Improvement of Lipophilic-Amine-based Thermomorphic Biphasic Solvent for Energy-Efficient Carbon Capture

Available online at www.sciencedirect.com

Energy Procedia 23 (2012) 92 – 101

TCCS-6

Improvement of lipophilic-amine-based thermomorphic biphasic solvent for energy-efficient carbon capture Jiafei Zhanga*, Yu Qiaoa, David W. Agara a

Faculty of Biochemical and Chemical Engineering; Technical University of Dortmund, Dortmund, D-44227, Germany

Abstract A series of novel lipophilic amine solvents have been studied for improving the energy efficiency of post-combustion carbon capture (PCC). More than thirty lipophilic amines with varying molecular structures, for instance primary amine Cycloheptylamine (CHpA), secondary amine Dibutylamine (DBA) and tertiary amine N-Ethylpiperidine (EPD), with respect to their phase separation temperatures, loading capacities, reaction kinetics and regenerabilities have been investigated by screening tests at temperatures between 25-90 °C. The further solvent selection was assessed by absorption at 30-40 °C with 50-500 mbar CO2 partial pressure and desorption at 75-85 °C with intensified regeneration measures in absence of steam stripping. Compared to monoethanolamine (MEA), selected lipophilic amines have performed excellently in CO2 loading, solvent regenerability and chemical stability. The low temperature (§80°C) required for desorption also permits the use of waste heat for reducing the operating cost.

©2012 2010The Published by Elsevier Ltd.Ltd. Selection peer-review responsibility [name © Authors. Published by Elsevier Selectionand/or and/or peer-review underunder responsibility of SINTEFofEnergi AS organizer] Keywords: post-combustion capture, biphasic solvent, CO2 absorption, novel technology

1. Introduction Solvent regeneration with high energy consumption at temperatures of 120-150 °C represents a significant shortcoming in the alkanolamine-based CO2 capture process. Development of novel absorbents and technologies to reduce the operational expense is thus essential for improving the process economics of PCC [1]. A chemical solvent with phase change potentially offers considerable advantages [2]. Since the absorbent is resolved into one phase highly concentrated in CO2 and another phase low in CO2, only the concentrated phase need be sent to the stripper. This is equivalent to a system operated with

* Corresponding author. Tel.: +49-231-755-2582; fax: +49-231-755-2698. E-mail address: [email protected].

1876-6102 © 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SINTEF Energi AS doi:10.1016/j.egypro.2012.06.072

