UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Advanced electrochemical reforming of methanol for hydrogen production Cloutier, Caroline R. 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_spring_cloutier_caroline.pdf [ 7.44MB ]
Metadata
JSON: 24-1.0059287.json
JSON-LD: 24-1.0059287-ld.json
RDF/XML (Pretty): 24-1.0059287-rdf.xml
RDF/JSON: 24-1.0059287-rdf.json
Turtle: 24-1.0059287-turtle.txt
N-Triples: 24-1.0059287-rdf-ntriples.txt
Original Record: 24-1.0059287-source.json
Full Text
24-1.0059287-fulltext.txt
Citation
24-1.0059287.ris

Full Text

Advanced Electrochemical Reforming of Methanol for Hydrogen Production by Caroline R. Cloutier M.A.Sc., Metals and Materials Engineering, University of British Columbia, 2006 B.A.Sc., Chemical Engineering, Environmental Option, Co-op, University of Ottawa, 1999.  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Chemical & Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) December, 2011  @ Caroline R. Cloutier, 2011  Abstract  The issue of efficient, low-cost, sustainable hydrogen (H2) production is one of the barriers to the adoption of a H2 economy.  In this thesis, the electrochemical production of H2 from liquid methanol  (CH3OH) in acidic aqueous media was studied in a proton exchange membrane (PEM) electrolyser in the static mode at low temperatures. A baseline study showing the influence of CH3OH concentration, catalyst, catalyst support, operating temperature and operating mode was established. A theoretical thermodynamic analysis of the system was carried out as a function of temperature, and the limiting current densities, kinetic parameters, including the Tafel slopes and current exchange density, and apparent activation energies were determined.  The effect of electrochemical promotion (EP) was  investigated to see if it can increase the efficiency and performance of H2 production through electrochemical processes.  The electrochemical promotion of electrocatalysis (EPOE) was investigated by carrying out the electrolysis in triode and tetrode operation.  It was shown to improve the PEM electrolysis in the  galvanostatic and potentiostatic modes. A decrease in electrolysis voltage or an increase in electrolysis current proportional to the current or potential imposed in the auxiliary circuit was observed when the auxiliary current or potential was opposite to the electrolyser circuit current or potential. The effect was observed using catalytic and non-catalytic non-precious electrolyser electrode materials.  It was  postulated that triode and tetrode operation enhanced the electro-oxidation rate through electrochemical pumping and spillover of protons. With this novel electrolysis configuration, electrolysis cost reduction may be achieved through the use of non-precious electrolyser anode materials and/or improving electrolyser performance.  The electrochemical promotion of catalysis (EPOC) was also investigated for the catalytic reforming of CH3OH at low temperature with Pt-Ru/C and Pt-Ru/TiO2. The synthesized Pt-Ru/TiO2 was characterized physico-chemically and electrochemically. Powder catalytic CH3OH reforming tests showed that both catalysts can be used to generate H2. EPOC experiments were conducted on gas diffusion electrodes (GDEs) in galvanostatic control. Under the experimental conditions, only supplying H+ to the catalyst ii  working electrode surface resulted in only in a Faradaic enhancement of the catalytic activity for the low temperature reforming of CH3OH, which appears to be a purely electrophilic behaviour.  iii  Preface The literature review, design of the experimental set-up, experimental data collection and analysis was conducted by Caroline R. Cloutier under the supervision of Dr. David P. Wilkinson. The material covered in Chapter 2 was published:  •  Cloutier, C. R., and Wilkinson, D. P., “Electrolytic Production of Hydrogen from Aqueous Acidic Methanol Solutions”, Int. J. H2 Energy, 35, (2010) 3967-3984.  Co-op student Jason Gao co-produced the data used in Fig. 2.6 (e) in Chapter 2 under the supervision of Caroline R. Cloutier.  Some of the material covered in Chapter 3 was published in a conference  proceeding and another paper has been submitted for publication:  •  Cloutier, C. R., and Wilkinson, D. P., "Triode Operation of a Proton Exchange Membrane (PEM) Electrolyser", ECS Transactions, 25 (23), (2010) 47-57.  •  Cloutier, C. R., and Wilkinson, D. P., “Electrochemical Promotion of Aqueous Acidic Methanol PEM Electrolysis in Triode and Tetrode Operation”, (submitted).  A manuscript based on the content of Chapter 4 is in preparation. The initial and final drafts of this thesis were prepared by Caroline R. Cloutier, with revisions edited and approved by Dr. David P. Wilkinson. The thesis proposal and the thesis manuscript were reviewed by the Ph.D. Committee, which included Dr. David Wilkinson, Dr. Elöd Gyenge, Dr. Walter Mérida, and Dr. David Dreisinger.  iv  Table of Contents  Abstract………………………………………………………………………............ii Preface………………………………………………………………………………..iv Table of Contents……………………………………………………………………v List of Tables…………………………………………………………………………x List of Figures………………………………………………………………………xii Nomenclature…………………………………………………………………...xvii List of Acronyms……………………………………………………………...…xxi Acknowledgements…………………………………………………………..…xxiv Dedication…………………………………….…………………………………xxvi Chapter 1: Introduction……………………..……………………………………1 1.1 Background……………………………………………...…………………………..…1 1.1.1 Alternative fuels..……….………………………………………………………………..……1 1.1.2 Technology gap ……………………………………..……………………………………..…3 1.1.3 Proposed solution………………………..………………………………………………….3 1.1.4 A sustainable closed loop system………….….…………………………………..............4  1.2 Thesis overview……………………………………………………..…………………6 1.2.1 Justification…………………………………..………………………………………….……6 1.2.2 Research objectives……………………………………………………………………..…7 1.2.3 Layout…………………………………………………………...……………………………8  1.3 Literature review………………..………………….…………………………………10 1.3.1 Heterogeneous catalysis……………..…………………………………………………….10 1.3.1.1 Steam reforming……………………...…………………………………………………11 1.3.1.2 Liquid phase reforming…………………………………………………………………12 1.3.1.3 Hydrogen spillover in heterogeneous catalysis………………………………………13 1.3.1.4 Catalyst local heating…………………………………………………………………14 1.3.2 Proton exchange membrane fuel cell………...………………………………………….15 1.3.2.1 Membrane electrode assembly……………………………………………………….16 1.3.2.2 Overvoltage………………………………………………………………………………17 1.3.2.3 Hydrogen proton exchange membrane fuel cell…………………………………….18 Hydrogen purity requirements……………….……..…………………………….…………19 1.3.2.4 Direct methanol fuel cell………………………………………………………………..20 Anodic methanol oxidation……………………………………………………………….....21 Methanol crossover…………………………………….………………………………..…22 1.3.2.5 Indirect methanol fuel cell………………………………………………………………23 1.3.3 Proton exchange membrane electrolysis………………………….……………………24 1.3.3.1 Proton exchange membrane water electrolysis……………………………………24 1.3.3.2 Depolarized proton exchange membrane water electrolysis……………………….26 Methanol depolarized electrolysis…………………………………………………………27 Applications of depolarized methanol electrolysis………………………………………32 v  1.3.4 Electrochemical promotion of heterogeneous catalysis………………………………34 1.3.4.1 Theory……………………………………………………………………………………35 1.3.4.2 Parameters………………………………………………………………………………39 1.3.4.3 Electrochemical promotion of methanol oxidation…………………………………41 1.3.4.4 Electrochemical promotion using protonic electrolytes……………….……………42 1.3.4.5 Applications for the electrochemical promotion of catalysis………………………44 1.3.5 Electrochemical promotion of electrocatalysis………………………………….………..47  Chapter 2: Electrocatalysis Baseline Study ………………………………….49 2.1 Synopsis………………………………………………………………………………49 2.2 Experimental………………………………………………………………………..…50 2.2.1 Materials……………………………………………………………………………………...50 2.2.2 Equipment……………………………………………………………………………………51 2.2.3 Electrochemical measurements……………………………………………………………54 2.2.4 Characterization…………………..………………………………………………………....54  2.3 Results and discussion……………………………………………………………57 2.3.1 Thermodynamic evaluation of methanol reforming at low temperatures……………...57 2.3.2 Electrolysis polarizations…………………………………………………………………62 2.3.2.1 Cell voltage stability……………………………………………………………………62 2.3.2.2 Effect of methanol concentration and anode catalyst………………………………64 2.3.2.3 Effect of dry purged cathode…………………………………………………………71 2.3.2.4 Effect of temperature……………………………………………………………………73 2.3.2.5 Anode Tafel kinetics……………………………………………………………………76 2.3.2.6 Activation energy………………………………………………………………………85  2.4 Summary……………………………………………………………………………….87  Chapter 3: Electrochemical Promotion of Electrocatalysis (EPOE).