experimental studies on co2 capture using absorbent in a polypropylene hollow fiber membrane

Mälardalen University Press DissertationsNo. 102

EXPERIMENTAL STUDIES ON CO2 CAPTURE USING ABSORBENTIN A POLYPROPYLENE HOLLOW FIBER MEMBRANE CONTACTOR

Yuexia Lu

2011

School of Sustainable Development of Society and Technology



Yuexia Lu

2011

School of Sustainable Development of Society and Technology



Yuexia Lu

Akademisk avhandling

som för avläggande av teknologie doktorsexamen i energi- och miljöteknik vidAkademin för hållbar samhälls- och teknikutveckling kommer att offentligen

försvaras fredagen den 17 juni 2011, 10.00 i Lambda, Mälardalens högskola, Västerås.

Fakultetsopponent: Umberto Desideri, University of Perugia (ITA)

Akademin för hållbar samhälls- och teknikutveckling



Yuexia Lu

Akademisk avhandling

som för avläggande av teknologie doktorsexamen i energi- och miljöteknik vidAkademin för hållbar samhälls- och teknikutveckling kommer att offentligen

försvaras fredagen den 17 juni 2011, 10.00 i Lambda, Mälardalens högskola, Västerås.

Fakultetsopponent: Umberto Desideri, University of Perugia (ITA)

Akademin för hållbar samhälls- och teknikutveckling

AbstractIn recent years, membrane gas absorption technology has been considered as one of the promisingalternatives to conventional techniques for CO2 capture due to its favorable mass transfer performance.As a hybrid approach of chemical absorption and membrane separation, it exhibits a number ofadvantages, such as operational flexibility, compact structure, high surface-area-to-volume ratio, linearscale up, modularity and predictable performance. One of the main challenges of membrane gasabsorption technology is the membrane wetting by absorbent over prolonged operating time, whichmay significantly decrease the mass transfer coefficients of the membrane module.

In this thesis, the experimental was set up to investigate the dependency of CO2 removal efficiency andmass transfer rate on various operating parameters, such as the gas and liquid flow rates, absorbenttype and concentration and volume fraction CO2 at the feed gas inlet. In addition, the simultaneousremoval of SO2 and CO2 was investigated to evaluate the feasibility of simultaneous desulphurizationand decarbonization in the same membrane contactor. During 14 days of continuous operation, it wasobserved that the CO2 mass transfer rate decreased significantly following the operating time, whichwas attributed to partial membrane wetting.

To better understand the wetting mechanism of membrane pores during their prolonged contactwith absorbents, immersion experiments for up to 90 days were carried out. Various membranecharacterization methods were used to illustrate the wetting process before and after the membranefibers were exposed to the absorbents. The characterization results showed that the absorbent moleculesdiffused into the polypropylene polymer during the contact with the membrane, resulting in theswelling of the membrane. In addition, the effects of operating parameters such as immersion time andabsorbent type on the membrane wetting were investigated in detail. Finally, based on the analysisresults, methods to smooth the membrane wetting were discussed. It was suggested that improvingthe hydrophobicity of polypropylene membrane by surface modification may be an effective way toimprove the long-term operating performance of membrane contactors. Therefore, the polypropylenehollow fibers were modified by depositing a thin superhydrophobic coating on the membrane surface toimprove their hydrophobicity. The mixture of cyclohexanone and methylethyl ketone was consideredas the best non-solvent to achieve the fiber surface with good homogeneity and acceptably highhydrophobicity. In the long-period operation, the modified membrane contactor exhibited more stableand efficient performance than the untreated one. Hence, surface treatment provides a feasibility ofimproving the system stability for CO2 capture from the view of long-term operation.

ISBN 978-91-7485-023-9ISSN 1651-4238

I

Abstract

In recent years, membrane gas absorption technology has been considered as one of the promising alternatives to conventional techniques for CO2 capture due to its favorable mass transfer performance. As a hybrid approach of chemical absorption and membrane separation, it exhibits a number of advantages, such as operational flexibility, compact structure, high surface-area-to-volume ratio, linear scale up, modularity and predictable performance. One of the main challenges of membrane gas absorption technology is the membrane wetting by absorbent over prolonged operating time, which may significantly decrease the mass transfer coefficients of the membrane module.

In this thesis, the experimental was set up to investigate the dependency of CO2 removal efficiency and mass transfer rate on various operating parameters, such as the gas and liquid flow rates, absorbent type and concentration and volume fraction CO2 at the feed gas inlet. In addition, the simultaneous removal of SO2 and CO2 was investigated to evaluate the feasibility of simultaneous desulphurization and decarbonization in the same membrane contactor. During 14 days of continuous operation, it was observed that the CO2 mass transfer rate decreased significantly following the operating time, which was attributed to partial membrane wetting.

To better understand the wetting mechanism of membrane pores during their prolonged contact with absorbents, immersion experiments for up to 90 days were carried out. Various membrane characterization methods were used to illustrate the wetting process before and after the membrane fibers were exposed to the absorbents. The characterization results showed that the absorbent molecules diffused into the polypropylene polymer during the contact with the membrane, resulting in the swelling of the membrane. In addition, the effects of operating parameters such as immersion time and absorbent type on the membrane wetting were investigated in detail. Finally, based on the analysis results, methods to smooth the membrane wetting were discussed. It was suggested that improving the hydrophobicity of polypropylene membrane by surface modification may be an effective way to improve the long-term operating performance of membrane contactors. Therefore, the polypropylene hollow fibers were modified by depositing a thin superhydrophobic coating on the membrane surface to improve their hydrophobicity. The mixture of cyclohexanone and methylethyl ketone was considered as the best non-solvent to achieve the fiber surface with good homogeneity and acceptably high hydrophobicity. In the long-period operation, the modified membrane contactor exhibited more stable and efficient performance than the untreated one. Hence, surface treatment provides the feasibility of improving the system stability for CO2 capture from the view of long-term operation.

Keywords: CO2 capture; Simultaneous removal of CO2 and SO2; Hollow fiber membrane contactor; Membrane gas absorption; Partial wetting; Surface modification.

III

Sammanfattning

En av de tekniker som under senare framhållits som ett lovande alternativ till konventionell CO2-avskiljning är membran-gas-absorptionstekniken på grund av god prestanda vad gäller masstransport. Det blandade angreppssättet med både kemisk absorption och membranseparation har en rad fördelar, såsom driftflexibilitet, kompakt konstruktion, högt yt-volymsförhållande, linjär uppskalning, modularitet och förutsägbar prestanda. En av de viktigaste utmaningarna för membran-gas-absorptionstekniken är vätningen av membranet med absorbenten under långa drifttider, vilket väsentligt kan minska membranmodulens masstransportkoefficienter.

I avhandlingen har en rad olika driftparametrars påverkan på CO2-reningsgraden och massöverföringshastigheten undersökts. Driftparametrar inkluderar gas- och vätskeflöden, typ av absorbent och koncentration och volymfraktion av CO2 vid gasinloppet. Avskiljning av SO2 och CO2 har dessutom undersökts för att utvärdera möjligheten att samtidigt, i samma membranenhet, avlägsna svavel och kol. Under 14 dagars kontinuerlig drift konstaterades det att massöverföringshastigheten för CO2 minskade avsevärt med drifttiden, vilket hänfördes till partiell vätning av membranet.

För att bättre förstå mekanismerna för vätning av membranporer under långvarig kontakt med absorbenter genomfördes doppningsexperiment i upp till 90 dagar. Olika metoder för karakterisering av membran användes för att illustrera vätningsprocessen före och efter det att membranfibrerna exponerades för absorbenterna. Resultaten av karakteriseringen visade att absorbentmolekylerna spreds in i polypropenpolymeren under kontakten med membranet, vilket ledde till att membranet svällde. Dessutom undersöktes effekterna av driftsparametrar såsom nedsänkningstid och typ av absorbent i detalj. Slutligen, på grundval av analysresultaten, diskuterades metoder för att underlätta vätningen av membran. Att förbättra polypropylenmembranets hydrofobicitet genom modifiering av ytan föreslogs kunna vara ett effektivt sätt att förbättra den långsiktiga driftprestandan för membranenheter. Därför modifierades de ihåliga fibrerna av polyproylen med ett tunt lager av en superhydrofob beläggning på membranets yta för att förbättra hydrofobiciteten. En blandning av cyklohexanon och metyletylketon ansågs vara det bästa icke-lösningsmedlet för att få en fiber yta med god homogenitet och acceptabelt hög hydrofobicitet. Under lång driftperiod, uppvisade den modifierade membranenheten stabilare och effektivare prestanda än den obehandlade. Därför erbjuder ytbehandling en möjlighet till att förbättra systemets stabilitet för CO2-avskiljning när det gäller långsiktig drift.

Nyckelord: CO2-avskiljning, samtidig avskiljning av CO2 och SO2, ihålig fiber, membran, gas absorption, vätning, ytmodifiering.

V

Acknowledgments

I would like to express my deep and sincere gratitude to my supervisor in MDH, Professor Jinyue Yan, for his continued instructive guidance, invaluable suggestions and unlimited encouragement and support during my studies. His wide knowledge and his logical way of thinking have been of great value for me. He has provided an excellent example as a successful scientist and professor.

I am deeply grateful to my supervisor in China, Professor Shandong Tu, who introduced me to the field of environmental engineering and provided me a lot of valuable opportunities to broaden my horizons. His understanding, encouragement and personal guidance have provided a good basis for the present thesis. I do appreciate his continuous guidance and support over these years.

I owe my sincere gratitude to my co-supervisor in MDH, Professor Erik Dahlquist, whose enthusiasm and valuable advice are contagious and motivational for me, even during tough times in my Ph.D. pursuit. In addition, I wish to express my warm thanks to him and his family for their friendly help in life during my stay in Sweden.

I am also grateful to my co-supervisor in China, Dr. Xinhai Yu for his guidance, patience and support. He is always ready to help me for experimental setup, constructive criticism, valuable advice, lots of good ideas and encouragement.

I would like to thank Bioenergy Group members: Dr. Eva Thorin, Dr. Hailong Li, Dr. Weilong Wang, Eva Nordlander, Han Song, Lilia Daianova, Ana Paz, Elena Tomas Aparicio and Johan Lindmark for their care and attentions during my stay in Sweden. I wish to thank Dr. Emma Nehrenheim and Dr. Niklas Hedin for their valuable comments. Thank all kind and helpful staffs of HST department.

I gratefully acknowledge Swedish Research Links Program, Mälardalen University and School of Mechanical and Power Engineering, East China University of Science and Technology (ECUST) for financial supporting of my research. I do appreciate China Scholarship Council and the Education Section of the Embassy of the People’s Republic of China in Sweden for the guidance and funding support.

Finally, I am forever indebted to my beloved parents and husband for their love, moral supporting, understanding, endless patience and encouragement when it was most required.

VII

List of papers

The author of this thesis is the main contributor to papers I-VIII. All these paper were written under the supervision of Professor Jinyue Yan and Professor Erik Dahlquist at Mälardalen University (Sweden), Professor Shan-tung Tu and Dr. Xinhai Yu at ECUST (China) who provided useful ideas, suggestions and comments.

Reprints are made with permission from the respective publishers.

This thesis is mainly based on the following five papers:

I. Y.X. Lv, J.Y. Yan, S.T. Tu, X.H. Yu, E. Dahlquist. CO2 capture by the absorption process in the membrane contactors. MATHMOD 2009 - 6th Vienna International Conference on Mathematical Modeling, Austria, February 11-13, 2009.

II. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Experimental investigation on CO2 absorption using absorbent in hollow fiber membrane contactor. International Scientific Conference on Green Energy with energy management and IT, Stockholm, March 12-13, 2008. (Journal of Nanjing University of Technology (Nature Science Edition), 2009, 31(5), 96-101.)

III. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Experimental studies on simultaneous removal of CO2 and SO2 in a polypropylene hollow fiber membrane contactor. International Conference on Applied Energy, ICAE 2011, Italy, May 16-18, 2011.

IV. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Wetting of polypropylene hollow fiber membrane contactors. Journal of Membrane Science. 2010, 362 (1-2): 444-452.

V. Y.X. Lv, X.H. Yu, J.J. Jia, S.T. Tu, J.Y. Yan, E. Dahlquist. Fabrication and characterization of superhydrophobic polypropylene hollow fiber membranes for carbon dioxide absorption. Applied Energy. 2010, In press: doi:10.1016/j.apenergy.2010.12.038.

