Sunday, January 12, 2020

State-of-the-art Assessment of the Research in Post-combustion CO2 Capture With Chemical Absorption


Global concentration of CO2 in the atmosphere is increasing rapidly. CO2 emissions  have an impact on  global climate change. Effective CO2 emission abatement strategies such as Carbon Capture and Storage (CCS) are required to combat this trend. There are three major approaches for CCS: Post-combustion capture, Pre-combustion capture and Oxyfuel  process. Post-combustion  capture offers  some advantages  as existing combustion technologies can still be used without radical changes on them. This makes post- combustion capture easier to implement as a retrofit option (to existing power plants) compared to the other two approaches. Therefore, post-combustion capture is probably the first technology that will be deployed.  This  paper aims to provide a state-of-the-art assessment of  the research work  carried out so   far in post-combustion capture with chemical absorption. The  technology  will  be  introduced  first, followed by required preparation of flue gas from power plants to use this technology. The important research programmes worldwide and the experimental studies based  on  pilot plants  will  be  reviewed. This is followed by an overview of various studies based on modelling and simulation. Key issues such as energy consumption and plant flexibility will be included. Then the focus is turned to review development   of different solvents and process intensification. Based on these, we try to  predict  challenges  and  potential new developments from different aspects such as new solvents, pilot plants, process heat integration (to improve efficiency), modelling and simulation, process intensification and  government  policy impact.

1. Introduction

1.1 Background

CO2 is the main greenhouse gas. CO2 emissions have an impact on global climate change. Global concentrations of CO2 in the atmosphere have  increased  from  pre-industrialisation  levels  of approximately 280 parts per million by volume (ppmv) in around 1860 to approximately 316 ppmv in 1958 and rapidly to approximately 369 ppmv  today (UNEP, 2005). Global  CO2 concentration is predicted to rise  to above 750 ppmv by 2100 if no action is taken to address the current situation.

Power generation from fossil fuel-fired power plants (e.g. coal and natural gas)  is  the single largest source of CO2 emissions (Freund, 2003). However, fossil fuel fired power plants play a vital role in meeting energy demands. For instance, coal-fired power plants could be operated flexibly in  meeting  with  varying  demand. With growing concerns over the increasing atmospheric concentration of anthropogenic greenhouse gases, effective CO2 emission abatement strategies such as Carbon Capture and Storage (CCS) are required to combat this trend.

CCS is a “process consisting of the separation of CO2  from industrial and energy-related sources, transport   to a storage location and long-term isolation from the atmosphere.”  (IPCC, 2005).  From  this  definition,  CCS consists of three basic stages: (a) Separation of CO2; (b) Transportation and (c) Storage. There are three major approaches for CCS: Post-combustion capture, Pre-combustion capture and Oxyfuel process (IPCC, 2005).

Post-combustion capture offers some advantages as existing combustion technologies can still be used without radical changes on them. This makes post-combustion capture easier to implement as a retrofit option to existing plants compared to the other two approaches. The advantage comes at the expense of  the efficiency of the power generation process. The separation stage (i.e. CO2 capture) is energy intensive and as such largely determines the cost of the CCS process. It represents about 75-80% of the total cost of CCS (Davison, 2007).

A number of separation technologies could be employed with post-combustion capture. These include: (a) adsorption; (b) physical absorption; (c) chemical absorption; (d) cryogenics separation and (e) membranes (see, for example, IPCC, 2005). Figure 1 shows  classification of various separation technologies available    for post combustion capture. A brief introduction of these technologies will set the background of this review paper.

Figure 1 Process technologies for post combustion CO2 capture adapted from (Rao and Rubin, 2002)

1.1.1 Adsorption

Adsorption is a physical process that involves the attachment of a gas or liquid to a solid surface. The adsorbent is regenerated by the application  of heat (temperature swing adsorption, TSA) or the reduction  of pressure (pressure swing adsorption, PSA). Adsorbents which could be applied to CO2 capture include activated carbon, alumina, metallic oxides and zeolites (IEA GHG, 1993, Zhao et al, 2007).

Current adsorption systems may not be suitable for application in large-scale power plant flue gas treatment. At such scale, the low adsorption capacity of most available adsorbents may pose significant challenges. In addition, the flue gas streams to be treated must have high CO2 concentrations  because of the generally low selectivity of most  available adsorbents. For instance,  zeolites  have a  stronger affinity  for water vapour. (IEA 2004, IEA 2007, Zhao et al, 2007)

1.1.2 Physical absorption

This involves the physical absorption of CO2 into a solvent based on Henry’s law. Regeneration can be achieved by using heat, pressure reduction or both. Absorption takes place at high  CO2 partial  pressures.  As such, the main energy requirements originate from the flue gas pressurization. Physical absorption is therefore not economical for gas streams with CO2 partial pressures lower than 15vol% (Chakravati et al, 2001, IEA, 2004). Typical solvents are Selexol (dimethyl ethers of polyethylene glycol) and Rectisol (methanol) (IEA GHG, 1993).

1.1.3 Cryogenics separation

Cryogenics separation separates CO2 from the flue gas stream by condensation. At atmospheric pressure, CO2 condenses at -56.6°C (IEA GHG, 1993). This physical process is suitable for  treating flue gas streams  with high CO2 concentrations considering the costs of refrigeration. This is also used for CO2 capture for oxyfuel process.

1.1.4 Membrane absorption

When membranes are used in gas absorption, membranes act as contacting devices between the  gas stream and the liquid solvent. The membrane may or may not provide additional selectivity. These offer some advantages over the conventional contacting devices such as packed columns as they are more compact and are not susceptible to flooding, entrainment, channelling or foaming. They, however, require that the pressures on the liquid and gas sides are equal to enable CO2 transport across  the  membrane. Their separation efficiency depends on the CO2 partial pressure. As such, they are suitable for high CO2- concentration applications (well above 20vol%) such as flue gas streams from oxyfuel and IGCC processes. (Favre 2007, IEA GHG 1993, IPCC, 2005).

1.1.5 Membrane-based separation

In membrane-based separation, selectivity is provided by the membranes themselves.  These  usually  consist of thin polymeric films and separate mixtures based on the relative rates at which constituent species permeate. Permeation rates would differ based on the relative sizes of the molecules or diffusion coefficients in the membrane material. The driving force for the permeation is the difference in partial pressure of the components at either side of the membrane. However, the selectivity of this separation process is low and thus a fraction of the CO2 is captured. In addition, the purity of the captured CO2 is low  for the same reason (IEA, 2004, IEA GHG, 1993). Multistage separation is employed to capture a higher proportion of CO2 incurring extra capital and operating cost (Chakravati et al, 2001, IEA, 2004, IEA GHG, 1993).

1.1.6 Chemical absorption

Chemical absorption involves the reaction of CO2 with a chemical solvent to form a weakly bonded intermediate compound which may be regenerated with the application of heat producing the original solvent and a CO2 stream (IPCC, 2005). The selectivity of this form of separation is relatively high.  In addition, a relatively pure CO2 stream could be produced. These factors make chemical absorption well suited for CO2 capture for industrial flue gases. More details will be described in Section 2.

