Here is the full text of the post I referred to. I didn't realise that you needed to sign into the group to even read what's there, but I'm not surprised that some applications were rejected - Alex runs a tidy group there.

Alex posted:-

For those who do not have online access to Physics Today, below is
the text of the article with the figures elided:

=======================================

Physics in Oil Exploration

Petroleum engineers use a variety of physical techniques to find and
exploit petroleum reservoirs in increasingly remote and complicated
geological environments.

Brian Clark and Robert Kleinberg

Perhaps the most memorable movie scene depicting the oil industry is
from the 1956 movie Giant, in which a gushing well sprays James Dean
with oil. The oil surging from the well turns Dean's character from a
poor Texas ranch hand into a millionaire. The dramatic image of oil
spraying from a well is so deeply imprinted in the public
consciousness that it appears even now in TV documentaries on energy.
It is a striking but inaccurate portrayal that reinforces the
public's misunderstanding of how one finds oil--drill until it spurts
out of the ground. In fact, this approach to finding oil has not been
used since the blowout preventer was introduced in the 1930s.
So how does one find oil without producing gushers? One possibility
is to examine pulverized rock cuttings for traces of oil. However,
accurately determining the depth from which the cuttings originated
or how much oil they contain may be difficult. Another approach is to
recover large pieces of rock taken with a specially designed coring
drill, but that is a very expensive operation.

In 1927, a small geophysical research company founded by two French
brothers developed a better approach. Conrad and Marcel Schlumberger,
physicist and engineer respectively, demonstrated that electrical
resistivity measurements could be made in a well and that the
readings were different for different geological layers. Further
investigation showed that formations saturated with oil had higher
resistivities than ones saturated with saline water, so a high
resistivity reading could indicate an oil-bearing formation. This
discovery was so useful that, by 1935, 70 teams of engineers with
resistivity-measuring equipment were deployed to oil-producing
regions worldwide. Such was the birth of the "well-logging" industry,
named for the strip of paper--the log--on which are recorded
measurements as a function of depth beneath the surface.1

Nowadays, oil and natural gas are located with the help of
sophisticated physical measurements of subsurface formations. These
measurements use electromagnetic fields and waves, acoustic waves,
neutron scattering, gamma-ray radiation, nuclear magnetic resonance
(NMR), infrared spectroscopy, and pressure and temperature sensors--
all in a narrow (15- to 50-centimeter diameter) hole, thousands of
meters deep, where the pressure and temperature are high and caustic
fluids are present. No one measurement is adequate to define the
structure and properties of an oil reservoir, and typically many
measurements are made simultaneously by combinations of instruments.
Significant advances have been made in instrumentation, modeling, and
inversion techniques: only a few can be highlighted here. Using
physics, the modern petroleum engineer can find smaller petroleum
reservoirs in more remote, complex, and difficult geological settings
than ever before.

Growing energy needs
A common misconception is that alternate energy sources will soon
replace hydrocarbons. But, according to the US Department of Energy,
20 years from now, hydrocarbons will satisfy a greater percentage of
the world's energy demand than they do today.2 As a consequence,
there is a need to develop new technologies for finding and producing
oil and natural gas. The DOE estimates that the world will need 59%
more energy in 2020 than in 1999. This projected increase is due
almost entirely to population growth and industrialization in
developing nations. Between 1999 and 2020, oil production will have
to increase by 60%, and natural gas production by 92%, to satisfy the
world's growing energy demands. Together, oil and natural gas
supplied 63% of the world's energy in 1999; they are projected to
supply 68% by 2020. (For details, see the table and the figure on
page 41 of this issue.)

Of perhaps equal significance is the uneven geographic distribution
of oil and gas reserves. For example, in 2000, two-thirds of oil
reserves were in the Middle East compared to 2% in the US.3 In the
future, the Middle Eastern nations' share of the oil market will be
even larger than it is today.

Increased energy demands and the asymmetric distribution of reserves
present daunting challenges, especially given the world's aging
reservoirs and price fluctuations that tend to inhibit long-term
investment. There is some debate as to when the production of
conventional, relatively cheap oil will peak and begin to decrease:
Some predict it will happen within a decade.4 As oil becomes more
expensive and difficult to find, new technologies become ever more
essential.