VIII

Other related publications not included in this thesis:

VI. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Influence of MEA and MDEA solutions on surface morphology of porous polypropylene membranes. First International Conference on Applied Energy, Hong Kong, January 5-7, 2009.

VII. Y.X. Lv, J.J. Jia, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Experimental study on PP membrane modification and performance evaluation for CO2 absorption in hollow fiber membrane contactors. International Conference on Applied Energy, 2010, Singapore, April 21-23, 2010.

VIII. Y.X. Lv. Experimental studies on CO2 absorption in hollow fiber membrane contactor. Mälardalen University Press. Licentiate Theses No. 121. ISBN: 978-91-86135-75-1.

IX. J.J Jia, Xinhai Yu, Y.X. Lv. Preparation and Characterization of Polypropylene Membranes Modified with Grafting Method. International Symposium on Polymer. Tianjin, August 18-22, 2009.

IX

List of Tables Table 2.1 Specifications of the hollow fiber membrane module .....................12Table 4.1 Outer surface roughness of studied membrane fibers......................32

Lists of FiguresFigure 1.1 Schematic representation of CO2 capture systems.............................1Figure 1.2 Schematic drawing of membrane gas absorption process .................2Figure 1.3 Photos of hollow fiber membrane contactor......................................2Figure 1.4 Membrane contactors for Industrial pertraction ................................3Figure 1.5 Three operating modes of hydrophobic hollow fiber membrane

contactor..........................................................................................................5Figure 1.6 Diagram of the thesis structure ..........................................................9Figure 2.1 Schematic drawing of gas absorption process for CO2/SO2 removal..

.........................................................................................................12Figure 2.2 Photos of overall experimental setup...............................................13Figure 2.3 Photos of main instruments used in the gas absorption system.......13Figure 2.4 Effects of liquid flow rate on removal efficiencies and mass transfer

rates of CO2 and SO2.....................................................................................16Figure 2.5 Long-term performance of membrane contactor over two weeks ...18

Figure 4.1 Flow chart of superhydrophobic modification of PP hollow fibers.27Figure 4.2 Modification instruments .................................................................28Figure 4.3 Schematic drawing of coating machine for PP membrane

modification ..................................................................................................28Figure 4.4 SEM image for studied membranes using cyclohexanone ..............29Figure 4.5 SEM image for studied membranes using MEK .............................30Figure 4.6 SEM image for studied membranes using the mixture of MEK and

cyclohexanone...............................................................................................31Figure 4.7 Three-dimensional AFM image for PP membrane ..........................32Figure 4.8 Two-dimensional AFM image for PP membrane.............................33Figure 4.9 Contact angles of the outer surface of the hollow fiber membrane. 34Figure 4.10 Influences of modification of the PP membrane on CO2 mass

transfer rate during long-time operation .......................................................35

XI

Nomenclature and Abbreviations

Symbols

η Removal Efficiency [%]

JCO2 Mass Transfer rate of CO2 [mol/m2h]

Qin Inlet Gas Flow Rate [m3/h]

Qout Outlet Gas Flow Rate [m3/h]

Cin CO2 Volumetric Fraction in the Gas Inlet [%]

Cout CO2 Volumetric Fraction in the Gas Outlet [%]

Tg Gas Temperature [K]

S Gas-liquid Mass Transfer Area [m2]

Z Difference between the highest and the lowest points

Ra Mean roughness [nm]

Rms Root mean square of Z [nm]

Δp Penetration pressure [KPa]

σL Surface tension of the liquid [mN/m]

θ Contact angle between the liquid phase and membrane [°]

dp Membrane pore diameter [µm]

XII

Abbreviations

CCS CO2 Capture and Storage

IEA International Energy Agency

MEA Monoethanolamine

MDEA Diethanolamine

AMP 2-Amino-2-methyl-1-propanol

PZ Piperazine

PE Polyethylene

PP Polypropylene

PTFE Polytetrafluoroethylene

PVDF Polyvinylidenefluoride

Teflon Polytetrafluoroethylene

HFMC Hollow Fiber Membrane Contactor

TCD Thermal Conductivity Detector

FE-SEM Field Emission Scanning Electron Microscope

AFM Atomic Force Microscope

XPS X-ray Photoelectron Spectroscopy

ATR-IR Attenuated Total Reflection-Infrared Spectroscopy

TGA Thermo-Gravimetric analysis

MEK Methyl ethylketone

DEA Diethanolamine

CORAL CO2 Removal Absorption Liquid

XIII

Table of ContentsAbstract ...................................................................................................................... I Sammanfattning ........................................................................................................... III Acknowledgments......................................................................................................... V List of papers.............................................................................................................. VII List of Tables ................................................................................................................ IX Lists of Figures ............................................................................................................ IX Nomenclature and Abbreviations ................................................................................. XI Table of Contents ...................................................................................................... XIII 1 Introduction .................................................................................................. 1

1.1 Background .................................................................................................. 1 1.1.1 Current status of CO2 capture and storage ......................................... 1 1.1.2 Advantages and disadvantages of gas-liquid membrane contactor ... 3 1.1.3 State of the art on absorbents in membrane gas absorption system ... 6 1.1.4 State of the art on membranes in membrane gas absorption system . 6

1.2 Objectives ..................................................................................................... 7 1.3 Methodology ................................................................................................ 8 1.4 Thesis outline ............................................................................................... 9

2 CO2/SO2 removal by membrane gas absorption ........................................ 11 2.1 Experimental setup ..................................................................................... 11

2.1.1 Materials and instruments ................................................................ 11 2.1.2 Flow chart of membrane gas absorption process ............................. 11 2.1.3 CO2 removal efficiency and mass transfer rate ................................ 13

2.2 Influences of various operating parameters on CO2 absorption ................. 14 2.2.1 Range of operating parameters ........................................................ 14 2.2.2 Effects of flow rates of liquid and gas on CO2 removal .................. 14 2.2.3 Effects of absorbent type and concentration on CO2 removal ......... 14 2.2.4 Effects of CO2 volume fraction on CO2 removal ............................. 15

2.3 Simultaneous removal of CO2 and SO2 by membrane gas absorption ....... 15 2.3.1 Influence of absorbent flow rate on simultaneous absorption of CO2

and SO2 ............................................................................................ 16 2.3.2 Influence of SO2 on CO2 absorption ................................................ 17

2.4 Long-term CO2 removal performance of membrane gas absorption ......... 17 2.5 Retrieval of absorption performance .......................................................... 18 2.6 Section summary ........................................................................................ 19

3 Wetting of PP hollow fiber membrane ....................................................... 21 3.1 Immersion experiments .............................................................................. 21 3.2 Characterization methods ........................................................................... 21

3.2.1 X-ray photoelectron spectroscopy measurement (XPS) .................. 22 3.2.2 Attenuated Total Reflection-Infrared Spectroscopy (ATR-IR) ........ 22 3.2.3 Thermogravimetric analysis (TGA) ................................................. 22 3.2.4 Contact angle measurement ............................................................. 22

XIV

3.2.5 Atomic Force Microscope (AFM) analyses ..................................... 23 3.2.6 Field Emission Scanning Electron Microscope (FE-SEM) ............. 23

3.3 Liquid breakthrough pressure ..................................................................... 23 3.4 Investigation on membrane wetting evolution ........................................... 23 3.5 Effects of membrane-absorbent interaction on PP membrane properties .. 24

3.5.1 Changes of hydrophobicity characteristics ...................................... 24 3.5.2 Membrane morphological changes in terms of SEM and AFM

images .............................................................................................. 24 3.6 Critical parameters affecting the breakthrough pressure ............................ 25 3.7 Methods for smoothing the membrane wetting .......................................... 25 3.8 Section Summary ....................................................................................... 26

4 Fabrication and characterization of superhydrophobic PP membrane ....... 27 4.1 Fabrication of superhydrophobic membrane surface ................................. 27 4. 2 Characterizations of modified membrane fibers ........................................ 29

4.2.1 SEM images of the modified membrane ......................................... 29 4.2.2 Membrane surface roughness changes before and after modification

......................................................................................................... 32 4.2.3 Hydrophobicity improvement by membrane modification .............. 33

4.3 Long-term performance of the modified membrane module ..................... 34 4.4 Section Summary ....................................................................................... 35

5 Conclusions ................................................................................................ 37 6 Future work ................................................................................................ 39 7 Reference .................................................................................................... 41

1

1 Introduction

1.1 Background

1.1.1 Current status of CO2 capture and storage

CO2 capture and storage (CCS) in geological formations has been recognized as a technically available option for mitigating atmospheric emissions of CO2 due to human activities. Current CO2 capture system mainly includes post-combustion capture, pre-combustion capture and oxy-fuel combustion capture processes, as shown in Figure 1.1. CO2 capture from flue gas produced by fossil fuels combustion is referred to be post-combustion capture, corresponding to the most widely applicable option in terms of industrial sections and is compatible to a retrofit strategy. Recent studies on post-combustion CO2 capture are mainly focused on chemical and physical absorption [1], solid adsorption [2], cryogenic distillation [3] and membrane techniques [4]. Among these technologies, chemical absorption with aqueous amines is considered to be the most well established CO2 capture option given its past commercial applications in other industries, such as natural gas processing, hydrogen and ammonia manufacturing. However, it suffers from high capital cost and a variety of operational problems, e.g. liquid channeling, flooding, entrainment and foaming. For the gas separation membrane, it is hard to ensure the membrane selectivity and permeability simultaneously.

Figure 1.1 Schematic representation of CO2 capture systems [5]

The main challenge for CO2 capture system is the large amount of energy consumption which reduces the net plant efficiency significantly [6]. In the CO2 chemical absorption process, the regeneration energy consumption for CO2-riched amine is estimated to be 15%-30% of the net power production of a coal-fired power plant when 90% CO2 is captured [7]. Under 2002 conditions, it is estimated that application of CCS to electricity production will increase electricity generation costs by about 0.01-0.05 US dollars per kilowatt hour, depending on the fuel, the specific technology, the location and the national circumstances [5]. Therefore, many researchers have examined the possibilities of

2

enhancing the process efficiency and reducing the effects of limitations and drawbacks by the process combination.

As an integration of chemical absorption technology and membrane separation technology, the membrane gas absorption technology was developed with the purpose of reducing the cost and improving the performance of post-combustion CO2 capture system. The schematic drawing of membrane gas absorption process is shown in Figure 1.2. The feed gas passes through the lumen or shell side of the hollow fibers, and the liquid absorbent flows countercurrent on the other side. Instead of depending on the membrane selectivity, the liquid flowing in the hollow fiber membrane contactor provides the selectivity and the porous, unselective membrane only acts as the contacting interface of the liquid and gas phases. The gases diffuse through the membrane pores to the other side of the membrane where they are absorbed in the absorbent and taken away from the contactor.

Figure 1.2 Schematic drawing of membrane gas absorption process

The essential element is the porous hydrophobic hollow fiber membrane contactor, as shown in Figure 1.3.

Figure 1.3 Photos of hollow fiber membrane contactor

3

1.1.2 Advantages and disadvantages of gas-liquid membrane contactor

In comparison with conventional absorption equipment, such as packed towers, bubble columns or spray towers, the membrane contactor has the following advantages [8]:

Operational flexibility. The liquid and gas phases flow in shell side and lumen side separately without direct contact. These two phases can be manipulated independently, thus avoiding the operational problems encountered in conventional columns.

Compact size. Due to the large surface area to volume ratio of membrane, the membrane contactor is less voluminous and more effective. Theoretically based performance comparisons between packed columns and membrane absorbers shown that the membrane absorber led to a near ten-fold reduction in absorber size for CO2 removal from flue gas [9][10]. Rangwala et al. [11] also noted a three-to-nine-fold increase in the overall mass transfer rate when using hollow fiber device over a conventional packed column. It was reported that hollow fiber contactors were about 30 times more efficient for gas absorption than conventional equipment and may reduce the size of the absorber and stripper units by 65% [12]. Yeon et al.[13] compared the performance of CO2 absorption in porous polyvinylidenefluoride (PVDF) hollow fiber membrane contactor with that in conventional packed column, and concluded that CO2 absorption rate per unit volume of the membrane contactor was 2.7 times higher than that of the packed column.