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

93

an extremely high CO2 capacity. Concerning the advantages provided by phase change in desorption, a new CO2 capture technology DMXTM process has also been developed by IFP Energies nouvelles to reduce the operating cost by phase splitting in decanter before steam stripping, but the activating solvents are still undisclosed [3]. High temperature required for solvent regeneration not only leads to the high energy consumption but also significant thermal degradation of conventional alkanolamines. Thermomorphic biphasic solvent (TBS) systems, comprising lipophilic amines as absorbing and activating component, have been proposed to meet this challenge [4]. The advantages of using TBS absorbent for CO2 capture, especially for low temperature regeneration at 80 °C and their high cyclic CO2 capacities, have already been demonstrated [5,6]. The excellent performance of such TBS absorbent makes it one of the most promising solvent systems for assessing technical viability in ongoing scale-up development work. In the lipophilic amine systems, the secondary amine Dipropylamine (DPA) and tertiary amine N,NDimethylcyclohexylamine (DMCA) were firstly disclosed and extensively studied with respect to their absorption and desorption characteristics. However, the slow absorption rate of DMCA and precipitation of protonated DPA ions and bicarbonate salts were the drawbacks for those solvents. Investigation on searching new amines has therefore been carried out. The structure influence on the activity of aminebased CO2 absorption and desorption has also been studied by other researchers [7-11]. Those works have demonstrated that the amino group attaching an alkyl group especially with an alpha branched carbon was found to be the most potential molecular structure for enhancing the absorption and desorption rates as well as loading capacity. However, those studies were limited to the aqueous single-phase system. Screening investigation to identify new thermomorphic biphasic lipophilic amines was thus conducted both theoretically and experimentally in this project. 2. Experiment 2.1. Chemicals More than thirty lipophilic amines with varying molecular structures have been studied in the screening experiments; some of them representatively are listed in Table 1. All the amines tested in this work were purchased from Merck, Sigma Aldrich, Acros Organics, ALFA AESAR and ABCR with high purity (•97%). 2.2. Test tube screening experiments The preliminary screening experiment was carried out in test tubes (Schott GL-18 with 18 mm O.D. and 180 mm length) containing 6 mL aqueous amine solutions at varying concentrations from 0.9 to 4.2 M. PTFE tubules with 1 mm I.D. and 240 mm length were used as gas inlet pipe through the rubber seal. Absorption was performed in water bath at 25 °C with CO2 gas flow rate at 20 mL/min, saturated with water vapour in order to make up the vaporised water, under atmospheric pressure. The weight of the tube is measured every 5 or 10 min to observe the amount of absorbed CO2, until the reaction is completely in equilibrium. The final CO2 rich loading was determined by the barium chloride method with titration. Desorption was carried out by heating in thermal oil bath stepwise from 40 to 90 °C with interval of 10 °C. In order to limit solvent vaporisation, cotton sponge was applied as demister for condensing water and volatile amine vapours. Magnetic stirrers were laid into each test tube to intensify CO2 release and also in oil bath bowl to homogenise the temperature in the heating system. The weight was also measured stepwise for indicating the mass of desorbed CO2 and the final lean loading was determined by barium chloride method.

94

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

Table 1. List of studied lipophilic amines No.

Chemical

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Hexylamine Heptylamine Octylamine Di-n-propylamine Diisopropylamine N-Ethylbutylamine Di-n-butylamine Diisobutylamine N-sec-Butyl-n-propylamine Triethylamine Tripropylamine Tributylamine N,N-Diisopropyl methylamine N,N-Diisopropyl ethylamine N,N-Dimethyl butylamine N,N-Dimethyl octylamine

Abbr. Type Sub. HxA HpA OtA DPA DIPA EBA DBA DIBA SBPA TEA TPA TBA DIMA DIEA DMBtA DMOA

I I I II II II II II II III III III III III III III

L L L L B L L B B L L L B B L L

No. Chemical 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Cyclohexylamine Cycloheptylamine Cyclooctylamine N-Isopropyl cyclohexylamine N,N-Dimethyl Cyclohexyl amine N,N-Diethyl Cyclohexyl amine Dicyclohexylamine 2,6-Dimethyl piperidine 3,5-Dimethyl piperidine 2,2,6,6-Tetramethyl piperidine N-Methyl piperidine N-Ethyl piperidine Benzylamine N-Methyl benzylamine N-Ethyl benzylamine N,N-Dimethyl benzylamine

Abbr. CHA CHpA COA IPCA DMCA DECA DCHA 2,6-DMPD 3,5-DMPD TMPD MPD EPD BzA MBzA EBzA DMBzA

Type Sub. I I I II III III II II II II III III I II II III

O O O O O O O P P P P P Z Z Z Z

Type: I - primary amine, II - secondary amine, III - tertiary amine; Substitute (Sub.): L - liner chain, B - branched chain, O - cycloalkane, P - piperidine, Z - benzyl group.

