Hydrocarbon Processing Copying and distributing are prohibited without permission of the publisher
Email a friend
  • Please enter a maximum of 5 recipients. Use ; to separate more than one email address.

Reduce CO2 in acid gas from amine-based TGTUs

03.01.2012  |  Spooner, B.,  Sulphur Experts Inc., Kemah, TexasDerakhshan, F.,  Sulphur Experts, Calgary, Alberta, Canada

Improve furnace temperature and sulfur recovery

Keywords: [tail gas treating] [amine] [CO2] [furnace] [sulfur recovery]

The purpose of an amine-based tail gas treating unit (TGTU) is to recycle any leftover sulfur components in the tail gas of a Claus sulfur recovery unit (SRU) to the front end of the plant, rather than incinerate them. The remaining sulfur components are converted to hydrogen sulfide (H2S) and removed from the gas using an amine solution.

Carbon dioxide (CO2) is also present in tail gas streams and will be partially co-absorbed with H2S. This CO2 co-absorption should be minimized. Any CO2 removed in the TGTU will be recycled to the SRU front end along with the H2S, which has negative consequences for the reaction furnace and for the overall sulfur recovery.

This article discusses the effect of CO2 on sulfur plants and how to minimize CO2 co-absorption or maximize CO2 slip through the TGTU’s amine absorber. A well-designed methyl diethanolamine (MDEA) TGTU should be able to achieve a minimum CO2 slip of 85%.

Effects of CO2 on the sulfur plant.

Slip is maximized through the TGTU to prevent CO2 from re-entering the sulfur plant. There are several reasons why it is advantageous to minimize CO2 in the SRU feed, the most important involving the reaction furnace temperature, carbonyl sulfide (COS) and carbon disulfide (CS2) formation, and sulfur plant capacity, as outlined below.

Reaction furnace temperature. CO2 does not burn in the reaction furnace and will, therefore, lower the furnace temperature. This will have negative effects on ammonia and aromatic destruction, which will, in turn, have consequences for the converter beds and the overall sulfur recovery.

Maintaining reaction furnace temperature is critical in sustaining smooth and trouble-free operation of a sulfur plant. When processing sour water stripper gas, ammonia destruction is of primary concern, and reaction furnace temperatures of 2,250°F or higher are required to avoid ammonia salt formation and plugging of downstream process equipment.

When processing acid gases containing aromatic hydrocarbons, reaction furnace temperatures of 1,920°F or higher are required to properly combust the benzene, toluene and xylene (BTX) components. Incomplete combustion of these compounds can lead to poisoning of the downstream catalyst bed. Finally, to maintain stable flame operation, a reaction furnace temperature of 1,700°F is required.

Although some cracking can occur in the reaction furnace, CO2 is generally considered an inert compound, and it cools the reaction furnace flame. As seen in Fig. 1, the dilution of an acid gas stream with increasing amounts of CO2 results in a rapid reduction of the reaction furnace temperature.

Fig. 1 shows that, for a simple acid gas containing varying amounts of H2S and CO2 and 1% hydrocarbons, significant problems with maintaining the required reaction furnace temperature start to occur at very low concentrations of CO2 in the amine acid gas (AAG). With a CO2 content of less than 5% (based on this example), it becomes difficult to maintain the required reaction furnace temperature for good ammonia destruction. At a CO2 content of 30%, a breakthrough of benzene, toluene, ethylbenzene and xylenes (BTEX) could start to occur; and with a CO2 content of over 50%, problems with maintaining a reaction furnace flame become pronounced.


  Fig. 1.  Calculated adiabatic reaction
  furnace temperatures.   

Although not necessarily detrimental to Claus plant operation, higher CO2 content in the acid gas will require modifications to the operation of the reaction furnace in order to achieve and maintain the required operating temperature. Sometimes, these modifications come at a high capital investment cost or at the detriment of the overall efficiency. These items could include the addition of air and feed gas preheaters, oxygen enrichment, split-flow reaction furnace designs, installation of high intensity or other specialty burners, co-firing with fuel gas or acid gas enrichment.

COS formation. Although the chemistry inside a reaction furnace is quite complex and chaotic, one certainty is that CO2 will partially “crack” in the furnace, resulting in the formation of CO, COS and (indirectly, through a drop in temperature) CS2. Fig. 2 shows the impact of CO2 on COS and CS2 formation rates, using the same simplified acid gas composition discussed previously.

