Equipment Costing
Equipment Costing
Equipment Costing
Department of Energy
Office of Energy Efficiency & Renewable Energy
Subcontract Report
Equipment Design and Cost NREL/SR-510-39946
Estimation for Small Modular May 2006
Biomass Systems, Synthesis
Gas Cleanup, and Oxygen
Separation Equipment
Task 2.3: Sulfur Primer
Nexant Inc.
San Francisco, California
NOTICE
This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof, nor any of their employees, makes any
warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not
infringe privately owned rights. Reference herein to any specific commercial product, process, or service by
trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States government or any agency thereof. The views and
opinions of authors expressed herein do not necessarily state or reflect those of the United States
government or any agency thereof.
Printed on paper containing at least 50% wastepaper, including 20% postconsumer waste
Contents
Section Page
i Methodology i
1 Introduction 1
2 Chemical Solvents 2
3 Physical Solvents 5
4 Catalytic Absorbents 10
5 Sulfur Recovery 11
Figures
Appendix
This deliverable is Subtask 2.3 of Task 2, Gas Cleanup Design and Cost
Estimates, as part of NREL Award ACO-5-44027, “Equipment Design and Cost
Estimation for Small Modular Biomass Systems, Synthesis Gas Cleanup and
Oxygen Separation Equipment”. Subtask 2.3 builds upon the sulfur removal
information first presented in Subtask 2.1, “Gas Cleanup Technologies for
Biomass Gasification” by adding additional information on the commercial
applications, manufacturers, environmental footprint, and technical specifications
for sulfur removal technologies. The data was obtained from Nexant’s
experience, input from GTI and other vendors, past and current facility data, and
existing literature.
Sulfur contaminants such as hydrogen sulfide (H2S), carbonyl sulfide (COS), and
mercaptans in syngas streams can create unacceptable levels of sulfur
emissions in power applications or poison catalysts used in chemical synthesis.
Sulfur contaminants are usually reduced to less than 300 ppm for power
generation and considerably lower (<1 ppm) for the synthesis of methanol,
ammonia, and Fischer-Tropsch (FT) liquids.
Three major sulfur removal technologies will be detailed here, with focus on H2S
removal: chemical solvents, physical solvents, and catalytic absorbents.
Chemical solvents create a chemical bond with H2S during contact with a gas
stream in an absorber column. The H2S loaded solvent is then sent to a stripper
column, where the chemical bond is broken; the solvent is reused and the H2S
sent elsewhere for sulfur recovery. Physical solvents use the high solubility of
H2S in the solvent to “dissolve” sulfur compounds. Their process layout is similar
to a chemical solvent system, using absorber and stripper columns. Catalytic
sorbents are usually single-use, fixed-bed solid catalysts that react with H2S to
remove it from the gas stream.
The main type of chemical solvent used for sulfur removal from gas streams are
amines. There are a number of different amines available for sulfur treatment;
proper selection depends on the compounds to be removed, the process
conditions, the level of treatment necessary, process metallurgy, and other
considerations. Amines commonly used for sulfur removal include:
Amine processes are proven technologies for the removal of H2S and CO2 (“acid
gases”) from gas streams. Amine systems generally consist of an absorber, a
stripper column, a flash separator, and heat exchangers. This is a low-
temperature process in which the gas to be treated enters the absorber at
approximately 110°F. In the absorber, acid gases are removed from the gas
stream by chemical reactions with the amine solution. The sweet (low sulfur) gas
stream exits at the top of the absorber.
Amine systems normally operate in the low to medium pressure range of 70-360
psi, although higher pressures can be accommodated with specific types of
amine solvents. However, in applications where the partial pressure of acid
gases is high, the economy of an amine system declines in comparison to
physical solvents. A sulfur removal level as low as 1 ppm can be achieved but at
the expense of operating cost due to the large solvent circulation rate required.
In general, amine systems provide flexibility, low cost, and high reliability to
industries seeking a proven sulfur removal technology. For low pressure sulfur
removal applications, amines are usually the technology of choice. Their
selectivity can be optimized to remove the compounds desired, while minimizing
process gas losses.
Design and construction of amine units is a standard practice for most major
engineering firms. Selection of amine vendors tends to be more specialized, with
a few firms available for packages that include process licensing and amine
supply. These firms include Dow (through UOP and Union Carbide subsidiaries),
BASF, Shell, Huntsman, and INEOS.