2.3. Bubble column screening experiments The further screening experiments were conducted in a 100mL glass bubble column containing 40 mL of the aqueous amine solution at 30 °C. Various amines with concentrations from 3 to 5 M and CO2 partial pressures of 5-50 kPa balanced with N2 were employed over contact times of 2-5 hrs, to ensure that equilibrium was achieved. Desorption was carried out by magnetic agitation (PTFE coated round stir bar, 30 mm length & 9 mm diameter) 1000 rpm at 75-80 °C, without gas tripping. The feed and reference gas flows were saturated by water and regulated by mass flow controllers (Bronkhorst EL-FLOW) so as to be constant in the absorption and desorption tests. During experiments, the outlet gas was monitored online by gas chromatography (GC) Agilent HP 6890 with a GS-GasPro capillary column 30.0 m x 320 ȝm and oven temperature 35 °C. The detailed experimental setup is presented in a previous publication [5]. After the reaction had been completed, the CO2 loading was ascertained by barium chloride method, total amine concentration was determined by acid-base titration and the blended amine compositions were determined by GC (CP-Volamine capillary column 60.0 m x 320 ȝm and oven step-heating procedure with temperatures from 60 to 250 °C) analysis. 3. Selection of lipophilic amines The screening of lipophilic amine solvents initially focused on observing the miscibility of organic and aqueous phases and temperature dependent phase transition behaviour. Based on the theoretical study and experimental measure on the physical and chemical properties, amines with high melting point, low boiling point, high viscosity or low chemical stability were excluded. In total over thirty available lipophilic amines were preliminarily selected for test tube experiment. The overall screening among those amines is based on the comprehensive performance in absorption and regeneration, including phase separation temperature, CO2 capacity, reaction rate and regenerability.

95

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

3.1. Phase transition Phase change behaviours such as lower critical solution temperature (LCST) for lean solvent and liquid-liquid phase separation (LLPS) for rich solvent, observed in absorption and desorption, respectively, are the most important phenomena in this study. Due to restricted miscibility, the LCST behaviour were found in mixtures of lipophilic amine and water [12,13], typically between 10 and 40 °C. The aqueous solubility of amine decreases with increasing temperature and thus a phase splitting occurs above the LCST. The prepared lean solvent is desired homogeneous in absorption at 30-40 °C and the loaded solution should exhibit LLPS in regeneration at 75-85 °C. The LCST of most lipophilic amines measured in the experiment is found below 20 °C, which are not suitable to be used as primary absorbent. Only a few amines, for example HxA, CHpA, A1 (secondary alkylamine), MPD and EPD, have shown approved phase change characteristics. 3.2. Reaction rate The absorption rate can be influenced by aqueous solubility, surface tension and basicity of amine. In this experiment, aqueous solubility is the key factor; because the fast reactions were found in the lipophilic amine solutions which are completely or partially aqueous miscible at room temperature, while the slow reactions were observed in those that are aqueous insoluble or slightly soluble. According to the reaction mechanism, carbamate formation (CF, eq. 1) is the main reaction for primary and secondary amines and takes place very fast [14]. Fig. 1 demonstrates the rapid reaction rate has been obtained in amine solutions such as HxA, CHA, A1, DPA, DIPA, SBPA, DMPDs, etc. Therefore, the highly reactive primary or secondary amines are typically considered as activators for absorption; while the moderate reactive tertiary or secondary amines can be used as regeneration promoter due to their remarkable regenerability enhanced by LLPS behaviour. 1,0

1,0 0,8

0,6

0,4

DPA DIPA DBA DIBA SBPA

0,2

CO2 loading [mol/mol]

CO2 loading [mol/mol]

0,8 0,6

0,4

HxA DMBA B1 DIMA DIEA

0,2

0,0

0,0 0

20

40

60

80

0

100

20

40

60

80

100

120

Time [min]

Time [min]

(a)

(b)

1,2 1,0

1,0

CO2 loading [mol/mol]

CO2 loading [mol/mol]

0,8

0,8

A1 CHA CHpA IPCA DMCA DCHA

0,6

0,4

0,6

0,4

2,6-DMPD 3,5-DMPD MPD EPD

0,2

0,2

0,0

0,0 0

20

40

60

Time [min]

(c)

80

100

120

0

20

40

60

80

100

Time [min]

(d)

