iron sulfide shown in Figure 3. Organic materials in the shale have an affinity for various toxic metals and radionuclides. Because the organic material also produces gas, horizontal boreholes used to extract the gas are often targeted in the richest organic layers of the shale, producing cuttings that are unusually enriched in metals. The drill cuttings produced from horizontal boreholes thousands of feet long can exceed hundreds of tons, compared to less than ten tons from a vertical borehole that may only penetrate a hundred feet of shale. Oxidation and weathering of these cuttings at the surface may produce mobile species that are much more water-soluble than the reduced forms.
Figure 3. Sulfide minerals in an outcrop specimen of black, organic-rich Marcellus Shale; the golden layer is iron sulfide, or pyrite, sometimes known as “fool’s gold.” The scale on the ruler is in inches (photograph by Dan Soeder, DOE).
The mobility of toxic metals and radionuclides was recently examined in three samples of upper, middle and lower Marcellus Shale from an outcrop near Berkeley Springs, WV, an old shale drill core from Morgantown, WV, and fresh cuttings from an active drill rig near Waynesburg, PA. The basic rock properties were assumed to be comparable across the three sites and the outcrop was assumed to have been exposed to the weather for thousands of years, so that nearly all of the leachable minerals should have been removed. The core was assumed to have the leachable minerals intact, although in oxidized forms since it had been kept dry but exposed to the air for 30 years. The cuttings were assumed to have all the leachable minerals present in a less-mobile, reduced state. Lab analyses included controlled atmosphere oxidation, and toxicity analysis following a leaching procedure (USEPA 1311). Radiation levels were also measured for alpha and beta decay. Results showed that trace metals above the minimum reporting limit in a composite shale sample included As, Ba, Cd, Co, Cr, Cu, Li, Mn, Mo, Ni, P, Pb, Sn, Sr, V, Zn, and Zr. The lowest concentrations were in the outcrop samples, as expected. The highest concentrations were in the core sample, which also contained the highest amount of organic material. It is not known if the high concentrations in the core were due to the oxidation state of the metals, or simply because the core contained more metals to start with as a result of the higher organic content. Future analyses on additional samples may resolve this uncertainty.
NETL researchers also plan to perform a field experiment on a hydraulically fractured, horizontal shale gas well to determine the fate of the injected water deep underground. Groundwater and soil gases will be monitored for a chemical tracer added to the fracturing fluid prior to injection. Monitoring will be concentrated along faults and other fractures that might provide conduits for fracture fluid to migrate upward. After several months, a vertical well will be drilled down through the production zone to sample formation water from every rock unit where it can be obtained. Analysis of the water samples for the tracer material will provide a definitive answer to where the “frac” fluid migrates underground. Researchers intend to try to have the well installed beneath the NETL site in Morgantown, where Marcellus Shale is known to exist. Ideally, the experiment should be run at several different sites on several different types of wells to provide a larger database of parameters.
Finally, NETL researchers plan to begin investigating the issue of “stray gas,” i.e., natural gas that appears in domestic water wells. Sometimes this gas appears to be associated with local drilling activity, and other times not. The circumstances and origin of the gas are very unclear. NETL scientists will work in cooperation with state researchers, the U.S. Environmental Protection Agency, and other experts to perform forensic analysis of stray gas, and ascertain the origin and transport processes that brought it into someone’s kitchen sink.
Enhanced Oil Recovery using CO2
The injection of carbon dioxide into a “depleted” oil reservoir can serve two purposes. First, the CO2 acts to re-pressurize the reservoir, driving the residual oil toward production wells. Second, if the CO2 can be made to remain behind in the reservoir, this separates or sequesters it from the atmosphere, eliminating it as a greenhouse gas and prevents it from contributing to global climate change.
Although this sounds good in theory, in practice there are several difficulties, not the least of which stems from the fact that a significant viscosity difference exists between the CO2 and the oil. Using a low viscosity fluid like CO2 to push a higher viscosity fluid like oil results in a phenomenon called “fingering.” It can be observed between air and honey in a honey jar by simply turning over a partially filled and capped jar of honey. The lower viscosity fluid breaks through the thicker liquid in long, narrow bubbles, instead of pushing it uniformly in one direction. The problem has been addressed in the past by placing a slug of saltwater or brine between the CO2 and the oil to act as a viscosity buffer, but this is not very effective.
NETL has been pursuing two research options to try to improve the underground mobility of CO2 in an oil reservoir. The first option is to increase the viscosity of the CO2 itself so that it can better push against the oil. The second option is to develop an improved viscosity buffer using CO2-based suds or foams created from residual brine that is already present in the reservoir. Either of these techniques could result in a dramatic increase in the rate of oil recovery, improving both the economics of oil recovery as well as the amount of oil recovered from domestic reservoirs undergoing CO2 flooding.
The design of direct CO2 thickeners seeks surfactant compounds that are CO2-soluble, water-insoluble, and oil-insoluble, which will form a thermodynamically stable, transparent, micellar solution when combined with dense CO2. The direct thickening surfactants must be designed to form long, cylindrical, viscosity-enhancing micelles in dense CO2.
