Speight J.G. Natural Gas: A Basic Handbook
Подождите немного. Документ загружается.
6.1
Water Removal
139
through a specialized boiler designed to vaporize only the water out
of
the solution. The boiling point differential between water
(lOO°C,
212°F)
and glycol
(204"C,
400°F)
makes it relatively easy to remove
water from the glycol solution, allowing it be reused in the dehydra-
tion process.
As
well as absorbing water from the wet gas stream, the glycol solu-
tion occasionally carries with it small amounts
of
methane and other
compounds found
in
the wet gas.
In
the past, this methane
was
simply vented out
of
the boiler.
In
addition to losing a portion
of
the
natural gas that was extracted, this venting contributes to air pollu-
tion and the greenhouse effect.
To
decrease the amount
of
methane
and other compounds that are lost, flash tank separator-condensers
work to remove these compounds before the glycol solution reaches
the boiler.
Essentially, a flash tank separator consists of a device that reduces the
pressure
of
the glycol solution stream, allowing the methane and
other hydrocarbons to vaporize
(flash).
The glycol solution then
travels to the boiler, which may also be fitted with air or water-cooled
condensers, which serve
to
capture any remaining organic
compounds that may remain
in
the glycol solution. The regeneration
(stripping)
of
the glycol is limited by temperature; diethylene glycol
and triethylene glycol decompose at or before their respective boiling
points. Such techniques as stripping
of
hot triethylene glycol with dry
gas (e.g., heavy hydrocarbon vapors, the
Drizu
process)
or vacuum dis-
tillation are recommended.
Water may be removed from gas streams at the same time as
hydrogen sulfide is removed. Moisture removal is necessary
to
pre-
vent harm
to
anhydrous catalysts and to prevent the formation
of
hydrocarbon hydrates (e.g.,
C,H,. 18H,O)
at low temperatures.
A
widely used dehydration and desulfurization process is the glycola-
mine process, in which the treatment solution is a mixture
of
ethano-
lamine and a large amount of glycol. The mixture is circulated
through an absorber and a reactivator in the same way as ethanola-
mine is circulated
in
the Girbotol process. The glycol absorbs mois-
ture from the hydrocarbon gas passing up the absorber; the
ethanolamine absorbs hydrogen sulfide and carbon dioxide. The
treated gas leaves the top
of
the absorber; the spent ethanolamine-
glycol
mixture enters the reactivator
tower, where heat drives
off
the
absorbed acid gases and water.
140
Chauter
6
Process
Classification
Solid adsorbent or solid-desiccant dehydration is the primary form
of
dehydrating natural gas using adsorption, and usually consists
of
two
or more adsorption towers, which are filled with a solid desiccant.
Typical desiccants include activated alumina or a granular silica gel
material. Wet natural gas is passed through these towers, from top to
bottom. As the wet gas passes around the particles of desiccant mate-
rial, water is retained on the surface
of
these desiccant particles.
Passing through the entire desiccant bed, almost all
of
the water is
adsorbed onto the desiccant material, leaving the dry gas
to
exit the
bottom
of
the tower.
Solid-adsorbent dehydrators are typically more effective than glycol
dehydrators, and are usually installed as a type of straddle system
along natural gas pipelines. These types
of
dehydration systems are
best suited
for
large volumes
of
gas under very high pressure, and are
thus usually located
on
a pipeline downstream
of
a compressor sta-
tion.
Two
or more towers are required because after a certain period
of
use, the desiccant in a particular tower becomes saturated with water.
To “regenerate” the desiccant, a high-temperature heater is used to
heat gas to a very high temperature. Passing this heated gas through a
saturated desiccant bed vaporizes the water in the desiccant tower,
leaving it dry and allowing for further natural gas dehydration.
Silica gel (SiO,) and alumina (A1,OJ have good capacities
for
water
adsorption (up to
8%
by weight). Bauxite (crude alumina,
A1,0,)
adsorbs up
to
6%
by weight water, and molecular sieves adsorb up to
15%
by weight water. Silica is usually selected for dehydration
of
sour
gas because
of
its high tolerance
to
hydrogen sulfide and
to
protect
molecular sieve beds from plugging by sulfur.