Petrophysics
Petroleum is a mixture of gaseous, liquid, and solid hydrocarbons.
The word derives from the Latin for rock "petra" and oil "oleum."
Petroleum is not located in underground caverns, but exists in the
microscopic pores of sedimentary rocks. Petroleum-bearing rocks, such
as those shown in figure 1, are heterogeneous and their origin,
composition, and microgeometry vary widely. Many interesting physical
quantities, including hydraulic permeability, resistivity, and grain
size span several orders of magnitude.5 (See the article by Po-zen
Wong, "The Statistical Physics of Sedimentary Rock," Physics Today,
December 1988, page 24*.)

Determining the relationships between macroscopic properties and the
microscopic structure of rocks and their components has been
difficult. Despite decades of study, researchers' understanding is
generally limited to empirical correlations based on laboratory
measurements. Only recently have researchers been able to calculate a
few of the macroscopic physical properties of rocks from first
principles.6

Figure 1
One common type of reservoir rock, depicted in figures 1a and 1b, is
quartz sandstone. Quartz grains, eroded by wind or water from
existing rock, are transported to a sandbar, river channel, or beach,
where they are buried and ultimately cemented together into an
underground rock formation. A typical sandstone reservoir might have
quartz grains 0.05-2 millimeters in diameter and a porosity (volume
fraction of pore space) of 0.1-0.3. A wide distribution of grain
sizes reduces the porosity, as does cementation or the inclusion of
clay minerals in the pores.

Since petroleum can only occupy the pore space, determining the
formation porosity is of primary importance. A number of techniques
including neutron and gamma-ray scattering, acoustic propagation, and
magnetic resonance have been developed to measure porosity. A widely
used well-logging instrument exploits Compton scattering. Gamma rays
emitted from a cesium-137 source are scattered by the rock formation
before reaching scintillation detectors located 5 to 30 cm away in
the borehole. Attenuation increases with electron density. The
electron density contrast between pore liquids on one hand and quartz
on the other makes the computation of porosity from attenuation
straightforward in many cases.

Carbonate (typically CaCO3) is another type of porous rock that can
hold petroleum. Carbonate rocks often contain fossils, shells, and
microscopic exoskeletons such as corals, making the microgeometry
much more varied than that of quartz sandstone, as one can see by
comparing figures 1b and 1c. Determining the mineral content of a
rock formation is useful in a variety of ways. For example, it
defines the electron density of the grain space, a key parameter of
the gamma-ray porosity measurement. A number of techniques are
sensitive to mineralogy.7 These include spectroscopy of gamma rays
arising from the inelastic scattering and capture of neutrons, as
illustrated in figure 2; spectroscopy of naturally occurring gamma
rays; and the observation of photoelectric absorption of gamma rays.

Figure 2
Water is always present in the pores of sedimentary rocks, which are
formed by rivers, lagoons, lakes, and oceans. Water found well below
Earth's surface is generally much more saline than seawater and
therefore electrically conductive. Oil and gas are insulators, so a
hydrocarbon-bearing rock formation has a relatively high electrical
resistivity. Sedimentary rocks can have resistivities ranging from
0.1 ohm-m to greater than 20 000 ohm-m depending on their porosity,
the proportions of hydrocarbons and water, and the salinity of the
water.

Many techniques have been developed for measuring resistivity. The
three most common methods used today are based on focused electrode
arrays, electromagnetic induction and electromagnetic propagation.
Electrical anisotropy, the invasion of drilling fluids into the rock
formation, the presence of geological bed boundaries, laminations
with sizes smaller than the spatial resolution of the measurement,
and other factors all complicate the determination of resistivity.
Solving Maxwell's equations in complex formation geometries,
developing efficient inversion techniques, and designing new hardware
remain very active research areas. With new techniques, it is
possible to measure formation resistivities of 100 ohm-m through
steel casings having resistivities of 2 % 10- 7 ohm-m.8 Such
measurements are used to search for bypassed oil in old fields and to
detect water replacing oil outside the well casing during production.