Fig. 1. Characteristics of CO2 absorption in 2.7M chain amines (a & b), cycloalkylamines (c) and cyclic amines (d) solutions

96

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

3.3. Absorption capacity CO2 loading capacity theoretically depends on the reaction mechanism, such as tertiary amine is typically able to achieve higher capacity than primary and secondary amines. Because tertiary amine can react with equivalent mole of carbon dioxide and water, the direct formation of bicarbonate (BF, eq. 2) is the only main reaction; while CF is the major and fast reaction in primary and secondary amine solutions, and the loading capacity only can be enhanced, when more carbamates get accumulated and carbamate reversion (CR, eq. 3) takes place, after CO2 loading approaching 0.5 mol/mol. However, it is very complicated in fact, due to the influences of sterically hindrance, aqueous solubility or insoluble salts formation. The loading capacity can be reduced, since CR or BF is inhibited by insoluble composition; or be enhanced if CR is promoted by aqueous environment. The high CO2 loadings were found in primary and secondary amine solutions for example CHpA, A1, DPA, 2,6-DMPD, etc.; while very low loadings were detected in tertiary amine solutions such as DIEA, TPA and TBA. Besides, the amine concentration also has negative influence on CO2 loading (mol/mol). Most amines exhibit high CO2 loading, >0.9 mol/mol, at 2 M; but it decrease to <0.8 mol/mol at 4 M (see Fig. 2). However, A1 is an exception, since it still keeps high loading capacity, 0.95 mol/mol, even at 5 M. Carbamate formation (CF)

CO2  2 RNH 2 œ RNH 3  RNHCOO 

Bicarbonate formation (BF)

CO2  RNH 2  H 2 O œ RNH 3  HCO3







(2)



CO2  RNHCOO  2 H 2 O œ RNH 3  2 HCO3

1,2

1,2

1,0

1,0

0,8

0,6

A1, 0.9 M A1, 1.8 M A1, 2.7 M A1, 4.2 M convert to 1 phase

0,4

0,2

CO2 loading [mol/mol]

CO2 loading / amine [mol/mol]

Carbamate reversion (CR)

(1)



(3)

0,8

0,6

DBA, 0.9 M DBA, 1.8 M DBA, 2.7 M DBA, 4.2 M convert to 1 phase precipitation

0,4

0,2

0,0

0,0 0

20

40

60

80

100

120

140

160

0

20

40

Time [min]

60

80

100

Time [min]

(a)

(b) 1,2

1,0

CO2 Loading [mol/mol]

0,6

0,4

DMCA, 0.9 M DMCA, 1.8 M DMCA, 2.7 M DMCA, 4.2 M convert to 1 phase

0,2

CO2 loading [mol/mol]

1,0 0,8

0,8

0,6

0,4

EPD, 0.9 M EPD, 1.8 M EPD, 2.7 M EPD, 4.2 M convert to 1 phase

0,2

0,0

0,0 0

80

160

240

Time [min]

(c) Fig. 2. CO2 absorption into A1 (a), DBA (b), DMCA (c) and EPD (d) solutions

0

40

80

Time [min]

(d)

120

160

97

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

3.4. Regenerability Solvent regeneration at high temperature is one of the most significant drawbacks in the alkanolaminebased CO2 capture process. However, lipophilic amine can solve this problem dramatically with thermomorphic LLPS. According to the comparison of regenerability, B1 (sterically hindered alkylamine), DMCA and EPD have performed the most remarkably and become the most considerable solvents as regeneration promoter for further studies. The regenerability of DMOA and MPD was also outstanding, unfortunately extremely slow absorption rate or high volatility, respectively, was observed. Fig. 3 illustrate that the regeneration rate became fast after the LLPS taking place and very high regenerability was obtained in tertiary amine solutions such as DMCA and EPD. The major advantage of using lipophilic amines with reaction involving LLPS to a multiphase system is that the regeneration can be carried out at a modest temperature, i.e. 80 °C or even lower, which is much less than the temperature employed in conventional industrial process using alkanolamines. Together with the high cyclic loading capacity of such amines, the TBS absorbents offer more degrees of freedom for cost reduction with respect to the quality of energy requirement.

