- Email: [email protected]
Regenerating Membrane Contactors for Solvent Absorption
Available online at www.sciencedirect.com
ScienceDirect Energy Procedia 114 (2017) 642 – 649
13th International Conference on Greenhouse Gas Control Technologies, GHGT-13, 14-18 November 2016, Lausanne, Switzerland
Regenerating membrane contactors for solvent absorption Colin A. Scholes*a, David deMontignyb, Sandra E. Kentisha, Geoffrey W. Stevensa a
Peter Cook Centre for Carbon Capture and Storage, Department of Chemical & Biomolecular Engineering, The University of Melbourne, 3010, Australia b Faculty of Engineering and Applied Science, University of Regina, S4S 0A2, Canada
Abstract
Membrane contactors are a hybrid technology that incorporates the advantages of both solvent absorption and membrane separation. Porous and asymmetric composite membrane contactors have been studied for CO2 absorption, and both configurations are susceptible to pore wetting. This results in a significant reduction in the mass transfer efficiency. As such, regeneration methods to remove entrained liquids from the contactors are of interest. In this work, four regeneration protocols are trialled for a porous poly tetrafluoroethylene (PTFE) contactor and a thin film composite poly(1-trimethylsilyl)-1-propyne (PTMSP) contactor. It was found that air and vacuum drying at elevated temperatures increased the overall mass transfer coefficient of both contactors compared to the wetted state, but did not return either to their original performance. In addition, both contactors experienced rapid rewetting of the pores. Prewashing with methanol before air drying at elevated temperature produced the greatest improvement in overall mass transfer for the regenerated contactors. This was attributed to methanol miscibility with the water in the pores reducing the capillary pressure experienced during drying, as well as methanol swelling the PTMSP layer. However, original functionality was not achieved for either contactors and both continued to experience wetting over time, though at a slower rate than with non-methanol wash regeneration protocols. © Published by Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license © 2017 2017The TheAuthors. Authors. Published by Elsevier (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. Peer-review under responsibility of the organizing committee of GHGT-13. Keywords: Membrane contactors; carbon dioxide; non-porous; pilot plant; regeneration
* Corresponding author. Tel.: +61-390358289; fax: +61-383444153. E-mail address: [email protected]
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. doi:10.1016/j.egypro.2017.03.1208
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
643
1. Introduction Membrane contactors are a combination of solvent absorption with membrane separation, which can be used to improve the efficiency of carbon dioxide capture from industrial flue gases. The approach involves the transfer of CO2 from the flue gas across a membrane to the permeate side where it is chemically absorbed into a solvent. The membrane acts as the interface between the gas and solvent phases, which enables much greater mass transfer area per unit volume to be achieved [1, 2], because of the compact membrane configuration compared to a traditional packed column [3]. Furthermore, the membrane physically separates the gas and solvent phases, so issues such as foaming and flooding are avoided. A similar approach can also be undertaken in the regeneration of the solvent, where CO2 liberated from the heated solvent passes through a membrane to the permeate side [4]. The separation of the solvent and CO2 product gas by the membrane again enables higher mass transfer area per unit volume than a traditional column, while also creating a high purity CO2 product stream. Steam is generally envisaged to act as a sweep gas on the permeate side to generate a strong driving force for stripping of CO2, this reduces the transfer of water from the solvent and hence reduce the load on the reboiler, as well as providing some of the thermal energy needed to regenerate the solvent. Hence, membrane contactors can be combined in a continuous process to undertake the efficient removal of CO2 from flue gas as well as solvent regeneration, as shown in Figure 1.
Figure 1: Membrane Contactor process for separating and purifying CO2 from flue gas.
The membrane material of the contactor is of paramount importance, as it strongly controls the CO2 mass transfer for both absorption and regeneration stages. Porous membrane contactors have been extensively studied [1, 5, 6], because the gas and solvent phases come into direct contact in the pores of the membrane. When the pores are gas filled this maximises CO2 mass transfer, however to ensure the gas phase does not enter the solvent, it is critical that the solvent be kept at a higher pressure than the gas [7]. As a consequence of this the solvent can wet the pores leading to significant reductions in CO2 mass transfer because of the slower diffusion of species through the solvent phase (Figure 2) [7-11]. Non-porous membranes were believed to overcome this, as the non-porous material physically separates the two phases and CO2 passes through the high permeance polymer as a gas [4, 12-14]. However, to maximise mass transfer the non-porous layer must be ultra-thin, and therefore is attached to a porous support. The pores in the porous support over time can become wetted because of the migration of water through the non-porous layer [12], and hence mass transfer is again significantly reduced. The performance of porous and nonporous membrane contactors have been trialled for CO2 separation from natural gas, syngas and flue gas through a number of pilot plant studies [15-17]. In all of these pilot trials the presence of pore wetting was observed and the difficulty in controlling the transmembrane pressure to limit the potential of wetting was highlighted.