1.2 Aim of this paper and its novelty

This paper aims to provide a state-of-the-art assessment of the research work carried out so far in post- combustion capture with chemical absorption. For beginners, this paper will give an introduction to the technology and a summary of literatures. For experienced researchers, this paper will review the recent progress and predict the future research directions based on the available achievements.

The differences between this paper and previous review reports such as IPCC (2005) and Davidson (2007) are: (a)  This paper is to provide an update of important  research programs  worldwide and major pilot  plant studies; (b) This paper is to provide a critical review of research  activities  in  modelling  and simulation; (c) This paper is to draw the readers’ attention on process intensification for post-combustion capture with chemical absorption; (d) This paper tries to predict challenges and future breakthroughs.

1.3 Outline of the paper

In Section 2, the technology is introduced. Then major research programs worldwide and experimental studies based on pilot plants are reviewed in Section 3. This is followed by an overview of various studies based on modelling and simulation in Section 4. Section 5 is a survey of various solvents and relevant problems such as degradation. Section 6 discusses new development in process intensification for post- combustion CO2 capture. In Section 7, we try to predict challenges  ahead  and  potential  new  developments from different aspects such as new solvents, pilot plants, heat integration (to improve efficiency), modelling and simulation, and process intensification. Conclusions are drawn in the end.

2. Post-combustion CO2 Capture with Chemical Absorption

Post-combustion CO2 Capture means that CO2 is removed after combustion of the fossil fuel.  In  other words, CO2 is captured from flue gases at power plants or other large point sources. The most commonly used solvent is monoethanolamine (MEA).

2.1 Preparation of the flue gas to use the technology

Prior to absorption, other acid gases such as SO2 and NO2 must be removed as  they  affect  the  performance of the system by  forming heat stable salts with solvent such as MEA. SO2 concentrations of  less than 10ppm are recommended (Davidson, 2007). SO2 removal is usually achieved in a Flue Gas Desulphurization (FGD) unit. NOx is removed using Selective Catalytic Reduction (SCR), Selective Noncatalytic Reduction (SCNR) or low NOx burners. Particulate matter such as fly ash is removed by either electrostatic precipitators (ESP) or bag house filters otherwise they would cause foaming in the absorber  and regenerator columns decreasing their performance.

The presence of oxygen increases the likelihood of corrosion in equipment. In addition, alkanolamines like MEA can easily be degraded in the presence of oxygen (Davidson, 2007). Oxygen levels of less than 1ppm are recommended for use with MEA when corrosion inhibitors are not employed (IEA GHG,  1993). The  Fluor Daniel ECONAMINETM Process makes use of inhibitors (IPCC, 2005).

Flue gases to the CO2 absorber must be cooled between 45–50°C (Rao et al. 2007, Ramezan et al. 2007).  This would improve absorption of CO2 and minimize solvent losses due to evaporation. This is achieved in     a direct contact cooler (DCC) where the flue gas is  cooled by a  spray of water,  this in  addition saturates  the flue gas to the absorber and thereby helps the water balance.

2.2 Process description

The conventional MEA absorption process is displayed in Figure 2 (IPCC, 2005). The cooled flue gas is contacted counter-currently with the lean solvent usually of CO2 loading of about 0.1–0.2 mol  CO2/mol  MEA yielding a rich solvent of about 0.4–0.5 mol CO2/mol MEA loading (Freguia and Rochelle, 2003). The scrubbed gas is then water washed of solvent and vented to the atmosphere. The lean solvent gradually heats up as it absorbs CO2. The temperature  inside the absorber is typically between 40°C and 60°C. The  rich solvent is heated in a cross heat exchanger by regenerated lean solvent from the Stripper (or regenerator). It is then pumped to the top of the Stripper where it is  regenerated  at  elevated  temperatures (100°C-120°C) and at slightly higher than atmospheric pressure (1.5 – 2 atm) (IPCC, 2005). Heat is supplied via the reboiler which is the major energy penalty of  the  process.  The  regenerated  solvent is then pumped back to the Absorber via the cross heat exchanger to reduce the temperature.

Figure 2 Process flow diagram for CO2 capture from flue gas by chemical absorption (IPCC, 2005)

Amine solvents like MEA degrade on contact with certain impurities such as  excess  oxygen,  sulphur  dioxide or nitrous oxides to form substances including heat stable salts. Part of the bottoms product from the stripper is sent to a reclaimer unit where the solvent is evaporated and returned leaving the non-  volatile solvent wastes which are purged from the system.

3. Important Research Programmes worldwide & Pilot Plants

This section seeks to give an overview of important research programs worldwide in post-combustion CO2 capture.

3.1 Luminant Carbon Management program

3.1.1 Participants and Purpose

The Luminant Carbon Management Program led by Professor Gary Rochelle in  the  Department  of  Chemical Engineering, University of Texas at Austin (USA) focuses on the technical obstacles to the deployment of post-combustion CO2 separation from flue gas by alkanolamine absorption/stripping and integrating the design of the capture process with aquifer storage/enhanced oil recovery processes (Rochelle, 2010)

3.1.2 Description of facilities

The schematic of the pilot plant facility is shown in Figure 3. Both the absorber and stripper columns are packed columns with internal diameters of 0.427m and total column height of 11m. Columns  consist  of  two 3.05m packed bed sections with a collector plate and redistributor between the  beds.  Both random and structured packings were used alternatively in the two columns. The facility has a capacity to process approximately 3 tons of CO2 per day (Dugas, 2006). The flue gas stream is prepared since it is not obtained directly from a power plant.

3.1.3 Activities

At this university pilot plant, several  studies  are carried out where  research looks into CO2 rate kinetics  and solubility measurements (Bishnoi and Rochelle, 2000), degradation of solvents (Chi and  Rochelle  (2002), Davis and Rochelle (2009)), systems modelling (Freguia and Rochelle (2003), Ziaii et al. 2009), pilot plant testing (Dugas (2006), Chen et al. (2006)) amongst others.

One of the main projects carried out was the Carbon Dioxide Capture by Absorption with Potassium Carbonate. This project ran from 2002 to 2007 with the aim to improve the process for CO2 capture by alkanolamine absorption/stripping by developing an alternative solvent, aqueous Potassium Carbonate (K2CO3) promoted by piperazine (NETL, 2008). Accomplishments include evaluation of three  solvents  – MEA, and two variants of the piperazine-promoted K2CO3. It was shown that the requirements for one of  the piperazine promoted K2CO3 solvents was much less than the conventional MEA due to increased absorption capacity and rates, as well as reduced heat of absorption – implying reduced regeneration requirements. A number of studies were also carried out on the process performance in terms of packing performance and absorber/stripper configurations among others. A  rate-based  model  of  the  absorber unit was developed. Absorber intercooling was found to improve absorption performance especially  for high absorption capacity solvents. (NETL, 2008)

Figure 3 Schematic of CO2 capture pilot plant from (Dugas, 2006)

3.2 International Test Centre (ITC) for CO2 capture

3.2.1 Participants and Purpose

The ITC is formed by the collaboration of University of Regina (Canada) led by Professor Malcolm Wilson and a consortium of government, industrial partners and its aim is to explore and develop new  cost effective technologies for CO2 capture.