0,6

DMCA, 0.9 M DMCA, 1.8 M DMCA, 2.7 M DMCA, 4.2 M convert to 2 phases

1,0

DBA, 0.9 M DBA, 1.8 M DBA, 2.7 M DBA, 4.2 M convert to 2 phases

Fraction of desorbed CO2 [mol/mol]

Fraction of desorbed CO2 [mol/mol]

0,8

0,4

0,2

0,0

0,8

0,6

0,4

0,2

0,0

20

30

40

50

60

70

80

90

20

100

30

40

50

60

70

80

90

100

80

90

100

o

Temperature [ C]

o

Temperature [ C]

(a)

(b) 1,0

Fraction of desorbed CO2 [mol/mol]

0,8

Fraction of desorbed CO2 [mol/mol]

B1, 0.9 M B1, 1.8 M B1, 2.7 M B1, 4.2 M convert to 2 phases

1,0

0,6

0,4

0,2

0,0

EPD, 0.9 M EPD, 1.8 M EPD, 2.7 M EPD, 4.2 M convert to 2 phases

0,8

0,6

0,4

0,2

0,0 20

30

40

50

60

70

80

90

100

Temperature [Ⱌ ]

(c)

20

30

40

50

60

70 o

Temperature [ C]

(d)

Fig. 3. Regeneration of loaded DBA (a), DMCA (b), B1 (c) and EPD (d) solutions

3.5. Detrimental phenomena Various phenomena, for example LLPS, gel formation, viscosity increasing, insoluble carbamate salts formation and precipitation of bicarbonates, were observed during the absorbents screening, however, only the miscibility gap with LLPS was of interest in this work and the remaining features were detrimental to performance. Those unfavourable phenomena such as gel and insoluble salts formations

98

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

are undesired in absorption; since they reduce the CO2 loading capacity by impediment of either reaction rate or carbamate reversing to bicarbonate; and also can damage the equipment by fouling. According to experimental observation, insoluble carbamate salts formation took place at the very beginning of absorption, typically in 5-15 min, before achieving CO2 loading of 0.5 mol/mol, because it is caused by CF in organic phase, especially in the gas-liquid interface. The solid salts are self-associated at the liquid surface or adhered on the wall of the reactor. In contrast, precipitation occurred close to the end of absorption during CR, due to the limited aqueous solubility of bicarbonate salts. In secondary amines, the isoalkyl group, for example DIPA, DIBA or IPCA, has shown high potential to induce insoluble carbamate salts formation; while n-alkyl groups, for instance DPA, DBA and EBA, can cause bicarbonate precipitation. However, some of the carbamate and bicarbonate solids can be dissolved in water by elevating temperature to 50-60 °C. Due to individual reaction mechanism of BF, precipitation and insoluble salts formation are avoided in tertiary amine solutions. 4. Technical viability of TBS absorbents Since a biphasic solvent would be inconvenient for absorption, the fine tuning of the phase transition behaviour has been investigated. The protonated lipophilic amine species was found to be able to play the role of a solubiliser for elevating the LCST. For example, the LCST of unloaded 5 M A1+B1 (4:1) solution is normally 15 °C, but can be increased to 40 °C for the lean solution with a CO2 loading 0.05 mol/mol, which demonstrates the technical feasibility for implementing the absorption with single phase solvent. In place of the high-energy-consumption steam stripping column, a continuous stirred-tank reactor (CSTR) has been employed for achieving the low temperature (§80 °C) regeneration with remarkable performance, when appropriate process intensification measures are adopted. 4.1. Classification After surveying more than thirty lipophilic amines in the test tube screening experiments; only less than ten proved to be comparable with, if not better than, conventional alkanolamines. Lipophilic amines such as DPA, 2,6-DMPD, A1, DMCA, MPD, EPD and B1 with their good performances were proposed as the most potential candidates which survived the rigorous selection process. Novel absorbent A1, has performed much better than the benchmark amines MEA and AMP (Fig. 4-a). Conversely, DMCA, EPD and B1 with low LLPS temperature exhibit an outstanding regenerability (>96%) at 80 °C. According to the operational performance, lipophilic amines can be classified into two categories: absorption activator, for instance, CHpA, MBzA and A1, with rapid reaction kinetics, and regeneration promoters, such as B1, DMCA and DMBzA, exhibiting outstanding regenerabilities. Aliphatic amines with various molecular structures have been studied for this purpose. Primary and secondary amines are typically regarded as activators; an alpha-branch has a positive influence on reaction kinetics, while a beta-branch can lead to unwanted precipitation. Tertiary amines are preferred for promoting regeneration by virtue of their low dielectric constants and thermally induced LLPS behaviour. 4.2. Blending