Electrical resistivity does not distinguish between oil and natural
gas, but neutron scattering does. Neutrons produced either by an
americium-beryllium source (4 MeV) or from an accelerator-induced
deuterium-tritium reaction (14 MeV) reach thermal energies primarily
by elastic scattering from hydrogen atoms. As the density of hydrogen
atoms increases, thermalized neutrons have a greater chance of being
absorbed, and so the count rates in thermal neutron detectors located
in the borehole 10-70 cm from the source decrease. While water and
oil have roughly the same hydrogen atom density, gas has a much lower
density, so neutron scattering differentiates between oil and gas.
Common rock solids such as quartz and limestone do not scatter
neutrons strongly, but interstitial water in clay does, thereby
interfering with the measurement.

Hydraulic permeability is another key physical feature of a reservoir
rock. The larger the hydraulic permeability, the easier it is to
extract petroleum from the formation--obviously an important
consideration. For simple porous media, including some quartz
sandstones, the hydraulic permeability is approximately proportional
to the square of pore size (the length scale associated with the
smallest dimension of the pore). Pore sizes can range from 0.1 to 100
micrometers. If there is clay in the pore space, the permeability
will be significantly reduced.

Petroleum geology
Petroleum accumulates in geological structures that are porous,
permeable, and capped by an impermeable barrier. Natural gas and oil
are less dense than water, and migrate upward through faults,
fractures, and permeable formations until they are prevented from
rising farther. Petroleum traps have permeability barriers on their
sides as well as above, and may be structural (for example, folds or
faults) or stratigraphic (for example, river channels overlaid with
shale).

Box 1
Satellite images and surface measurements of Earth's magnetic and
gravitational fields are used to locate favorable areas for petroleum
traps, but seismic reflection surveys are the most important
technique for locating potential reservoirs.9 Acoustic energy in the
bandwidth from 10 to a few hundred Hertz is transmitted into the
earth, and waves reflected from subsurface structures are collected
by large arrays of surface sensors. The collected data yield three-
dimensional images mapping a volume that may extend many kilometers
deep and cover an area of hundreds of square kilometers. Image
resolution, however, is limited by the acoustic wavelength to
approximately 10 meters. Series of seismic surveys taken over time
(called four-dimensional seismic surveys) show how reservoirs evolve.
Such surveys may be used to locate bypassed pockets of oil for later
access by additional drilling. As discussed in box 1, 4D seismic
surveys can also be used for environmental monitoring.

Once a potential petroleum trap has been identified, a well must be
drilled to determine if the structure actually contains hydrocarbons.
Hydrocarbon-bearing zones vary in thickness from a few centimeters to
hundreds of meters. The thickest reservoirs, which tend to be in the
Middle East, are very economical to exploit. Most reservoirs,
however, are 1-10 m thick, though with much greater lateral extent.
Producing oil or gas from such thin, pancake-shaped formations has
not been particularly efficient; recovery efficiencies of 30% or
lower are quite common.

Various factors control recovery efficiency, one being reservoir
geometry. Often distinct layers of gas, oil, and water, stratified by
density, are present in a single permeable formation. Once brought to
the surface, oil is easily transported and is a versatile material
for fuel or chemical refining. It is therefore more profitable to
produce than gas, which provides less energy per liter extracted and
requires a pipeline or chilled tanker for transportation. Highly
saline water is considered a hazardous waste, and therefore must be
properly disposed of if brought to the surface. Hence, energy
companies generally try to produce oil while leaving gas and water in
place. Steel casing is cemented into the borehole to isolate gas,
oil, and water from each other. Shaped explosive charges similar to
those used in armor-piercing ordnance perforate the casing in the oil
layer and allow oil to flow into the well. Initially, mostly oil is
produced. But as the pressure in the oil layer decreases near the
well, water and gas begin to move into the low-pressure zone.
Eventually, more distant oil cannot reach the well, and water, gas,
or both, are produced instead. The presence of gas and water layers
in a formation thus limits the oil drainage to a circular area
surrounding the well.

Microscopic processes also limit recovery. For example, capillary
forces between water and rock can effectively close off small throats
in a rock formation (such as that seen in figure 1b) so that oil
cannot pass through them. Oil in large pores may be extracted, but
oil in smaller pores is blocked; eventually only water is produced.