3.2.2 Description  of facilities

Infrastructure in the ITC consists of bench-scale CO2 separation units as well as a multi-purpose pilot plant unit (1 ton of CO2 per day) at the University of Regina. A 250kW steam  boiler is used to generate the flue  gas which is then treated in a CO2 absorption unit. The absorption column is composed of three 0.3 m- diameter sections for a total height of 10 m.

In  2000, the ITC re-commissioned a semi-commercial  (4 ton of CO2 per day)  demonstration unit adjacent  to SaskPower’s 875 MW Boundary Dam power station. The unit captures CO2 from  part  of  the flue gas from this coal-fired power plant  (Wilson et  al, 2004). This facility consists of three units in series as shown  in Figure 4:

i.A baghouse unit for flyash removal
ii.A scrubbing unit for removal of SO2 down to 2ppm.
iii.Chemical absorption-based CO recovery unit (Fluor’s Econamine FGSM technology).

Figure 4 Schematic of Boundary Dam CO2 Pilot Plant (Wilson et al, 2004)

3.2.3 Activities

Kinetics of the reactive absorption of carbon dioxide in high CO2-loaded, concentrated aqueous MEA solutions were studied in Aboudheir, et al. (2003) by experiments. Kinetics of the reactive absorption of carbon dioxide with mixed solvents MEA and MDEA were again studied in Edali et al. (2009). Pilot plant studies of the CO2 capture performance of aqueous  MEA and mixed MEA/MDEA solvents at the University of Regina CO2 capture technology development plant and the Boundary Dam CO2 capture demonstration plant were compared in Idem et al. (2006). Uyanga and Idem (2007) studied the degradation of  solvent  MEA in  the presence of SO2. Kittel  et  al. (2009) studied corrosion of  MEA unit for CO2 capture through  pilot plant experiment.


3.3.1 Participants and Purpose

This European Commission-funded and IFP-run project involves  capturing  and  providing  geological  storage for 30% of the emissions released by large industrial facilities around Europe (conventional power stations, principally), i.e. for 10% of Europe’s CO2 emissions. CASTOR, which started in February 2004, was a 4-year program and counts members from 11 EU countries, including: (a) 16 industrial  firms  (Dong Energy, Vattenfall, Repsol, Statoil, Gaz de France, Rohoel, Alstom power, RWE, etc); (b) 12 research  institutes (IFP, BGRM, Imperial College, TNO, BGS, IFP, etc.). The CASTOR project had a total budget €15.8 million with a contribution of €8.5 million from European Commission (FP6 – sixth framework program) (IEAGHG, 2010).

Its specific goals involve halving the cost of capturing (from €40-60/ton CO2 to €20-30/ton CO2) and separating CO2, developing the  geological-storage  concept’s  efficiency, safety and security while limiting  its environmental impact, and testing it in real-life, industrial-scale facilities.

3.3.2 Description of facilities

Figure 5 Simplified flow diagram of the CASTOR pilot plant at Esbjergværket (Knudsen et al, 2009)

A gas-fired mini plant with full absorption/desorption cycle was built at the University of Stuttgart. This facility consists of  an absorber with  0.125m diameter absorber and 4m  height and a  2.5m high stripper.  An industrial-scale pilot plant facility (Figure 5) was launched alongside a power plant run by Dong Energy (formerly ELSAM) in Esbjerg, Denmark on 15 Mar. 2006. This plant has a capacity to capture  about  24tonnes of CO2 per day.

3.3.3 Activities

Studies were carried out on the selection of solvents as well as solvent degradation for the larger CASTOR pilot plant (Notz et al, 2007). Since Mar. 2006, four 1000-hour test campaigns have been carried out with MEA solvent and new proprietary solvents – CASTOR-1 and CASTOR-2. In January to February 2006, a 1000-hour preliminary test campaign was conducted using 30wt% MEA being the reference solvent.  Another 1000-hour test was repeated from mid December 2006 to February 2007 to improve on certain problems encountered in the first test whilst collecting data (Knudsen et al, 2009). Results show that it is possible to run  the post combustion plant continuously  whilst achieving roughly  90% CO2 capture levels. In addition, one of the proprietary solvents, CASTOR-2, operated with lower steam requirement and L/G ratio than the conventional MEA solvent.

From CASTOR project, it was concluded that future investigations would involve tests  of  solvents  developed in the EU CESAR project. In addition, the effect  of  process  modifications  on  steam requirements as well as environmental effects would be investigated (Knudsen et al, 2009).

3.4 iCap

iCap is a 4-year project supported by the European Commission under the 7th Framework Program. It consists of a consortium of 15 partners  including eight Research &  Technical Development providers and  six power companies as well as an Australian research institute and a Chinese university.  The  project started on January 1, 2010.

iCap aims to remove barriers that cause bottlenecks in post-combustion and pre-combustion CO2 capture. Targets include halving the efficiency penalty of CO2 capture for power plants and reducing the associated CO2 avoidance cost to 15€/tonne CO2. These aims should accelerate the commercial development of large scale near zero emission power generation technology based on CCS (iCap, 2010).


CAPRICE is funded by the European Union and was scheduled to last two years. This project began on 1 January 2007. TNO, a Dutch organization, is running it. CAPRICE stands for CO2 capture  using  Amine  Process International  Cooperation and Exchange and involves  pooling information and research findings   on amine-enabled CO2 capture with non-European CSLF countries.

More specifically, findings from the European CASTOR project’s MEA chapter will be compared with those  of the ITC at the University of Regina (Canada). It counts: (a) 10 research centres (University of Regina, Alberta Research Council, ITC, Energy Inet, IFP, Trondheim University, Stuttgart University, Tsinghua University, Topchiev Institute of Petrochemical Synthesis, and Salvador  University);  (b)  3  power  generation companies (E-ON, Dong Energy and Vattenfall) (CAPRICE, 2010).


3.6.1 Participants and Purpose

This  4-year long FP7 (Seventh Framework  Programme)-funded  project was launched in 2008 and aims for  a breakthrough in the development of low-cost post-combustion CO2 capture technology to provide economically feasible solutions for both new power plants and retrofit of existing power plants which are responsible for the majority of all anthropogenic CO2 emissions. CESAR focuses on post-combustion as it is the only feasible technology for retrofit and current power plant technology. The primary objective is to decrease the cost of capture  down to 15  €/ton CO2.  The  consortium  consists of 3 research  organizations, 3 universities, 1 solvent supplier, 1 membrane producer (SME), 3 equipment suppliers, 2 oil and gas companies and 6 power generators (= industrial commitment) (CO2cesar, 2010) .

3.6.2 Description of facilities

This project employs the Esberg pilot plant used in the CASTOR project.

3.6.3 Activities

Novel activities and innovations CESAR focuses at are (CO2cesar, 2010):
  • Novel (hybrid) solvent systems
  • New high flux membranes contactors
  • Improved modelling and integration studies on system and plant level
  • Testing of new solvents and plant modifications in the Esbjerg pilot plant. In the  Esbjerg  Pilot  Plant novel technologies  are  assessed and compared with mainstream techniques to provide a  fast track towards further scale-up and demonstration.
3.7 The joint UK-China Near Zero Emissions Coal (NZEC) initiative

The project is managed by UK consultants AEA (for the Department Energy and Climate Change in the UK),  in partnership with the Administrative Centre for China's  Agenda 21 (ACCA21), and involves  a consortium  of 28 industrial and academic partners from the UK and China. Its objectives are (a) knowledge sharing     and capacity building; (b) future technology perspectives; (c) case studies for carbon dioxide capture; (d) carbon dioxide storage potential; and (5) policy assessment.  The Project started  in 2007 and was finished  at the end of 2009 (NZEC, 2009).