x x x x

The ideal chemical solvent for CO2 absorption is desired to match the following criteria [2,15]: Fast kinetics – reduces height requirements for the absorber and/or solvent circulation flow rates; High absorption capacity – directly influences solvent circulation flow rate requirements; Low regeneration costs – cut energy consumption; High thermal stability and low solvent degradation – reduce solvent waste;

99

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

x Low solvent expense – should be easy and cheap to produce; x Low environmental impact; x Technical feasibility. Due to the difficulties in obtaining a single lipophilic amine as a perfect solvent to fulfil all the screening criteria, i.e. with both high reactivity and good regenerability, blended solvent of activator + promoter, has been developed to overcome the drawbacks of the individual amines and combine the advantages of several components. The blended solutions such as B1+A1, EPD+A1 and DMCA+A1 (3M, 3:1) were formulated for the investigation. Those blended lipophilic solvents have successfully achieved both the rapid absorption rate thank to the activator and low residual loading due to the promoter. Fig. 4-b indicates the superiority of the lipophilic amine solvent blend lies not only in the high loading capacity but also in the good regenerability in comparison with the benchmark absorbents; where the cyclic loading capacity of the newly developed solvent blend “p-A1” (regeneration promoted A1) was 130% higher than that for MDEA+MEA solution. 4,0

1: 87.8%

1 2

Regenerability at 80 Ⱌ

3

3: 85.6%

2,5

Regenerability at 80 Ⱌ

a

b: 98.1%

2: 88.1%

3,0

a: 96.5% 3,5

4

4: 46.4% 2,0 1,5

1. A1, 2. AMP, 3. A1, 4. MEA,

1,0 0,5

5M 30 wt% 3M 30 wt% (~5 M)

0,0

Loading [mol-CO2/L-soln]

Loading [mol-CO2/L-soln]

3,5

3,0

c: 97.3%

2,5

d: 99.2%

b c d

e: 87.6%

2,0

e

1,5 1,0 0,5

a. p-A1, b. DMCA+A1, c. EPD+A1, d. B1+A1, e. MDEA+MEA,

4M 3M 3M 3M 3M

100

140

0,0

0

20

40

60

80

Time [min]

(a)

100

120

140

0

20

40

60

80

120

Time [min]

(b)

Fig. 4. Characteristics of single (a) and blended (b) amine solutions in absorption at 19.4 kPa CO2 & 30 °C and desorption at 80 °C

4.3. Solubilisation Thermal-induced phase transition is a distinctive property of TBS system. The homogeneous solution converts to heterogeneous upon heating after achieving LCST; contrarily, the regenerated biphasic solution reverses to a single phase by cooling. Most of the aliphatic amines exhibit the phase transition behaviour in the aqueous solutions, but very few of them can be employed as absorbent, because of the and high volatile loss of lower alkylamines (C5-C7) due to low boiling point and low LCST of fatty alkylamines (C8-C10) which is typically close to the freezing point of aqueous solutions. In order to improve the solubility property of lipophilic amines, aqueous solubiliser was considered for the regulation of phase transition behaviour. Varying organic solvents, for example alkanols, alkanolamines and diamines, were hence studied by screening tests. The ideal solubiliser should meet the following criteria: x Effectively elevate the LCST of lipophilic amine solution with only small amount of additive; x No negative influence on absorption or desorption; x Highly chemical stability, no significant degradation; x Low volatile loss. Among all the tested solubilisers, AMP has exhibited the best performance on the influence of LCST.

