Hybrid membranes to limit amine evaporation in membrane contactors for CO capture 2 Marta Westad Hauge Chemical Engineering and Biotechnology Submission date: June 2017 Supervisor: Liyuan Deng, IKP Co-supervisor: Luca Ansaloni, IKP Zhongde Dai, IKP Norwegian University of Science and Technology Department of Chemical Engineering Abstract In order to meet the global demand of energy from combustion processes with fossil fuels withoutinterferingwiththeatmosphericlevelsofCO , carboncaptureandstorage(CCS) 2 is considered a viable solution. Separation of CO from combustion gas performed by 2 absorption/stripping in columns is today a costly and energy demanding process. One way to reduce the energy requirement of CO absorption is to use DEEA MAPA blends, 2 but they are volatile and their use at industrial level is thus problematic. Membranebasedabsorption,definedasmembranecontactorsallowoperatingwithvolatile absorbents without emitting harmful chemicals. This require a membrane that allows high CO and low absorbent permeability. Membrane contactors also have the poten- 2 tial to reduce the capital cost of absorption as effective membrane modules increase the surface area of absorption, and thereby reduce the process volume. This thesis has investigated two different AF2400 membranes with addition of ZIF-8 and XT-RGO nanoparticles, and in particular how these nanocomposite membranes affect characteristics that are important to consider in membrane contactor absorption. The membranes were tested together with a volatile 3rd generation CO absorbent named 2 3D3M and with MEA as a reference absorbent. It was discovered that the two nanocom- posite membranes decreased the permeability of the two amine absorbents substantially compared to the pure AF2400 membrane. However, the nanocomposite membranes also revealed a lower CO permeability compared to pure AF2400. A simple membrane con- 2 tactor model was proposed to evaluate the membrane module performance in terms of evaporation prevention. Other highly CO permeable membranes were also tested to- 2 gether with 3D3M and two other 3rd generation absorbents - 3DEA2M and 3HEPP2M, where only AF2400 proved to be stable with the absorbents. Figure 0.1: Schematic illustration of a membrane contactor. i Sammendrag Karbonfangst er en teknologi som muliggjør˚a tilfredsstille det globale energibehovet med forbrenning av fossile brennstoff uten˚a forstyrre atmosfæreniv˚aet av CO . I dag blir CO 2 2 separert fra forbrenningsgasser ved absorpsjon/stripping i kolonner, som er en kostbar og energikrevende prosess. En m˚ate ˚a redusere energikravet til absorpsjon er ˚a benytte seg av en DEEA/MAPA-absorbentmix, men disse er svært volatile aminer som skaper utfordringer ved implementering til industriskala. Membranbasert absorpsjon, kalt membrankontaktorer, tillater bruk av volatile absorben- ter uten hensyn til utslipp. Dette krever at membranen tillater en høy permeabilitet av CO og lav permeabilitet av absorbenter. Membrankontaktorer kan ogs˚a redusere in- 2 vesteringskostadene til absorpsjon siden effektive membranmoduler øker overflaten mel- lom gassfase og absorbent, og reduserer dermed prosessvolumet. Denne masteroppgaven har undersøkt to ulike nanokompsoittmembraner med polymeren AF2400 og nanopartiklene ZIF-8 og XT-RGO for karakteristikker som er viktige i mem- branabsorpsjon. Membranene ble testet med en volatil CO -absorbent kalt 3D3M og 2 medMEAsomreferanseabsorbent. Detblekonkludertmedatbeggenanokomposittmem- branene senket permeabiliteten til begge absorbentene betraktelig sammenlignet med ren AF2400-membran. Samtidig viste det seg at nanokomposittmembranene senket CO - 2 permeabiliteten i forhold til ren AF2400. En enkel membrankontaktormodell ble brukt for˚a evaluere membranmodulen med tanke p˚a redusering av utslipp. Andre membraner medhøyCO -permeabilitetbleogs˚atestetsammenmed3D3Mogtoandre3. generasjon- 2 sabsorbenter - 3DEA2M og 3HEPP2M, hvor kun AF2400 viste seg˚a være kompatibel. iii Preface This Master’s Thesis is written spring 2017 at the Norwegian University of Science and Technology, as a part of the Environmental Engineering and Reactor Technology research group. The work has been a part of the 3GMC project delivered by CLIMIT, and supervised by Dr. Liyuan Deng with Dr. Luca Ansaloni and Dr. Zhongde Dai as co- supervisors. First of all, I would like to thank Liyuan for offering me a thesis I have found interesting and rewarding to work with. I will also like to thank Zhongde for encouragement and fun chats in the lab. Last, but not least I will thank Luca for being available 24/7, answering my stupid and not so stupid questions and helping me out in the lab. I also appreciate all the interesting people I have met in the lab that have helped me or entertained me in spare moments. Speaking of entertainment, I want to thank my co-students in all floors of chemistry block 4 for making my lunches and cafeteria dinners fun, and offering me a break from my thesis work. I especially want to thank Vilde and Natalie for membrane chats, Eirik for keeping me in company at my desk and ˚Asmund for making me laugh. The Bible says you’re supposed to honor your mother and father, and in case I might lose my inheritance if I don’t, I hereby thank my parents for parenting me. I also have to thank ˚Asmund’s father for proofreading my thesis, and then just not to exclude I have to thank his mother also. I hereby declare that this is an independent work according to the exam regulations of the Norwegian University of Science and Technology. v CONTENTS Contents 1 Introduction 1 2 Theory 3 2.1 Membrane transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2 Nanocomposite membranes . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2.1 Polymer phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2.2 Nanofillers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.3 Pervaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.4 Membrane contactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.4.1 Mass transfer in membrane contactors . . . . . . . . . . . . . . . 14 2.4.2 CO absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2 2.4.3 Absorbent chemicals . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.4.4 Hollow fiber coating . . . . . . . . . . . . . . . . . . . . . . . . . 17 3 Experimental methods 19 3.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.2 Membrane preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.3 Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.3.1 Membrane morphology . . . . . . . . . . . . . . . . . . . . . . . . 19 3.3.2 Membrane compatibility study . . . . . . . . . . . . . . . . . . . . 19 3.3.3 Amine evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.3.4 Titration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.5 Ion chromatography . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.6 Mixed gas permeation . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.7 Membrane contactor study . . . . . . . . . . . . . . . . . . . . . . 22 3.4 Dip coating of hollow fibers . . . . . . . . . . . . . . . . . . . . . . . . . 23 4 Experimental results and discussion 25 4.1 Nanocomposite membrane morphology . . . . . . . . . . . . . . . . . . . 25 4.2 Membrane compatibility study . . . . . . . . . . . . . . . . . . . . . . . . 27 4.3 CO permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2 4.4 Amine evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.4.1 CO /amine selectivity . . . . . . . . . . . . . . . . . . . . . . . . 34 2 4.5 Membrane contactor results . . . . . . . . . . . . . . . . . . . . . . . . . 36 4.5.1 CO /amine selectivity . . . . . . . . . . . . . . . . . . . . . . . . 38 2 4.6 Dip coating characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . 39 5 Modelling 43 5.1 Model theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5.2 Simulation results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 6 Conclusion 51 7 Further research 52 vii LIST OF FIGURES A Titration error i B Membrane preparation ii C Selectivity calculation iii D Morphology of hollow fibers viii E Sensitivity analysis of simple membrane model ix List of Figures 0.1 Schematic illustration of a membrane contactor. . . . . . . . . . . . . . . i 2.1 Illustration of a membrane separation mechanism. . . . . . . . . . . . . . 3 2.2 The Robeson upper bound for CO /N separation. . . . . . . . . . . . . 5 2 2 2.3 Structural formula of a Teflon AF2400 copolymer . . . . . . . . . . . . . 7 2.4 Robeson plot of different AF2400 + nanofiller membranes. . . . . . . . . 8 2.5 Structural image of a ZIF-8 crystal. . . . . . . . . . . . . . . . . . . . . . 9 2.6 Chemical structure of reduced graphene oxide. . . . . . . . . . . . . . . . 10 2.7 Molecular sieving mechanisms of CO and amine in graphene and ZIF-8. 11 2 2.8 Illustration of a vacuum pervaporation principle. . . . . . . . . . . . . . . 12 2.9 Flow scheme of a membrane contactor setup. . . . . . . . . . . . . . . . . 13 2.10 Hollow fiber module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.1 Pervaporation flow sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.2 Flow sheet of the mixed gas permeation setup. . . . . . . . . . . . . . . . 22 3.3 Membrane contactor flow sheet. . . . . . . . . . . . . . . . . . . . . . . . 23 4.1 Cross section S(T)EM images of AF2400/3wt% XT-RGO. . . . . . . . . 25 4.2 Cross section S(T)EM images of AF2400/3wt% ZIF-8. . . . . . . . . . . 26 4.3 Surface S(T)EM image of a AF2400/3wt% ZIF-8 membrane. . . . . . . . 26 4.4 Development of weights of membrane pieces in a 3D3M solution. . . . . . 27 4.5 Development of weights of membrane pieces in a 3DEA2M solution. . . . 28 4.6 Development of weights of membrane pieces in a 3HEPP2M solution. . . 28 4.7 Development of weights of membrane pieces in a MEA solution. . . . . . 29 4.8 Development of weights of membrane pieces in a H O solution. . . . . . . 29 2 4.9 CO permeabilities of nanocomposite AF2400 membranes. . . . . . . . . 31 2 4.10 CO /N separation factors of nanocomposite AF2400 membranes. . . . . 32 2 2 4.11 Prevention of amine emission with addition nanoparticles to AF2400. . . 33 4.12 CO /amine selectivity at 50 ◦C. . . . . . . . . . . . . . . . . . . . . . . . 35 2 4.13 CO permeabilityAF2400/XT-RGOandAF2400/ZIF-8membraneinmem- 2 brane contactor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 4.14 DifferenceinCO permeabilityinMGandMCstudyforAF2400/XT-RGO 2 membrane. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.15 CO /amine selectivity in membrane contactor. . . . . . . . . . . . . . . . 38 2 4.16 AF2400 thickness layer of big sized polypropylene hollow fibers. . . . . . 39 4.17 AF2400 thickness layer of medium sized polypropylene hollow fibers. . . . 40 viii