The second technique employs slightly CO2-soluble, very water-soluble, “foam-forming” surfactants capable of generating low mobility, CO2-in-brine emulsions or foams within the limestone or sandstone formation. In this process, as the low viscosity, transparent, CO2-surfactant solution is injected into the formation, the surfactant will partition (dissolve) into the brine phase, where it can then stabilize thin, continuous lamellae that bridge across micron-scale pore throats while compartmentalizing droplets of high pressure CO2. Because the targeted reservoirs contain both oil and brine, there is already enough of an aqueous phase in the reservoir to generate mobility control foams if a suitable CO2 soluble surfactant can be identified. The ability to dissolve surfactant in the CO2 would reduce or eliminate the need for alternate injections of brine or aqueous surfactant solutions. In addition, NETL has developed a multiphase model that can simulate the experimental observation and ultimately predict the phenomena observed in the field.
The first surfactant thickener for CO2 was successfully designed, based on a divalent cation and two sets of highly fluorinated twin tails. NETL has also successfully designed the first CO2 surfactant thickener with a monovalent Na+ cation and a twin tail, one tail being hydrocarbon-based, the other a fluorocarbon. These are both illustrated below. Future research will attempt to make completely non-fluorinated surfactants that can thicken CO2. NETL continues to be the only research group in the world making progress in this area.
Figure 4. The effect of surfactants on micelle aggregate structure in CO2 has been investigated by high-pressure small-angle-neutron-scattering (HP-SANS) and high-pressure falling cylinder viscometry. Three counterions, Na+, CO2+, and Ni2+ of the partially fluorinated, di-chain anionic surfactant bis(1H,1H,5Hoctafluoro-n-pentyl) sodium sulfosuccinate (di-HCF4) have been studied. Results from HP-SANS and viscosity indicate that the Na+ version of the surfactant forms compact essentially spherical aggregates while the CO2+ and Ni2+ salts appear to form extended rod-shaped aggregates. Both are illustrated in this figure.
Equation of State
Oil companies are drilling extremely deep wells to produce oil and natural gas from porous rock formations that are several miles below the seafloor of the Gulf of Mexico. The temperature and pressure conditions associated with these “ultradeep” formations can reach as high as 500 °F and 35,000 psi. Again, extracting oil and natural gas from such formations was commercially impossible until relatively recently, and is still considered unconventional. NETL is conducting research on how best to drill at these depths in its new Advanced Drilling Research Laboratory, which will be reported here sometime later this year, and also how to accurately predict the density and viscosity of hydrocarbons under such extreme conditions so that the amount of recoverable petroleum in these formations and ease of production can be determined. This is a difficult task, because little data exist on the density or viscosity of hydrocarbons at these high temperatures and pressures, and current mathematical models cannot be applied with confidence. However, given this uncertainty, oil companies are naturally reluctant to attempt production at many potentially productive sites.
NETL’s equation of state research is attempting to predict the behavior of petroleum liquids under the high pressure and temperature conditions encountered at these great depths. We are measuring hydrocarbon density and viscosity at temperatures ranging from 50 to 250°C, and pressures up to 280 MP. We will use the data to develop accurate models for both of these important fluid properties.
For density measurements, a non-sampling variable volume high-temperature, high-pressure cell will be used. The internal volume of the cell is determined using a rod with a magnetic end piece connected to the piston. As the piston moves in response to a volumetric change for a given pressure and temperature, the magnetic end piece travels through a linear variable differential transformer (located outside the high temperature zone), which precisely tracks the location of the internal floating piston.
A rolling ball viscometer will be used to measure viscosity. This consists of a non-sampling, variable-volume, high temperature/high pressure cell, as shown in Figure 5. A sample is added in the cell at the horizontal position, where it is heated and pressurized to the desired conditions. The cell is then tilted to a particular angle, and the ball starts rolling down. The time taken by the ball to pass each observation window is noted and correlated to the viscosity of the fluids.
Figure 5. An exploded view of a high-pressure, high-temperature variable-volume rolling ball viscometer; small sapphire windows will allow us to determine the ball’s position as it rolls through the tilted cell.
A successful model will allow more confident predictions of the amount of petroleum that can be recovered, and the rate at which it can be produced from porous rock at great depths. The modeling is also expected to account for the behavior of natural saltwater or brine present along with petroleum in the pores of these deep underground formations. Data from a brine study might provide useful modeling tools for another technology: the potential sequestration of CO2 in deep, hot, high-pressure saline aquifers. These models could have the potential to predict the mutual solubility of CO2 and saltwater in both the CO2-rich phase and the brine-rich phase.
NETL plans to develop a database from the literature and new measurements of the viscosity and density of pure components and mixtures of CO2, hydrocarbons and water, and to use this data to develop new equation of state (EOS) correlations for density and viscosity. These mathematical tools will also be applicable to individual components like pentane or hexane, and multi-component reservoir fluids such as crude oil or natural gas liquids.