…….88 3.1 Synopsis………………………………………………………………………………88 3.2 Overview………………………………………………………………………………89 3.3 Experimental…………………………………………………………………………93 3.3.1 Materials and membrane electrode assembly…………………………………………...93 3.3.2 Glass cell for triode and tetrode operation……………..…………………………………95 3.3.3 Electrochemical measurements……………………………………………………………97 3.3.4 Stability of carbon containing electrolyser ring electrodes………………….…………101  3.4 Triode or tetrode operation evaluation expressions…………………………103 3.4.1 Triode or tetrode operation evaluation in the galvanostatic mode………………….103 3.4.2 Triode or tetrode operation evaluation in the potentiostatic mode……………………105  3.5 Results and discussion…………………………………………………………....106 3.5.1 Galvanostatic characterization…………………………………………………………106 3.5.1.1 Triode and tetrode operation in galvanostatic control…………………………….106 3.5.1.2 Parallel and reverse operation in galvanostatic control…….……………..………107 3.5.1.3 Electrolyser ring electrode material in galvanostatic control………………………109 3.5.1.4 Electrolyser voltage ratio, power enhancement ratio and power gain ratio in galvanostatic control……………………………………………………………………………111 3.5.1.5 Benefits of starting in triode operation in galvanostatic control…………………114 3.5.2 Potentiostatic characterization……………………..……………………………………116 3.5.2.1 Triode configuration in potentiostatic control……….…...………………………….116 3.5.2.2 Tetrode configuration in potentiostatic control……..……………………...............118 3.5.2.3 Counter electrode potential measurements in potentiostatic control……………119 3.5.2.4 Electrolyser ring electrode material in potentiostatic control..………..…………121 3.5.2.5 Electrolyser rate enhancement ratio, power gain ratio and Faradaic efficiency in potentiostatic control………………………………………………………..………………123 3.5.3 Durability investigation………………………..…………………………………………126 3.5.3.1 Long-term triode and tetrode operation in galvanostatic control…………………126 3.5.3.2 Methanol concentration change over time in long-term triode or tetrode vi  galvanostatic control….....................................................................................................128 3.5.3.3 Long-term tetrode operation in mixed potentiostatic and galvanostatic control...130 3.5.3.4 Methanol concentration change over time in long-term tetrode operation in mixed potentiostatic and galvanostatic control……………………………………………………132 3.5.4 Mechanism investigation……………………………………….....................................134 3.5.4.1 Electrolytic contact of auxiliary electrodes in potentiostatic control ..……………134 3.5.4.2 Effect of acid concentration and conductivity on triode operation in potentiostatic control………………………………………………………………………………………….136 3.5.4.3 Effect of triode operation on membrane conductivity in potentiostatic control….138 3.5.4.4 Effect of electrolyser working electrode ring geometry in potentiostatic control..139 3.5.4.5 Effect of proton flux lines direction in potentiostatic control….……………………140 3.5.4.6 Effect of the electrochemical surface area of the carbon fibre paper electrolyser working electrode in potentiostatic control….……………………….………………...........143 3.5.4.7 Effect of triode operation using non-noble non-carbon containing working electrode materials in potentiostatic control…………….…………………………………...144 3.5.4.8 Proposed mechanism…………………………………………………………………153  3.6 Summary………………………………………………….…………………………156  Chapter 4: Catalysis and Electrochemical Promotion of Catalysis (EPOC)……………………………………………………………………………157 4.1 Synopsis………………………………………………………………………..…….157 4.2 Experimental…………………………………………………………………………160 4.2.1 Materials……..……………………………………………………………………………160 4.2.2 Catalyst synthesis…….……………………………………………………………………160 4.2.3 Pt-Ru/TiO2 physicochemical characterization…………….…………………………162 4.2.3.1 Scanning electron microscopy……………………………………………………162 4.2.3.2 Energy-dispersive X-ray spectroscopy………………………………………………162 4.2.3.3 Transmission-electron microscopy…….……………………………………………163 4.2.3.4 X-ray diffraction…………………………………………………………………...……163 4.2.3.5 X-ray photoelectron spectrometry……………………………………………………164 4.2.3.6 Inductively coupled plasma optical emission spectrometry……………………164 4.2.3.7 Brunauer, Emmett and Teller surface area and Barett-Joyner-Halenda pore size and volume analysis……………………………..…………………………………………164 4.2.4 Pt-Ru/TiO2 electrochemical characterization……………..…………………………165 4.2.5 Apparatus………………………………………………………………………………...