6.2
Liquids
Removal
Natural gas coming directly from a well contains many natural gas
liquids that are usually removed. In most instances, natural gas liq-
uids (NGLs) have a higher value as separate products, and it
is
thus
economical
to
remove them from the gas stream. The removal
of
nat-
ural gas liquids usually occurs
in
a relatively centralized processing
plant, and uses techniques similar to those used to dehydrate natural
gas. The justification for building a liquid recovery
(or
a liquid
removal) plant depends on the
(1)
the specification for natural gas
sold
to
the consumer and
(2)
the price differential between the
enriched gas (containing
the higher molecular weight hydrocarbons)
and lean gas with
the
added value
of
the extracted liquid.
6.2
Liquids
Removal
141
Condensates are most often removed from the gas stream at the well-
head through the use
of
mechanical separators.
In
most instances, the
gas flow
into
the separator comes directly from the wellhead, because
the gas-oil separation process is not needed. The gas stream enters the
processing plant at high pressure
(600
psi or greater) through an inlet
slug catcher where free water is removed from the gas, after which it
is directed to a condensate separator. Extracted condensate is routed
to on-site storage tanks.
There are
two
basic steps to the treatment
of
natural gas liquids in the
natural gas stream. First, the liquids must be
extracted
from the nat-
ural gas. Second, these natural gas liquids must be separated them-
selves, down
to
their base components (see Fractionation). These
two
processes account for about
90%
of the total production
of
natural
gas liquids.
There are three principle techniques for removing natural gas liquids
from the natural gas stream:
1.
The absorption method
2.
The cryogenic expander process
3.
A
membrane process
The extraction
of
natural gas liquids from the natural gas stream pro-
duces both cleaner, purer natural gas, as well as the valuable hydro-
carbons that are the natural gas liquids themselves.
6.2.1
Absorption
The absorption method
of
extraction is very similar to using absorp-
tion for dehydration. The main difference is that, in the absorption
of
natural gas liquids, absorbing oil is used instead
of
glycol. This
absorbing oil has an affinity for natural gas liquids
in
much the same
manner as glycol has an affinity for water. Before the oil has picked
up any natural gas liquids, it is termed
lean
absorption oil.
The oil absorption process involves the countercurrent contact of the
lean (or stripped) oil with the incoming wet gas
with
the temperature
and pressure conditions programmed to maximize the dissolution of
the liquefiable components
in
the oil. The rich absorption oil (some-
times referred to
as
fat oil), containing natural
gas
liquids,
exits
the
absorption tower through the bottom. It is now a mixture
of
absorption
142
Chavter
6
Process Classification
oil, propane, butanes, pentanes, and other higher boiling hydrocarbons.
The rich oil is fed into lean oil stills, where the mixture is heated
to
a
temperature above the boiling point of the natural gas liquids but below
that
of
the oil. This process allows
for
the recovery
of
about 75% by
volume of the butanes, and 85 to
90%
by volume of the pentanes and
higher boiling constituents from the natural gas stream.
The basic absorption process can be modified to improve its effective-
ness, or to target the extraction
of
specific natural gas liquids.
In
the
refrigerated oil absorption method, where the lean
oil
is cooled
through refrigeration, propane recovery can be upwards
of
90%
by
volume, and about
40%
by volume
of
the ethane can be extracted
from the natural gas stream. Extraction
of
the other, higher boiling
natural gas liquids can be close to
100%
by volume using this process.
The absorption method,
on
the other hand, uses an absorbing oil to
separate the methane from the natural gas liquids. While the gas
stream is passed through an absorption tower, the absorption oil
(lean
oil)
soaks up a large amount
of
the natural gas liquids. The absorption
oil
(enriched oil),
now containing natural gas liquids, exits the base of
the tower after which the enriched oil is fed into distillers, where the
blend is heated
to
above the boiling point
of
the natural gas liquids,
while the oil remains fluid. The absorption oil is recycled while the
natural gas liquids are cooled and directed to a fractionator tower.