1
GPSA Engineering Data Book, 10th ed., “Hydrocarbon Treating,” Volume II, 1987
Task 2: Gas Cleanup Design and Cost Estimates 3
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
Amine treating technologies have been the process of choice for the natural gas
industry for decades, due to the limited affinity of amines for hydrocarbons.
Petroleum refiners also have extensive experience with amines for similar
reasons, using the technology for hydrocarbon gas sweetening and off-gas
treatment. The first commercial amine applications began in the 1920’s2, with
hundreds of units in active operation today. Operators of Integrated Gasification
Combined Cycle (IGCC) units operating on coal, coke, and biomass also have
extensive experience with amine technology. In fact, amines have been the
process of choice for acid gas removal in IGCC projects, and it continues to be
selected for new projects3.
Environmental Footprint
The major direct emissions from the amine process are an acid gas stream (H2S
and CO2, predominantly), a vent stream from the flash tank, waste water from
heat exchangers, and sour water from the stripper column. Standard processes
are usually integrated into a plant containing an amine unit to handle these
emissions:
Besides these direct emissions, indirect emissions are generated in this process
due to the need for electricity (for pumps) and steam (for reboilers and the
stripper column). Since the generation of these energy sources create their own
emission streams, these emission rates should be considered when determining
the overall environmental footprint of the process. The amount of the footprint
will vary considerably depending on the gas being treated and the depth of
treatment required. In general, amine processes tend to use less electricity but
more steam than units applying physical solvents.
2
Bullin, J., and Polasek, J, “Selective Absorption Using Amines”, Bryan Research and Engineering, Bryan,
TX, 2001.
3
Korens, N., Simbeck, D., Wilhelm, D., “Process Screening Analysis of Alternative Gas Treating and
Sulfur Removal for Gasification”, SFA Pacific, prepared for the US Department of Energy, National
Energy Technology Laboratory, December 2002.
Task 2: Gas Cleanup Design and Cost Estimates 4
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
SECTION 3: PHYSICAL SOLVENTS
This removal technology uses an organic solvent to remove acid gases from gas
streams by physical absorption without chemical reaction. The driving force of
this process is the high solubility of acid gases in the organic solvent. In most
cases, solubility increases as the temperature decreases and the pressure
increases. Thus, physical absorption is a low-temperature, high-pressure
process, with high partial pressure of acid gases required for the economy and
efficiency of this process. The two most common physical solvent processes are
Selexol and Rectisol, which use the dimethyl ether of polyethylene glycol and
chilled methanol, respectively, as the active solvents.
SELEXOL
The solubility of sulfur compounds present in gas process streams, coupled with
high solvent stability and excellent operation at high pressure, gives Selexol
some advantages over other sulfur removal processes. Besides removal of H2S
and CO2, hydrogen cyanide (HCN), mercaptans, COS, and heavy hydrocarbons
can all be removed to some extent by the Selexol solvent. Process developers
concerned about removal of these compounds from gas streams may find
physical solvents superior to chemical ones. Depending on the level of CO2 and
H2S removal desired, a Selexol system can be modified to meet a range of
different needs.
The Selexol process was first commercially used in the late 1960’s. The
technology is currently owned and licensed by Dow, with commercial packages
available from a variety of engineering companies including UOP, Udhe, and
Coastal. Roughly 60 Selexol plants are in commercial application worldwide6.
Selexol technology is used in high pressure natural gas, IGCC, and petroleum
refinery applications. At the refinery level, Selexol units can be seen upstream of
reforming and partial oxidation (POX) units. This technology is currently a main
competitor of amine processes for IGCC applications. Some of the key
considerations for process selection for IGCC use are the gasifier technology
chosen, feedstock type, and syngas end-use.
Environmental Footprint
As can be seen in Figure 2, the process configuration for a physical solvent unit
is similar to that of an amine system. In general, the direct and indirect emission
sources will be very similar to that of an amine unit, with differences in the
relative amount of emissions and where the waste streams are emitted. Key
differences of the Selexol process relative to amine units include:
5
Gerhard Ranke, “Advantages of the Rectisol-Wash Process in Selective H2S Removal from Gas
Mixtures,” office communication, 30 January 2005.
6
Gas Processes 2004, Hydrocarbon Processing, Gulf Publishing Company, 2004.
Task 2: Gas Cleanup Design and Cost Estimates 6
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
• Different acid gas and off-gas compositions. Since CO2 has lower
solubility than H2S, the flash drum off-gas will usually contain a greater
amount of CO2. The acid gas stream will also contain a greater amount of
mercaptans and heavy hydrocarbons, if these species are present in the
process gas stream.