Easy scale-up. The modularity of membrane contactor makes it easier to design a membrane plant to operate over a wide range of capacities. The operation can be linearly scaled up according to the practical requirements by increasing or decreasing the membrane modules. Chemical firm KoSa Netherlands BV in Netherlands launched a full scale industrial installation in 1998, using a series of polypropylene hollow fiber membrane contactors to extract organic compound from waste water [14]. The photo is shown in Figure 1.4.

Figure 1.4 Membrane contactors for Industrial pertraction

4

Because of its excellent mass transfer properties, membrane gas absorption technology has currently been identified as a technically viable option by the International Energy Agency’s (IEA) working group on CO2 capture [15], and has been considered as one of promising alternatives to conventional and potential large scale application technology for the recovery and removal of CO2 [16]. Kvaerner Oil & Gas and W.L. Gore & Associates GmbH have been developing a membrane gas absorption process for the removal of acid gases from natural gas and exhaust of the offshore gas turbines [17]. Feasibility study has also demonstrated that the CO2 can be produced economically from flue gas on a large scale [18]. A development project is currently underway in which a pilot plant producing around 150 kg CO2/h will be built. Successful completion of the development will then pave the way towards a demonstration on a large scale (10 tonnes CO2/h) for the greenhouse [19].

Although the membrane contactor offers many advantages over conventional contacting equipment, additional mass transfer resistance is introduced due to the existence of membrane phase. The membrane pores can be theoretically filled with either gas for the hydrophobic membrane or liquid for the hydrophilic membrane, corresponding to non-wetting mode and overall-wetting mode, respectively. For membrane gas absorption system, it is essential to avoid a strong increase in mass transfer resistance in a liquid filled membrane pore compared to a gas filled pore. However, in practical application, the aqueous solutions with organic absorbents can penetrate into partial pores of the hydrophobic membrane and the membrane contactor is operated under partial wetting mode. Figure 1.5 depicts these three operating modes.

Wetting phenomenon of the membrane leads to the increase of overall mass transfer resistance and deterioration of membrane performance [20]. Zhang et al. [21] found that the flux dropped severely after two days of operation due to the appearance of DEA droplets in gas phase even the membrane contactor was operated for a short period. Keshavarz et al. [22] considered the effect of chemical reactions in the wetted pores on the prediction of membrane wetting fraction, and their theoretical results showed that the absorption flux was decreased significantly by membrane wetting, even in very low fractions. Theoretical calculation and experimental data of Lu et al. [23] and Kreulen et al. [24] concluded that the wetting of membrane pores significantly affected the mass transfer coefficients of the membrane module, and thus the membrane phase resistance increased sharply and the operation performance declined rapidly. Mahmud et al. [25][26] stripped the air from water with porous polypropylene hollow fiber membrane modules, and attributed the difference between theoretical prediction and experimental results to the partial wetting of membrane pores. They found that the membrane mass transfer resistance of the partially wetted pores was two orders of magnitude higher than that of air-filled pores. Mavroudi et al. [27] studied the membrane wetting in gas absorption membrane systems and found that the membrane mass transfer resistance accounted for 20-50% of the total resistance in the case of liquid-filled membrane pores. Therefore, the long-term compatibility of the membranes with absorbents or the membrane wetting by the absorbents has been a big concern and drawback for the practical application of membrane contactors.

5

Membrane pore

Gas

Liquid

Gas-liquid contact interface

Gas

(a) Non-wetting mode

Membrane pore

Gas

Liquid


Gas

(b) Overall wetting mode

Membrane pore

Gas

Liquid


Gas

(c) Partial wetting mode

Figure 1.5 Three operating modes of hydrophobic hollow fiber membrane contactor

6

1.1.3 State of the art on absorbents in membrane gas absorption system

For the membrane gas absorption system, the liquid absorbents provide the selectivity and the porous membrane only acts as the physical barrier between the gas and liquid phases. Therefore, the proper solvent selection is a determining and critical step to reduce the capital and energy costs for CO2 absorption in membrane contactor. The commercial absorbents should be qualified with the following features: high capacity and solubility rate of CO2 for high mass transfer rate, high chemical and thermal stability, low vapor pressure to minimize the losses, easy regeneration for recycle, commercially available at low cost, low corrosiveness, low viscosity and non-toxic [28][29]. In recent years, various liquid absorbents including pure water and aqueous solutions of NaOH, KOH, monoethanolamine (MEA), diethanolamine (DEA) and N-methyldiethanolamine (MDEA) are used as absorption liquids in polyethylene (PE) or polypropylene (PP) or polytetrafluoroethylene (PTFE) porous hydrophobic hollow fiber membrane contactors [30][31][32][33], in which the MDEA and MEA aqueous solutions in PP hollow fiber membrane contactor are the most widely used for CO2 absorption. The use of piperazine (PZ) as the activator for those amines is also the subject of interest. The mass-transfer flux was enhanced when 2-Amino-2-methyl-1-propanol (AMP) and PZ were added into MDEA solution as the activators to activate the gas-liquid mass transfer process [34].

In a gas-liquid membrane contactor, the liquid phase contacts directly with the hollow fiber membrane, so the compatibility between the absorbent and the membrane should also be considered. The membrane should not be damaged physically or chemically during the contact with the liquid absorbents, so as to ensure long-term stability of the membrane module. But most absorbent solutions, especially the organic compounds, can penetrate into the membrane pores and result in partial membrane wetting gradually with time. The polypropylene membranes suffered from significant changes in surface morphology after being exposed to water and MEA [35][36]. Therefore, Yan et al. [37] developed a new liquid absorbent, aqueous potassium glycinate (PG) with high surface tension, to eliminate the wetting problem of commercial PP porous membrane contactor. Kumar et al. [38] proposed a new chemical absorbent based on amino acid salts for CO2 absorption with no wetting effect on polyolefin fibers. Feron et al. [39] also employed novel absorption liquids named CO2 Removal Absorption Liquid (CORAL) to remove CO2 from various feed gases in porous polyolefin membrane module.

1.1.4 State of the art on membranes in membrane gas absorption system

For the membrane gas absorption process, the membrane does not provide the selectivity. It only acts as the gas-liquid contact interface. Generally, hydrophobic porous membranes can be used as a gas-liquid membrane contactor. The most widely used hydrophobic polymers include PP, PVDF, PTFE, PE membranes. PP, PTFE and PE membranes are usually prepared by stretching and thermal methods, while PVDF asymmetric membranes are prepared via phase inversion method. PP and PE hollow fiber membranes are widely used in industry due to their low cost and commercial availability in various sizes, but their hydrophobicity is lower than PTFE, PVDF membranes and other fluorine-containing

7

polymers. Khaisri et al. [40] compared the CO2 absorption performance by MEA solutions in three hollow fiber membrane contactors, and they found that PTFE showed the best absorption performance and operation stability, followed by PVDF membranes, PP showed the least. Only porous PTFE membrane shows good gas absorption performance and stability without membrane wetting by MEA aqueous solutions for more than 6,600 hours due to its higher hydrophobicity [41]. However, the application of PTFE is limited for its high production cost and is unavailable in small diameters. The hydrophobicity of polymer membranes can be improved by various modification methods, such as pretreatment with fluorocarbonic material [41], surface grafting [42], interfacial polymerization [43] and solvent casting [44].

In addition to the above polymers, the grafted ceramic hollow fiber membrane by aluminum oxides exhibits high chemical-thermal stability, mechanical strength and high hydrophobicity, showing great potential applications for CO2 removal in a membrane contactor [45][46]. But the price of ceramic membranes is much higher than that of PE and PP membranes. The techno-economic analysis of different membrane materials for CO2 capture by membrane gas absorption process is still unknown.

1.2 Objectives

As aforementioned, CO2 removal by absorbents in a hollow fiber membrane contactor has attracted great attentions in recent years, and exciting experimental and theoretical results have been reported. But there are few reports on the simultaneous removal of acid gases in a gas-liquid membrane contactor, and the wetting process is still arguing. There is a need to investigate the wetting mechanism, and then find an effective way to eliminate or reduce the influence of membrane wetting from the perspective of long-term stability of system operation performance. The main objectives of this thesis are as follows:

To investigate the effects of various operating parameters on CO2 absorption performance in PP hollow fiber membrane contactor in the laboratory;

To investigate the feasibility of simultaneous removal of CO2 and SO2 and effects of SO2 on CO2 absorption;

To study the long-term system performance stability;

To better understand the wetting phenomenon during the operation by various characterization methods;

To smooth or eliminate membrane wetting by increasing the gas phase pressure or by improving the membrane surface hydrophobicity via membrane surface modification.

The detailed objectives of each chapter are as follows:

(1) The start-of-art on membrane gas absorption technology for CO2 capture from flue gas.

Paper I. Y.X. Lv, J.Y. Yan, S.T. Tu, X.H. Yu, E. Dahlquist. CO2 capture by the absorption process in the membrane contactors. MATHMOD 2009 - 6th Vienna International Conference on Mathematical Modeling, Austria, February 11 - 13, 2009.

8

(2) Experimental equipment was set up in the laboratory of ECUST to investigate the effects of various operating parameters on the CO2 removal efficiency and mass transfer rate, such as mixed gas flow rate, the CO2 volume concentration at the feed gas inlet, liquid flow rate, the absorbent type and concentration.

Paper II. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Experimental investigation on CO2 absorption using absorbent in hollow fiber membrane contactor. International Scientific Conference on Green Energy with energy management and IT, Stockholm, March 12-13, 2008. (Journal of Nanjing University of Technology, 2009, 31(5), 96-101.).

(3) The feasibility of simultaneous removal of CO2 and SO2 from coal-fired flue gas by membrane gas absorption technology was studied. The wetting phenomenon of PP membrane was observed in 14 days of continuous experiments. The performance deterioration of the experimental system as a function of operation time was discussed.

Paper III. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Experimental studies on simultaneous removal of CO2 and SO2 in a polypropylene hollow fiber membrane contactor. International Conference on Applied Energy, ICAE 2011, Italy, May 16-18, 2011.

(4) Various membrane characterization methods were used to investigate the interaction between the membrane and the absorbent. The influences of contact time, absorbent concentration and absorbent type on membrane morphological changes were investigated. In addition, methods to smooth the membrane wetting were proposed.

Paper IV. Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Wetting of polypropylene hollow fiber membrane contactors. Journal of Membrane Science. 2010, 362 (1-2): 444-452.

(5) The PP hollow fibers were modified by depositing a rough layer on the surface to improve the hydrophobicity of membrane and enhance the system duration performance. Performance comparison between the original membrane contactor and the modified membrane contactor was carried out.

Paper V. Y.X. Lv, X.H. Yu, J.J. Jia, S.T. Tu, J.Y. Yan, E. Dahlquist. Fabrication and characterization of superhydrophobic polypropylene hollow fiber membranes for carbon dioxide absorption. Applied Energy. 2010, In press: doi:10.1016/j.apenergy. 2010.12.038.

1.3 Methodology In this thesis, experimental equipment was set up to investigate the effects of various operating parameters on CO2 absorption in a PP hollow fiber membrane contactor using three aqueous solutions as absorbents. The influence of SO2 on CO2 removal during the simultaneous removal of these two acid gases was also studied. After a few days of operation, it was found that the liquid penetrated into the gas phase through the membrane pores, indicating that membrane contactor was gradually wetted by the liquid. Therefore, immersion experiments for up to 90 days were carried out to investigate the interaction between the absorbents and the membrane. The effects of operating parameters on

9

membrane wetting were studied, including the immersion time, absorbent type and concentration. According to previous experimental results, the modified porous PP fibers with high hydrophobicity may be an alternative for stabilizing long-term operation performance and eliminating the influence of membrane wetting. Hence, the PP membrane fibers were modified by depositing a rough layer on the surface to improve their non-wettability. The system performance comparison between the unmodified membrane contactor and the modified membrane contactor was conducted to evaluate the availability of surface modification.

1.4 Thesis outline The schematic drawing of the thesis structure is illustrated in Figure 1.6. This thesis describes the experimental setup in the laboratory and the experiments carried out to further understand the absorption performance of CO2 removal in hollow fiber membrane contactor, including the selection of operating parameters, the wetting evolution, and solution to improve the long-term operation performance of membrane gas absorption technology.