3.8 The Cooperative Research Centre for Greenhouse Gas Technologies (CO2CRC)

3.8.1 Participants and Purpose

CO2CRC comprises participants mainly from Australia and New Zealand. CO2CRC works through 3 main research programmes in CCS: Capture research, Storage research, Demonstration and pilot projects. The programme commenced in July 2003 and was expected to last 7 years. It has since been extended up till 2015 (CO2CRC, 2009a).  Capture research  involves a  number of post  combustion capture technologies  such as solvent-based systems, innovative membrane systems and pressure swing  adsorption  (PST) systems.

3.8.2 Description of facilities

CO2CRC employs a 28-metre high absorber column in its International Power Capture Plant built at the Hazelwood coal-fired power station. The plant can capture up to 50 tonnes of CO2 per day and  is  the  largest post-combustion capture plant at a power plant in Australia (CO2CRC, 2009a).

3.8.3 Activities

The Latrobe Valley (post-combustion capture) project involving CO2CRC, Loy Yang Power, International Power and Australia's Commonwealth Scientific and Industrial Research Organisation (CSIRO) consists of a $5.6 million PCC research project with the aim of reducing CO2 emissions from brown coal power stations. Research on three post combustion capture technologies – solvents, membranes and adsorbents is being carried out.

The CO2CRC H3 Capture Project, based at Hazelwood power station, utilises the International Power  Capture Plant. At present, new solvents (such as BASF PuraTreatTM)  (CO2CRC,  2009b)  are  being  researched with studies conducted at industrial scale. These studies generally relate to  investigating  process and energy efficiency improvements. The International Power Capture plant is used for these studies and can capture up to 50 tonnes of CO2 per day. (CO2CRC, 2009a)

3.9 Investigations at Mitsubishi Heavy Industries, Hiroshima R&D Centre Japan

3.9.1 Participants and Purpose

The Mitsubishi Heavy Industries (MHI)  in  conjunction with the Kansai  Electric Power  Company (KEPCO)  has conducted research and development of capturing CO2 since 1990 (Kishimoto et al, 2009). They have developed a sterically hindered amine KS-1 solvent which has been shown to have less heat requirements than the conventional MEA solvent as well as reduced solvent degradation. This solvent is used in their proprietary KM-CDR process (Kishimoto et al, 2009). MHI has deployed four commercial CO2 recovery  plants which recover CO2 from natural gas fired flue gas. Tests have been carried out  on  a  1  tonne  CO2/day as well and recently on a 10 tonne CO2/day pilot plant to demonstrate that the process is  applicable to coal-fired power plants as well (Davidson, 2007, Kishimoto et al, 2009).

3.9.2 Description of facilities

MHI has successfully deployed four commercial CO2 capture plants, currently operating in Malaysia, Japan and two locations in India. The latter two plants have the highest capacity of 450 tonnes CO2/day. These plants operate in the chemical and fertilizer industries which recover CO2 from natural gas fired flue gas (Kishimoto et al, 2009).

A 1 ton CO2/day pilot plant has been constructed in the MHI Hiroshima R&D centre to demonstrate CO2 capture from coal-fired flue gas and to carry out various tests for the treatment of impurities (Davidson, 2007).

In Japan, a testing plant capable of recovering 10 tons of CO2 per day using MHI's KM-CDR process was reported to have operated at a coal-fired power generation plant in Nagasaki – the Matsushima Thermal Power Station of Electric Power Development Co., Ltd. (J-POWER) for over 4000 hours (Kishimoto et al, 2009).

3.9.3 Activities

Currently, MHI is working to resolve certain key issues it has identified with the deployment of post combustion CO2 capture (Kishimoto et al, 2009):
  1. Reduction in energy consumption of such plants
  2. Efficient integration with other environmental control systems
  3. Minimizing the decrease in net electrical output of the power plant
3.10 Summary

Although many research programs or investigations have been carried out globally, all the pilot plants for post-combustion CO2 capture with MEA are small scale (for power plants up to 2.0 MWe). Consequently,  the packed column sizes are modest in size (up to 1.1 m in diameter) (Feron et al., 2007, Dugas, 2006). Therefore, post-combustion CO2 capture with chemical absorption (although viewed as best technology available) has still not been fully demonstrated.

4. Modelling, Simulation and relevant studies

4.1 Level of complexity in modelling

In modelling absorbers and stripper, two approaches are commonly used: the equilibrium-based approach   and the rate-based approach. The former approach assumes a theoretical stage in which liquid  and  gas  phases attain equilibrium and the performance of each stage is adjusted using a tray efficiency correction factor (Schneider et al., 1999). This is usually sufficient to model  non-reactive  systems.  In  amine absorption, chemical reactions are involved and such equilibrium is rarely attained. As such, the rate-based approach is more appropriate. In the rate-based approach, actual rates of multi-component mass and heat transfer as well as chemical reactions are considered directly (Noeres, 2003).

To model such a reactive absorption process,  simple or complex representations can be used for mass  transfer and reaction aspects. Differences between various forms of models are indicated in Figure 6 (Kenig  et al., 2001).

Figure 6 Different levels of reactive absorption model complexity (Kenig et al., 2001)

Starting from Model 1 in Figure 6, the packed column is modelled as a distillation column containing different equilibrium stages. Moving rightwards (Model 2), the model accuracy is increased by considering the bulk phase reaction kinetics. Moving upwards (Models 3, 4 and 5),  the  mass  transfer  rate  is  considered. Thus, the models are called rate-based models. The mass transfer across the gas-liquid interface can be described using the two-film theory. At its lowest level of complexity (Model 3 in Figure 6), the chemical reactions of the rate-based model are assumed to be at equilibrium. Model 3 can be accurate only when the reaction rate between CO2 and the solvent is very fast. In Model 4, an enhancement factor is used   to estimate actual absorption rates. However, the enhancement factor used is strictly valid for the pseudo first-order reaction regime (Kucka et al., 2003). Chemical reactions are assumed to  be completed  in the liquid film while the bulk fluid remains in chemical equilibrium. Model 5, the most complex  of  all,  considers mass transfer resistances, electrolyte thermodynamics, the reaction system as well as the column configurations. With Model 5, the acceleration of mass transfer due to reactions in the liquid film is taken  into account (Kucka et al., 2003). Therefore, Model 5 is the most accurate and will give more realistic predictions.

4.2 Basic Theory behind modelling - Two-film theory and Penetration theory

To describe such a process accurately, it is necessary to develop mathematical models taking into account the column hydraulics, mass transfer resistances and reactions. The influence of chemical reactions on mass transfer cannot be neglected. From Figure 7, the model will include ideally mixed vapour and liquid bulk phases and two film regions adjacent to the interface.