Another absorption method often used is the refrigerated oil absorp-
tion method, where the lean oil is chilled rather than heated, a fea-
ture that has the potential to enhance recovery rates.
6.2.2
Cryogenic Expander Process
In
this process, a turboexpander is used to produce the very low tem-
peratures needed to achieve high recovery
of
light components, such
as ethane and propane. Essentially, cryogenic processing consists
of
lowering the temperature
of
the gas stream
to
about -85°C
(-120°F).
There are several ways
to
do this, but the turboexpander process (in
which external refrigerants are used to chill the gas stream) is the
most effective. The quick drop in temperature that the expander is
capable
of
producing condenses the hydrocarbons in the gas stream,
but maintains methane in its gaseous form.
In
the process, the natural gas
is
first dehydrated using a molecular
sieve followed by cooling (Figure
6-3).
The separated liquid containing
6.2
Liquids
Removal
143
Figure
6-3
Drying using a molecular sieve.
most of the heavy fractions is then demethanized, and the cold gases
are expanded through a turbine that produces the desired cooling
for
the process. The expander outlet is a two-phase stream that is fed to
the top
of
the demethanizer column. This serves as a separator
in
which:
(1)
the liquid is used as the column reflux and the separator
vapors combined with vapors stripped in the demethanizer are
exchanged with the feed gas, and
(2)
the heated gas, which is partially
recompressed by the expander compressor, is further recompressed to
the desired distribution pressure
in
a separate compressor. This process
allows for the recovery
of
90-95%
by volume
of
the ethane originally
in the gas stream.
In
addition, the expansion turbine is able to convert
some
of
the energy released when the natural gas stream is expanded
into recompressing the gaseous methane effluent, thus saving energy
costs associated with extracting ethane.
The cryogenic method is better at extraction
of
the lighter liquids,
such as ethane, than is the alternative absorption method.
6.2.3
Membrane Processes
These separation processes are versatile, designed to process a wide
range
of
feedstocks, and offer
a
simple solution for removal
and
recovery
of
higher boiling hydrocarbons (natural gas liquids) from
144
Chapter
6
Process Classification
natural gas (Foglietta,
2004).
The process
is
based
on
high-flux mem-
branes that selectively permeate higher boiling hydrocarbons
(compared to methane) and are recovered as a liquid after recompres-
sion and condensation. The residue stream from the membrane is
partially depleted of higher boiling hydrocarbons, and is then sent to
the sales gas stream. Gas permeation membranes are usually made
with vitreous polymers that exhibit good selectivity but, to be effec-
tive, the membrane must be very permeable with respect to the sepa-
ration process (Rojey et al.,
1997).
6.3
Nitrogen
Removal
A
significant fraction of many natural gas reserves is sub-quality (low
heating value) due to the high nitrogen content of the gas. Gas con-
taining more than about
6%
nitrogen must be treated to remove the
nitrogen.
In
many cases where nitrogen occurs in natural gas, the
reserves cannot currently be exploited because of the lack of suitable
nitrogen removal technology.
The separation of methane and nitrogen is challenging for any tech-
nology because these gases are similar in size, boiling point, and
chemical nature. Conventional processes such as cryogenic distilla-
tion and pressure swing adsorption are used (Table
6-Z),
but these
technologies are not widespread because they are cost prohibitive.
Most plants that practice nitrogen rejection were built for dual use,
such as production of helium and carbon dioxide for enhanced oil
recovery
(EOR)
applications.
In
these cases, the costs of separating
nitrogen are shared by several products, making the process feasible.
Once the hydrogen sulfide and carbon dioxide are processed to
acceptable levels, the stream
is
routed to
a
nitrogen rejection unit
(NRU),
where it is further dehydrated using molecular sieve beds.