Figure 1.6 Diagram of the thesis structure

The thesis is composed of the following eight chapters and based on five scientific papers:

Chapter 1 Present the research background of CO2 removal by membrane gas absorption process, determine the objectivities, methodology and structure of this thesis.

Chapter 2 Set up the experimental system in the laboratory; investigate the influences of various operating parameters on membrane absorption performance, including the flow rates of gas and liquid phases, absorbent type and concentration, and CO2 volume fraction at the inlet gas.

Investigate the feasibility of simultaneous removal of CO2 and SO2 using absorbent in the same membrane contactor. In addition, wetting phenomenon

10

of PP membrane fibers was observed in prolonged operation.

Chapter 3 Propose absorption-swelling wetting mechanism based on the results of immersion experiments; observe the changes of membrane surface morphologies and roughness via various characterization methods.

Chapter 4 Fabricate the superhydrophobic modification on the surface of PP hollow fibers; investigate the effects of modification on membrane fiber characteristics, membrane-absorbent interaction and the long-term performance of membrane module.

Chapter 5 Draw the final conclusions and summaries.

Chapter 6 Present further proposed work.

11

2 CO2/SO2 removal by membrane gas absorption

The experimental apparatus for CO2 recovery in a PP hollow fiber membrane contactor using MEA, MDEA and deionized water as the absorbents was set up. The dependency of CO2 absorption performance on various operating parameters was investigated, including the absorbent type and concentration, the CO2 volume fraction, the flow rates of gas and liquid phases. In addition, 14 days of long-term experiments were carried out to study the system operation stability.

In order to realize the co-capture of SO2 and CO2 in the real engineering applications, some of the important issues were also investigated, including the influence of SO2 on CO2 absorption, dependency of CO2 and SO2 removal efficiencies on the flow rates of flue gas and liquid absorbent.

This chapter presents the contributions of Paper II and Paper III.

2.1 Experimental setup

2.1.1 Materials and instruments

a. Absorbents: 99.5% grade MEA and MDEA purchased from Shanghai Bangcheng Chemical Co., Ltd. were chosen as the model absorbents in this study due to their commercial applicability and reasonable CO2 absorption capacity, representing for the cases of primary and tertiary amines respectively. While deionized water was the representative absorbent for physical absorption.

b. HDMF-100-1 type porous polypropylene hollow fiber membrane module was provided by Tianjin Blue Cross Membrane Technology Co., Ltd. The specifications are listed in Table 2.1.

c. Mass Flow Controller D07 of Sevenstar Electronics Co., Ltd MFC D07 was used to precisely control the inlet gas flow rate.

d. 9790 III gas chromatograph of FuLi Analytical Instrument Co., Ltd was used to analyze the inlet and outlet gas compositions.

e. Stainless steel peristaltic pump of Tian Li Liquid Industrial Equipment Factory was used to pump the liquid into the lumen side of the hollow fibers from solvent container.

f. Mass Flow Meter of Sevenstar Electronics Co., Ltd was used to measure the outlet gas flow rate.

g. Gas cylinders provided feed gas.

2.1.2 Flow chart of membrane gas absorption process

The schematic drawing of CO2/SO2 removal by liquid absorbents in a hollow fiber membrane contactor is shown in Figure 2.1. A gas mixture containing CO2/SO2 in balance of N2 with various volume ratios was fed into the system from compressed gas cylinders.

12

The gas flow rate was adjusted by Mass Flow Controller. Then the gases were introduced into the static mixer to be mixed uniformly. Pressure gauges at the inlet and outlet of the membrane module measured the gas pressures. The gas flow rate at the outlet was obtained by using a mass flow meter. The outlet gas compositions were analyzed on-line by a 9790 III gas chromatograph using a thermal conductivity detector (TCD). A stainless steel peristaltic pump was used to pump the liquid into the lumen side of the hollow fibers from solvent container, and the flow rate of liquid was controlled by a rotational flow meter. Under the same operating conditions, five samples of the gas and absorbent samples were taken and the average value was calculated. All experiments were carried out at atmospheric pressure (0.1 MPa) and at room temperature (20 °C).

Table 2.1 Specifications of the hollow fiber membrane module

Parameter Value

Module outer diameter (mm) 50

Module inner diameter (mm) 42

Module length (mm) 360

Fiber inner diameter (μm) 380

Fiber outer diameter (μm) 500

Fiber length (mm) 300

Fiber porosity 0.65

Pore size (μm) 0.16

Figure 2.1 Schematic drawing of gas absorption process for CO2/SO2 removal

13

Some photos of overall experimental setup are shown in Figure 2.2 and Figure 2.3.

Figure 2.2 Photos of overall experimental setup

Figure 2.3 Photos of main instruments used in the gas absorption system

2.1.3 CO2 removal efficiency and mass transfer rate

Removal efficiency and mass transfer rate of CO2 (SO2) were used to evaluate the separation properties of hollow fiber membrane module, which can be calculated by Eq.(2-1) and Eq.(2-2) :

14

in in out out

in in

Q C Q CQ C

(2-1)

273.150.0224

in in out outgas

g

Q C Q CJ

T S

(2-2)

Where η denotes the CO2 (SO2) volumetric removal efficiency, %; JCO2 is the CO2 mass transfer rate, mol/(m2·h); Qin and Qout represent the inlet and outlet gas flow rate respectively, m3/h; Cin and Cout are the CO2 (SO2) volumetric fraction in the gas inlet and outlet respectively, %; Tg is the gas temperature, K; S represents the gas-liquid mass transfer area and herein equals to the effective membrane area, m2.

2.2 Influences of various operating parameters on CO2

absorption

2.2.1 Range of operating parameters

(a) Absorbent: 0.05-0.25 mol /L MEA, 0.05-0.25 mol/L MDEA, deionized water

(b) CO2 volume fraction in feed gas: 10%-40%

(c) Gas flow rate: 75-200 mL/min

(d) Liquid flow rate: 17-68 mL/min

2.2.2 Effects of flow rates of liquid and gas on CO2 removal

With the increase of liquid flow rate, the absorbent was supplied at a higher speed and the consumed absorbent was replaced by more fresh absorbent, resulting in a lower average CO2 concentration in the liquid phase. The driving force for CO2 transfer was increased, leading to a more efficient gas removal. Therefore, both CO2 removal efficiency and mass transfer were increased with the increase in liquid flow rate.

The increase of gas flow rate decreased the gas retention time in the contactor and increased the CO2 concentration at the gas-liquid interface, resulting in an increase in the mass transfer rate for the absorbents. Increase of gas flow rate reduced the thickness of gas boundary layer and enhanced the gas mass transfer, which was favorable for the CO2 removal. However, it simultaneously decreased the residence time of gas in the membrane contactor, which was unfavorable for the CO2 removal. Therefore, higher gas flow rate led to the increase in mass transfer rate but the decrease in removal efficiency.

2.2.3 Effects of absorbent type and concentration on CO2 removal

Under the same experimental conditions, the use of chemical aqueous solutions (MEA and MDEA) enhanced the mass transfer of CO2 compared with deionized water; the CO2 removal efficiency of MEA was higher than that of MDEA especially due to the higher rate of MEA reacting with CO2.

Higher removal efficiency and mass transfer rate could be effectively achieved by increasing the solvent concentration. The reason was that, with the increase in absorbent concentration, the effective component absorbing CO2 in the liquid boundary layer

15

increased, resulting in higher CO2 transfer rate into the liquid. As CO2 entered the liquid and reacted with the corresponding solvent, the CO2 concentration decreased in liquid-gas boundary layer. It enhanced the CO2 solubility rate and increased the CO2 removal efficiency. The CO2 removal efficiency could be as high as 95% with the MEA concentration of 0.25 mol/L.

However, it should be noted that, even through higher solvent concentration leads to high removal efficiency and higher mass transfer rate, it also increases the risk of membrane wetting especially for long-term operation. The membrane wetting will be discussed in detail in other sections. Therefore, the absorbent concentration should be compromised between removal efficiency and the wetting to ensure a stable and efficient absorption of CO2 over a long life of the hollow fiber membrane. In the next chapter, we will further investigate the wetting evolution and find ways to smooth the membrane performance deterioration.

2.2.4 Effects of CO2 volume fraction on CO2 removal

With the increase of CO2 volume fraction, more liquid was consumed with higher CO2 concentration at the membrane interface. The liquid was insufficient relative to higher CO2 concentration at a constant liquid flow rate, which resulted in a decrease of CO2 removal efficiency. At the same time, the CO2 concentration gradient at the liquid-gas boundary layer increased with increasing CO2 volume fraction. The CO2 driving force of mass transfer in the gas was enhanced, which led to the increase of CO2 diffusion mass transfer rate. Therefore, more CO2 was absorbed in the liquid by permeating the membrane module.

2.3 Simultaneous removal of CO2 and SO2 by membrane gas absorption

In addition to CO2 removal, the gas-liquid membrane contactors have also been used to remove other acid gases, such as H2S [47], SO2 [48] and NO2 [49]. Recently, the application of the porous hollow fiber membrane contactor for simultaneous absorption of acid gases has gained an increasing attention as a relatively new concept. The co-capture of acid gases may reduce the equipment investment and operation costs. Experimental and mathematical modeling studies were conducted on the simultaneous removal of CO2 and H2S in hollow fiber modules using aqueous solutions DEA [50], potassium carbonate [51], MEA [52], blends of AMP and DEA [53], respectively. However, the simultaneous removal of two major pollutants (CO2 and SO2) from coal-fired flue gas by membrane gas absorption technology has seldom been reported in the literatures so far. Accordingly, we investigated the feasibility of simultaneous removal of CO2 and SO2 in a PP hollow fiber membrane contactor using 0.5 mol/L MEA aqueous solution as the absorbent. The SO2 and CO2 removal efficiencies and mass transfer rates at variable liquid flow rates are shown in Figures 2.4. The flow rate of feed gas was 150 mL/min. Feed gas I (CO2-SO2-N2 system: Cin, CO2: 20%, Cin, SO2: 16000 ppm, N2 balanced) and feed gas II (CO2- N2 system: Cin, CO2: 20%, N2 balanced) which are both in the composition range of flue gas were used as the simulated flue gas, respectively.

16

2.3.1 Influence of absorbent flow rate on simultaneous absorption of CO2

and SO2

When the liquid flow rate was increased from 9 to 45 mL/min, CO2 removal efficiency was increased sharply, while the SO2 removal efficiency was increased rapidly in the beginning and then stabilized at 100% after the liquid flow rate was increased to 34 mL/min. At low liquid flow rate, the absorbent available at the membrane interface was insufficient, thus suppressing the gas transfer. With the increase of liquid flow rate, the liquid disturbance was enhanced, which resulted in a higher speed of gas diffusing into the liquid. The consumed absorbents at the membrane boundary layer could diffuse into the liquid phase at a higher speed due to the increase in liquid flow rate. Therefore, a higher solvent concentration gradient could be maintained at the gas-liquid interface, enhancing the CO2 and SO2 removal efficiency. When the liquid flow rate was higher than 34 mL/min, the SO2 was completely removed from the feed gas.

5 10 15 20 25 30 35 40 45 5040

50

60

70

80

90

100

40

50

60

70

80

90

100

Feed gas I: CO2-SO2- N2

Feed gas II: CO2- N2


Liquid flow rate (mL min-1)

SO2 re

mov

al e

ffici

ency

(%)

CO

2 rem

oval

effi

cien

cy /[

%]

(a)

5 10 15 20 25 30 35 40 45 500

1

2

3

4

5

6

0

1

2

3

4

5

6

J SO

2×1

04 /[m

ol(m

2 .s)-1

]

J CO

2×1

04 /[m

ol(m

2 .s)-1

]

Liquid flow rate (mL min-1)

Feed gas II: CO2- N2



(b)

Figure 2.4 Effects of liquid flow rate on removal efficiencies and mass transfer rates of

CO2 and SO2

17

As shown in Figure 2.4b, when the liquid flow rate was increased from 9 to 45 mL/min, the mass transfer of CO2 was correspondingly improved from 2.22 to 4.95 mol/m2s. With the increase in liquid flow rate, the thicknesses of gaseous and liquid-phase boundary layers decreased, leading to the enhancement of the mass transfer rate. However, the SO2 mass transfer rate was increased slightly from 0.35 to 0.41 mol/m2s under the same condition. This was mainly due to the low SO2 amount in the feed gas.