644
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
Here, we investigate the potential to rejuvenate membrane contactors after pore wetting has occurred, and determine the potential to return both porous and non-porous membrane contactors back to original functionality. This is achieved by subjecting used wetted membrane contactors to a range of treatments designed to dry out the pores. The treated membrane contactors are then tested again for CO2 absorption from flue gas to enable the overall CO2 mass transfer coefficient to be measured and hence enable comparison with a fresh contactor. The objective is to establish a potential methodology to enable membrane contactors to be reused within a CO2 capture process.
Figure 2: Membrane contactor pores that are non-wetted, partially wetted, fully wetted and during regeneration.
2. Experimental Two membrane contactors are studied here; porous polytetrafluoroethylene (PTFE), as used in the CO2CRC’s H3 Capture project [15], and composite membrane componsed of a poly(1-trimethylsilyl)-1-propyne (PTMSP) layer on a porous PP support (Table 1). Both membrane contactors have previously been studied for CO2 capture and demonstrated significant wetting of the pores. Their performance in CO2 absorption is reported in detail elsewhere [12, 15]. Table 1: Specifications of the two membrane contactors undergoing regeneration.
Membrane Material Configuration
PTFE Porous
No. of fibers Inner fiber diameter (m) Outer fiber diameter (m) Average pore size (µm) Porosity (%) Effective Fiber length (m) External mass transfer area (m2)
19 0.0016 0.002 0.16 22.5 0.147 0.0175
PTMSP on PP Asymmetric non-porous 6 0.0004 0.000804 0.2 (PP support) 61 (PP support) 0.302 0.00458
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
To undertake rejuvenation, the following four protocols were trialled to remove the build-up of water in the pores. x Heated to 80 oC and air blown across the contactor for 4 hours. x Heated to 100 oC and air blown across the contactor for 4 hours. x Heated to 100 oC and exposed to vacuum for 4 hours. x Prewashed with methanol and then heated to 80 oC and air blown across the contactor for 4 hours. The regenerated PTFE membrane contactor was trialled within the CO2CRC’s H3 project membrane pilot plant (Figure 3). As such, the vacuum protocol was not applied to this contactor because of the difficulty in achieving a vacuum onsite. The pilot plant is a custom built plant using flue gas from a Victorian brown coal fired power station [15]. The flue gas was taken after the blower discharge of a direct contact cooler (DCC). The average CO2 composition was 11.99 mol%, with the majority N2 (80.95 %) as well as minor components (O2 – 7.06 mol%, SOx – 45 ppm, NOx – 130 ppm). The flue gas pressure was cooled to ambient conditions to remove the majority of the water before being heated back to ~ 45 oC. The contactor was operated with the solvent on the shell side. The solvent was PuraTREAT (BASF) that was drawn directly from a neighbouring solvent absorption plant onsite at a lean loading of ~0.17 at 35 oC. The lean solvent pressure was greater than the flue gas pressure throughout the contactor, and the differential pressure monitored at both ends of the contactor. The enriched solvent was returned to the solvent absorption plant for regeneration. Composite PTMSP membrane contactors were tested under laboratory conditions on instrumentation that has been reported elsewhere [12]. The solvent was 30 wt% monoethanolamine (MEA) and flowed on the shell side, to ensure the solvent contacted only the non-porous layer side of the membrane. The laboratory simulated flue gas was based on 10 mol% of CO2 in N2, operated at 35 oC.
Figure 3: Membrane Contactor Pilot Plant, as part of the CO2CRC H3 Capture project (image courtesy of the CO2CRC).