…166 4.2.5.1 Powder catalysis………………………………………………………………………166 4.2.5.2 GDE catalysis and electrochemical promotion of catalysis……………………….171  4.3 Results and discussion……………………………………………………………173 4.3.1 Pt-Ru/TiO2 physicochemical characterization…………….…………………….………173 4.3.1.1 SEM images………..………….……………………………………………….………173 4.3.1.2 EDX measurements……………...……………………………………………....…...174 4.3.1.3 TEM images……….……..………………………………………………………..…...175 4.3.1.4 XRD measurements……………...……………………………………………….…..177 4.3.1.5 XPS measurements…………………………………………………………………...180 4.3.1.6 ICP-OES measurements…………..………………………………………………….181 4.3.1.7 BET surface area and BJH pore size and volume analysis……………………….182 4.3.2 Pt-Ru/TiO2 electrochemical characterization……………...……………………………184 4.3.3 Powder catalysis…………………………….……………………………………………..192 4.3.4 GDE catalysis and electrochemical promotion of catalysis………………………..….199  4.4 Summary……………………………………………………………………………214  Chapter 5: Conclusions…………………………………………………………216 5.1 Electrocatalysis baseline study……………………………………….…………216 5.2 Electrochemical promotion of electrocatalysis (EPOE).……………………218 5.2.1 Galvanostatic and potentiostatic characterization……………………………………218 vii  5.2.2 Durability investigations…………………………………………………………………219 5.2.3 Mechanism investigations………………………………………………………………220  5.3 Electrochemical promotion of catalysis (EPOC).……………………………..221 5.3.1 Catalyst synthesis and characterization…………………………………………………222 5.3.2 Powder catalysis…………………………………………………………………………...222 5.3.3 GDE catalysis and electrochemical promotion of catalysis…………………………...223  5.4 Potential application, significance and impact of research findings……...224 5.4.1 Potential applications……………………………………………………………………...224 5.4.2 Significance and impact…………………………………………………………………..228  5.5 Recommendations for further research………………...………………………230 5.5.1 Electrocatalysis baseline study…………………………………………………………..230 5.5.2 Electrochemical promotion of electrocatalysis (EPOE)….…………………………….231 5.5.3 Electrochemical promotion of catalysis (EPOC).……………………………………….234  References……………………………….…………………………………….….236 Appendix A: Publications, presentations, and posters…………………………..250 A.1 Publications………………………………………...…………………………..…………….250 A.2 Presentations………………………………………………………...………………………250 A.3 Posters……………………………………………………………...………………………...250  Appendix B: Methanol electro-oxidation mechanism………………………….…251 B.1 Methanol electro-oxidation parallel pathways………………………………………..…..251 B.2 Methanol electro-oxidation mechanism on Pt……………………………………….……251 B.3 Methanol electro-oxidation mechanism on Pt-Ru…………………………………..……252 B.4 References……………………………………………………………………………….…..252  Appendix C: Efficiency and economic comparison of H2 production…………254  C.1 Objective…………………………………………………………………………..………….254 C.2 System definition………………………………………………………………………...…..254 C.3 Block diagrams………………………………………………………………………………254 C.4 Feedstock processing………………………………………………………………...…….256 C.4.1 Methanol production…………………………………………………………………….256 C.4.2 Sodium borohydride production………………………………………………………..257 C.5 Hydrogen production………………………………………………………………...……...257 C.5.1 Electrolysis……………………………………………………………………………….258 C.5.2 Catalytic steam reforming………………………………………………………………259 C.5.3 Sodium borohydride hydrolyser………………………………………………………..260 C.6 Efficiency analysis…………………………………………………………..……………….260 C.7 Economic analysis………………………………………………………...………………...262 C.7.1 Capital cost……………………………………………………………………………….264 C.7.2 Feedstock cost…………………………………………………………………………..265 C.7.3 Energy cost………………………………………………………………………………265 C.7.4 Carbon cost………………………………………………………………………………266 C.7.5 Economic analysis summary…………………………………………………………...267 C.8 Conclusion…………………………………………………………………………..………..268 C.9 References…………………………………………………………………………………...269  Appendix D: Supplemental EPOC and EPOE theory…………………….…...…..271 D.1 EPOC rules…………………………………………………………..……………...……….271 D.2 Chemisorption on a transition metal……………………………………………………….273 D.3 Kirchhoff’s first law for triode and tetrode operation………………….………………….275 D.3.1 Triode fuel cell……………………………………………………………………………275 D.3.2 Triode electrolyser……………………………………………………………………….276 D.3.3 Tetrode electrolyser……………….…………………………………………………….276 D.4 References…………………………………………………………………………………...277 viii  Appendix E: Experimental procedures…………………………………………...