Natural gas origin and sources
There are different theories as to the origins of fossil fuels. The most widely accepted theory of the origin
of natural gas assumes that natural gas hydrocarbons come from organic matter (the remains of land and
aquatic plants, animals and microorganisms) that was trapped within sediments as they were deposited and
transformed over long periods of time into their present form. Two main mechanisms, namely, biogenic
and thermogenic are responsible for the degradation of fossil organic material in sediments (Rojey et al.,
1997). Biogenic gas is formed at shallow depths and low temperatures due to the action of bacteria on the
organic debris accumulating in the sediments. In contrast, thermogenic gas is formed at deeper depths by
degradation of organic matter, called kerogen, accumulated in fine-grained sediments, especially clays and
shales. This degradation occurs through the combined effects of temperature and pressure. Thermogenic
gas is believed to be produced by two mechanisms, namely, direct thermal cracking of sedimentary organic
matter and secondary thermal cracking of oil that is formed in the first stage. The former is called the
primary thermogenic gas that coexists with oil, while the latter is called secondary thermogenic gas that
coexists with insoluble solid matter, called pyrobitumen. Both mechanisms involve thermal cracking with
some degree of sustained pressure, mainly through the weight of the sedimentary formation. Little
information is available on the time required to generate thermogenic gas other than the general assumption
that it is a long time. Natural gas comes from both “conventional” (easier to produce) and “unconventional”
(more difficult to produce) geological formations.
Conventional gas is typically “free gas” trapped in multiple, relatively small, porous zones in various
naturally occurring rock formations such as carbonates, sandstones, and siltstones. Conventional natural
gas generally occurs in deep reservoirs, either associated with crude oil (associated gas1) or in reservoirs
that contain little or no crude oil (nonassociated gas2). Natural gas from coal (also known as coal-bed
methane), tight gas sands, gas shales, geopressurized aquifers, and gas hydrates 3 are often referred to as
unconventional gas resources. The common characteristic of the different types of unconventional gas
resources is that they contain large quantities of natural gas, but it is usually more difficult to produce this
gas as compared to conventional reservoir rocks. New technologies are continually being developed to
allow more accurate estimations of the amount of gas in these unconventional reservoirs and to stimulate
these rocks to produce the gas.
Natural gas composition and classification
Natural gas is a complex mixture of hydrocarbon and nonhydrocarbon constituents and exists as a gas under
atmospheric conditions. Virtually hundreds of different compounds may be present in natural gas in varying
amounts. Even two wells producing from the same reservoir may produce gases of different composition
as the reservoir is depleted.
While natural gas is formed primarily of methane (CH4), it can also include significant quantities of ethane
(C2H6), propane (C3H8), butane (C4H10), pentane (C5H12), as well as traces of hexane (C6H14) and heavier
hydrocarbons. Many natural gases often contain nitrogen (N2), carbon dioxide (CO2), hydrogen sulfide
(H2S), and other sulfur components such as mercaptans (R-SH),4 carbonyl sulfide (COS), and carbon
disulfide (CS2). Trace quantities of argon, hydrogen, and helium may also be present. Trace quantities of
metallic substances are known to exist in natural gases including arsenic, selenium, mercury, and uranium.
According to the proportion of hydrocarbons heavier than methane, different types of natural gas (dry, wet,
and condensate) can be considered. Natural gas is considered “dry” when it is almost pure methane, having
had most of the other commonly associated hydrocarbons removed. When other hydrocarbons are present,
the natural gas is “wet,” where it forms a liquid phase during production at surface conditions. “Condensate”
gases have a high content of hydrocarbon liquids and form a liquid phase in the reservoir during production,
during the depletion process. Natural gases are commonly classified according to their liquid content as
either lean or rich and according to their sulfur content as either sweet or sour. The lean and rich terms refer
to the amount of potentially recoverable liquids. The term usually applies to ethane and heavier components
but sometimes applies instead to propane and heavier components (if ethane is not regarded as a valuable
liquid component). To quantify the liquids content of a natural gas mixture, the industry uses GPM, or
gallons of liquids recoverable per 1000 standard cubic feet (Mscf5) of gas. Lean natural gas has liquid
content less than 2 GPM. Moderately rich natural gas has between 2 and 5 GPM, and very rich natural gas
has greater than 5 GPM (Ewan et al., 1975).