645
646
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
For both contactors, the overall mass transfer coefficient (K) for CO2 transport from the gas phase into the solvent phase was determined from the CO2 molar flux (N) by: N=
(ଢ଼ ିଢ଼౫౪ )ୋ
= KC
(1)
where G is the inert gas flowrate, A the mass transfer area, Y the mole ratio of CO2 in the gas phase, and CLM represents the log mean average of the concentration driving force between the bulk gas and solvent phases. 4. Results and Discussion 4.1 Porous PTFE Contactor The overall CO2 mass transfer coefficients for the porous PTFE contactor after the three regeneration protocols are provided in Figure 4 as a function of time. Also included in the figure are the overall mass transfer coefficients for the PTFE contactor performance in the laboratory, where 22% pore wetting was observed, as well as in the pilot plant, where >100% pore wetting occurred [15]. For the three regeneration protocols trialled, it is very clear that none were able to return the PTFE contactor to its original state. The 80 and 100 oC air blown protocols have almost identical result, giving an overall mass transfer coefficient that is half that observed for a fresh contactor under laboratory conditions. This implies that the temperature at which the pore water is evaporated has little difference on the regenerated contactor performance. The reduced performance is attributed to changes occurring within the pore morphology as a result of water evaporation. A strong capillary force exists upon the water droplets within the pores [5], as the water evaporates this capillary force generates a strong stress on the pore wall (Figure 2). As a result, the pores become deformed and possibly reduce the porosity, as some of the pores are effectively sealed. Hence, this change in the morphology means that the porous PTFE contactor exposed to air drying cannot be regenerated back to the original state. The rapid wetting observed with time after both of these air blown protocols is attributed to the hydrophobic PTFE surface being coated with residual salts and surfactants from the dried PuraTREAT solvent. These deposits facilitate wetting and therefore water rapidly re-enters the pores. Hence, air blown drying of porous contactors is unable to regenerate a wetted membrane contactor back to its original performance. Prewashing the contactor with methanol and then air drying improves the overall mass transfer coefficient performance of a regenerated porous contactor compared to the air dried only protocols. However, the performance remains significantly less than that observed for a fresh contactor under laboratory conditions. The improvement compared to the air dried only protocols is attributed to methanol miscibility with the water in the pores, which results in a reduction in surface tension [9]. This lowers the capillary force acting on the liquid water within the pore and hence reduces the stress on the pore walls when the droplets evaporate. As such, the pore deformation is less and hence the regenerated contactor can achieve a greater mass transfer than when air dried only. However, the overall mass transfer coefficient still decreases rapidly with time following regeneration, again probably the result of residual surfactants and other adsorbed species on the membrane pore surface facilitating wetting. Critically, while prewashing with methanol clearly improves the wetted contactor performance, it is not capable of regenerating the contactor to its original performance.
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
647
Figure 4: Regeneration attempts on porous PTFE contactor based on membrane pilot plant trials, after exposure to PuraTREAT (BASF) solvent.
4.2 Asymmetric PTMSP contactor The overall CO2 mass transfer coefficient for the asymmetric PTMSP contactor is provided in Figure 5 for the four regeneration protocols trialled as a function of time, as well as the laboratory performance where the porous support is wetted and non-wetted [12]. The two air blown protocols and the prewashed methanol protocols have almost identical behaviour after 30 mins of testing. These overall mass transfer coefficients are ~84% of the nonwetted performance of the fresh contactor, which is significantly higher than that observed for the porous PTFE contactor, ~43% (Figure 4). Hence, air drying and methanol treatment are able to more successfully regenerate the asymmetric PTMSP contactor, compared to the PTFE contactor. This is attributed to the different solvents in each study, commercial PuraTREAT compared to MEA. However, the air dried protocol contactors performance rapidly decreases to the wetted scenario, while the prewash with methanol protocol contactor was able to retain reasonable overall mass transfer coefficient performance with time. This difference is attributed to methanol swelling the PTMSP layer and reversing any aging effects that have occurred within PTMSP film [18]. Hence, the methanol wash is able to return the contactor to a state that enabled a higher CO2 flux. The vacuum protocol on the PTMPS contactor resulted in the lowest overall mass transfer coefficient (Figure 5), ~50% of the non-wetted original contactor, which is similar to the results observed for the PTFE contactor (Figure 4).This reduced performance is attributed to the vacuum for drying generating a stronger capillary force on the residual water than in the air blown protocols. This resulted in a higher wall stress on the pores during droplet evaporation and hence results in greater deformation of the porous support. As a consequence, the regenerated contactor was unable to achieve high mass transfer coefficients.
648
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
Figure 5: Regeneration attempts of asymmetric composite PTMSP membrane contactor, after exposure to MEA solution.
5. Conclusion Membrane contactors have the potential to replace traditional solvent columns for the efficient separation and purification of CO2 from flue gas. However, both porous and asymmetric contactors experience wetting that significantly reduces their mass transfer performance and impacts their CO2 separation efficiency. The ability to regenerate wetted contactors is therefore important. Of the four regeneration protocols studied here, it was determined that a methanol prewash achieved the highest overall mass transfer coefficient in regenerated porous and asymmetric contactors. This is because methanol miscibility with the water in the pores reduced the capillary force stresses induced as the pores dry. The methanol also reversed aging within the PTMSP layer of the asymmetric contactor. However, none of the regeneration protocols trialled here were able to restore the contactor performance back to that of an original pre-wetted contactor, and there was clear evidence that pore wetting occurred on a faster timescale in a regenerated contactor than a fresh contactor. Acknowledgements Colin Scholes would like to thank the Victorian State Government for funding. Colin Scholes, Sandra Kentish and Geoff Stevens would like to acknowledge funding provided by the Australian Government through its Cooperative Research Centre program as well as the Particulate Fluids Processing Centre of the University of Melbourne. References [1] F. Porcheron, D. Ferre, E. Favre, P.T. Nguyen, O. Lorain, R. Mercier, L. Rougeau, Hollow fiber membrane contactors for CO2 capture: from lab-scale screening to pilot-plant module conception. Energy Procedia 2011:4:763-770.