As shown in Fig. 2, the increasing CO2 concentration of the acid gas results in ever-increasing formation rates of CS2. CS2 formation has been shown to decrease with increasing temperatures, although it is not clear if less CS2 is formed at higher operating temperatures or if CS2 is formed and then quickly hydrolyzed. However, the addition of CO2 cools the reaction furnace temperatures, which results in increased CS2 formation. CS2 is especially unwanted since it binds up two sulfur molecules.


  Fig. 2.  COS and CS2 formation as a
  function of CO2 content in acid gas.   

COS and CS2 formation in the reaction furnace is important because, once formed, COS and CS2 do not participate in the modified Claus reaction. These compounds must, therefore, be converted or hydrolyzed back to H2S downstream of the furnace, either in the first Claus converter bed (operating at high temperatures, typically 600°F to 630°F) and/or utilizing special and often expensive catalyst; or they can be converted back to H2S in the TGTU hydrogenation reactor. Since amine-based TGTUs do not pick up COS and CS2 in the absorber, any unconverted COS and CS2 will result in increased sulfur emissions and reduced recovery efficiencies.

Overall gas capacity. In a world of low-sulfur fuels, the demand for raising the existing processing capacity of refinery sulfur plants is becoming increasingly important. Since CO2 flows straight through the sulfur plant and does not participate in the Claus or modified Claus reactions, it takes up space and reduces the amount of sulfur-bearing gases that could otherwise be processed. A sulfur plant already operating at or slightly above its design capacity would quickly run into trouble with an increase in the CO2 content of the feed gas.

Apart from the issues related to furnace temperature and increased COS and CS2 formation, operating a sulfur plant at higher-than-design throughput would create greater operating pressures that could result in:

• Problems with air blowers not being able to deliver sufficient air at the higher operating pressures

• Heat exchangers not being able to sufficiently cool process gases, resulting in additional sulfur vapor losses

• Problems with mass velocities through condenser tubes, resulting in liquid sulfur carryover (both of which would require additional reducing gases in the TGTU hydrogenation bed).

To address some of the reduced capacity issues, processing companies would have to consider technologies such as oxygen enrichment or the construction of additional/larger sulfur plants, which would have a significant impact on the operating or capital budget of refineries.

Effects of CO2 on TGTU quench water.

 High CO2 tail gas streams serve to lower the pH of the quench water, which can decrease the strength of the protective iron sulfide (FeS) film. The water will, therefore, be darker in color and could cause fouling and plugging of the quench system. Ideally, the quench water has a pH of 7 to 8; however, in plants with high CO2 levels, the pH is typically between 6 and 7. Some plants try to correct this by regular caustic addition to the (partly) circulating quench water.

Since a significant quantity of the quench water is not recycled, caustic injection is, at best, a temporary solution. Regular caustic injection can result in strong pH fluctuations that destabilize the protective FeS film.

CO2 removal with MDEA.

CO2 does not react directly with the MDEA molecule; instead, it dissolves and reacts in the water portion of the solution:

CO2 + H2O t H2CO3 (carbonic acid)

H2CO3 t H+ + HCO3 (bicarbonate)

H+ + R1R2R3N t R1R2R3NH+


CO2 + H2O + R1R2R3N t R1R2R3NH+ + HCO3

The reaction between CO2 and water (carbonic acid formation) is a “slow” step; it takes time to occur. Once the carbonic acid is formed, however, the MDEA reacts with it quite quickly, and the bond will not be broken again until the amine is regenerated. The removal of CO2 with a tertiary amine like MDEA is, therefore, kinetically limited by the reaction rate in the first step.

H2S, on the other hand, reacts directly with the MDEA molecule:

H2S + R1R2R3N t R1R2R3NH+ + HS

The reaction between H2S and the amine is a very fast or instantaneous reaction, which means that H2S removal is almost always equilibrium-limited. Each contact stage or tray in an absorber reaches the H2S equilibrium between gas and liquid.

The difference in chemistry between H2S and CO2 removal is the key to understanding how H2S can be removed with a minimum amount of CO2. The best strategy to minimize CO2 removal with MDEA is to prevent the CO2/water reaction from occurring by:
1. Optimizing the amine temperature
2. Optimizing the amine strength
3. Optimizing the amine circulation rate
4. Optimizing the amine feed point of the absorber
5. Choosing a more selective solvent—e.g., formulated MDEA or sterically hindered amine.