2.3.2 Influence of SO2 on CO2 absorption

The influence of SO2 existence on CO2 absorption was investigated, and the corresponding results are also shown in Figure 2.4. It was clear that the removal efficiency and mass transfer rate of CO2 in CO2-SO2-N2 system were slightly lower in comparison with CO2 -N2 system at a given liquid flow rate. When desulphurization and decarbonization were carried out simultaneously by the MEA absorbent in the membrane contactor, the SO2 competed with CO2 and consumed a part of effective component in the absorbent, especially when the absorbent was relatively insufficient at low flow rate. A slight influence of SO2 on the CO2 absorption in the liquid was related to the affinity of SO2 towards the absorbent and the competition with CO2. The small amount of SO2 in the flue gas had a limited impact on the absorption of CO2. However, the influence of SO2 could not be negligible when the membrane gas absorption system was operated under recycle mode, that was, regenerated MEA rather than fresh MEA was used as the absorbent. Because the heat stable alkanolamine sulfates accumulated in the system and decreased the efficiencies of CO2 and SO2 absorption. Hence, fresh MEA solutions should be added to the system under that condition. The alkanolamine sulfates could be regenerated according to methods of Huang et al. [54].

2.4 Long-term CO2 removal performance of membrane gas absorption

Even though the PP membrane contactor used in gas absorption is intensively hydrophobic, in the practical experiments, we found that some absorbent drops accumulated at the outlet of gas phase, indicating that the absorbent solutions penetrated into partial pores of the hydrophobic membrane, thus the membrane contactor was operated under partial-wetting mode. Most of experimental and theoretical investigations on membrane contacting systems have been done very close to system start up (after steady condition) to prevent difficulties of wetting. However, it is clear that the study of long time operation of such systems is unavoidable, especially for industrial applications. Therefore, the system was operated continuously for 2 weeks using 0.5 mol/L MEA aqueous solution as the absorbent.

Figure 2.5 depicts the CO2 mass transfer rate as a function of operation time. CO2 mass transfer rate was stable within the first 3 days, indicating that the membrane contactor was still as fresh as the new one. Liquid droplets were detected in the outlet of gas phase after 3 days of operation, illustrating that some liquid molecules penetrated the membrane pores and the membrane contactor was operated under partial wetting mode. Then the CO2 mass transfer rate was continuously decreased to about 59% of the initial value after 14 days of operation. This may be attributed to the increase of membrane mass transfer resistance

18

resulting from partial membrane wetting. Membrane wetting is mainly influenced by membrane properties, absorption liquid properties and the mutual interaction between the absorption liquid and the membrane material. However, the corresponding causes of membrane wetting have always been disputed. Chapter 3 will detail the wetting mechanism and membrane morphological changes due to wetting.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

1

2

3

4

5

6

Time/(day)

J CO

2×1

04 /[m

ol(m

2 .s)-1]

Figure 2.5 Long-term performance of membrane contactor over two weeks

2.5 Retrieval of absorption performance

After 2 weeks of continuous operation, slight over-pressure was applied on the gas side to retrieve the membrane performance. At given liquid flow rate of 27 mL/min, the CO2 mass transfer rate could be retrieved to 86% of the original amount in a fresh membrane module by increasing the gas pressure temporarily. It can be explained by the fact that the intrusion liquid was pushed from the membrane pores back to the liquid-side, consequently reducing the membrane mass transfer resistance and enhancing the CO2 absorption. Such method could possibly be performed frequently to keep the performance at a reasonable level. The disadvantage of such method is the limited controllability and operational freedom. Special care should be taken during the applying of over-pressure, so as to avoid bubble formation or critical wetting.

Another effective way to smooth the membrane wetting is to improve the non-wettability of membrane contactor by surface modification. In our laboratory, the PP hollow fibers were modified by depositing a thin superhydrophobic coating on the membrane surface to improve their hydrophobicity. Even though an increase in the membrane thickness and a decrease in the surface porosity due to such treatment may offset some advantages of its hydrophobic surface, it still provides the feasibility of improving the performance stability of PP membrane contactor from the view of long-term operation. The detailed information on membrane modification can be found in Chapter 4.

19

2.6 Section summary

A variety of experiments were carried out to investigate the dependency of CO2 removal efficiency and mass transfer rate on operating parameters. The co-capture of CO2 and SO2 was also studied by the same membrane gas absorption system. In addition, the long-term system performance and the retrieval of the gas removal efficiency were conducted to gain a better understanding of the system duration performance.

The experimental results can be concluded as follows:

A PP hollow fiber membrane contactor was used to separate the CO2 from N2 using deionized water, MDEA and MEA as the absorbent. Results showed that membrane contactors were more efficient and compact than conventional absorption towers for acid gas removal. CO2 removal efficiency can be as high as 98% using MEA aqueous amine solution. The CO2 removal efficiency increased with the increase of the liquid flow rate and solvent concentration, while the CO2 mass transfer rate increased with the increase of liquid flow rate, CO2 volume fraction in the feed gas, solvent concentration and gas flow rate. The absorbent concentration should be compromised between absorption efficiency and the membrane wetting to ensure a stable and efficient removal of CO2 with a long life of the hollow fiber membrane.

The simultaneous experimental results indicated that the membrane contactor could eliminate these two sour gases simultaneously and effectively. Absorption of SO2 and CO2 was enhanced by the increase in liquid flow rate and decrease in gas flow rate. It was observed that a small amount of SO2 in the flue gas had a slight influence on the absorption of CO2, showing the feasibility of simultaneous removal of SO2 and CO2 in the same membrane contactor.

The membrane contactor was continuously operated for two weeks to evaluate its duration performance. The results showed that the CO2 mass transfer rate decreased significantly following the operating time caused, which was mainly caused by the increase in membrane transfer resistance under partial-wetting mode. The membrane wetting by the absorbents was a great concern for the practical application of membrane gas absorption technology, which resulted in economically unviable operation.

21

3 Wetting of PP hollow fiber membrane

In Chapter 2, we mentioned that membrane wetting by absorbents had been a big concern and a primary drawback for the long-term applications of membrane contactors. The corresponding causes for the membrane wetting have always been disputed. Wang et al. [55]

attributed the membrane wetting to possible chemical reactions between the PP membrane and DEA solution. Porcheron and Drozda [56] attributed partial wetting of membrane pores to a possible plasticizer effect of CO2 inside the membrane contactor fibers. Whereas, Mavroudi et al. [57] reported that the membrane wetting was related to the fact that the absorbent modified the surface hydrophobicity and penetrated into the membrane pores by the necking phenomenon. Thus, further investigation in this regard should be conducted to explain the wetting mechanism. In addition, for the hollow fiber membrane contactor, there are few reports on the property changes induced by absorbents as a function of time. The main objectives of this chapter are to experimentally investigate the interaction between the absorbents and the membrane, and to characterize the effects of membrane wetting on membrane properties. Based on the experimental results, methods are presented to eliminate or smooth membrane wetting phenomenon. This chapter is based on Paper IV.

3.1 Immersion experiments In order to investigate the membrane wetting evolution and the corresponding causes, the PP fibers were immersed in different absorbents, then the fibers were characterized by various characterization methods. The immersion experimental assumes that the immersed membranes will undergo similar exposure conditions as those used in the membrane contactor.

MEA and MDEA of 99.5 wt.% analytical grade were dissolved in deionized water to prepare the aqueous solutions. Then the prepared 30 wt.% MEA, 30 wt.% MDEA solutions and deionized water were loaded in three conical flasks, respectively. The 40 mm long PP hollow fibers were immersed into the absorbent in each conical flask. At the end of immersion days, five pieces of fibers were taken out and washed by deionized water for several times. Subsequently, these fibers were dried at 70 °C under vacuum for 10 h to for further characterization. Several times of washing step and 10 h of vacuum drying at 70 °C treatment can completely get rid of the remnant amine solution on the fiber surface.

3.2 Characterization methods

The membrane wetting evolution process and corresponding mechanism were investigated by x-ray photoelectron spectroscopy (XPS), real time attenuated total reflection Fourier transform infrared spectroscopy (ATR-IR), and thermogravimetric analysis (TGA). The property changes of the studied membrane fibers, in terms of hydrophobicity, morphological structure and surface toughness, were characterized by contact angle goniometer measurement, atomic force microscopy (AFM) and field emission scanning electron microscope (FE-SEM), respectively.

22

3.2.1 X-ray photoelectron spectroscopy measurement (XPS)

X-ray photoelectron spectroscopy is a quantitative spectroscopic technique that measures the elemental composition, empirical formula, chemical state and electronic state of the elements that exist within a material. XPS experiments were carried out on a RBD upgraded PHI-5000C ESCA system (Perkin Elmer) with Al K radiation (h=1486.6 eV). In general, the X-ray anode was run at 250 W and the high voltage was kept at 14.0 kV with a detection angle at 54. The pass energy was fixed at 23.5, 46.95 or 93.90 eV to ensure sufficient resolution and sensitivity. The base pressure of the analyzer chamber was about 510-8 Pa. The sample was directly pressed to a self-supported disk (1010 mm) and mounted on a sample holder then transferred into the analyzer chamber. The whole spectra (0~1100 eV) and the narrow spectra of all the elements with much high resolution were both recorded by using RBD 147 interface (RBD Enterprises, USA) through the AugerScan 3.21 software. Binding energies were calibrated by using the containment carbon (C1s = 284.6 eV). The data analysis was carried out by using the RBD AugerScan 3.21 software provided by RBD Enterprises. Where necessary, the observed XPS bands were curve-fitted using 80% Guassian plus 20% Lorentzian line shape, during the curve fitting.

3.2.2 Attenuated Total Reflection-Infrared Spectroscopy (ATR-IR)

The Attenuated Total Reflection (ATR) is a Fourier Transform Infrared Spectroscopy (FTIR) sampling technique that provides high quality data in conjunction with good reproducibility of any IR sampling technique. ATR-FTIR experiments were carried out using Thermo Nicolet 5700 FTIR spectrophotometer with an ATR accessory (American Thermo Electron Scientific Instruments Corp.) to analyze the membrane composition variation before and after the immersion process. The operating wave number range was 4000-400 cm−1. After the crystal area was cleaned and the background was collected, enough membrane fibers were placed onto and cover the small crystal area to achieve accurate results. Once the fibers had been placed on the crystal area, the pressure arm should be positioned over the sample area. The software was used at ‘Preview Mode’ which allows the quality of the spectrum to be monitored in real-time while fine tuning the exerted force.

3.2.3 Thermogravimetric analysis (TGA)

TG experiments were carried out using a high temperature thermal analyzer (PerkinElmer Pyris Diamond TG) in the presence of flowing N2 (100 mL/min) from 40 °C to 500 °C. The heating rate was 10 °C/ min.

3.2.4 Contact angle measurement

The characterization based on the contact angle measurement can provide information on the hydrophilicity and hydrophobicity characteristics of the membrane surface. The contact angle between distilled water and the external surface of hollow fiber membrane was measured using an OCA 40 Plus produced by Beijing Eastern-dataphy Instruments Co., Ltd to evaluate the surface properties of PP hollow fiber membranes. A droplet of distilled water was placed on the membrane surface by means of a 0.40 mL syringe. The contact angles were calculated from a digital video image of the drop on the membrane using an

23

image-processing program, which allowed the estimation of the contact angle from the height and width of the drop. To minimize experimental error, all the contact angle data were an average of five measurements on different locations of individual membrane surface. All measurements were carried out at room temperature.

3.2.5 Atomic Force Microscope (AFM) analyses

The surface morphologies of PP fibers were studied by a NanoScope IIIa MultiMode AFM (Digital Instrument) using a tapping mode. The images were obtained in the area of 5.00 μm × 5.00 μm. Various roughness parameters such as the root mean square of Z (the difference between the highest and the lowest points within the given area) values (Rms) and mean roughness (Ra) were used, which were calculated in accordance with the methods of Stamatialis et al. [58] and Chung et al. [59].