Drilling operations
Wells for fresh drinking water have been drilled to depths of a few
hundred feet for hundreds of years. In 1859, the Pennsylvania Rock
Oil Co of Connecticut drilled the first oil well, in Titusville,
Pennsylvania, and struck the pay zone at 61 feet. The site was chosen
based on surface oil seeps long known to Native Americans. Since that
first well, almost every corner of the continental US has been
explored, as have many other countries. The search for new petroleum
reservoirs has now moved into deep offshore waters worldwide. Rigs in
the Gulf of Mexico are drilling in water depths approaching 3 km, and
then as much as another 10 km into the earth, where pressures can
reach 200 MPa (30 000 pounds per square inch or 2000 atmospheres) and
temperatures exceed 175°C. Capital and operating expenses of
deepwater drilling rigs can reach several hundred thousand dollars
per day, so every effort is made to ensure efficient, safe
operations.

One of the most surprising developments is that wells are now being
drilled horizontally.10 An excellent illustration of so-called
extended-reach horizontal drilling is in the British Petroleum (BP)
Wytch Farm field, located in an environmentally sensitive area on the
southern coast of England. Part of this field is onshore and has been
exploited using vertical wells. However, a large portion of the field
extends several kilometers offshore. An artificial island could have
been built for drilling and production, but that would have been
costly and environmentally controversial. Instead, extended-reach
wells were drilled from existing, secluded onshore facilities. Using
recent advances in drilling and logging technology, BP saved $180
million while minimizing visual impact. Figure 3 shows the profile of
a recent Wytch Farm well.

Figure 3
The drilling of extended-reach wells has many economic and
environmental advantages. A vertical well can produce hydrocarbons
only from a circular area of limited extent. A horizontal well may
have to penetrate kilometers to reach hydrocarbons, but once it has
reached the pay zone, increasing the drainage area is very cost
effective. The ability to tap petroleum resources within a several-
kilometer radius allows drilling and production facilities to be
concentrated in one small area. Such a concentration is especially
advantageous when drilling from an offshore platform, a very
expensive installation, or in environmentally sensitive areas, such
as Alaska's North Slope.

The mechanical technology that allows engineers to drill great
lateral distances while precisely steering the drill bit is evolving
rapidly, but discussion of this topic is beyond the scope of this
article. Instead, we focus on the key measurement technologies
required to hit a small target zone and to maintain the well bore in
a thin hydrocarbon layer.

Until the 1980s, wells were drilled blindly: Directly measuring
properties of rock formations was impossible while the drill bit was
in the hole. To locate hydrocarbons, it was necessary to remove the
drilling assembly from the borehole and to lower well-logging
equipment into place. This approach is suitable for vertical wells,
but not for extended-reach or horizontal wells. Geological layers are
rarely truly horizontal: Generally they are dipping or folded, have
faults (as seen in figure 3), or are pinched off. Hence, keeping the
well bore in a hydrocarbon layer requires continuous measurements of
formation properties and fluid saturations.

The advent of microelectronics brought sophisticated electronics and
sensors small enough to fit within the drilling assembly itself. The
first measurements-while-drilling (MWD) system contained
magnetometers to determine hole direction and accelerometers to
determine hole inclination--information necessary for directing the
drill bit.

MWD systems have to be entirely self-contained. They derive power
from batteries or turbines; have sensors, electronics, acquisition
systems, microprocessors, and memory; perform data processing; and
have unique telemetry systems. There are no electrical connections to
the surface. Instead, MWD tools transmit data to the rig using
pressure pulses in the fluid column inside the drill pipe, or by
sending low frequency electromagnetic waves through the earth. The
maximum data rate is a few bits per second, so a high level of signal
processing and data compression must be done downhole by the MWD
apparatus.

MWD systems must maintain their accuracy for hundreds of hours, while
exposed to shocks exceeding 100 g; abrasion by rock as they rotate at
100-200 RPM; pressures up to 170 MPa (25 000 psi or 1700 atm); and
temperatures up to 175°C. The equipment must transmit extremely
high
torsional and axial loads to the drill bit, and must, without being
eroded, provide a passageway for high-velocity drilling fluids.

By the early 1990s, MWD technology used electromagnetic propagation
to measure resistivity, along with Compton scattering to measure
porosity and neutron scattering to differentiate oil from gas. The
MWD measurements, made by sensors located near the drill bit, are
similar to measurements developed for conventional well logging but
are adapted for the much harsher drilling environment. Information
gleaned from MWD measurements is transmitted to the surface in real
time and allows a horizontal well to be steered to hit a thin
hydrocarbon layer and then stay within it.