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649 [2] O. Falk-Pedersen, M.S. Gronvold, P. Nokleby, F. Bjerve, H.F. Svendsen, CO2 capture with membrane contactors. Int. J. Green Energy 2005:2:157-165. [3] B.W. Reed, M.J. Semmens, E.L. Cussler, Membrane contactors, in: R.D. Noble, S.A. Stern (Eds.) Membrane Separations Technology. Principles and Applications, Elsevier Science, New York, 1995. [4] C.A. Scholes, S.E. Kentish, G.W. Stevens, J. Jin, D. deMontigny, Non-porous membranecontactors for desorption of CO2 from monoethanolamine at elevated temperatures. Separation and Purification Technology 2015:156:841-847. [5] V.Y. Dindore, D.W.F. Brilman, F.H. Geuzebroek, G.F. Versteeg, Membrane-solvent selection for CO2 removal using membrane gas-liquid contactors, Separation and Purification Technology. 2004:40:133-145. [6] Y. Lv, X. Yu, J. Jia, S.-T. Tu, J. Yan, E. Dahlquist, Fabrication and characterization of superhydrophobic polypropylene hollow fiber membranes for carbon dioxide absorption. Applied Energy 2012:90:167-174. [7] Y. Lv, X. Yu, S.-T. Tu, J. Yan, E. Dahlquist, Wetting of polypropylene hollow fiber membrane contactors. Journal of Membrane Science 2010:362:444-452. [8] L. An, X. Yu, J. Yang, S.-T. Tu, J. Yan, CO2 capture using a superhydrophobic ceramic membrane contactor. Energy Procedia 2015:75:22872292. [9] J.-G. Lu, Y.-F. Zheng, M.-D. Cheng, Wetting mechanism in mass transfer process of hydrophobic membrane gas absorption. Journal of Membrane Science 2008:308:180-190. [10] R. Wang, H.Y. Zhang, P.H.M. Feron, D.T. Liang, Influence of membrane wetting on CO2 capture in microporous hollow fiber membrane contactors. Separation and Purification Technology 2005:46:33-40. [11] J. Franco, D. deMontigny, S. Kentish, J. Perera, G. Stevens, A study of the mass transfer of CO2 through different membrane materials in the membrane gas absorption process. Sep. Sci. Technol. 2008:43:225-244. [12] C.A. Scholes, S.E. Kentish, G.W. Stevens, D. deMontigny, Comparison of non-porous and porous membrane contactors for CO2 absorption into monoethanolamine. International Journal of Greenhouse Gas Control 2015:42:66-74. [13] K. Li, W.K. Teo, Use of permeation and absorption methods for CO2 removal in hollow fibre membrane modules. Sep. Purif. Technol. 1998:13:79-88. [14] P.T. Nguyen, E. Lasseuguette, Y. Medina-Gonzalez, J.C. Remigy, D. Roizard, E. Favre, A dense membrane contactor for intensified CO2 gas/liquid absorption in post-combustion capture. J. Membr. Sci. 2011:377:261-272. [15] C.A. Scholes, A. Qader, G.W. Stevens, S.E. Kentish, Membrane gas-solvent contactor trials of CO2 absorption from flue gas. Separation Science and Technology 2014:49:2449-2458. [16] C.A. Scholes, M. Simioni, J. Franco, A. Qader, S.E. Kentish, G.W. Stevens, Membrane gas-solvent contactor trials of CO2 absorption from syngas. Chemical Engineering Journal 2012:195-196:188-197. [17] H. Herzog, O. Falk-Pedersen, The Kvaerner membrane contactor: lessons from a case study in how to reduce capture costs, in: 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, 2000. [18] A. Hill, S. Pas, T.J. Bastow, M.I. Burgar, K. Nagai, L.G. Toy, B.D. Freeman, Influence of methanol conditioning and physical aging on carbon spin-lattice relaxation times of poly(1-trimethylsilyl-1-propyne). Journal of Membrane Science 2004:243:37-44.