3.2.6 Field Emission Scanning Electron Microscope (FE-SEM)

Compared to standard SEM, the FE-SEM provides high resolution imaging at low accelerating voltages which is of big advantage for polymer and therefore for membrane characterization [60]. The hollow fiber samples were positioned on a metal holder and gold coated using a sputter coating operated under vacuum for 40 s. The FE-SEM pictures were observed under a JEOL JSM-7401F field emission scanning electron microscope to investigate the morphology. The accelerating voltage used was 5 kV, and all samples were imaged at 20 000 ×. Images from three different areas of each sample surface were recorded, and the pores in each traced and shaded on a sheet of clear film to provide sufficient contrast for image analysis. The images were captured by a video recorder attached to a Macintosh computer system. The images were digitized and analyzed by Image-Pro-Plus Version 5.0 to quantify the shape and size of the pores, as well as pore size distribution. The pore parameters listed in the paper were the averages of three different examined areas.

3.3 Liquid breakthrough pressure

The wettability of a hydrophobic porous membrane by the absorbent can be quantified by the value of breakthrough pressure of the absorbent through the membrane. Hydrophobic porous membranes may not permit the aqueous liquid to enter into the pores until the pressure difference between liquid and gas phase stream in membrane pores is larger than the breakthrough pressure, which can be estimated by the Laplace-Young equation:

4 cos L

p

pd

(3-1)

Where Δp is the penetration pressure, σL is the surface tension of the liquid, θ is the contact angle between the liquid phase and membrane and dp is the effective pore diameter.

3.4 Investigation on membrane wetting evolution

Wang et al. [55] immersed the PP fibers in 30 wt. % DEA absorbent. According to their XPS analysis, the presence of the C-N in XPS band was attributed to a possible chemical

24

reaction between the membrane and DEA, and thus they attributed the wetting to the chemical reaction between the PP membrane and DEA solution. However, it seems unreasonable to confirm the existence of the chemical reaction only by the presence of the C-N in the PP polymer surface region because the DEA itself contains the C-N. We assume that the C-N may be from the MEA absorbent absorbed in the membrane. In our XPS characterization, a peak of 285.4 eV corresponding to C-N was also observed when the membrane was immersed in the 30 wt. % MEA absorbent for 10 days or 90 days. In order to further verify our assumption, FTIR-ATR was used to investigate the composition variation before and after the immersion process. The distinct absorption peak could be seen at 3367.5 cm-1 and another characteristic peak of the immersed membrane was the scissoring vibration at 1601.3 cm-1, which was attributed to the NH2 groups in MEA, further illustrating that the PP membrane had absorbed a significant fraction of MEA within the cross-linked network. The TGA experiments were carried out to further study the interaction between PP fiber and the amine absorbent. By correlation of the results of XPS, ATR-IR, and TG, the diffusion of absorbent into the PP polymer can be confirmed during the immersion in the amine absorbents.

3.5 Effects of membrane-absorbent interaction on PP membrane propertiesAs the diffusion of the absorbent molecules in the PP polymer, a swollen gel can be triggered and the properties of the fiber may be changed [61].The hydrophobicity of the membrane surface is a dominant parameter that influences the membrane wetting, and it can be quantified by determining the pore parameters and contact angle of the surface layer. In this section, the membrane properties, in terms of contact angle measurement, SEM images and AFM images, are studied.

3.5.1 Changes of hydrophobicity characteristics

The characterization based on the contact angle measurement can provide information on the hydrophilicity and hydrophobicity characteristics of the membrane surface. The membrane samples were immersed in a 30 wt.% MEA, 30 wt.% MDEA and deionized water for different days, respectively. The contact angle of the untreated membrane was also measured as the reference sample. The contact angle decreased from 121.6° to 90.3° with the immersion of the PP fibers in 30 wt.% MDEA for 60 days, suggesting a remarkable decrease of the membrane surface hydrophobicity. With further increasing the immersion time over 60 days, the decrease of the contact angles slowed down. The contact angle of fiber immersed in deionized water was always being the highest, while fiber immersed in MEA was higher than that of MDEA at the end of the same immersion period. This is exactly the same order as the surface tension of deionized water, MEA and MDEA at the same concentration.

3.5.2 Membrane morphological changes in terms of SEM and AFM images

SEM and AFM images can be used to illustrate the changes of membrane morphology due to interaction between the membrane and the absorbent. It was found that the immersed membrane surface morphology, in terms of surface roughness, was apparently higher than

25

that of the blank membrane, and fiber immersed in MDEA showed the more obvious changes compared with that of MEA and deionized water. From the SEM images, it can be obviously observed that the membrane surface morphologies suffered significant changes after being immersed in various absorbents for a certain period compared with the blank sample. Some slit-like membrane pores in the blank sample shrank longitudinally and became elliptical or even circular after being immersed in the absorbents. The changes indicated that exposure of PP membrane to the selected solvents caused important changes of the membrane structure and therefore a deterioration of the membrane performance.

3.6 Critical parameters affecting the breakthrough pressure Hydrophobic porous membranes may not permit the aqueous liquid to enter into the pores until the applied liquid pressure is larger than the critical liquid breakthrough pressure. Based on Laplace-Young equation, how these three critical parameters affected the liquid breakthrough pressure was discussed, including the surface tension of the liquid, the contact angle between the liquid phase and membrane and membrane pore radius. It was concluded that the change of the contact angle played a more important role in the wetting than that of the pore sizes.

3.7 Methods for smoothing the membrane wetting In accordance with Laplace-Young equation, the following methods are proposed to eliminate or smooth the membrane wetting:

Membranes with high hydrophobicity: Different polymeric materials such as PTFE, PE, PP and PVDF have been used for CO2 absorption in hollow fiber membrane contactor [30] [62], in which porous PTFE membrane is the most hydrophobic membrane and shows good gas absorption performance as well as good stability without membrane wetting [41]. However, the application of PTFE is limited for its high production cost and lack of commercial availabilities. More efforts should be taken by the membrane manufacturers to reduce the cost of PTFE and make it commercially available in a wider size range.

Develop new absorbents with high surface tension to avoid the membrane resistance increase caused by membrane wetting. Yan et al. [37] developed a new liquid absorbent, aqueous potassium glycinate (PG), to eliminate the wetting problem of commercial PP porous membrane contactor. Feron and Jansen [39] patented novel absorption liquids CORAL to remove CO2 from various feed gases in porous polyolefin membrane module without wetting phenomenon. However, a new commercial absorbent for removal of CO2 requires both a high net cyclic capacity and high reaction/absorption rate for CO2, as well as high chemical stability, low vapor pressure and low corrosiveness [28]. There is still a long way to go to find available solvents that can meet all the requirements above.

Improve the membrane non-wettability through hydrophobic modification of membrane surface. A super-hydrophobic membrane which can show the contact angle of 152o with the deionized water was investigated by our group, which will be detailed in Chapter 4.

26

3.8 Section Summary In this section, the hollow fibers were immersed in 30 wt.% MEA, 30 wt.% MDEA, and deionized water for different days to simulate the actual experimental conditions during CO2 absorption in hollow fiber membrane contactor. The experiments were carried out to explore the effect of absorbent on the membrane wetting by correlation of the results of FE-SEM, AFM, XPS, Contact Angle Goniometer and ATR-FTIR, so as to better understand the membrane performance deterioration phenomenon and the membrane wetting mechanism during the CO2 absorption process in the hollow fiber membrane contactor. In addition, the effect of operating parameters, such as the operation period and absorbent type, on the membrane wetting was also experimented.

The diffusion of the absorbent into the PP polymer and the sorption-induced swelling were confirmed when the membrane fibers contacted with the amine absorbents.

The interaction between the membrane and the absorbent altered the surface properties and reduced the hydrophobicity of the membrane, which in turn increased the degree of wetting of the membrane pores. With the diffusion of the absorbent molecules into the membrane, the contact angle decreased remarkably and then leveled off. The average equivalent diameter increased with the immersion time, whilst the length/width ratio decreased because of the higher lateral wall area along the long axial.

The extent of pores deformation of 30 wt.% MEA aqueous solution was severer compared with that of deionized water due to the decrease of surface tension with the increase of solvent concentration. The change in morphology of MDEA was larger than that of MEA, which was attributed to a lower degree of intrusion of the MEA solution relative to that of MDEA due to the higher surface tension of the former.

It was suggested that improving the membrane surface hydrophobicity was an effective way to overcome the problem of wetting.

27

4 Fabrication and characterization of superhydrophobic

PP membrane

In previous chapter, we mentioned that improving the membrane surface hydrophobicity may be an effective way to overcome the problem of wetting. This chapter presents the fabrication and characterization of a super hydrophobic PP membrane by depositing a thin super hydrophobic coating on the membrane surface. The effects on modification on the membrane fiber characteristics and membrane-absorbent interaction were investigated. In addition, CO2 absorption in the polypropylene hollow fiber membrane contactors with and without surface modification was investigated in a continuous experiment for 20 days. This chapter is related with Paper V.

4.1 Fabrication of superhydrophobic membrane surface

The flow chart to fabricate the superhydrophobic membrane surface is shown in Figure 4.1.

Figure 4.1 Flow chart of superhydrophobic modification of PP hollow fibers

The flask containing 10 g of granulated polypropylene and 30 mL of xylene was placed in the oil bath and heated to 130 °C while it was constantly stirred by a magnetic stirring bar to prepare uniform polypropylene solution. After the polypropylene was dissolved in the xylene, a small amount of the non-solvent was added in the solution to ensure that no polypropylene granule came out of the solution. Methyl ethyl ketone (MEK), cyclohexanone, and the mixtures of MEK and cyclohexanone with the weight ratio of 1:1 were used as the non-solvents, respectively, to accelerate the nucleation rate and create smaller aggregates. Then the prepared uniform hot solution was deposited onto the original PP hollow fiber membrane fibers which were positioned on the spin coater, rotating at 2000 rpm for 40 s. The coated membrane fibers were cleaned with alcohol to eliminate to residual solution and then immediately placed in the vacuum oven at 70 °C for three hours. Then the modified PP membrane fibers were assembled in the module and the module ends were sealed with epoxy resins.

28

The photo of some modification instruments is shown in Figure 4.2.

Figure 4.2 Modification instruments

The schematic drawing of the rotating machine for modification is shown in Figure 4.3.

Figure 4.3 Schematic drawing of coating machine for PP membrane modification

29

4. 2 Characterizations of modified membrane fibers

4.2.1 SEM images of the modified membrane

The non-solvent type was varied sequentially to establish optimum conditions for the chemical treatment. MEK and cyclohexanone were used as the non-solvent. The corresponding SEM images are shown in Figure 4.4 and Figure 4.5.

Figure 4.4 SEM image for studied membranes using cyclohexanone (a) ×20000, untreated PP membrane surface; (b) ×20000, treated PP membrane surface; (c) ×5000, treated PP membrane surface; (d) ×20000, treated PP membrane surface; (e) ×2000, treated PP membrane surface; (f) ×5000, treated PP membrane surface. Non-solvent is cyclohexanone, the concentration of the PP is 14mg/mL

30

From Figure 4.4a, it was clear that some slit-like pores can be observed for the untreated fibers. After the modification, the nano-scale polypropylene aggregates were evenly dispersed on the PP membrane. When cyclohexanone was used as the non-solvent, high surface roughness was achieved due to a large number of spherulites. However, the use of cyclohexanone application led to an increase in the inhomogeneity of the coating due to an increased precipitation rate, thus offsetting the benefit of a rougher and more porous surface. As indicated by the red mark in Fig. 4.4f, several obvious large cracks were found and some membrane pores were blocked on the surface of the modified PP fibers using cyclohexanone as the non-solvent. In addition, cyclohexanone had a very low evaporation velocity due to its high boiling point (155 °C), resulting in longer evaporation time to obtain the superhydrophobic surface than MEK.

As shown in Figure 4.5, when MEK was used as the non-solvent, small PP crystallites were presented on the modified PP fiber surface with narrow cylindrical bridges and large pores. But in the practical operation, a part of MEK volatilized due to its lower boiling point (79.6 °C) when MEK was added into the heated PP solution at about 120 °C, thus affecting the effective utilization of MEK as a precipitator.

Figure 4.5 SEM image for studied membranes using MEK (a) ×20000, untreated PP membrane surface; (b) ×20000, treated PP membrane surface; (c) ×5000, treated PP membrane surface; (d) ×5000, treated PP membrane surface. Non-solvent is MEK, the concentration of the PP is 14 mg/mL

31

In order to shorten the preparation time and benefit the manufacturing process, the mixture of cyclohexanone and MEK (weight ratio of 1:1) was also investigated as the non-solvent. The corresponding SEM images are shown in Figure 4.6. The modified fiber surface featured gel-like porous coating with good homogeneity and crystallites distribution.