The sweet and sour terms refer to the H2S content. Strictly speaking, “sweet” and “sour” refer to both acid
gases (H2S and CO2) but are usually applied to H2S alone. A sweet gas contains negligible amounts of H2S,
whereas a sour gas has unacceptable quantities of H2S. The terms are relative, but generally, sweet means
the gas contains less than 4 ppmv of H2S. Carbon dioxide can be tolerated to much higher levels, say 3–4
mol%, as long as the heating value of the sales gas is satisfactory.
Natural gas phase behavior
Natural gas is a naturally occurring hydrocarbon mixture that is found underground and at elevated
conditions of pressure and temperature. Therefore, there is an essential need to know a priori how the gas
fluid will behave under a wide range of pressure and temperature conditions, particularly in terms of its
volumetric and thermophysical properties that are required in simulating reservoirs, evaluating reserves,
forecasting production, designing production facilities, and designing gathering and transportation systems.
In fact, an accurate knowledge of hydrocarbon fluid phase behavior is crucial in designing and operating
the gas engineering processes efficiently and optimally. This means, having advanced predictive tools for
the characterization of hydrocarbon phase behavior with the highest accuracy possible is the key to
mastering the economics of natural gas systems.
The natural gas phase behavior is a plot of pressure versus temperature that determines whether the natural
gas stream at a given pressure and temperature consists of a single gas phase or two phases, gas and liquid.
The phase behavior for natural gas with a given composition is typically displayed on a phase diagram, an
example of which is shown in Figure 1-1. The left-hand side of the curve is the bubble point line and divides
the single-phase liquid region from the two phase gas–liquid region. The right-hand side of the curve is the
dew point line and divides the two-phase gas–liquid region and the single-phase gas region. The bubble
point and dew point lines intersect at the critical point, where the distinction between gas and liquid
properties disappears. The maximum pressure at which liquids can form is called the cricondenbar (PCC),
and the maximum temperature at which liquids can form is called the cricondentherm (T CC). However,
there is something very interesting going on within the region Tc < T < Tcc, where we will be moving from
a 0% liquid to another 0% liquid condition (both on the dew point curve) in an isothermal compression.
This different behavior of a vapor under compression is called retrograde (contrary to expectation)
condensation. It is also important to see that a similar behavior is to be expected within the region Pc < P <
Pcc. In this case, we talk about retrograde vaporization since we will be moving from a 100% liquid to
another 100% liquid condition (both on the bubble point curve) in an isobaric heating. The natural gas phase
behavior is a function of the composition of the gas mixture and is strongly influenced by the concentration
of the heavier hydrocarbons, especially hexane plus.6 The presence of heavier hydrocarbons will increase
the phase envelope, and failure to include them in a phase calculation will underpredict the phase envelope.
There is also an essential need for proper characterizing the heavy-ends. In fact, although some different
fluid descriptions match to some extent the behavior of the reservoir fluids at reservoir conditions, they
exhibit larger variations once surface simulators are used and the fluids are subjected to process conditions.
Natural Gas Liquids Recovery
Most natural gas sources contain hydrocarbon liquids that must be removed to meet the hydrocarbon dew
point and heating value specifications of the pipeline gas before they can be used by the consumers. The
removal of heavy hydrocarbons is necessary to ensure pipeline transportation safety. The hydrocarbon
liquids consist of two components, the natural gas condensate (C5þ) and the natural gas liquids (NGLs)
(C2–C4). These components can be sold at a premium over natural gas for the equivalent heating value. The
C5þ condensate is separated in the condensate stabilization unit that has been discussed in Chapter 5. The
C3–C4 liquids are valued as a liquid fuel. The C2 component can be sold as petrochemical plant feedstocks.
Separation of the NGL components are discussed in this chapter. The richness of the NGL components can
be identified by the term “GPM,” that is, gallons per mole of hydrocarbon (C2þ) liquid. The value of “GPM”
and the higher heating value of the hydrocarbon component are shown in Table 8-1.
The C2+ NGL recovered from an NGL recovery unit is termed “Y-grade” NGL. The acceptable Y-Grade
is a mixture of NGLs composed principally of ethane, propane, butane, pentanes, and natural gasoline,
which typically meets the specifications given in Table 8-2. The Y-grade liquids must be free from sand,
dust, gums, gum-producing substances, oil, glycol, inhibitor, amine, caustics, chlorides, oxygenates, heavy
metals, and any other contaminants or additive to the product added to enhance the ability to meet
specifications. This chapter covers the production of NGL from hydrocarbon dew pointing to propane and
ethane recovery, including their history, and various technologies and design options as well as NGL
fractionation.