100

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

Fig. 5 proves AMP also has successfully elevated the LCST of the DMCA+A1 solutions at different proportions. Fig. 6 demonstrated the limited influence was found on the absorption and desorption performance of lipophilic amine solutions. However, increase the mass fraction of AMP has positive influence on the LCST of the DMCA+A1 solution. But it is limited at 10 wt%, since the regenerability is significantly depressed when over 20 wt% of AMP is added in the lipophilic amine solution. 50

30

n emulsio 10

5

2

3,0

20

Absorption at 40 Ⱌ 1

5

2,5

4

1,3. TBS 2,4. TBS + 9wt.% AMP 5. 30wt.% MEA

2,0

3 1,5 1,0

Single phase

1

PTT: phase transition temperature CST: critical solution temperature

0

Primary TBS solution: 2M DMCA + 2M A1 CO2 partial pressure: 19.6 kPa

3,5

CO2 loading (mol/L)

Temperature (Ⱌ )

40

4,0

PTT of DMCA+A1 with 9wt% AMP CST of DMCA+A1 with 9wt% AMP PTT of DMCA+A1 CST of DMCA+A1 ion emuls Two phases

0,5

Desorption at 75 Ⱌ

3 2

4

0,0 3M(2:1)

3M(1:1)

4M(2:1)

4M(1:1)

0

20

40

Concentration (mol/L)

60

80

100

120

on

the

140

160

Time (min)

Fig. 5. Influence of concentration, proportion and AMP addition on phase transition

Fig.6. Influence of AMP characteristics of TBS solution

ab-/desorption

4.4. Solvent and process performance Experiments for optimising solvent formulations were initially conducted in a bubble column. Blended solvents such as A1+B1 with varying concentrations and proportions have been developed, possessing rapid absorption rates, high loading capacities, low regeneration temperatures and good regenerabilities. Fig. 7 illustrates that the net CO2 loading capacity of the preferred A1+B1 solution is higher than for the benchmark conventional 30wt% MEA and DMX-1 [3] solutions at varying CO2 partial pressures. After identifying the optimal parameters, this solvent was studied in a scaled-up 1 m high packed absorption column. The regenerated TBS solution becomes homogeneous upon cooling to the LCST of ca. 40°C, and the loaded solution separates into two immiscible phases upon heating to 80 °C. In the improved TBS process (see Fig. 8), deep regeneration is enhanced by LLPS, which permits exploitation of the low value heat at §90 °C from the other processes and cutting the energy consumption by more than 35% [16] compared to the conventional alkanolamine solvent systems. 30-40 Ⱌ

500

Lean Solvent

400

Treated Gas 300

CO2 (Transport / Storage)

PCO2 (mbar)

200

100 90 80 70 60 50

Org.

80 Ⱌ

A1+B1, 4+1M, 30Ⱌ A1+B1, 4+1M, 80Ⱌ DMX-1, 40Ⱌ [3] MEA 30wt%, 40Ⱌ MEA 30wt%, 120Ⱌ

Flue Gas Aq.

40 0

1

2

3

4

5

Absorber

Rich Solvent

Regenerator

Loading (mol-CO2/kg-sol.)

