Information for Prospective Students

Research in Energy Resources Engineering

Hydrocarbons in oil and gas reservoirs are deep underground in porous rock at high temperatures and pressures. Oil is either produced under its own pressure or water is injected to displace the oil. Even so, more than half the oil is never recovered--of the more than 400 billion barrels of oil discovered in the U.S., about 300 billion barrels cannot be recovered by conventional techniques such as water flooding. This imperfect recovery occurs because injected water flows preferentially through high permeability pathways, bypassing the large oil-bearing zones. Moreover, pore-sized ganglia of oil are trapped by surrounding water and are not recovered from the water-flooded regions.

Where should wells be drilled to ensure that the most oil is contacted? How can trapped oil be mobilized by the injection of hydrocarbon gas, carbon dioxide, steam or foaming agents? How will horizontal wells improve recovery? How can we make the best use of the data we have? The answers to these questions are of enormous economic importance. Finding the answers involves research that is a fascinating mix of fluid mechanics, thermodynamics, applied mathematics and geology, coupled with sound engineering principles.

The configuration of oil, gas and water in the interstitial spaces (about 10-100µm across) between rock grains and porous rock is controlled by surface tension with perturbative influences from buoyancy and viscous forces. These forces, together with the effects of phase behavior, influence the distribution of oil at the microscale, and hence the overall recovery. We tend to understand these processes by a combination of experiments in small rock samples and artificial porous media, such as bead packs and two dimensional etched-glass networks, phase behavior measurements, computer modeling and analytical theory. A knowledge of flow in porous media need not be applied exclusively to oil field problems--the transport of contaminants, such is spilled fuels and organic solvents underground, or the flow of radioactive material through fractured rock, are governed by the same physical principles.

Major flow pathways and reservoirs may extend over several miles and are determined by the connectivity of high permeability zones and major faults and fractures. The permeability may vary by orders of magnitude between different rock types. Thus, a measurement on a small core sample taken while drilling a well is unlikely to be representative of the whole field. However, it is possible to generate the statistical ensemble of possible permeability distributions in the reservoir that faithfully reflect the measured values and their possible spatial distribution. This technique, known as geostatistics, combines mathematical models with geological insight.

One cannot experiment with oil reservoirs. Decisions on the type of recovery process and the positioning of the wells have to be correct the first time. This requires extensive use of computer simulation. The average properties of flow in porous media are described by nonlinear differential equations, whose parameters are spatially variable. Their solution in a complex geological environment, combined with the intricate phase properties of crude oil and gas at high pressure, involves a number of mathematical challenges that make most gas dynamics or aeronautical engineering problems seem straightforward. We have developed novel gridding techniques, advanced numerical schemes and fast solution methods. Computations often require supercomputers or massively parallel machines. Many of these reservoir engineering techniques apply to environmental clean-up, as well as the recovery of geothermal steam and hot water from underground.

As oil and gas developments move to ever more remote and difficult locations, it becomes more important to maximize the recovery of the resource and minimize the cost. Several of our graduate students have undertaken research in optimization techniques, and have applied intricate and novel algorithms to the computation of optimum strategies.