…278 E.1 Catalyst electrode preparation…………………………………...………..…….………...278 E.1.1 Cathode carbon sublayer composition calculation…………………………………..278 E.1.2 Anode and cathode catalyst ink composition calculation…………………………...279 E.1.3 Catalyst ink preparation procedure…………………………………………………….281 E.1.4 Electrode spraying procedure………………………………………………………….282 E.2 Platinization procedure…………………………………………………………..………….283 E.3 Silicon ring preparation………………………………….…………………….…………….284 E.4 Nafion membrane conditioning………………………………..…………………………...285 E.5 Determination of CH3OH concentration……………………..…………………………….286 E.6 RDE catalyst ink preparation………………………………...……………………………..289 E.7 Pt-Ru/TiO2 preparation…………………………………………….………………………..290 E.8 References…………………………………………………………...………………………294  Appendix F: Electrochemical techniques………………………………………..…295 F.1 Cyclic voltammetry………………………………………………………………….……….295 F.1.1 Pt-C catalyst electrochemical surface area determination…………………………..296 F.1.2 Pt-Ru/C catalyst electrochemical surface area determination………………………300 F.2 Electrochemical impedance spectroscopy………………………………………………303 F.3 Determination of the open circuit voltage…………………………………………………305 F.4 Current - potential transients…………………….………………………………………....306 F.5 Rotating disk electrode……………………………..…………………………………..…...309 F.6 References……………………………………………………………………………………311  Appendix G: Thermodynamic data and sample calculations…………..……….313 G.1 Thermodynamic data………………………………………………………………………..313 G.2 Thermodynamic sample calculations…………………………………………………..….313 G.2.1 Liquid phase……………………………………………………………………………..313 G.2.2 Gas phase……………………………...………………………………………………..315 G.2.3 System efficiency calculation…………………………………………………………..320 G.2.4 Effect of pressure in the gas phase……………………………………………………321 G.2.5 Effect of anode and cathode pressure in the liquid phase………………………….322 G.3 References…………………………………………………………………………………...323  Appendix H: Electrochemical data and sample calculations..………………….324 H.1 Power consumption………………………………………………………………………….324 H.2 CO2 emissions and H2 production rate……………………………………………...…….324 H.3 Proton diffusion limited H2 production – back envelope calculation……..…………….326 H.4 Overall system efficiency…….……………………………………………………...……...328 H.5 References…………………………………………………………………………………330  ix  List of Tables Table 1.1 Table 1.2 Table 1.3 Table 1.4 Table 1.5  Methanol properties………………………………………………………………………..2 Relevant methanol electrochemical reforming membrane reactors..……………….28 Vapour phase methanol oxidation EPOC studies with oxide conductors……………42 Vapour/gaseous phase EPOC studies using protonic conducting electrolytes……..43 EPOC studies in aqueous systems…………………………………………………….44  Table 2.1  Electrochemical characterization of supported and unsupported catalysts studied…………………………………………………………………………………….56 Literature Tafel kinetic parameters for the electro-oxidation of methanol and water on Pt and Pt-Ru catalysts……………………………….……………………………………78 Tafel kinetic parameters from current work for the electro-oxidation of methanol and water on Pt and Pt-Ru catalysts………………………………….………………………80 Activation energy data for the electro-oxidation of methanol on Pt and Pt-Ru catalysts…………………………………………………………………………………….86  Table 2.2 Table 2.3 Table 2.4 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8 Table 4.9 Table 4.10  Table 4.11  Table C.1 Table C.2  List of possible half-cell reactions and their standard potentials..……………………92 IR corrected on-set potentials for CFP samples with and without Pt black at varying CH3OH concentrations………………………………………..…………………………102 Effect of auxiliary WEaux potential (Uaux, we) on the rate enhancement ratio (ρe) (Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, Uelec, we = 0.52 V vs. SHE)…………… ……….123 IR corrected on-set potential for different electrolyser and auxiliary WEaux materials at varying CH3OH concentrations………………………………………….…………...148 Effect of auxiliary WEaux potential (Uaux, we) on the rate enhancement ratio (ρe) using Au plated 316 SS components (2 mg/cm2 Pt/C CE, Uelec, we = 0.52 V vs. SHE)…150 Scherrer’s crystallite size for selected peaks from the carbon, Pt-Ru/C, TiO2 and PtRu/TiO2 XRD patterns.……………………………………………………….