• Greater electricity use. The higher pressures and lower temperatures
usually seen in a Selexol system require additional process electricity.
• Lower steam use. In general, physical solvents are easier to regenerate
in the stripper column, making the steam requirements lower.
As with amine units, treatment processes for the waste streams generated are
standard commercial equipment. Sulfur recovery system configurations can vary
due to the different acid gas stream compositions.
RECTISOL
While many different configurations of the Rectisol process exist, the flow
scheme is, in general, similar to that displayed in Figure 2. The typical difference
is an additional column or process unit to aid in the regeneration of the chilled
methanol. In addition, refrigeration is required to keep the methanol solvent
between –40 and –80°F for process use. These additional steps add greater
cost and complexity to the sulfur removal scheme.
The German firm Lurgi GmbH first developed the Rectisol process back in the
1950’s. They remain the main licenser of the technology for commercial
applications. Over 50 units have been licensed and built worldwide7.
Gasification plants with processes that are sensitive to sulfur, such as ammonia,
methanol, or Fischer-Tropsch synthesis reactors, have employed Rectisol
technology. While typically higher in cost than other sulfur removal technologies,
7
Information from Lurgi, available at www.lurgi.de.
Task 2: Gas Cleanup Design and Cost Estimates 7
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
any process looking to obtain very deep sulfur removal will usually consider
applying the Rectisol process.
Environmental Footprint
The direct and indirect emissions from the Rectisol process are similar to that of
Selexol, with main two differences for indirect emissions. The first is due to the
use of a compressor to circulate refrigerant to keep the process cold. The
addition of this unit to the Rectisol process adds greater electricity requirements
to the overall process. In addition, some Rectisol process designs require
multiple absorber or stripper columns for contaminant removal and solvent
regeneration. This will lead to greater steam and power requirements relative to
other treatment processes.
The technologies available for sulfur removal are not limited to the ones outlined
above. Two major commercially available technologies use a mix of chemical
and physical sorbents to take advantage of the unique properties inherent in
each type of solvent. Each process uses a proprietary hindered amine/physical
solvent blend to allow selective removal of H2S and other contaminants, such as
organic sulfur compounds and CO2.
The first is Sulfinol, developed and licensed by Shell. Since the 1960’s over 200
Sulfinol units have been installed and operated worldwide8. Sulfinol technology
has been used in natural gas and refinery applications for selective H2S removal.
The other is FLEXSORB, developed by ExxonMobil in the 1970’s. Similar in
nature to Sulfinol, roughly 50 commercial plants are currently in operation9.
8
Hydrocarbon Processing, ibid.
9
Fedich, R., Woerner, A., and Chitnis, G., “Selective H2S Removal”, Hydrocarbon Engineering, Volume
9, Number 5, May 2004.
Task 2: Gas Cleanup Design and Cost Estimates 8
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
There does not appear to be any physical solvents in commercial operation for
the removal of large amounts of tars, although Dow has stated that they have
successfully removed “trace” amounts of aromatics from syngas streams with
Selexol10. While aromatics are very soluble in physical solvents, aliphatic
hydrocarbons have a much lower solubility—C6 compounds (hexanes) are
roughly as soluble as H2S, but this factor quickly drops off when dealing with
lighter hydrocarbons. For example, methane is roughly 100 times less soluble
than H2S, and ethane is roughly 20 times less soluble.
Using physical solvents in this mode would not be a course of action that is
recommended at this time. More detailed investigation could be performed,
requiring vendor input and small scale testing. There are a number of concerns
about using physical solvents for this application:
1. The syngas stream would need to be cooled to ~100°F or lower for feed to
the physical solvent system. Cooling down the gas coming right out of the
gasifier could condense oils, chlorides, ammonia, fatty acids, alkali
species, and others which could create plugging problems, especially if
particulates were present (due to agglomeration). Typical syngas streams
from biomass gasification have a tar dew point at around ~450-500°F.
One potential way to reduce the plugging risk would be to partially cool
above the dew point, then have a water wash/quench to remove water
soluble compounds. While this would reduce the plugging potential, the
condensate would contain a considerable amount of dissolved material.
This step would also likely lead to a lower process efficiency than using
catalytic tar cracking.