649
ScienceDirect Energy Procedia 114 (2017) 642 – 649
13th International Conference on Greenhouse Gas Control Technologies, GHGT-13, 14-18 November 2016, Lausanne, Switzerland
Regenerating membrane contactors for solvent absorption Colin A. Scholes*a, David deMontignyb, Sandra E. Kentisha, Geoffrey W. Stevensa a
Peter Cook Centre for Carbon Capture and Storage, Department of Chemical & Biomolecular Engineering, The University of Melbourne, 3010, Australia b Faculty of Engineering and Applied Science, University of Regina, S4S 0A2, Canada
Abstract
Membrane contactors are a hybrid technology that incorporates the advantages of both solvent absorption and membrane separation. Porous and asymmetric composite membrane contactors have been studied for CO2 absorption, and both configurations are susceptible to pore wetting. This results in a significant reduction in the mass transfer efficiency. As such, regeneration methods to remove entrained liquids from the contactors are of interest. In this work, four regeneration protocols are trialled for a porous poly tetrafluoroethylene (PTFE) contactor and a thin film composite poly(1-trimethylsilyl)-1-propyne (PTMSP) contactor. It was found that air and vacuum drying at elevated temperatures increased the overall mass transfer coefficient of both contactors compared to the wetted state, but did not return either to their original performance. In addition, both contactors experienced rapid rewetting of the pores. Prewashing with methanol before air drying at elevated temperature produced the greatest improvement in overall mass transfer for the regenerated contactors. This was attributed to methanol miscibility with the water in the pores reducing the capillary pressure experienced during drying, as well as methanol swelling the PTMSP layer. However, original functionality was not achieved for either contactors and both continued to experience wetting over time, though at a slower rate than with non-methanol wash regeneration protocols. © Published by Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license © 2017 2017The TheAuthors. Authors. Published by Elsevier (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. Peer-review under responsibility of the organizing committee of GHGT-13. Keywords: Membrane contactors; carbon dioxide; non-porous; pilot plant; regeneration
* Corresponding author. Tel.: +61-390358289; fax: +61-383444153. E-mail address: [email protected]
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. doi:10.1016/j.egypro.2017.03.1208
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
643
1. Introduction Membrane contactors are a combination of solvent absorption with membrane separation, which can be used to improve the efficiency of carbon dioxide capture from industrial flue gases. The approach involves the transfer of CO2 from the flue gas across a membrane to the permeate side where it is chemically absorbed into a solvent. The membrane acts as the interface between the gas and solvent phases, which enables much greater mass transfer area per unit volume to be achieved [1, 2], because of the compact membrane configuration compared to a traditional packed column [3]. Furthermore, the membrane physically separates the gas and solvent phases, so issues such as foaming and flooding are avoided. A similar approach can also be undertaken in the regeneration of the solvent, where CO2 liberated from the heated solvent passes through a membrane to the permeate side [4]. The separation of the solvent and CO2 product gas by the membrane again enables higher mass transfer area per unit volume than a traditional column, while also creating a high purity CO2 product stream. Steam is generally envisaged to act as a sweep gas on the permeate side to generate a strong driving force for stripping of CO2, this reduces the transfer of water from the solvent and hence reduce the load on the reboiler, as well as providing some of the thermal energy needed to regenerate the solvent. Hence, membrane contactors can be combined in a continuous process to undertake the efficient removal of CO2 from flue gas as well as solvent regeneration, as shown in Figure 1.
Figure 1: Membrane Contactor process for separating and purifying CO2 from flue gas.
The membrane material of the contactor is of paramount importance, as it strongly controls the CO2 mass transfer for both absorption and regeneration stages. Porous membrane contactors have been extensively studied [1, 5, 6], because the gas and solvent phases come into direct contact in the pores of the membrane. When the pores are gas filled this maximises CO2 mass transfer, however to ensure the gas phase does not enter the solvent, it is critical that the solvent be kept at a higher pressure than the gas [7]. As a consequence of this the solvent can wet the pores leading to significant reductions in CO2 mass transfer because of the slower diffusion of species through the solvent phase (Figure 2) [7-11]. Non-porous membranes were believed to overcome this, as the non-porous material physically separates the two phases and CO2 passes through the high permeance polymer as a gas [4, 12-14]. However, to maximise mass transfer the non-porous layer must be ultra-thin, and therefore is attached to a porous support. The pores in the porous support over time can become wetted because of the migration of water through the non-porous layer [12], and hence mass transfer is again significantly reduced. The performance of porous and nonporous membrane contactors have been trialled for CO2 separation from natural gas, syngas and flue gas through a number of pilot plant studies [15-17]. In all of these pilot trials the presence of pore wetting was observed and the difficulty in controlling the transmembrane pressure to limit the potential of wetting was highlighted.