Figure 7 Modelling the conventional process based on the two-film theory

The two-film theory assumes that the liquid and vapour phases both consist of film and bulk regions. Heat  and mass transfer resistances are assumed to be restricted to these laminar film regions (Danckwerts, 1970). The penetration theory originally proposed by Higbie assumes that every element on the surface  of  the  liquid is exposed to the vapour phase for the same length of time, before it is replaced by liquid of the bulk composition. The exposure time  encompasses the effects of the hydrodynamic properties of the system and  is used to define their effect on the mass transfer coefficient (Danckwerts, 1970).

4.3 Current status of modelling Post-combustion with CO2 capture

The performances of the Absorber and the Stripper, the two major components in the conventional CO2 capture process, have been studied by a number of researchers through modelling and simulation.

In Lawal et al. (2009a), a dynamic rigorous model was developed for the Absorber.  This  assumed  rate- based mass transfer with reactions at equilibrium (Model 3). Process analysis based on this  model found    that the Absorber performance can be maintained during part load operation by maintaining the ratio of the flowrates of the lean solvent and flue gas to the Absorber. In Kvamsdal et al. (2009), the Absorber was modelled dynamically assuming rate-based mass transfer and counting the impact of reaction with an enhancement factor (Model 4 in Figure 6). The dynamic model of  Absorber  was then used to investigate  two transient operation scenarios: start-up and load-reduction. The authors also pointed out that a dynamic model for the whole CO2 capture process (i.e. with the stripper and heat exchange units) is required to evaluate different operational challenges.

In Lawal et al. (2009b), a dynamic model for the Stripper was developed. This assumed rate-based mass transfer and reaction at equilibrium (Model 3). It was used to analyse the impact of the reboiler duty on the CO2 loadings in the solvent at the bottom of the Stripper. In Ziaii et al. (2009), only the Stripper was  modelled dynamically using rate-based mass transfer and assuming reaction at equilibrium (Model 3  in Figure 6). This model was subsequently used to minimise the energy consumption of the Stripper.

The limitation of the publications so far (Lawal et al. (2009a), Kvamsdal et al. (2009), Ziaii et al. (2009)     and Lawal et al. (2009b)) is that process dynamic analysis was carried out with individual dynamic models  for Absorber or Stripper independently without considering their possible interaction when  operation  together as a plant.

In Lawal et al. (2010), dynamic models for the absorber and regenerator columns were developed (Model 3  in Figure 6). The gPROMS (Process Systems Enterprise Ltd.)  advanced  process  modelling environment  was used. These models were then linked together with recycle (including heat exchanger). A study of the dynamic responses of a post-combustion CO2 capture plant was carried out based on modelling and simulation. The study gives insights into the operation of the absorber-regenerator  combination  with  possible disturbances arising from integrated operation with a power generation plant. The importance of appropriate water balance in the absorber column is shown.

All the above models were developed based on the two-film theory (for mass  transfer  calculations).  Tobiesen et al. (2010) developed a rigorous absorber model based on penetration theory (for mass transfer calculation). Experimental data from a laboratory pilot plant absorber were used for model validation. The model is fairly accurate.

In summary, an accurate dynamic model of the whole CO2 capture process is required to study the start-up, shutdown and operation under different power plant loads and process disturbances.

4.4 Studies based on modelling and simulation

Based on modelling (the two-film theory) and simulation, Aroonwilas and Veawab (2007) studied the performance of different amines such as MEA, MDEA and mixture of MEA/MDEA for CO2 capture in a 500 MW supercritical coal-fired power plants. It was found that when the MEA and MDEA are blended at the appropriate ratio and used as solvent for CO2 capture, the energy consumption for regenerating CO2 is reduced significantly. Compared with MEA only, the whole power plant with CO2 capture can improve thermal efficiency around 3%. However, this study was carried out  at steady  state  and when power  plant is operated at full load.

Chalmers and Gibbins (2007) studied post-combustion CO2 capture with MEA for pulverised coal power plant under part load operating conditions. This study was again carried out at steady state based on simplified models. It pointed out that further work is required to better understand transient behaviour of power plants with CO2 capture especially from one load changing to another.

Kvamsdal et al. (2010) studied a 450 MWe natural gas combined  cycle (NGCC)  sea  floating power  plant (for offshore use) with post-combustion CO2 capture with MEA. It is concluded that the NGCC  plant will have 9% efficiency penalty due to adding CCS with 90% CO2  captured and compression of  CO2  to 1.47  MPa. Another key issue of water balance for the whole CO2 capture process was solved by adjusting operating or design conditions.

4.5 Commercial products and studies based on commercial products

Aspen Plus® provides two rigorous multistage separation models for modelling absorber  and regenerator (or stripper). These are RadFrac and RateFrac sub-models (Aspentech,  2008).  RadFrac  model  is  equilibrium stage column model (equivalent as Model 2 in Figure 6). RateFrac model considers actual multi-component heat and mass transfer (equivalent as Model 3 or Model 5 in Figure 6 depending upon whether reaction kinetics is used).

Freguia and Rochelle (2003) used RateFrac to simulate the absorber and stripper. These were linked with heat exchanger to form a whole CO2 capture process. In the absorber, the reactions involving CO2 were described with kinetics. In the stripper model, all the reactions were set to equilibrium due to higher operating temperature. Effects of process design and operating variables (such as solvent circulating rate, absorber height, stripper height and stripper pressure)  on energy requirement were studied. Intercooling   in the absorber was also explored and found that the reboiler duty was reduced by 3.8% for inlet stream with 10 vol% CO2.

Kvamsdal and Rochelle (2008) used Aspen RateSepTM in Aspen Plus® process simulator (a  second  generation rate-based multistage separation unit operation) and a dynamic rate-based gPROMS model (Model 4 in Figure 6) to study the position of temperature bulge (which is caused by combing chemical absorption and water vaporisation) in the absorber and its impact on absorption of CO2.

Zhang et al. (2009) used Aspen RateSepTM in Aspen Plus® again for simulation study.  Experimental  data from the pilot plant tests in the University of Texas at Austin were used for model validation  of  the absorber only (Dugas, 2006). These experimental data include 48  runs  at 24 operating conditions  since  two runs were executed at each operating condition. It also discussed the  importance  of  film discretisation.

Dugas et al. (2009) used Aspen RateSepTM in Aspen Plus® to  simulate the  absorber.  Experimental  data from pilot plant tests in CASTOR project were used for model validation. It claimed that the model successfully represented both gas phase temperature and CO2 profiles in the absorber.

5. Solvents

The ideal chemical solvent possesses (Davidson, 2007):
  • High reactivity with respect to CO2 – which would reduce height requirements for the absorber and/or reduce solvent circulation flow rates.
  • Low regeneration cost requirements –based on a low heat of reaction with CO2
  • High absorption capacity – which directly influences solvent circulation flow rate requirements.
  • High thermal stability and reduced solvent degradation – reduced solvent waste due to thermal and chemical degradation
  • Low environmental impact
  • Low solvent costs – should be easy and cheap to produce
5.1 Amine-based solvents

Amines have been used for around 75 years for the treatment of industrial gas streams with the alkanolamines being the most popular group of solvents (Booth, 2005). Amines could be classified as primary, secondary or tertiary based on the degree of substitution of the nitrogen atom. MEA consists of one alkanol chain and two hydrogen atoms bonded to a nitrogen atom and thus  it  is  classified  as  a primary amine with the molecular formula (C2H4OH)NH2. Diethanolamine (DEA) consists of two alkanol chains and one hydrogen atom bonded to the nitrogen atom and is a secondary amine with the molecular formula (C2H4OH)2NH. In the same way, triethanolamine (TEA) is a tertiary  amine  (C2H4OH)3N  (Booth, 2005).