3. High concentrations of tars and oils in the acid gas stream would create
problems for a sulfur recovery unit, including black sulfur and VOC
emissions from the incinerator. It is not clear at this time how much
experimental data exists for removing a high concentration of tars from the
physical solvents during the stripping step, although Lurgi has done some
work in this area11.
10
Jotaro Itoh, Dow Chemical Japan, “Chemical and Physical Absorption of CO2”, RITE International
Seminar, 14 January 2005.
11
Dennis Leppin, Gas Technology Institute, office communication, June 2005.
Task 2: Gas Cleanup Design and Cost Estimates 9
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
4. Hydrocarbons in physical solvents have caused significant foaming and
operational problems in the past, especially with Selexol. While some
steps can be taken to minimize this, it can become a major operational
issue.
5. It is unclear due to limited data how removing large quantities of tars and
oils will impact the ability of the physical solvent to remove H2S. As the
physical solvent becomes more loaded (with preferential solubility to
aromatic compounds over H2S and CO2), greater recycle rates and larger
vessels would likely be necessary.
Catalytic absorbents, such as zinc oxide (ZnO) or copper oxide (CuO), are often
used as a polishing step for sulfur removal in gas streams where the incoming
sulfur level is 10 to 20 ppm or lower. For hydrocarbon and IGCC applications,
catalytic absorbents are usually not a primary sulfur removal process. They are
typically applied downstream of either a chemical or physical solvent to assure
the sulfur level entering other process equipment, such as chemical synthesis
reactors, is very low. Since ZnO is the technology most commonly deployed, it
will be focused upon here.
H2S in a gas stream reacts with the ZnO absorbent in a fixed bed reactor, where
it is converted to zinc sulfide. According to technology providers, sulfur removal
down to 20 to 50 ppb is attainable with ZnO12. Since the sulfur specifications for
alcohols and FT liquids are <1 ppm in the syngas stream, ZnO could be used to
assist in achieving these requirements.
While catalysts that could be regenerated were used for many years prior to the
application of the current breed of catalysts, regulations limiting air sulfur
emissions have spurred the use of hot ZnO catalysts that are disposed of after
their useful lifetime. Spent catalyst usually contains a sulfur content that is low
enough to enable it to be landfilled or used as an agricultural fertilizer.
12
Considerable information in this section provided by Ronald Huber, Sud-Chemie Technical Service
Engineer, personal correspondence, 2 June 2005.
Task 2: Gas Cleanup Design and Cost Estimates 10
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
The main rationale for the use of ZnO catalysts is for systems where very low
sulfur content is a key priority. For chemical synthesis, these systems can be
placed downstream of amine or Selexol units to act as a secondary removal step
(a “polishing” step), or downstream of a Rectisol unit (a “guard” step) to prevent
sulfur breakthrough. The lifespan of the catalyst between changeouts is longer in
guard operation than as a polishing step.
A catalyst bed for polishing or guard sulfur removal duty is a conventional unit
process that has been effectively been applied for decades. Processes for
chemical synthesis from natural gas or syngas, such as ammonia, methanol, and
FT liquids production, have all employed catalytic absorbents. The current hot
oxide catalysts have been recently employed since the 1980’s, with the current
number in service estimated from the high hundreds to the low thousands. Major
vendors for this technology include Sud-Chemie, BASF, Synetix, Unicat, and
Haldor Topsoe.
Environmental Footprint
Removing sulfur from a natural gas or syngas process stream is only part of the
story. The residual sulfur present in an acid gas stream must then be recovered
to prevent environmental and safety harms, as well as meet operator permit
requirements. Two main technologies are commercially available to recover
sulfur: the Claus process (partial combustion) for high levels of sulfur, and
catalytic Redox processes, for relatively low levels of sulfur.
CLAUS PROCESS
Technology Description
In the Claus process, a high H2S concentration stream is the feedstock for
recovery to elemental sulfur. Roughly 1/3rd of the H2S is burnt (partial
combustion) to form sulfur dioxide (SO2). The remaining H2S reacts with the
synthesized SO2 over an alumina or bauxite catalyst to produce elemental sulfur.
Depending on their concentrations, the unreacted components (tail gas), such as
Task 2: Gas Cleanup Design and Cost Estimates 11
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
residual SO2, CO2, and H2S, are either emitted, thermally oxidized, or further
treated in an additional recovery process. A graphical depiction of the process
can be seen in Figure 3.