8.3.4 Turboexpander NGL recovery processes
The term “turboexpander” refers to an expander/compressor machine as a single unit. It consists of two
primary components: the radial inflow expansion turbine and a centrifugal compressor integrated as a single
assembly (see Figure 8-7).
The expansion turbine is the power unit and the compressor is the driven unit. In cryogenic NGL recovery
processes, the turboexpander achieves two different but complementary functions. The main function is to
generate refrigeration to cool the gas stream. This is done by the expansion turbine end that expands the
gas isentropically by extracting the enthalpy from the gas stream, causing it to cool. The other function is
the use of the extracted energy to rotate the shaft to drive the compressor end of the turboexpander, which
recompresses the residue gas stream.
The first turboexpander unit was built in 1964 for NGL recovery in the city of San Antonio, Texas. The gas
is supplied at 700 psig pressure and is letdown in pressure to about 300 psig to the demethanizer. Methanol
injection was used for hydrate inhibition. Until this time, Liquefied Petroleum Gas (LPG) recovery was
mainly achieved with refrigerated lean oil, which is described in the later section.
The first turboexpander process patent was issued to Bucklin (Fluor) in 1966. The flow schematic is shown
in Figure 8-8. The concept was to use turboexpander, which is a more efficient method of cooling, instead
of the J–T valve. The feed gas is cooled by the cold demethanizer overhead, and separated in the HP cold
separator. The separator vapor is letdown in pressure using the turboexpander and fed to the top of the
demethanizer as a reflux. The HP cold separator liquid is letdown in pressure to the Low Pressure (LP) cold
separator. The separator liquid is further letdown and used to cool the feed gas before it is fed to the lower
part of the demethanizer. The demethanizer bottom product is heated with steam that controls the methane
content (see Table 8-2). Earlier NGL recovery units did not have the benefits of the compact BAHX and
would require multiple shell and tube heat exchangers to achieve the chilling requirements. Even with
extensive heat integrated scheme as shown in Figure 8-9, high NGL recovery could not be achieved. The
other limitation of earlier expander designs is the low expansion ratio, which required two expanders
operating in series to achieve a high expansion ratio. A higher expansion ratio would generate more cooling
that is necessary for high NGL recovery.
The process flow schematic of a typical turboexpander NGL recovery unit built in the 1970s is shown in
Figure 8-9. The process uses two expanders–compressor sets operating in series and several separators to
produce various liquid streams that are fed to different locations in the demethanizer.
The process also requires propane refrigeration for feed gas cooling. The NGL recovery process could
achieve about 70% ethane recovery and was typically based on a relatively rich feed at about 950 psig
pressure. Most of these NGL recovery plants are still operating today, but they are processing much leaner
gas as the reservoirs are depleted. The leaner gas has an impact on the NGL recovery level. The heat
exchanger surface areas are no longer adequate and the expanders are operating at lower efficiency.
Consequently, ethane recovery drops to about 55%. A picture of the earlier NGL recovery unit is shown in
Figure 8-10. The main contributors to the success of today’s NGL recovery plants are the turboexpanders
and the brazed aluminum exchangers. The application of turboexpanders to the natural gas industry began
in the early 1960, which was followed by the development of BAHXs.
8.3.4.1 Turboexpander
Today’s turboexpander designs can yield very high adiabatic efficiencies (over 85%). However, there are
aerodynamic limitations for both the expander and compressor. Machinery efficiencies will drop if the gas
composition or flow rate were different than the design points. In most instances, when the changes are
temporarily, they can be managed by the expander bypass J–T valves. During the expander bypass
operation, the NGL unit temperature profile will increase and operation of the turboexpander will be
inefficient; consequently, NGL recovery and plant capacity will be reduced. If the flow conditions are
expected to be permanent, rewheeling of the expander and compressor can be economically justified.