Amine temperature. As with most chemical reactions, the higher the temperature, the faster the CO2 reaction takes place. In a TGTU, lower lean amine temperature will minimize CO2 pickup because it reduces the reaction between CO2 and water. It should be noted that high amine temperatures (above 140°F) will also slip amounts of CO2. Unfortunately, at high temperatures, H2S will also be slipped due to equilibrium limitations. Therefore, operating the absorber at high temperatures cannot be used as an operating strategy.

A low absorber temperature slows down the kinetics of the CO2 reaction. H2S removal occurs through a different mechanism that is much less affected by temperature. Therefore, the amine temperature should be kept as low as possible, although typically not lower than the inlet gas temperature.

Since the gas is coming from the TGTU quench tower, it is saturated with water. An amine temperature lower than the gas temperature would condense water and dilute the amine solvent. It should be noted that the normal temperature guideline for amine systems—maintain 5°F to 10°F temperature difference between gas and amine—is not applicable for amine-based TGTUs. Normal operation is to maintain the lean amine temperature at or within 1°F to 2°F of the inlet gas. An ideal temperature range for both the amine and the gas is 90°F to 100°F.

Amine strength. Since water is one of the reactants, the more water there is in the absorber, the more CO2 will be absorbed into the amine solution. MDEA can be successfully operated at up to 50% strength, with the other 50% consisting of water. Above this strength, the viscosity of the solution becomes too high and will negatively affect the mass transfer of H2S into the amine.

The CO2 hydrolysis should be minimized by limiting the water content of the MDEA solution. Solvent strength should be maintained between 45% and 50% while utilizing a proper filtration program to ensure that the viscosity of the amine does not rise too high. Minimizing SO2 breakthrough from the Shell Claus Offgas Treatment (SCOT) or SCOT-type cobalt-molybdenum (CoMo) reactor will reduce the heat-stable amine salt buildup in the system and minimize the chance of a viscosity increase due to excess heat-stable salt formation.

Amine circulation rate. The CO2 reaction with water (to form bicarbonate) is a kinetically limited reaction. This means that CO2 builds up at the gas/liquid interface and only reacts with the water and amine as it slowly diffuses to the bulk of the amine solution.

In a trayed tower, higher circulation rates increase the height of the liquid on each tray, as the weir creates a flow obstruction. The amine “stacks up” against it and provides a larger surface area for CO2 absorption. In a packed tower, greater amine flows create higher holdup of amine into each section of packing. The more liquid in the tower, the higher the CO2 removal will be, as there is more gas/liquid interface for the CO2 to be absorbed. Over-circulation of the amine is the single largest contributor to poor CO2 slip in TGTUs.

Furthermore, CO2 increases the loading of the amine, which takes up valuable acid gas holding capacity, especially in low-pressure TGTU applications. Under certain conditions, it is possible that CO2 will force the amine to release previously absorbed H2S. Optimizing (i.e., reducing) the amine circulation rate will always result in a decrease in H2S and an increase in CO2 in the treated gas, which is the goal.

In Fig. 3, the effects of circulation rate are shown. As the amine rate is decreased from 200 gallons per minute (gpm) to 150 gpm, the CO2 in the treated gas increases by 1,000 ppm, whereas the H2S decreases from 45 ppm to 40 ppm.


  Fig. 3.  Effect of an increase in circulation
  rate on H2S and CO2 removal.  

The circulation rate should be initially targeted for a rich H2S loading of 0.05 mol/mol, and slowly decreased with a final target loading of 0.1 mol/mol. This should be reinforced by the use of a reliable amine plant simulator.

Amine feedpoint into absorber. TGTU absorbers often have multiple inlet points for the lean amine solution. Several inlet points are normal and can be used to increase or decrease the interfacial contact area between the gas and the amine solution. A shorter contact time (less interfacial contact area) will result in less CO2 absorption.

The contact time between the CO2 and amine depends on the internals of the absorber and the height at which the amine is injected. In a trayed absorber, contact time on each tray is determined by the weir height and the number of trays. If packing is used, contact time depends on the height and size of the packing. In either case, lowering the injection point of the lean amine will reduce the contact time and interfacial contact area between the gas and amine, thus increasing the amount of CO2 slip.

Neither the weir height of trays nor the total height of packing is a parameter that can be adjusted during normal operation. Changes to either require shutting down the system and performing large-scale maintenance. This is why multiple feedpoints are built into the design of most TGTU absorbers.