In
the nitrogen rejection unit, the gas stream is routed repeatedly
through a column and a brazed aluminum plate fin heat exchanger,
where the nitrogen is cryogenically separated and vented.
Another type of nitrogen rejection unit involves separation of
methane and heavier hydrocarbons from nitrogen using an absorbent
solvent. The absorbed methane and heavier hydrocarbons are flashed
off from the solvent by reducing the pressure
on
the processing stream
in multiple gas decompression steps. The liquid from the flash regen-
eration step
is
returned to the top
of
the methane absorber as lean
6.3
Nitrogen
Removal
145
Table
6-2
Processes Currently Used or Under Development for
Removal of Nitrogen from Natural Gas (Membrane Technology and
Research, Inc.,
1999)
Cryogenic Condensation
Typically High methane recovery.
distillation and distillation
high- Significant pretreatment
(Proven
at cryogenic flow-rate and compression costs.
commercially)
temperatures applications
High capital costs.
Pressue swing Adsorption of Generally Pretreatment required.
Adsorption methane small to High capital and
(PSA)
(Limited medium flow compression costs. High
Commercial rates operating costs. Moderate
success) methane recovery.
~~~~ ~ ~
Lean Oil Absorption of Suitable for High capital costs.
Absorption methane in high Processing costs
(New process) chilled nitrogen significant. Need to
hydrocarbon oil content absorb bulk of methane
streams increases equipment size
and compression
requirements.
Nitrogen Selective
Absorption absorption of
(Research nitrogen in
stage) chelating
solvent
No
methane
recompression needed.
Stability of solvent
suspect.
solvent. Helium, if any, can be extracted
from
the gas stream through
membrane diffusion
in
a pressure swing adsorption
(PSA)
unit.
Generally, nitrogen removal from natural gas requires liquefaction
and fractionation of the entire gas stream, which may affect process
economics.
In
many cases the nitrogen-containing natural gas is
blended with
a
gas having a higher heating value and sold at a
reduced price depending upon the thermal value (Btu/ft3).
A
membrane system can produce pipeline-quality gas and nitrogen-
rich fuel from raw natural gas. The process relies
on
proprietary mem-
branes that are significantly
more permeable
to
methane, ethane, and
other hydrocarbons than to nitrogen. Gas containing
8-18%
nitrogen
146
Chapter
6
Process Classification
is compressed and passed across a first set
of
membrane modules. The
permeate, which contains
4%
nitrogen, is sent to the pipeline; the
nitrogen-rich residue gas is passed
to
a second set
of
membrane mod-
ules. These modules produce a residue gas containing 50% nitrogen
and a nitrogen-depleted permeate containing about
9%
nitrogen. The
residue gas is used as fuel; the permeate is mixed with the incoming
feed gas for further recovery. The membrane process divides the gas
into two streams. The first stream is product gas containing less than
4%
nitrogen, which
is
sent to the pipeline; the second stream, which
contains
30-50%
nitrogen, is used
to
fuel the compressor engines.
In
some cases, a third stream is produced; this stream, which contains
60-85'30 nitrogen, is flared
or
reinjected.
In
a typical two-step process, the feed gas containing
10
to
15%
nitrogen is compressed to 800-1,200 psi using a gas-powered com-
pressor. The gas then passes through the first set
of
methane-
permeable membrane modules. The product gas, which contains
4%
nitrogen, is sent to the pipeline. The residue is sent to a second set
of
modules, but the permeate from these modules is too rich
in
nitrogen
to be delivered to the pipeline, and the gas is recirculated
to
the front
of
the compressor for further treatment. The residue gas from the
second set
of
modules, which contains about 50% nitrogen, is used to
fuel the compressor engines. The process achieves 80-9OYo recovery
of
the fuel gas Btu value in the pipeline product. Recovery values as
high as 95% or higher can be achieved depending
on
the composi-
tion
of
the inlet gas
(http://www.mtrinc.com/Pages/NaturalGas/
ng.html#).