8.8 NGL fractionation
NGL produced from the NGL recovery unit is typically sent to a fractionation center for further processing
into individual products. A simplified NGL fractionation flow schematic is shown in Figure 8-22. A picture
of a fractionation center is shown in Figure 8-23. NGL is fractionated by heating the NGL stream and
processing through a series of distillation columns. Fractionation takes advantage of the differing boiling
points of the various NGL components. As the NGL stream is heated, the lightest (lowest boiling point)
NGL component boils off first and is separated. The overhead vapor is condensed and a portion is used as
reflux and the remaining portion is routed to product storage. The heavier liquid mixture at the bottom of
the first column is routed to the second column where the process is repeated and a different NGL
component is separated as product. This process is repeated until the NGL is separated into its individual
components. In some cases, where isobutane is desired, the mixed butane stream can be further separated
with a deisobutanizer or butanes splitter to produce the required product for refinery consumption. There
are many options in configuring the NGL fractionation train and the optimum arrangement may vary
depending on the NGL compositions, and the product specifications.
8.8.1 Fractionation column design and operation
Fluctuation in feed conditions and compositions, which may happen on a daily basis, has significant impacts
on the NGL fractionation unit operation. Their impacts on the design can be evaluated on the process
simulators and the optimum operating variables can be assessed. The simulation software can now be
integrated to the distributed control system allowing real-time optimization.
Three distillation books published by Kister (1989, 1992, 2006) can be used as references and guidelines
in the design and operation of distillation columns. These books provide design parameters for equipment,
discuss limitations of the design methods, and suggest solutions in troubleshooting columns. They also
contain process calculations on column hydraulic and tower performance, tray and packing design details,
and the methods to maximize column operating efficiency.
13.4.5.3 Cryogenic recoveries (turboexpander processes)
Expansion with turboexpanders is now the main process employed for recovering natural gas liquids.
Turboexpanders can be controlled for various objectives: inlet pressure, demethanizer pressure, or residue
pressure are the most common. Guide vanes are manipulated to control the speed of the expander. A Joules–
Thompson (JT) valve is always included to allow rapid unloading of the expander. One split range controller
typically operates the guide vanes and JT valve so that the JT valve will open when the manipulation of the
guide vanes has been exhausted.
The compressor driven by the turboexpander in either inlet compression or residue compression mode
requires a recycle valve to maintain a minimum flow for surge protection. Depending on the exact cryogenic
recovery processing scheme, additional controls may be required for heat exchanger circuit flow splits,
chillers, separator levels, and pressure profiles. Heat exchanger flow splits are typically configured as flow
ratios. This ratio may be overridden to prevent “cold spins” or prevent temperatures below the critical
temperature in the cold separator upstream of the turboexpander.
13.4.5.4 Demethanizer
The demethanizer is integral to the turboexpansion process. The various feeds to the column are created at
multiple points in the process. Side reboil heat sources and sometimes the bottom reboiler heat are integral
to the heat exchanger circuit. Seldom are the side reboiler temperatures controlled. Manipulating the heat
to the bottom reboiler controls a bottom temperature, preferably pressure compensated.
An online chromatograph monitors the methane and/or carbon dioxide content. Ideally, this output would
reset the bottom temperature. Demethanizers are good candidates for model predictive control due to the
disturbances caused by the side reboilers, inlet flow rates, and inlet compositions. Minimizing the pressure
of the demethanizer based on turboexpander constraints and residue compression constraints is a major
opportunity for increasing liquids recoveries.
13.4.6 NGL fractionation
NGL fractionation consists of deethanization, depropanization, debutanization, and butane splitting (or
deisobutanization). The control schemes for each are analogous. The major control points for these
fractionators are reboiling heat, reflux, and pressure. Again, nonlinear level control is recommended for
feed tanks and bottom surge levels. Reboiling heat is manipulated to control the bottom composition. The
composition is cascaded to a sensitive temperature below the feed tray. Preferably the temperature is
pressure compensated. Reflux flow is manipulated to control the bottom composition. The composition is
cascaded to a sensitive temperature above the feed tray. Preferably the temperature is pressure compensated.
Minimum reflux schemes are employed to assure that reboiling load is not increased due to excessive reflux
and conversely, excessive reboiling leading to greater reflux rates for a given separation. Internal reflux
calculations and multivariable control schemes can achieve minimum energy consumption for a given
separation. Pressure should be minimized on these towers subject to constraints such as flooding, condenser
temperature, and bottom hydraulics. Flooding is indicated by delta pressure measurements across the tower.
Reflux is more difficult at lower temperatures as the available duty of the condenser may be limited. There
must be sufficient head on the bottom of the tower to allow liquids to feed downstream towers or satisfy
minimum head requirements for pumps. Again, multivariable control schemes handle the pressure
minimization issue elegantly.