It should be noted that differences exist in the selectivity of packed and trayed towers. These differences are due to the hydraulics of the internals and the corresponding relationship with mass transfer. A trayed tower has the liquid phase, which is highly agitated; packed towers are opposite in that the liquid flows over the packing relatively smoothly. The gas flows are turbulent in both. Either type of internals can result in decent CO2 slip, but the choice between the two must be carefully considered and researched in the design stage. Since the net liquid holdup in a packed bed (1% to 6% of tower volume) is lower than in a trayed tower (8% to 12% of tower volume), the lower liquid hold-up can result in less interfacial contact area and, therefore, lower CO2 absorption.

A TGTU absorber typically has three lean amine injection points: one at the top tray or top height of packing, and two more at successively lower intervals. Balance is achieved by injecting at a high enough point to remove all of the H2S necessary, but no higher. This procedure is best predicted first on a simulator, as shown in Fig. 4. Using this type of chart, it is apparent where appropriate injection points into the absorber exist, depending on the required H2S specification.


  Fig. 4.  Simulated acid gas removal
  tray-by-tray in a TGTU absorber.   

If possible, the amine injection should be lowered by one feedpoint, and the treated gas H2S content should be measured. If it is acceptable and simulations agree, the amine injection can be lowered to another feedpoint. The main concern here is the H2S content of the treated gas.

Choosing a more selective solvent.

The use of an amine with high selectivity can engender a number of benefits for the TGTU, as discussed in the following two options.

Hindered amine. The most selective solvent for TGTU applications is a hindered amine, which is normally a secondary amine with a bulky group that hinders the direct reaction with CO2. These molecules combine the low CO2 reaction rate of a tertiary amine with the base strength of a secondary amine. The high base strength is particularly useful at the low pressure of a TGTU absorber because it makes it possible to have a significantly higher rich amine loading than with MDEA. The higher loading allows for a reduction in amine circulation rate, which further improves the selectivity.

Formulated MDEA. Enhanced MDEA formulations for TGTUs have been available for 25 years. These are normally the solvents that contain “pH suppressants” such as phosphoric acid. The reduction in pH allows for easier and deeper regeneration of the amine, especially for H2S. The same effect of improved regeneration is achieved in MDEA solutions containing between 0.5 wt% and 1.0 wt% heat-stable salts.

By lowering the H2S lean loading, the H2S in the treated gas will also drop. This could enable the lean amine to be introduced into a lower feedpoint in the absorber, thus increasing the CO2 slip, as described above. Normally, however, the aim of the improved regeneration is to obtain a lower H2S level in the treated gas. TGTUs are, therefore, often designed with more trays in the regenerator than what is normally seen in amine units; this allows for deep H2S stripping. HP

The authors 

  Ben Spooner, a senior process engineer, has been working in the amine industry as an operator and engineer since 1998. He joined Amine Experts in 2003 and has worked in over 25 countries and on hundreds of amine systems, providing expert assistance and advice regarding plant operations, troubleshooting, optimization and operator training. Mr. Spooner is one of the primary speakers at Amine Experts’ world-recognized Amine Treating Seminar, which has been presented in dozens of locations around the globe. He holds a BSc degree in petroleum engineering from the University of Alberta. 

  Farsin Derakhshan is a professional engineer with over 16 years of experience. He joined Sulphur Experts in 1996, initially working out of the German office, and since then he has been directly involved in all aspects of Sulphur Experts’ process engineering consulting work. Now residing in Canada, Mr. Derakhshan is an experienced and well-traveled sulfur plant engineer, providing expert advice and consulting services to clients around the world. His specialty areas include sulfur plant troubleshooting and process optimization. Mr. Derakhshan is also Sulphur Experts’ regional engineer for Europe and the Middle East. He is technically responsible for all projects in those regions, and he is also a guest speaker at the internationally recognized Sulphur Recovery seminars. He holds a BSc degree in mechanical engineering from the University of Calgary. 

Have your say
  • All comments are subject to editorial review.
    All fields are compulsory.

Rajive bansal

One of the positive feature of lowered temperature is reduction in nox formation.
Also, addition of so2 from tgtu will reduce air requirement and in turn reduce nox formation.

Related articles


Boxscore Database

A searchable database of project activity in the global hydrocarbon processing industry


Should the US allow exports of crude oil? (At present, US companies can export refined products derived from crude but not the raw crude itself.)



View previous results

Popular Searches

Please read our Term and Conditions and Privacy Policy before using the site. All material subject to strictly enforced copyright laws.
© 2013 Hydrocarbon Processing. © 2013 Gulf Publishing Company.