A variety of ethanolamine derivatives can be produced by replacing one or more of the ethanol groups by other hydrocarbon groups. An example of this is  the tertiary  amine methyldiethanolamine (MDEA)  where  a methyl group replaces one of the ethanol groups. MDEA has the molecular formula C2H4OH)2N(CH3) (Booth, 2005).

Primary and secondary alkanolamines react  rapidly with CO2  to form carbamates. Tertiary alkanolamines  do not possess a hydrogen atom attached to the nitrogen atom. They therefore facilitate  the  CO2  hydrolysis reaction to form bicarbonates. The heat of reaction involved with  bicarbonate  formation  is lower than that of carbamate formation and thus tertiary amines like MDEA are often  blended  with  primary or secondary amines to reduce solvent regeneration costs (Vaidya and Kenig, 2007).

Sterically hindered amines are primary and secondary amines modified to reduce regeneration costs. Examples include 2-Amino-2-methyl-1-propanol and 2-piperidineethanol (Vaidya and Kenig, 2007).
MEA and popular alternative solvents for chemical absorption are discussed.

5.2 MEA

Chemical absorption of CO2 is preferred for post combustion capture of CO2 from pulverized fuel power plants because it is able to capture CO2 in low partial pressures. MEA solvent is relatively cheap  and chemical absorption process with MEA is backed up by  commercially  available  and  proven  technology (Rao et al., 2004).

Davidson (2007) highlighted some problems encountered using MEA such as (a) degradation of solvents in the oxidising environment of flue gas; (b) energy consumption for regeneration of solvents; (c) corrosion. Alternative solvents to MEA should have higher capacity for CO2 capture and lower energy consumption.

5.2.1 Reaction kinetics of the reactive absorption of CO2

Vaidya and Kenig (2007) reviewed the reaction kinetics of CO2 absorption in alkanolamines. In the paper it was shown that there have been varied predictions of the reaction rate coefficients for absorption with MEA. Aboudheir et al. (2003) explained that these discrepancies are due to (among other factors) the assumption of a pseudo-first order reaction with respect to both CO2 and MEA. More detailed reaction mechanisms are needed to accurately describe the kinetics. Three mechanisms were described in Vaidya  and Kenig (2007):
  • Zwitterion mechanism
  • Termolecular mechanism
  • Base-Catalyzed Hydration Mechanism
5.2.2 Degradation of MEA solvent

Davidson (2007) discussed three main degradation routes: (a) carbamate polymerisation; (b) oxidative degradation; (c) thermal degradation.

Carbamate polymerisation is insignificant at temperatures below 100 °C. Thermal  degradation takes place  at temperatures above 205 °C. Most degradation comes from the presence of oxygen in the flue gas. Davidson (2007) explains that four carboxylic acids (formate, glycolate, oxalate and acetate) are major  amine degradation products while nitrites, nitrates and ethylenediamine were also found in significant quantities.

Sexton and Rochelle (2009) described catalysts and inhibitors  for MEA  oxidation. They carried out studies  at 55°C and found that dissolved metals catalyze the oxidation process in the order copper > chromium/nickel > iron > vanadium. They also identified effective degradation inhibitors such as ethylenediaminetetracetic acid (EDTA) and explained that  some  expected  inhibitors  such  as formaldehyde, formate and sodium sulphite actually increased MEA losses.

Davis and Rochelle (2009)  focused on the regeneration in the stripper unit. At 135°C the degradation  rate  is 2.5 to 6% per week. CO2 loading of the solvent was found to have a first order effect and amine concentration had a slightly higher than first order effect on increasing the degradation rate. The paper also suggests that MEA degradation is significantly reduced if temperatures are kept below 110°C.

5.2.3 Effects of Sulphur Dioxide (SO2)

SO2 reacts with MEA to form heat stable salts. By varying the SO2 concentrations in the range of 6  -  196ppm, Uyanga and Idem (2007) demonstrated that an increase in SO2 concentration would result in an increase in MEA degradation. Their study also suggests that an increase in CO2 loading in the liquid phase produced an inhibition  effect to MEA degradation because this would reduce the amount of SO2 and O2  that could react with the MEA solution to induce degradation. It may therefore  be  of  advantage  to  operate the absorption process with higher lean CO2 loading. However,  in  doing so,  consideration has  to be made regarding the corrosive effect of more CO2 in the system (Davidson, 2007).

5.2.4 Corrosion of MEA solvent

Davidson (2007) explained that factors that influenced corrosion rates in  amine  plants  include  CO2  loading, amine type and concentration, temperature, solution velocity and degradation products. MEA is quite corrosive compared to the secondary or tertiary amines used for gas treating. (Kittel et al, 2009). Corrosion is found to reduce in the following order MEA>AMP>DEA>MDEA (Davidson, 2007).

5.3 MEA and MDEA Blends

Other solvents used for CO2 chemical absorption include methyldiethanolamine (MDEA). MEA can react more quickly with CO2 than MDEA can, but MDEA has a higher CO2 absorption capacity and requires lower energy to regenerate CO2 (Davidson, 2007, Aroonwilas and Veawab, 2007).

Aroonwilas and Veawab (2007) studied the performance of different amines such as MEA and MDEA for coal-fired power plants. It was found that when the MEA and MDEA is mixed at the appropriate ratio and used as solvent for CO2 capture, the energy consumption for regenerating CO2 is reduced significantly. Compared with MEA only, the whole power plant with CO2  capture  can  improve  thermal  efficiency around 3%. However, this study was carried out at steady state and when power plant is operated at full load.

Other solvents include the sterically-hindered amines (KS-1, KS-2 and KS-3) developed by  the  Kansai  Electric Power Co. (Davidson, 2007). These solvents are claimed to offer lower energy consumption and solvent loss. These solvents however, have higher costs (Reddy et al., 2003).

5.4 Ammonia

Ammonia has been identified as a possible alternative to MEA solvent as it has a number of desirable characteristics. It is a relatively cheap solvent that is commercially available. It has a relatively high CO2 absorption capacity compared to most other solvents based (among other factors) on its low molecular weight. It absorbs CO2 with a low heat of reaction and thus the regeneration energy requirements are also low. It is not as corrosive as MEA and has a lower susceptibility to degradation in the presence of oxygen  and other contaminants when compared with MEA (Davidson (2007), Darde et al. (2010), Kozak et al.  (2009).