Figure 4.6 SEM image for studied membranes using the mixture of MEK and cyclohexanone

(a) ×20000, untreated PP membrane surface; (b) ×20000, treated PP membrane surface; (c) ×5000, treated PP membrane surface; (b) ×20000, treated PP membrane surface. Non-solvent is MEK : Cyclohexanone = 1:1, the concentration of the PP is 14 mg/mL

The above modification process led to surface morphological changes of PP hollow fiber membranes, as well as the changes of membrane specifications, in terms of membrane porosity, membrane thickness and pore size distribution. According to the results of mercury intrusion porosimetry, the values of total porosity for the fibers before and after the modification (the mixture of cyclohexanone and MEK as non-solvent) were 0.65 and 0.45, respectively, indicating that the surface porosity of the polypropylene membrane fiber was significantly decreased by chemical modification process. In addition, the fiber thickness was increased by 7 μm due to the additional coating on the surface. The average equivalent pore diameter was 0.23 μm for the untreated PP fibers, higher than that of 0.20 μm for the modified fibers. The decrease in pore size after modification may result from the blockage of many pores during the deposition of chemical solution onto PP fibers.

32

4.2.2 Membrane surface roughness changes before and after modification

The three-dimensional and two-dimensional AFM images for studied membrane are shown in Figure 4.7 and Figure 4.8, respectively, to evaluate the roughness changes after the membranes are modified. The lumpy aggregates protruded out of the surface due to the chemical coating.

Figure 4.7 Three-dimensional AFM image for PP membrane

(a) untreated; (b) treated with Cyclohexanone; (c) treated with MEK;(d) treated with MEK : Cyclohexanone = 1:1. The concentration of granulated PP is 14 mg/mL

Table 4.1 Outer surface roughness of studied membrane fibers

Membrane type Ra (nm) Rms (nm)

Untreated PP 27 33

Modified PP (non-solvent: cyclohexanone) 321 404

Modified PP (non-solvent: MEK) 88 166

Modified PP (non-solvent: cyclohexanone : MEK = 1:1) 195 274

The AFM parameters to evaluate the surface roughness of the studied membrane fibers are shown in Table 4.1. The relatively high mean roughness values were obtained by surface modification compared with that of untreated membranes. The modified membrane fibers using the cyclohexanone as the non-solvent exhibited the highest Rms value of 404 nm,

33

approximately 12.2 times of untreated fibers. This was attributed to the fact that a large number of spherulites were aggregated on the fiber surface, as shown in Figure 4.4. In contrast, low Rms value of 166 nm was achieved for the fibers using MEK as the precipitator. The fibers modified using the mixture of cyclohexanone and MEK as the non-solvent showed the Rms of 274 nm.

Figure 4.8 Two-dimensional AFM image for PP membrane

(a) untreated;(b) treated with MEK; (c) treated with Cyclohexanone. (d) treated with MEK: Cyclohexanone = 1:1. The concentration of granulated PP is 14 mg/mL

4.2.3 Hydrophobicity improvement by membrane modification

The hydrophobicity of the membrane surface is a dominant parameter for the membrane wetting, which can be quantified by determining the contact angle of the surface layer. The images and results of the studied membrane are shown in Figure 4.9. When the PP membranes were deposited with a thin super hydrophobic coating, the contact angle was dramatically increased, indicating that the hydrophobicity of membrane surface was greatly improved. By correlation of the results of SEM images, surface roughness and contact angle measurement, although the fibers modified by cyclohexanone showed the favorably highest roughness and resultantly highest contact angle, the unfavorable cracks on the fiber surface and longer evaporation time were concomitantly found. By weighing the homogeneity,

34

hydrophobicity and modification process efficiency, the best non-solvent in this study was the mixture of cyclohexanone and MEK (weight ratio of 1:1).

Figure 4.9 Contact angles of the outer surface of the hollow fiber membrane

(a) untreated: 122°; (b) treated with MEK: 149°; (c) treated with Cyclohexanone: 162°; (d) treated with MEK : Cyclohexanone = 1:1: 158°. The concentration of granulated PP is 14 mg/mL

4.3 Long-term performance of the modified membrane module By the correlation of results of previous contact angle measurement and SEM images, the modified membrane may be sufficiently hydrophobic for long-term operation and suitable for the membrane contactor. To verify that, a 20 days of long-term system operation with untreated and modified membrane contactors was carried out, as shown in Figure 4.10. The points in this figure have been connected for visualization. The absorption system was almost the same as used in Section 2.2 except that the gas phase passed through the lumen side and the absorbent flowed countercurrent through the shell side. Although the modification treatment enhanced the hydrophobicity, it concomitantly brought about a decrease in porosity and an increase in thickness, which was unfavorable for the CO2 absorption. Therefore, during the first 6 days, the CO2 mass transfer rate of the membrane without modification was higher than that of the modified membrane. When the operating time was more than 7 days, the CO2 mass transfer rate of the modified membrane was gradually stabilized, indicating that the membrane contactor was operated at a stable condition. In addition, the CO2 mass transfer rate of modified membrane contactor was obviously higher than that of the untreated membrane contactor. It indicated that the modified membrane contactor offered the possibility of improving the long-term performance deterioration due to the membrane wetting.

35

0 5 10 15 200.0

0.5

1.0

1.5

2.0

2.5

3.0

Modified membrane contactor Untreated membrane contactor

J CO

2×1

04 /[m

ol(m

2 .s)-1]

Time/[day]

Figure 4.10 Influences of modification of the PP membrane on CO2 mass transfer rate during long-time operation

4.4 Section Summary

The modification of PP membrane surface can be achieved by depositing a rough layer on the surface to improve its hydrophobicity. Three types of non-solvents were used to accelerate the nucleartion rate and create smaller aggregates. Weighing the coating homogeneity, hydrophobicity and modification process efficiency, the mixture of cyclohexanone and MEK system was considered as the best non-solvent.

The modification process led to surface morphological changes of PP hollow fiber membranes. Some nano-scale polypropylene aggregates were evenly dispersed on the membrane surface. In addition, the mean roughness value of membrane fibers was remarkably improved after the modification, thus achieving great improvement in the hydrophobicity of fiber surface.

The surface modification also introduced the decrease in membrane surface porosity and average pore size, and the increase in membrane thickness, which may result in additional mass transfer resistance and offset the superiority of surface modification. Therefore, the CO2 mass transfer rate of the modified membrane was a little lower than that of untreated in the first 6 days.

Beyond 7 days, the modified membrane contactor exhibited good stability and durability compared with the untreated membrane contactor. This was mainly attributed to the improved hydrophobicity via modification treatment. This chapter provides the feasibility of improving the properties and durability of PP hollow fiber membrane contactor by surface treatment for CO2 capture from power stations.

37

5 Conclusions CO2 capture from flue gas produced by fossil fuels combustion is corresponding to the most widely applicable option in terms of industrial sectors and is compatible to a retrofit strategy. The main challenge for CO2 capture systems is the large amount of energy consumption which reduces net plant efficiency significantly. Therefore, many researchers have examined the possibilities of enhancing the efficiency of these processes to reduce the effect of their drawbacks. Membrane gas absorption technology is such a hybrid technology that combines membrane separation technology and chemical absorption technology with advantages of these two separation processes.

The experimental apparatus for CO2 removal was set up in a PP hollow fiber membrane contactor using MEA, MDEA and deionized water as the absorbents. Experimental results showed that the CO2 removal efficiency increased with the increase of the liquid flow rate and solvent concentration, while the CO2 mass transfer rate increased with the increase of liquid flow rate, CO2 volume fraction in the feed gas, solvent concentration and gas flow rate.

In order to achieve the co-capture of SO2 and CO2 in the real engineering applications, simultaneous removal of CO2 and SO2 was also carried out by membrane gas absorption process. Experimental results indicated that the membrane contactor could eliminate these two sour gases simultaneously and effectively. The existence of SO2 had a slight influence on CO2 absorption when the SO2 amount was low in the feed gas.

The duration performance of membrane gas absorption system showed that the CO2 removal efficiency and mass transfer rate were decreased significantly with the operating time. Membrane wetting by absorbents was the main drawback for CO2 capture in a membrane contactor, especially for long-term industrial applications.

To better understand the wetting mechanism and propose solutions, PP membrane fibers were immersed in the absorbents for up to 90 days. Membrane swelling caused by absorbent intrusion rather than previously reported chemical reaction should be responsible for the membrane wetting. The membrane surface morphology, surface toughness and membrane hydrophobicity suffered significant changes by the interaction between the membrane and absorbent. More efforts should be taken to develop new commercial absorbents and improve membrane hydrophobicity for commercial applications of membrane gas absorption technology.

The results of duration experiments showed that the membrane performance could be temporarily retrieved by increasing the gas pressure or improving the hydrophobicity of PP fibers. Superhydrophobic surface of PP hollow fibers could be achieved by solvent casting techniques. Although additional membrane mass transfer resistance was introduced due to the existence of chemical coating, it still offered the possibility of enhancing the long-term performance of the membrane contactor and improving the membrane wetting phenomenon from the view of long-term operation performance.

39

6 Future work In this thesis, experiments were carried out to investigate how the operating parameters affect the mass transfer and removal efficiency, to study the feasibility of co-capture of CO2 and SO2, to find out what is the main drawback of membrane gas absorption system and finally to propose solutions to eliminate the wetting problem. Although increasing interest in the membrane gas absorption technology has been given in recent years and exciting results have been reported, most of the researches are still at the laboratorial stage and there is a long way to go before this technology is used for commercial applications. The following work is recommended to be studied in future researches:

Long-term stability of the membrane gas absorption operation is important from an economic point of view for its future commercial application. To avert the increase in mass transfer resistance due to membrane wetting, more efforts should be made to develop new absorbents with suitable surface tension. In addition, the compatibility between the membrane material and selected absorbent should be taken into consideration.

In order to replace conventional techniques by membrane gas absorption process, the detailed economical evaluation of membrane gas absorption process for acid gas capture should be carried out before and after the membrane modification, and then compared with conventional chemical absorption technology.

Time-dependent absorption mathematical models will be integrated into conventional mass transfer models to simulate the absorption process. The currently improved mathematical models only consider the non-uniformity or the pore size, time-related attributes-relationship between pore size and time should be considered.

To commercialize the membrane gas absorption technology, there is a need to analyze the acid gas removal based on real gas stream conditions as flue gas streams. Most researches consider a pure CO2 or CO2/N2 mixture system, the effect of impurities on absorption, such as flash ash, O2, H2S, SO2, NOX which is more practical for industrial application (important for the overall design and plant cost), is seldom considered. A small quantity of impurities may adversely impact the cost and suitability of absorption process. For example: relatively high oxygen content represents a particular challenge for amine absorption due to degradation of amine by the oxygen; flash ash deposits on process equipment, blocking flow, damaging pumps and seals; SO2 and NO2 degrade the solvent, forming heat stable salts and depleting the absorption capacity of the solvent. Therefore, the effects of impurities on the membrane gas absorption technology should be considered for commercial applications.

The regeneration process should be taken into consideration due to higher energy consumption in the absorbent stripper. The regeneration in vacuum hollow fiber membrane contactor may be an effective way.

41

7 Reference

[1] B.P. Mandal, S.S Bandyopadhyay. Absorption of carbon dioxide into aqueous blends of 2-amino-2-methyl-1-propanol and monoethanolamine. Chemical Engineering Science, 61 (2006), 5440-5447.

[2] M.L. Gray, K.J. Champagne, D. Fauth, J.P. Baltrus, H. Pennline. Performance of immobilized tertiary amine solid sorbents for the capture of carbon dioxide. International Journal of Greenhouse Gas Control, 2 (2008), 3-8.

[3] P.S. Northrop, J.A. Valencia. The CFZ™ process: A cryogenic method for handling high CO2 and H2S gas reserves and facilitating geosequestration of CO2 and acid gases. Energy Procedia, 1 (2009), 171-177.