Fig. 7. Vapour-Liquid Equilibrium of CO2 in aqueous amine solutions

Fig. 8. Simplified process flow sheet for TBS system

Jiafei Zhang et al. / Energy Procedia 23 (2012) 92 – 101

101

5. Conclusions Novel lipophilic amine has been introduced and proposed as an alternative solvent for PCC. According to screening tests on the physicochemical properties and ab/desorption characteristics, outstanding alkylamines were selected. The blended absorbents such as DMCA+A1 and A1+B1 have shown better performance compared to conventional alkanolamines in various aspects, especially in loading capacity and solvent regenerability. The technical feasibility of TBS system for CO2 ab/desorption has also been studied with respect to improvement of solvent formulation, regulation of phase change behaviour and optimisation of operating conditions. The temperature induced LLPS potentially enables an extensive regenerability at only 80 °C, thus facilitating a more flexible and expedient thermal integration of the CO2 capture process into existing heat recovery networks and reducing the dominant operating expenditure.

Acknowledgements This work was supported by Shell Global Solutions International B.V.

References [1] Wang M., Lawal A., Stephenson P., Sidders J., Ramshaw C. Post-combustion CO2 capture with chemical absorption: A stateof-the-art review. Chem. Eng. Res. Des., 2011, 89 (9): 1609-1624. [2] Svendsen H.F., Hessen E.T., Mejdell T. Carbon dioxide capture by absorption; challenges and possibilities. Chem. Eng. J., 2011, 171 (3): 718-724. [3] Raynal L., Alix P., Bouilon P., et al. The DMXTM process: an original solution for lowering the cost of post-combustion carbon capture. Energy Procedia 2011, 4: 779-786. [4] Agar D.W., Tan Y., Zhang X. Separating CO2 from gas mixtures. Patent WO/2008/015217. [5] Zhang J., Nwani O., Tan Y., Agar D.W. Carbon dioxide absorption into biphasic amine solvent with solvent loss reduction. Chem. Eng. Res. Des., 2011, 89 (8): 1190-1196. [6] Zhang J., Agar D.W., Zhang X., Geuzebroek F. CO2 absorption in biphasic solvents with enhanced low temperature solvent regeneration. Energy Procedia 2011, 4: 67-74. [7] Singh P., Niederer J.P.M., Versteeg G.F. Structure and activity relationships for amine-based CO2 absorbents – I. International Journal of Greenhouse Gas Control, 2007, 1 (1): 5-10. [8] Singh P., Versteeg G.F. Structure and activity relationships for CO2 regeneration from aqueous amine-based absorbents. Process Safety and Environmental Protection, 2008, 86 (5): 347-359. [9] Singh P., Niederer J.P.M., Versteeg G.F. Structure and activity relationships for amine-based CO2 absorbents – II. Chem. Eng. Res. Des., 2009, 87 (2): 135-144. [10] Chowdhury, F.A., Okabe, H., Yamada, H., Onoda, M. Fujioka, Y. Synthesis and selection of hindered new amine absorbents for CO2 capture. Energy Procedia 2011, 4: 201-208. [11] Robinson K., McCluskey A., Attalla M. The effect molecular structural variations has on the CO2 absorption characteristics of heterocyclic amines. Energy Procedia 2011, 4: 224-231. [12] Yalkowski S.H. Solubility and Solubilization in Aqueous Media, American Chemical Society, Washington, DC; 1999. [13] Zhang J., Misch R., Tan Y., Agar D.W. Novel thermomorphic biphasic amine solvents for CO2 absorption and low-temperature extractive regeneration. Chem. Eng. Technol. 2011, 34 (9): 1481-1489. [14] Astarita G. Gas Treating with Chemical Solvents. Wiley-Interscience Publication, New York; 1983. [15] Davidson R.M. Post-combustion Carbon Capture from Coal Fired Plants – Solvent Scrubbing. IEA Clean Coal Centre, CCC/125; 2007. [16] Geuzebroek F., Schneider A., Last T., Zhang X. Benchmark study on the energy consumption of novel solvents for CO2 capture, Presentated in TCCS-5 Conference, 16-17 June, 2009, Trondheim, Norway.