The Chilled Ammonia Process (CAP) is being developed for CO2 capture. In this process, CO2 is absorbed in the absorber at low temperatures. This would minimize solvent losses due to its relatively  volatile  compared with the amine solvents. First the flue gas is cooled to 0 to 20°C (preferably 0 – 10°C) it is then contacted in the absorber with lean solvent typically composed of 28wt% ammonia, CO2 loading between

0.25 and 0.67mol CO2/mol ammonia and water (Darde et al, 2010). The CO2 loading in the lean solvent should be high enough to prevent excessive solvent evaporation and low enough to maximize the capture efficiency of the plant (Darde et al, 2010). The CO2 rich stream is typically a slurry  as  solid  products (typically of ammonium bicarbonate) are formed. (Darde et al, 2010)

The stripper operating temperature ranges from 50 to 200°C (preferably 100 to 150°C) and the operating pressure from 2-136atm..This produces a high pressure stream rich in CO2. Water vapour and ammonia could be recovered by cold washing possibly  with a weak acid  for higher efficiency. Energy requirements  for regeneration are significantly lower than those for MEA absorption.

The Aqua Ammonia Process is another process that employs ammonia solvent and is proposed to capture SO2, NOx and CO2 from the flue gas (Davidson, 2007).

5.5 Piperazine promoted K2CO3

The promotion of potassium carbonate (K2CO3) with amines appears to be a particularly effective way to improve overall solvent performance. K2CO3 in solution with catalytic amounts of piperazine (PZ) has been shown to exhibit a fast absorption rate comparable to 30% wt% MEA. Heat of absorption is significantly lower than that for aqueous amine systems (Cullinane and Rochelle,  2005).  This could  translate  to 29-  33% regeneration energy savings for certain conditions when compared with MEA (Davidson, 2007).

5.6 Concentrated aqueous piperazine

Concentrated, aqueous piperazine (PZ) has been investigated as a novel amine solvent for CO2 chemical absorption. The CO2 absorption rate of aqueous PZ is much faster than MEA. Thermal degradation is negligible in concentrated aqueous PZ up to a temperature of 150 °C. This is a significant advantage over MEA systems. Industrial study showed that PZ will use 10 to 20% less energy than MEA (Freeman et al., 2010).

6. Process intensification (PI)

6.1 Motivations for using process intensification in post-combustion chemical absorption

It should be recognised that the flue gas volumetric flowrate involved for a typical coal-fired power plant is extremely large. For example, an electrical output of 500 MWe corresponds to a combustion  rate  of roughly 1500 MWh. Assuming that coal’s calorific value is based on the heat of formation of  CO2 and  with 10% excess air, the flue gas flow is roughly 550m3/s. Hence, with a flue duct velocity of  15  m/s in  order to limit flue pressure drops to reasonable values, the duct cross sectional area must be substantial (around 65m2). The CO2 chemical absorption system need only be a fairly  crude  affair  which  would provide, say, the equivalent of 2 to 3 transfer units which corresponds roughly to a 10 fold CO2 reduction. This is based on the assumption of a low CO2 partial pressure over the CO2 rich solvent flowing to the  stripper system.

With respect to the absorption stage, as shown above, large gas ducts are needed whether or not carbon capture is involved. It is, therefore, worth considering the installation of a simple liquid spray system to implement the absorption stage. This leaves us with the issue of the stripping duty and the lean/rich solvent cross heat exchanger where PI may be employed to good effect.

6.2 Principles and benefits

Process intensification can be achieved with a novel technology called Higee (an abbreviation for high gravity). This was first proposed by Ramshaw and Mallinson (1981) when they invented a rotating packed bed (RPB) to enhance distillation and absorption efficiencies. It takes advantage of centrifugal  fields  through RPB to generate high gravity, and therefore boosts the mass transfer coefficients, resulting in an order of magnitude reduction in equipment size.

PI is a strategy of making major (i.e. orders of magnitude) reductions in the volume of processing plant without compromising its production rate. This is achieved by reducing individual process elements and, where possible, combining process functions in multi-functional modules. In order to achieve this, new technology is exploited, often where high centrifugal acceleration fields and fine channels are used to intensify heat and mass transfer rates as described below.

The benefits of this approach are potentially profound:
  1. Piping and structural support costs can be dramatically reduced, leading to lower installation factors and reduced capital costs.
  2. Process inventories are slashed, giving improved intrinsic safety when toxic or flammable material is being treated.
  3. Residence times are reduced to seconds rather than the hours encountered in conventional plant. As a consequence, plant response times are much faster. This can be particularly important for reactors because it facilitates rapid grade changes and just-in-time production. This may allow cost savings to be realised by reducing warehouse inventories of  a multi- grade product.
  4. The improvement in heat/mass transfer coefficients may be used to reduce thermodynamically parasitic differences in temperature/concentration so that the process thermodynamic efficiency is improved.
  5. In some situations, a much smaller plant may be more compatible with the environment.
Approximately two thirds of process unit operations involve the contact of two or more phases. Thus  we  have boilers, condensers, absorbers and distillation columns etc. For these unit  operations,  the  fluid  dynamic intensity is controlled by the interfacial slip velocity which is  dependent on  the buoyancy force Δρ.g where Δρ is the interfacial density difference and g is the applied  acceleration.  For a given  system, this implies that higher levels of acceleration can result in greater flooding rates, enhanced settling and greater heat/mass coefficients. This key idea suggests that the operation of many process units in a centrifugal acceleration environment such as RPB or a spinning disc, could be a  powerful  means of realising PI. This has indeed proved to be the case, as has been shown in many publications (Ramshaw and Mallinson, 1981). Thus distillation or absorber heights of ~ 1.5cm (gas  film  controlled)  and  4 to  6cm (liquid film controlled) can be realised in a RPB while achieving very high mass fluxes in view of the enhanced flooding rates. Gas-liquid reactions can be performed in seconds on spinning discs provided the chemical kinetics are not limiting. Stripping and evaporation  can  be operated in one rotating element  as  they both benefit from the high acceleration field. This is a good example of a multi-functional application. This latter example is particularly relevant to the stripper stage of a carbon capture process.

When the process operation does not involve multiple phases and with relatively clean systems, it may be possible to exploit laminar flow in narrow (0.1 to 2 mm) channels as exemplified by the printed circuit heat exchanger/reactor marketed by Heatric Ltd. Very high transfer coefficients can be achieved in view of the short conduction lengths involved. As the narrow channels also permit large specific areas, the space-time performance of these units is impressive.

6.3 Recent progress

A recent laboratory study by Lin and Liu (2007) of CO2  absorption  has shown that the height of transfer   unit (HTU) for an 8 cm diameter RPB was 1 – 4 cm, at modest rotational speed of less than 1,000 rpm. The inlet CO2 concentration was 1% mole fraction of CO2-N2 mixture and NaOH (0.2 kmol/m3) was the absorbent. This compares with the 10-60 cm of conventional  packed beds for similar ratings. In a more  recent study by Cheng and Tan (2009) using high voidage packed beds, similar findings were reported with amine solutions. These studies indicate that process intensification technology has the potential to reduce   the size of CO2 capture plants and reduce both the capital and operating costs.

The standard concentration of the MEA solution used for carbon capture in conventional equipment  is 30wt%. This is largely determined by the need to limit solution viscosity and the corrosion potential with carbon steel construction. However, recent work by Jassim et al. (2009) using a RPB  and  stronger  solutions, has shown that much more efficient absorption can be achieved despite the higher solution viscosity. This effect is probably due to the accelerated chemical kinetics associated with MEA solution concentrations in the range 50-100wt%. A further advantage of this approach is that the CO2 loading/m3 of solvent solution is much higher,  thereby resulting in lower liquid  flows between the absorber  and stripper. In order to overcome corrosion problems, the rotating stripper (including reboiler) would have to be  fabricated in stainless steel. However, this would not present too much of a cost penalty in view of the small size of the unit compared to the carbon steel conventional columns. A comparison of the size of rotating units with their equivalent columns is shown in Figure 8 (Trent, 2004).