[4] L. Zhao, E. Riensche, R. Menzer, L. Blum, D. Stolten. A parametric study of CO2/N2 gas separation membrane processes for post-combustion capture. Journal of Membrane Science, 325 (2008), 284-294.

[5] IPCC. In: Metz, Davidson, Coninck, editors. IPCC special report on carbon dioxide capture and storage. Cambridge: Cambridge University Press, IPCC, 2005.

[6] J. David, H. Herzog. The Cost of Carbon Capture. In: Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, 2000, 973-978.

[7] P. Singh, G.F. Versteeg. Structure and activity relationships for CO2 regeneration from aqueous amine-based absorbents. Process Safety and Environmental Protection, 86 (2008), 347–359.

[8] A. Gabelman, S.T. Hwang. Hollow fiber membrane contactors. Journal of Membrane Science, 159 (1999), 61-106.

[9] P.H.M. Feron, A.E. Jansen. Capture of carbon dioxide using membrane gas absorption and reuse in the horticultural industry. Energy conversion and management, 36 (1995), 411-414.

[10] R. Klaassen, P.H.M. Feron, A.E. Jansen. Membrane contactors in industrial applications. Chemical Engineering Research and Design, 83 (2005), 234-246.

[11] H.A. Rangwala. Absorption of carbon dioxide into aqueous solutions using hollow fiber membrane contactors. Journal of Membrane Science, 112 (1996), 229-240.

[12] H.J. Herzog. What future for carbon capture and sequestration. Environmental Science & Technology, 35 (2001), 148A-153A.

[13] S.H. Yeon, K.S. Lee, B. Sea, Y.I. Park, K.H. Lee. Application of pilot-scale membrane contactor hybrid system for removal of carbon dioxide from flue gas. Journal of Membrane Science, 257 (2005), 156-160.

[14] Available from: www.Liqui-cel.com., 2004 [15] CO2 Capture Project (CCP) website. Available from:

http://www.co2captureproject.org /index.htm [16] J.L. Li, B.H. Chen. Review of CO2 absorption using chemical solvents in hollow

fiber membrane contactors. Separation and Purification Technology, 41 (2005), 109-122.

42

[17] C. Hanish. Exploring options for CO2 capture and management. Environmental Science and Technology, 33 (1999), 66-70.

[18] P.H.M. Feron, A.E. Jansen. In: Eliasson B, Riemer P, Wokaun A (Eds.), Techno-economic assessment of membrane gas absorption for the production of carbon dioxide from flue gas, Greenhouse Gas Control Technologies, Pergamon, Oxford ,1999, 53-58.

[19] R. Klaassen, P.H.M. Feron, A.E. Jansen. Membrane contactors in industrial applications. Chemical Engineering Research and Design, 83 (2005), 234-246.

[20] 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, 46 (2005), 33-40.

[21] H.Y. Zhang, R. Wang, D.T. Liang, J.H. Tay, Theoretical and experimental studies of membrane wetting in the membrane gas-liquid contacting process for CO2 absorption, Journal of Membrane Science, 308 (2008), 162-170.

[22] P. Keshavarz, J. Fathikalajahi, S. Ayatollahi. Gas Analysis of CO2 separation and simulation of a partially wetted hollow fiber membrane contactor. Journal of Hazardous materials, 152 (2008), 1237-1247.

[23] J.G. Lu, Y.F. Zheng, M.D. Cheng, Wetting mechanism in mass transfer process of hydrophobic membrane gas absorption. Journal of Membrane Science, 308 (2008), 180-190.

[24] H. Kreulen, C.A. Smolders, G. F.Versteeg, W.P.M.van Swaaij. Determination of mass transfer rates in wetted and non-wetted microporous membranes. Chemical Engineering Science, 48(1993), 2093-2102.

[25] H. Mahmud, A. Kumar, R.M. Narbaitz. The air-phase mass transfer resistance in the lumen of a hollow fiber at low air flow. Chemical Engineering Journal, 97 (2004), 69-75.

[26] H. Mahmud, A. Kumar, R.M. Narbaitz. A study of mass transfer in the membrane air-stripping process sing microporous polypropylene. Journal of Membrane Science, 179 (2000), 29-41.

[27] M. Mavroudi, S.P. Kaldis, G.P. Sakellaropoulos. A study of mass transfer resistance in membrane gas-liquid contacting process. Journal of Membrane Science, 272 (2006), 103-115.

[28] S. Ma’mun, H.F. Svendsen, K.A. Hoff, O. Juliussen, Selection of new absorbents for carbon dioxide capture, Energy Conversion and Management, 48 (2007), 251-258.

[29] 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, 40 (2004), 133-145.

[30] Y.S. Kim, S.M. Yang. Absorption of carbon dioxide through hollow fiber membranes using various aqueous absorbents. Separation and Purification Technology, 21 (2000), 101-109.

[31] B.P. Mandal, M. Guha, A.K. Biswas, S.S. Bandyopadhyay. Removal of carbon dioxide by absorption in mixed amines: modeling of absorption in aqueous MDEA/MEA and AMP/MEA solutions. Chemical Engineering Science, 56 (2001), 6217-6224.

43

[32] S. Atchariyawut, R. Jiraratananon, R. Wang. Separation of CO2 from CH4 by using gas-liquid membrane contacting process. Journal of Membrane Science, 304 (2007), 163-172.

[33] Y.W Gong, Z. Wang, S.C. Wang. Experiments and simulation of CO2 removal by mixed amines in a hollow fiber membrane module. Chemical Engineering and Processing, 45 (2006), 652-660.

[34] J.G. Lu, Y.F. Zheng, M.D. Cheng and L.J. Wang, Effects of activators on mass-transfer enhancement in a hollow fiber contactor using activated alkanolamine solutions. Journal of Membrane Science, 289 (2007), 138-149.

[35] A.M. Barbe, P.A. Hogan, R.A. Johnson. Surface morphology changes during initial usage of hydrophobic microporous polypropylene membranes. Journal of Membrane Science, 172 (2000), 149-156.

[36] Y.X. Lv, X.H. Yu, S.T. Tu, J.Y. Yan, E. Dahlquist. Wetting of polypropylene hollow fiber membrane contactors. Journal of Membrane Science, 362 (2010), 444-452.

[37] S.P. Yan, M.X. Fang, W.F. Zhang, S.Y. Wang, Z.K. Zu, Z.Y. Luo, K.F. Cen. Experimental study on the separation of CO2 from flue gas using hollow fiber membrane contactors without wetting. Fuel Processing Technology, 88 (2007), 501-511.

[38] P.S. Kumar, J.A. Hogendoorn, P.H.M. Feron, G.F. Versteeg. New absorption liquids for the removal of CO2 from dilute gas streams using membrane contactors. Chemical Engineering Science, 57 (2002), 1639-1651.

[39] P.H.M. Feron, A.E. Jensen. CO2 separation with polyolefin membrane contactors and dedicated absorption liquid: performance and prospects. Separation and Purification Technology, 27 (2002), 231-242.

[40] S. Khaisri, D. deMontigny, P. Tontiwachwuthikul, R. Jiraratananon. Comparing membrane resistance and absorption performance of three different membranes in a gas absorption membrane contractor. Separation and Purification Technology, 65 (2009), 290-297.

[41] N. Nishikawa, M. Ishibashi, H. Ohta, N. Akutsu, H. Matsumoto, T. Kamata, H. Kitamura. CO2 removal by hollow fiber gas-liquid contactor, Energy Conversion and Management, 36 (1995), 415-418.

[42] Z.K. Xu, J.L. Wang, L.Q. Shen, D.F. Men, Y.Y. Xu. Microporous polypropylene hollow fiber membrane: Part I. Surface modification by the graft polymerization of acrylic acid. Journal of Membrane Science, 196 (2002), 221-229.

[43] J.M. Dickson, R.F. Childs, B.E. McCarry, D.R. Gagnon. Development of a coating technique for the internal structure of polypropylene microfiltration membranes. Journal of Membrane Science, 148 (1998), 25-36.

[44] H.Y. Erbil, A.L. Demirel, Y. Avci, O. Mert. Transformation of a simple plastic into a superhydrophobic surface. Science, 299 (2003), 1377-1379.

[45] S. Koonaphapdeelert, K. Li. The development of ceramic hollow fiber membranes for a membrane contactor. Desalination, 200 (2006), 581-583.

[46] S. Koonaphapdeelert, Z.T. Wu, K. Li. Carbon dioxide stripping in ceramic hollow fibre membrane contactors. Chemical Engineering Science, 64 (2009), 1-8.

[47] H. Kreulen, G.F. Versteeg, C.A. Smolders, W.P.M. Van Swaaij. Selective removal of

44

H2S from sour gas with microporous membranes. Part I. application in a gas-liquid system, Journal of Membrane Science,73 (1992), 293-304.

[48] H.H. Park, B.R. Deshwal, I.W. Kim, H.K. Lee. Absorption of SO2 from flue gas using PVDF hollow fiber membranes in a gas-liquid contactor, Journal of Membrane Science, 319 (2008), 29-37.

[49] H.H. Park, B.R. Deshwal, H.D. Jo, W.K. Choi, I.W. Kim, H.K. Lee. Absorption of nitrogen dioxide by PVDF hollow fiber membranes in a G-L contactor. Desalination, 243 (2009), 52-64.

[50] P. Keshavarz, J. Fathikalajahi, S. Ayatollahi. Mathematical modeling of the simultaneous absorption of carbon dioxide and hydrogen sulfide in a hollow fiber membrane contactor. Separation and Purification Technology, 63 (2008), 145-155.

[51] R. Faiz, M. Al-Marzouqi. Insights on natural gas purification: Simultaneous absorption of CO2 and H2S using membrane contactors. Separation and Purification Technology, 76 (2011), 351-361.

[52] R. Faiz, M. Al-Marzouqi. Mathematical modeling for the simultaneous absorption of CO2 and H2S using MEA in hollow fiber membrane contactors. Journal of Membrane Science, 342 (2009), 269-278.

[53] B.P. Mandal, S.S. Bandyopadhyay. Simultaneous absorption of carbon dioxide and hydrogen sulfide into aqueous blends of 2-amino-2-methyl-1-propanol and diethanolamine. Chemical Engineering Science, 60 (2005), 6438-6451.

[54] C.H. Huang, T.W. Xu, X.F. Yang. Regenerating fuel-gas desulfurizing agents by using bipolar membrane electrodialysis (BMED): effect of molecular structure of alkanolamines on the regeneration performance. Environmental Science & Technology, 41 (2007), 984-989.

[55] R. Wang, D.F. Li, C. Zhou, M. Liu, D.T. Liang. Impact of DEA solutions with and without CO2 loading on porous polypropylene membranes intended for use as contactors. Journal of Membrane Science, 229 (2004), 147-157.

[56] F. Porcheron, S. Drozdz. Hollow fiber membrane contactor transient experiments for the characterization of gas/liquid thermodynamics and mass transfer properties. Chemical Engineering Science, 64 (2009), 265-275.

[57] M. Mavroudi, S.P. Kaldis, G.P. Sakellaroporlos. Reduction of CO2 emissions by a membrane contacting process. Fuel, 82 (2003), 2153-2159.

[58] D.F. Stamatialis, C.R. Dias, M.N. Pinho. Atomic force microscopy of dense and asymmetric cellulose-based membranes. Journal of Membrane Science, 160 (1999), 235-242.

[59] T.S. Chung, J.J. Qin, A. Huan, K.C. Toh. Visualization of the effect of die shear rate on the outer surface morphology of ultrafiltration membranes by AFM. Journal of Membrane Science, 196 (2002), 251-266.

[60] K.J. Kim, A.G. Fane, C.J.D. Fell, T. Suzuki and M.R. Dickson. Quantitative microscopic study of surface characteristics of ultrafiltration membranes. Journal of Membrane Science, 54 (1990), 89-102.

[61] E. S. Tarleton, J. P. Robinson, M. Salman. Solvent-induced swelling of membranes - Measurements and influence in nanofiltration. Journal of Membrane Science, 280 (2006): 442-451.

45

[62] S. H. Yeon, B. Sea, Y. I. Park, K. H. Lee. Determination of mass transfer rates in PVDF and PTFE hollow fiber membranes for CO2 absorption. Separation Science and Technology, 38 (2003) 271-293.

experimental studies on co2 capture using absorbent in a polypropylene hollow fiber membrane

Documents