Stanford University

News Service



CONTACT: Stanford University News Service (415) 723-2558


STANFORD - In the Beverly Hillbillies, Jed Clampett struck "bubblin' crude" by inadvertently shooting into an oil seep. Life rarely is so easy.

Non-fiction prospectors have to drill to depths of as much as 15,000 feet, and the oil they find usually flows through reservoirs at no more than a foot per day.

Stanford University petroleum engineers Tom Hewett, John Fayers and Khalid Aziz hope to improve the efficiency of oil production by better understanding the flow of oil in horizontally drilled wells, a technology that has become increasingly popular over the past five years.

They have submitted a proposal to the Department of Energy to develop a computer simulation of flow in these wells. Oil companies could use the simulation to decide which locations and configuration of wells would maximize production in an oil field.

Traditionally, producers have drilled oil wells straight downward to reach a reservoir. They "complete" the well with a perforated metal pipe placed vertically through the reservoir to drain the oil. The problem is that most oil deposits are relatively flat.

"If you look at road cuts where you see exposed geology, you see bedding - the sediments have been laid down over time, so the layering is generally horizontal," Hewett said. "If you have a horizontal geologic zone, and you punch a vertical well through, you just access a small portion of it. You then have to drain large distances to the well bore, which causes the flow to be slower."

Companies try to get around the problems by placing multiple wells and injecting water into some to push the oil toward others. But each additional well can cost more than $1 million.

Now, producers increasingly are trying horizontal wells, a method pioneered by the Soviet Union in the 1950s and pursued most actively by the French company Elf Aquitaine.

In a horizontal well, the drill curves forward as it goes down, producing a well shaped like a cross-section of a mixing bowl - a curve that flattens out to a horizontal line on the bottom.

"The drillers have gained the capability of placing long horizontal sections to within plus or minus 50 feet, even at depths of 10,000 to 12,000 feet," Hewett said . "The [drill] motor is shaped at an angle, so when it drills forward it always travels at an angle to the drill pipe. It can be steered by rotating the pipe."

As of the beginning of October, 3,500 horizontal wells had been drilled in the United States. Enthusiasts predict that, internationally, 50 percent of wells will be horizontal within 10 years.

"There is no major company that is not actively trying this out," Hewett said.

Horizontal wells can produce oil four to five times more quickly than vertical wells because they overcome the problem of draining oil through long distances of rock. The segment of the well that is in direct contact with, and drains, the oil may be as long as 5,000 feet.

But while most of the technical difficulties in drilling horizontal wells have been overcome, "a good engineering understanding of exactly how a horizontal well operates - and how to predict how it will operate before you drill it - has lagged," Hewett said.

"In any particular geological region, there are always new surprises about what you encounter. Typically, the first horizontal well costs two or three times as much as a vertical well, but after drilling four or five of them, you may get that down to 10 or 15 percent more than a vertical well."

To cut that cost to the point where increased production makes it worthwhile, horizontal wells must be placed properly. That's where the proposed new computer simulation would come in.

"Currently the simulations are crude," Hewett said. "[Engineers] just did a quick modification of the way they model vertical wells, without accounting for the fact that flow in horizontal wells is a very different phenomenon."

To build the new simulation, the Stanford team must study flows in a physical model.

"We probably will work with a major oil company that has an existing facility that we can use, a 'surface flow loop,' " he said. "You pump fluids through the pipe, and flow fluids into perforations along its length to see how oil and water enter the pipe and segregate or flow together."

The team also will incorporate data on rock properties from actual wells into their model.

Using the computer simulation, the team will "think of more innovative ways of using the horizontal well geometry to more completely access and drain a reservoir - for example, putting horizontal wells along the boundary and pushing the oil to the center." Hewett said. "Once we get the modeling capability, we can try out some of these ideas without having to drill the holes and spend money."


This story was written by Peter Chen, a science writing intern at the Stanford News Service.


This is an archived release.

This release is not available in any other form. Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images, some of which may be available to you online. Direct your request by EMail to

© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300. Terms of Use | Copyright Complaints