Figure 8 Three very small size RPBs equivalent to one large packed column (photo courtesy of The  Dow Chemical Company)

7. Challenges ahead and potential breakthroughs

7.1 Solvents

The characteristics of an ideal solvent are presented in  Section  5. In the same section, recent developments  in new solvents are also reviewed. The most commonly used solvent MEA in post-combustion chemical absorption of CO2 has the following disadvantages (Resnik et al., 2004):
  • Low carbon dioxide loading capacity (kg CO2 absorbed per kg solvent)
  • Solvent degradation due to SO2 and O2 in flue gas
  • High equipment corrosion rate
  • High energy consumption
Development of new solvents has achieved some progress such as ammonia (Darde et al., 2010), piperazine promoted K2CO3 (Cullinane and Rochelle, 2005), and concentrated aqueous piperazine (Freeman et  al., 2010). Future efforts should be directed towards developing better solvents. Future progress in solvents should be (a) to reduce energy consumption; (b) to avoid damage to environment and human being (when it  is vented with treated gas); (c) solvent degradation and corrosion to the packed column acceptable. More important is to combine all these in one solvent.

7.2 Pilot plants - the roadmap to commercial scale

A number of pilot plant scale studies have been carried out. Although these projects have so far  been  below 10MWe scale, their costs usually run to millions of dollars (Herzog et al, 2009). Processing flue gas from real fossil fuelled power plants has  been carried out  in a  number of plants by treating  a slip stream  of the flue gas produced. However, to evaluate the effect of the significant energy requirements of post combustion process, demonstration projects, with scales of 100s of MWe and likely costing over a billion dollars, are required (Herzog et al, 2009).

The first demonstration project may be in the UK resulting from the government’s  CCS  competition  (Herzog et al, 2009). The UK government CCS competition has finalised in Mar. 2010 that E.ON UK and Scottish Power will share a £90 million pot (BBC News, 2010). These two firms now compete to build the UK’s first CCS coal-fired power plant. The undisclosed amount of money each firm won, which has been drawn from the £90 million pot, will support engineering and design work for the CCS projects.

7.3 Modelling and Simulation

7.3.1 Dynamic validation

Currently all the dynamic models reported were only validated at  steady state  (rather than dynamically).  This is mainly due to lack of experimental data for CO2 capture pilot plants running at transient conditions. The dynamic validation is vital to gain understanding for operation and control design of such a process.

7.3.2 Use of commercial software for simulation study

Several studies were reported to study the post-combustion CO2 capture process for process design and operation (Freguia and Rochelle (2003), Kvamsdal and Rochelle (2008), Zhang et al. (2009), Dugas et al. (2009)). These studied were all conducted by experts in modelling and simulation. The challenge is how to make these commercial tools easy to use by practising engineers.

7.4 Process Intensification

As described in Section 6, it will be seen that PI has the potential to make large cost savings in a carbon capture process. It is recommended that the development of a rotating unit which combines the function of stripper including reboiler is actively considered.

7.5 To reduce energy consumption of CO2 capture

7.5.1 By developing new solvents

Generally for a given solvent, it can only save less than 10% of energy consumption by optimising design  and operating conditions (Freguia and Rochelle, 2003). However, a  new solvent  has a  potential to save 20  to 30% (even more) of energy consumption (Darde et al., 2010).

7.5.2 By process heat integration

When the temperature profile inside the absorber is plotted, it is easy to find a temperature bulge, which is caused by combing chemical absorption and water vaporisation. The temperature bulge is not helpful  for  CO2 chemical absorption. Freguia and Rochelle (2003) indicated that when the absorber deals with 3 vol% CO2 in the flue gas and MEA solvent is used, the temperature bulge is not significant. On the other hand, when the absorber deals with 10 vol% CO2 in the flue gas, the temperature bulge is significant. Intercooling means to take away heat from middle of the absorber to the reboiler in stripper (which has no effect on the rich end of the column). For the case with 10 vol% CO2 in  the flue  gas, reboiler  duty reduced  3.8%  (Freguia and Rochelle, 2003). Plaza et al. (2010) carried out a similar simulation study with Piperazine- promoted Potassium Carbonate as solvent which is intercooling in the absorber  with  different  configurations. It is also found that the effect of intercooling is related to the position of the temperature bulge.

Another way to reduce energy consumption of CO2 capture can be achieved by better process integration of CO2 capture plant with the power generation plant. Lucquiaud and Gibbins (2009) studied effective integration between these two parts and pointed out that with different solvent used in CO2 capture plant, there will be different requirements to steam pressure, temperature and flowrate extracted  from  the  crossover pipe between the intermediate pressure and low pressure turbines.

7.6 Simultaneous removal of SO2 and CO2

Resnik et al. (2004) carried out an experimental study to use ammonia as solvent to capture CO2, SO2 and NOx simultaneously from flue gas. The benefits are: (a) A single process to capture all the acidic gases is expected to reduce the total cost and complexity of emission control systems; (b)  The CO2 loading capacity by NH3 can approach 1.20 kg CO2 / kg NH3, while CO2 loading capacity by MEA is only 0.4 kg CO2 / kg MEA. In other words, ammonia’s CO2 loading capacity is three times that of MEA; (c) Cycling tests results demonstrated that a 64% reduction in regeneration energy is possible due to higher CO2 loading capacity of aqueous ammonia solution and lower heat of reaction.

7.7 Policy impact

It is sensible to assume that the government will legislate at some point so that retrofitting combined  cycle gas turbine (CCGT) plant with CO2 capture technologies is mandatory (The Press and Journal, 2010). Although capture efficiencies are low compared with coal fired applications, it begs the question of flue gas re-cycling on gas turbines to increase CO2 concentration and reduce NOx.

8. Conclusions

Post-combustion capture with chemical absorption is probably the first technology that will be deployed. A state-of-the-art assessment of the research work carried out so far in post-combustion capture with chemical absorption is provided in this paper. These include important research programs globally,  experimental studies with pilot plants, and studies through modelling and simulation. The use  of process intensification  has also been discussed. Based on these recent developments, we tried to predict future challenges and potential breakthroughs. More efforts in the future should be directed to reduce energy consumption in post-combustion CO2 capture with chemical absorption. Another concern is the solvent’s damage to the environment when it is vented with treated gas. The impact of government  policies  cannot  be  ignored before this post-combustion CO2 capture with chemical absorption process can be commercialised.

Source: M. Wanga A. Lawala, P. Stephensonb, J. Siddersb, C. Ramshawa and H. Yeunga


The 10 largest coal producers and exporters in Indonesia:

  1. Bumi Resouces
  2. Adaro Energy
  3. Indo Tambangraya Megah
  4. Bukit Asam
  5. Baramulti Sukses Sarana
  6. Harum Energy
  7. Mitrabara Adiperdana 
  8. Samindo Resources
  9. United Tractors
  10. Berau Coal

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...