Challenges

Box 2
Most giant oil fields, those containing more than 5 ´ 108 barrels
of
oil, were discovered between 1950 and 1970. Fields discovered in the
past 30 years or so have tended to be small, geologically complex, or
in remote or difficult environments such as in the Arctic or below
deep water. Hence, improving the recovery factor in existing fields
is crucial. Physics will play an increasingly important role in
optimizing production, through such techniques as borehole NMR (see
box 2).

Illumination of the reservoir regions between well bores will be
essential for understanding the complex formations whose exploitation
is becoming increasingly common. Breakthroughs in measurement
techniques and signal processing will be needed before hydrocarbon
accumulations can be mapped in three dimensions with high spatial
resolution. Significant advances in the understanding of both
multicomponent flow in porous media and the physics of wettability
are crucial for the development of technology to minimize bypassed
oil.

Well-logging instruments, which give snapshots of fluid distributions
when a well is first drilled and then infrequently thereafter, will
be supplemented by permanently implanted sensors. Sensor arrays, some
of which will be based on fiber optics, will provide spatially and
temporally continuous data streams. That information will be used to
control downhole valves that regulate the rate and location of fluid
extraction from the reservoir. Simple versions of such arrays have
already been tested, at Wytch Farm and other places.

As the supply of cheap oil and gas is depleted and the remainder is
increasingly concentrated in the Middle East, consumer nations may
need to develop local resources, even if they are of lower grade.
Japan seeks to exploit gas hydrates, a form of ice containing
methane. Gas hydrates are abundant below the sea floor near Japan's
population centers. Canada has made major investments in the
development of tar sands, a plentiful, low-grade resource. New ways
of thinking about reservoir characterization and production will be
needed to efficiently utilize such resources.

References
1. J. Hearst, P. Nelson, F. Paillet, Well Logging for Physical
Properties: A Handbook for Geophysicists, Geologists and Engineers,
2nd ed., Wiley, New York (2000). S. M. Luthi, Geological Well Logs:
Their Use in Reservoir Modeling, Springer-Verlag, New York (2001).
2. Energy Information Administration, Office of Integrated Analysis
and Forecasting, International Energy Outlook 2001, rep. no. DOE/EIA-
0484(2001), US Department of Energy, Washington, DC (March 2001),
available online at http://www.eia.doe.gov/oiaf/ieo/index.html.
3. Energy Information Administration, Office of Energy Markets and
End Use, International Energy Annual 1999, rep. no. DOE/EIA-0219(99),
US Department of Energy, Washington, DC (February 2001). Data on
reserves by region are available online at
http://www.eia.doe.gov/emeu/international/petroleu.html.
4. C. J. Campbell, J. H. Haherrere, Sci. Am. March 1998, p. 78.
5. A. H. Thompson, A. J. Katz, C. E. Krohn, Adv. Phys. 36, 625 (1987).
6. F. M. Auzerais et al., Geophys. Res. Lett. 23, 705 (1996).
7. D. V. Ellis, Science 250, 82 (1990).
8. A. A. Kaufman, Geophysics 55, 29 (1990).
9. R. E. Sheriff, L. P. Geldart, Exploration Seismology, Cambridge U.
Press, New York (1995).
10. G. A. Cooper, Sci. Am. May 1994, p. 82.
11. O. Eiken, "Seismic Monitoring of a Unique CO2 Injection Site at
Sleipner," online at http://www.physicstoday.org/vol-55/iss-
4/pdf/Eiken.ppt.
12. R. L. Kleinberg, in Encyclopedia of Nuclear Magnetic Resonance,
vol. 8, D. M. Grant, R. K. Harris, eds., Wiley, New York (1996), p.
4960.
13. S. Godefroy, J.-P. Korb, M. Fleury, R. G. Bryant, Phys. Rev. E
64, 021605 (2001).
14. J. S. Schweitzer, in Remote Geochemical Analysis: Elemental and
Mineralogical Composition, C. Pieters, P. Englert, eds., Cambridge U.
Press, New York (1993).

Related readings from Physics Today
December 1988, page 24

Brian Clark is a physicist with Schlumberger in Sugar Land, Texas.
Robert Kleinberg is a physicist with Schlumberger-Doll Research in
Ridgefield, Connecticut. Both are Schlumberger Fellows.

© 2002 American Institute of Physics