The Claus process is the oldest commercial process covered in this paper, with
development dating back to the late 19th century. Today, Claus processes are
the main step used for elemental sulfur production worldwide—in fact, 90 to 95%
of the sulfur recovered in the United States was from the Claus process. 38
companies operate 108 Claus processes in the United States, recovering nearly
9 million tons per year of sulfur14. The petroleum and natural gas industries are
the main users of the technology, with IGCC applications making up a small but
growing segment of the user population.
13
US Environmental Protection Agency, AP42, 5th Edition, “Compilation of Air Pollutant Emissions
Factors Volume 1: Stationary Point and Area Sources, 1995.
14
US Geologic Survey, Mineral Commodity Summary, Sulfur, January 2005.
Task 2: Gas Cleanup Design and Cost Estimates 12
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
In operations where the sulfur recovery is more than 20 TPD, a Claus unit is
generally the most economical approach. For efficient process operation, the
concentration of H2S in the acid gas stream entering the Claus unit should be a
minimum of ~40%; lower concentrations can be handled with some modifications
to the process including acid gas bypass of the reaction furnace and the addition
of oxygen to the air used for combustion.
Environmental Footprint
The main direct outlet streams from the Claus process are elemental sulfur,
spent catalyst, tail gas, and wastewater. The sulfur is of very high purity, and can
be used for industrial processes such as the production of sulfuric acid. The tail
gas stream contains the major emissions from the process. Thermal oxidation of
this stream could lead to emissions of SO2, CO2, CO, NOx, and other pollutants.
When a tailgas treatment unit is added, the emissions of sulfur oxides are
reduced and overall sulfur recovery can be 99% or more. The wastewater
generated from the sulfur condensers typically requires downstream treatment in
a standard treatment unit.
The Claus process has a relatively small amount of indirect emissions coming
predominantly from the number of reheaters used to maintain process heat. The
majority of the process heat comes from the exothermic reactions involving H2S.
LO-CAT® is an oxidation process that uses iron catalyst held in a chelating agent
to oxidize H2S to elemental sulfur15. H2S is the only acid gas being removed in
this process. A LO-CAT® process consists of 3 sections that include an
absorber, an oxidizer for catalyst regeneration, and a sulfur handling unit. Figure
4 illustrates a typical LO-CAT® unit. When the gas stream comes in contact with
15
Information provided by GTP/Merichem, personal communications and www.gtp-merichem.com
Task 2: Gas Cleanup Design and Cost Estimates 13
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
the LO-CAT® solution in the absorber, H2S in the gas stream is converted to
elemental sulfur. The spent catalyst along with the elemental sulfur exits the
absorber and then enters the oxidizer where the spent catalyst is regenerated by
contact with oxygen. The elemental sulfur is then concentrated into a sulfur
slurry, which moves to the sulfur handling unit to recover any entrained catalyst
solution.
As mentioned above, GTP and LGI are the two active manufacturers of catalytic
sulfur removal technology. GTP first commercialized the LO-CAT® technology in
1978, with the SulFerox process coming to market in the early 1980’s. 300 to
400 units are currently in operation in a number of different industries that require
recovery of low levels of sulfur16. These include natural gas sweetening, syngas
operations, air purification, biogas/landfill gas units, and refinery “off-gas”
streams.
Environmental Footprint
Due to the simpler make-up of the catalytic recovery process relative to the Claus
process, the direct and indirect emissions are lower. Similar to the Claus
process, sulfur, tail gas, and spent catalysts are all process products. The sulfur
removed from this process is not directly suitable for industrial use; the product is
a “cake” that contains up to 70% water. The sulfur recovered contains a small
amount of entrained residual catalyst and is considered low-value sulfur that is
suitable for agricultural purposes but is undesirable as a chemical feedstock.
16
Combination of information provided by both GTP/Merichem and Shell.
Task 2: Gas Cleanup Design and Cost Estimates 14
Subtask 2.3: Sulfur Primer
United States Department of Energy/National Renewable Energy Laboratory
The catalyst can either be landfilled or used as an agricultural fertilizer. Finally,
the tail gas stream, depending on composition, can either be vented or burned in
a thermal oxidizer. While this stream may still be a source of SO2, CO2, and CO,
the concentrations tend to be lower than in a Claus process.
16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
OF ABSTRACT OF PAGES
a. REPORT b. ABSTRACT c. THIS PAGE
Unclassified Unclassified Unclassified UL
19b. TELEPHONE NUMBER (Include area code)
F1146-E(12/2004)