644
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
Here, we investigate the potential to rejuvenate membrane contactors after pore wetting has occurred, and determine the potential to return both porous and non-porous membrane contactors back to original functionality. This is achieved by subjecting used wetted membrane contactors to a range of treatments designed to dry out the pores. The treated membrane contactors are then tested again for CO2 absorption from flue gas to enable the overall CO2 mass transfer coefficient to be measured and hence enable comparison with a fresh contactor. The objective is to establish a potential methodology to enable membrane contactors to be reused within a CO2 capture process.
Figure 2: Membrane contactor pores that are non-wetted, partially wetted, fully wetted and during regeneration.
2. Experimental Two membrane contactors are studied here; porous polytetrafluoroethylene (PTFE), as used in the CO2CRC’s H3 Capture project [15], and composite membrane componsed of a poly(1-trimethylsilyl)-1-propyne (PTMSP) layer on a porous PP support (Table 1). Both membrane contactors have previously been studied for CO2 capture and demonstrated significant wetting of the pores. Their performance in CO2 absorption is reported in detail elsewhere [12, 15]. Table 1: Specifications of the two membrane contactors undergoing regeneration.
Membrane Material Configuration
PTFE Porous
No. of fibers Inner fiber diameter (m) Outer fiber diameter (m) Average pore size (µm) Porosity (%) Effective Fiber length (m) External mass transfer area (m2)
19 0.0016 0.002 0.16 22.5 0.147 0.0175
PTMSP on PP Asymmetric non-porous 6 0.0004 0.000804 0.2 (PP support) 61 (PP support) 0.302 0.00458
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
To undertake rejuvenation, the following four protocols were trialled to remove the build-up of water in the pores. x Heated to 80 oC and air blown across the contactor for 4 hours. x Heated to 100 oC and air blown across the contactor for 4 hours. x Heated to 100 oC and exposed to vacuum for 4 hours. x Prewashed with methanol and then heated to 80 oC and air blown across the contactor for 4 hours. The regenerated PTFE membrane contactor was trialled within the CO2CRC’s H3 project membrane pilot plant (Figure 3). As such, the vacuum protocol was not applied to this contactor because of the difficulty in achieving a vacuum onsite. The pilot plant is a custom built plant using flue gas from a Victorian brown coal fired power station [15]. The flue gas was taken after the blower discharge of a direct contact cooler (DCC). The average CO2 composition was 11.99 mol%, with the majority N2 (80.95 %) as well as minor components (O2 – 7.06 mol%, SOx – 45 ppm, NOx – 130 ppm). The flue gas pressure was cooled to ambient conditions to remove the majority of the water before being heated back to ~ 45 oC. The contactor was operated with the solvent on the shell side. The solvent was PuraTREAT (BASF) that was drawn directly from a neighbouring solvent absorption plant onsite at a lean loading of ~0.17 at 35 oC. The lean solvent pressure was greater than the flue gas pressure throughout the contactor, and the differential pressure monitored at both ends of the contactor. The enriched solvent was returned to the solvent absorption plant for regeneration. Composite PTMSP membrane contactors were tested under laboratory conditions on instrumentation that has been reported elsewhere [12]. The solvent was 30 wt% monoethanolamine (MEA) and flowed on the shell side, to ensure the solvent contacted only the non-porous layer side of the membrane. The laboratory simulated flue gas was based on 10 mol% of CO2 in N2, operated at 35 oC.
Figure 3: Membrane Contactor Pilot Plant, as part of the CO2CRC H3 Capture project (image courtesy of the CO2CRC).