6.4
Acid
Gas
Removal
In
addition to water and natural gas liquids removal, one
of
the most
important parts
of
gas processing involves the removal
of
hydrogen
sulfide and carbon dioxide. Natural gas from some wells contains sig-
nificant amounts
of
hydrogen sulfide and carbon dioxide and is usu-
ally referred
to
as
sour
gas.
Sour gas is undesirable because the sulfur
compounds it contains can be extremely harmful, even lethal, to
breathe, and the gas can be extremely corrosive. The process for
removing hydrogen sulfide from sour gas is commonly referred to as
sweetening
the gas. Acid gas removal (i.e., removal
of
carbon dioxide
and hydrogen sulfide
from
natural gas streams)
is
achieved by absorp-
tion and/or adsorption.
6.4
Acid
Gas
Removal
147
As currently practiced, acid gas removal processes involve the selec-
tive absorption
of
the contaminants into a liquid, which is passed
countercurrent to the gas. Then the absorbent is stripped
of
the gas
components (regeneration) and recycled to the absorber. The process
design will vary and,
in
practice, may employ multiple absorption
columns and multiple regeneration columns.
Liquid absorption processes (which usually employ temperatures below
50"C,
120°F) are classified either as
physicar solvent processes
or
chemical
soZvent processes.
The former processes employ an organic solvent, and
absorption is enhanced by low temperatures, or high pressure, or both.
Regeneration of the solvent is often accomplished readily (Staton et al.,
1985).
In
chemical solvent processes, absorption
of
the acid gases is
achieved mainly by use
of
alkaline solutions, such as amines or carbon-
ates (Kohl and Riesenfeld,
1985).
Regeneration (desorption) can be
brought about by use
of
reduced pressures and/or high temperatures,
whereby the acid gases are stripped from the solvent.
Adsorbers are widely used to increase a low gas concentration prior to
incineration, unless the gas concentration is very high in the inlet air
stream. Adsorption also is employed to reduce problem odors from
gases. There are several limitations to the use
of
adsorption systems,
but
it
is generally felt that the major one is the requirement for mini-
mization
of
particulate matter and/or condensation
of
liquids (e.g.,
water vapor) that could mask the adsorption surface and drastically
reduce its efficiency.
6.4.1
Olamine
Processes
The primary process (Figure
6-4)
for sweetening sour natural gas is
quite similar to the processes of glycol dehydration and removal of
natural gas liquids by absorption.
In
this case, however, amine
(oh-
mine)
solutions are used to remove the hydrogen sulfide (the
amine
process).
The sour gas is run through a tower, which contains the ola-
mine solution. There are two principle amine solutions used, monoet-
hanolamine
(MEA)
and diethanolamine (DEA). Either
of
these
compounds, in liquid form, will absorb sulfur compounds from nat-
ural gas as it passes through. The effluent gas is virtually free
of
sulfur
compounds, and thus loses its sour gas status. Like the process
for
the
extraction
of
natural gas liquids and glycol dehydration, the amine
solution used can be regenerated for reuse.
148
Chaater
6
Process Classification
Figure
6-4
Schematic
of
the amine (olamine) process.
Amine derivatives such as ethanolamine (monoethanolamine, MEA),
diethanolamine (DEA), triethanolamine (TEA), methyldiethanola-
mine (MDEA), diisopropanolamine (DIPA), and diglycolamine (DGA)
are the most widely used
olamines
in commercial applications
(Table
6-3;
Katz,
1959;
Kohl and Riesenfeld,
1985;
Maddox et al.,
1985;
Polasek and Bullin,
1985;
Jou et al.,
1985;
Pitsinigos and
Lygeros,
1989;
Speight,
1993;
Mokhatab et al., 2006). They are
selected according
to
their relative ability to interact with and remove
carbon dioxide and/or hydrogen sulfide.
The chemistry can be represented by simple equations for low partial
pressures of the acid gases:
2RNH,
+
H,S
+(RNH,),S
2RHN,
+
CO,
+
H,O +(RNH,),CO,