645
646
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
For both contactors, the overall mass transfer coefficient (K) for CO2 transport from the gas phase into the solvent phase was determined from the CO2 molar flux (N) by: N=
(ଢ଼ ିଢ଼౫౪ )ୋ
= KC
(1)
where G is the inert gas flowrate, A the mass transfer area, Y the mole ratio of CO2 in the gas phase, and CLM represents the log mean average of the concentration driving force between the bulk gas and solvent phases. 4. Results and Discussion 4.1 Porous PTFE Contactor The overall CO2 mass transfer coefficients for the porous PTFE contactor after the three regeneration protocols are provided in Figure 4 as a function of time. Also included in the figure are the overall mass transfer coefficients for the PTFE contactor performance in the laboratory, where 22% pore wetting was observed, as well as in the pilot plant, where >100% pore wetting occurred [15]. For the three regeneration protocols trialled, it is very clear that none were able to return the PTFE contactor to its original state. The 80 and 100 oC air blown protocols have almost identical result, giving an overall mass transfer coefficient that is half that observed for a fresh contactor under laboratory conditions. This implies that the temperature at which the pore water is evaporated has little difference on the regenerated contactor performance. The reduced performance is attributed to changes occurring within the pore morphology as a result of water evaporation. A strong capillary force exists upon the water droplets within the pores [5], as the water evaporates this capillary force generates a strong stress on the pore wall (Figure 2). As a result, the pores become deformed and possibly reduce the porosity, as some of the pores are effectively sealed. Hence, this change in the morphology means that the porous PTFE contactor exposed to air drying cannot be regenerated back to the original state. The rapid wetting observed with time after both of these air blown protocols is attributed to the hydrophobic PTFE surface being coated with residual salts and surfactants from the dried PuraTREAT solvent. These deposits facilitate wetting and therefore water rapidly re-enters the pores. Hence, air blown drying of porous contactors is unable to regenerate a wetted membrane contactor back to its original performance. Prewashing the contactor with methanol and then air drying improves the overall mass transfer coefficient performance of a regenerated porous contactor compared to the air dried only protocols. However, the performance remains significantly less than that observed for a fresh contactor under laboratory conditions. The improvement compared to the air dried only protocols is attributed to methanol miscibility with the water in the pores, which results in a reduction in surface tension [9]. This lowers the capillary force acting on the liquid water within the pore and hence reduces the stress on the pore walls when the droplets evaporate. As such, the pore deformation is less and hence the regenerated contactor can achieve a greater mass transfer than when air dried only. However, the overall mass transfer coefficient still decreases rapidly with time following regeneration, again probably the result of residual surfactants and other adsorbed species on the membrane pore surface facilitating wetting. Critically, while prewashing with methanol clearly improves the wetted contactor performance, it is not capable of regenerating the contactor to its original performance.
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
647
Figure 4: Regeneration attempts on porous PTFE contactor based on membrane pilot plant trials, after exposure to PuraTREAT (BASF) solvent.
4.2 Asymmetric PTMSP contactor The overall CO2 mass transfer coefficient for the asymmetric PTMSP contactor is provided in Figure 5 for the four regeneration protocols trialled as a function of time, as well as the laboratory performance where the porous support is wetted and non-wetted [12]. The two air blown protocols and the prewashed methanol protocols have almost identical behaviour after 30 mins of testing. These overall mass transfer coefficients are ~84% of the nonwetted performance of the fresh contactor, which is significantly higher than that observed for the porous PTFE contactor, ~43% (Figure 4). Hence, air drying and methanol treatment are able to more successfully regenerate the asymmetric PTMSP contactor, compared to the PTFE contactor. This is attributed to the different solvents in each study, commercial PuraTREAT compared to MEA. However, the air dried protocol contactors performance rapidly decreases to the wetted scenario, while the prewash with methanol protocol contactor was able to retain reasonable overall mass transfer coefficient performance with time. This difference is attributed to methanol swelling the PTMSP layer and reversing any aging effects that have occurred within PTMSP film [18]. Hence, the methanol wash is able to return the contactor to a state that enabled a higher CO2 flux. The vacuum protocol on the PTMPS contactor resulted in the lowest overall mass transfer coefficient (Figure 5), ~50% of the non-wetted original contactor, which is similar to the results observed for the PTFE contactor (Figure 4).This reduced performance is attributed to the vacuum for drying generating a stronger capillary force on the residual water than in the air blown protocols. This resulted in a higher wall stress on the pores during droplet evaporation and hence results in greater deformation of the porous support. As a consequence, the regenerated contactor was unable to achieve high mass transfer coefficients.
648
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649
Figure 5: Regeneration attempts of asymmetric composite PTMSP membrane contactor, after exposure to MEA solution.
5. Conclusion Membrane contactors have the potential to replace traditional solvent columns for the efficient separation and purification of CO2 from flue gas. However, both porous and asymmetric contactors experience wetting that significantly reduces their mass transfer performance and impacts their CO2 separation efficiency. The ability to regenerate wetted contactors is therefore important. Of the four regeneration protocols studied here, it was determined that a methanol prewash achieved the highest overall mass transfer coefficient in regenerated porous and asymmetric contactors. This is because methanol miscibility with the water in the pores reduced the capillary force stresses induced as the pores dry. The methanol also reversed aging within the PTMSP layer of the asymmetric contactor. However, none of the regeneration protocols trialled here were able to restore the contactor performance back to that of an original pre-wetted contactor, and there was clear evidence that pore wetting occurred on a faster timescale in a regenerated contactor than a fresh contactor. Acknowledgements Colin Scholes would like to thank the Victorian State Government for funding. Colin Scholes, Sandra Kentish and Geoff Stevens would like to acknowledge funding provided by the Australian Government through its Cooperative Research Centre program as well as the Particulate Fluids Processing Centre of the University of Melbourne. References [1] F. Porcheron, D. Ferre, E. Favre, P.T. Nguyen, O. Lorain, R. Mercier, L. Rougeau, Hollow fiber membrane contactors for CO2 capture: from lab-scale screening to pilot-plant module conception. Energy Procedia 2011:4:763-770.
Colin A. Scholes et al. / Energy Procedia 114 (2017) 642 – 649 [2] O. Falk-Pedersen, M.S. Gronvold, P. Nokleby, F. Bjerve, H.F. Svendsen, CO2 capture with membrane contactors. Int. J. Green Energy 2005:2:157-165. [3] B.W. Reed, M.J. Semmens, E.L. Cussler, Membrane contactors, in: R.D. Noble, S.A. Stern (Eds.) Membrane Separations Technology. Principles and Applications, Elsevier Science, New York, 1995. [4] C.A. Scholes, S.E. Kentish, G.W. Stevens, J. Jin, D. deMontigny, Non-porous membranecontactors for desorption of CO2 from monoethanolamine at elevated temperatures. Separation and Purification Technology 2015:156:841-847. [5] V.Y. Dindore, D.W.F. Brilman, F.H. Geuzebroek, G.F. Versteeg, Membrane-solvent selection for CO2 removal using membrane gas-liquid contactors, Separation and Purification Technology. 2004:40:133-145. [6] Y. Lv, X. Yu, J. Jia, S.-T. Tu, J. Yan, E. Dahlquist, Fabrication and characterization of superhydrophobic polypropylene hollow fiber membranes for carbon dioxide absorption. Applied Energy 2012:90:167-174. [7] Y. Lv, X. Yu, S.-T. Tu, J. Yan, E. Dahlquist, Wetting of polypropylene hollow fiber membrane contactors. Journal of Membrane Science 2010:362:444-452. [8] L. An, X. Yu, J. Yang, S.-T. Tu, J. Yan, CO2 capture using a superhydrophobic ceramic membrane contactor. Energy Procedia 2015:75:22872292. [9] J.-G. Lu, Y.-F. Zheng, M.-D. Cheng, Wetting mechanism in mass transfer process of hydrophobic membrane gas absorption. Journal of Membrane Science 2008:308:180-190. [10] R. Wang, H.Y. Zhang, P.H.M. Feron, D.T. Liang, Influence of membrane wetting on CO2 capture in microporous hollow fiber membrane contactors. Separation and Purification Technology 2005:46:33-40. [11] J. Franco, D. deMontigny, S. Kentish, J. Perera, G. Stevens, A study of the mass transfer of CO2 through different membrane materials in the membrane gas absorption process. Sep. Sci. Technol. 2008:43:225-244. [12] C.A. Scholes, S.E. Kentish, G.W. Stevens, D. deMontigny, Comparison of non-porous and porous membrane contactors for CO2 absorption into monoethanolamine. International Journal of Greenhouse Gas Control 2015:42:66-74. [13] K. Li, W.K. Teo, Use of permeation and absorption methods for CO2 removal in hollow fibre membrane modules. Sep. Purif. Technol. 1998:13:79-88. [14] P.T. Nguyen, E. Lasseuguette, Y. Medina-Gonzalez, J.C. Remigy, D. Roizard, E. Favre, A dense membrane contactor for intensified CO2 gas/liquid absorption in post-combustion capture. J. Membr. Sci. 2011:377:261-272. [15] C.A. Scholes, A. Qader, G.W. Stevens, S.E. Kentish, Membrane gas-solvent contactor trials of CO2 absorption from flue gas. Separation Science and Technology 2014:49:2449-2458. [16] C.A. Scholes, M. Simioni, J. Franco, A. Qader, S.E. Kentish, G.W. Stevens, Membrane gas-solvent contactor trials of CO2 absorption from syngas. Chemical Engineering Journal 2012:195-196:188-197. [17] H. Herzog, O. Falk-Pedersen, The Kvaerner membrane contactor: lessons from a case study in how to reduce capture costs, in: 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, 2000. [18] A. Hill, S. Pas, T.J. Bastow, M.I. Burgar, K. Nagai, L.G. Toy, B.D. Freeman, Influence of methanol conditioning and physical aging on carbon spin-lattice relaxation times of poly(1-trimethylsilyl-1-propyne). Journal of Membrane Science 2004:243:37-44.
649