………179 ICP-OES results for H2 reduced Pt-Ru/TiO2 before and after ball milling and PtRu/C………………………………………………………………………………………182 BET surface areas of the Pt-Ru/TiO2 materials synthesized, as well as commercial TiO2 and Pt-Ru/C materials……………………………………………………………183 On-set potentials for the forward and reverse CH3OH oxidation on different catalytic surfaces (2 M CH3OH in 0.5 M H2SO4).…………………….………………………..187 Peak potentials for the forward and reverse CH3OH oxidation on different catalytic surfaces (2 M CH3OH in 0.5 M H2SO4).…....………..………………………………...187 List of CV studies on unsupported and supported Pt, Pt-Ru and TiO2 available in the literature………………………………………………………………………………….188 Literature powder catalysis CH3OH reforming production rates at low temperature (0.5 g Pt-Ru/TiO2, 10 wt% metal, Pt:Ru 1:1 a/o, 60 ml of 2 M CH3OH)……...…….196 Powder catalysis CH3OH reforming production rates (0.5 g catalyst, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 20 ml of 2 M CH3OH in 0.5 M H2SO4, 75 ±1oC)…..………………....196 Imposed currents or potentials on selected chemical reactions subjected to EPOC……………………………………………………………………………………..200 Average H 2 unpromoted production rates for GDE catalysis of CH 3OH at low temperature (2.01 cm2, 4 mg/cm2 Pt-Ru/C or Pt-Ru/TiO2, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75±1oC)……………………………………………………….204 Faradaic H2 production rate, measured H2 production rates, Faradaic efficiencies and EPOC efficiencies for the GDE catalytic reforming of CH3OH at low temperature (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4, 75±1oC)…………………………………………………………………...210 High level comparison of four H2 generation process schemes………………….…258 Efficiency comparison of four H2 generation process schemes…………………….261 x  Table C.3 Table E.1 Table E.2 Table E.3 Table E.4 Table F.1 Table G.1 Table G.2 Table H.1  Cost comparison of four H2 generation process schemes…………………………..268 Cathode carbon sublayer calculation spreadsheet…………………………………...279 Anode catalyst ink composition calculation spreadsheet………….…………………280 Cathode catalyst ink composition calculation spreadsheet……………………….…281 Precursor salt addition calculation spreadsheet for the impregnation method….…291 Total metal area results comparison for electrode degradation evaluation………..300 Thermodynamic data………………………………………………………………….…313 Empirical coefficients for Shomate equations…………………………………………317 Direct methanol reformer electrochemical active surface area as a function of proton diffusion coefficient in Nafion, based on H2 production for a 100 kW PEMFC at a voltage of 0.6 V………………………………………………………………………….328  xi  List of Figures Figure 1.1 Figure 1.2 Figure 1.3 Figure 1.4 Figure 1.5 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5  Figure 2.6  Figure 2.7  Figure 2.8 Figure 2.9 Figure 2.10  Figure 2.11  Figure 2.12  Figure 2.13  (a) Schematic diagram of conventional methanol synthesis combined with a DMR and a PEMFC (b) Schematic diagram of a future sustainable system comprising a DMR and PEMFC…………………………………………………………………………………………...5 Schematic diagram of a methanol electrochemical reforming membrane cell (Modified from patent DE 197007384 [112])…………………………………………………………...29 Schematic diagram of an alcohol electrolyser (Modified from Hu et al. [7] with permission from Elsevier)………………………………….……………………………........30 Schematic diagram of a metal catalyst-electrode deposited on (a) an oxide-conducting solid electrolyte, and (b) a proton-conducting solid electrolyte (Modified from Vayenas et al. [18] with permission from Springer)……………………………………………………...37 Schematic diagram of a triode fuel cell electrical circuit…...……………………………...47 Schematic diagram of (a) the electrochemical glass cell, (b) the MEA components (c) the MEA holder.………… ………………………………………………….…………….......52 Picture of (a) the electrochemical glass cell, (b) the MEA holder………………………...53 Theoretical thermodynamic efficiency and theoretical cell voltage as a function of temperature (1 atm) for a DMR and PEM water electrolyser……………………………..61 Electrolysis cell voltage stability with respect to current density (16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC)………………………………………………...62 IR corrected electrolysis cell voltage repeatability as a function of geometric current density in the stable and unstable regions (2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 2 mg/cm2 Pt/C anode and cathode, 23 ±2oC)………………………………………………………………………………………......63 Effect of methanol concentration on the IR corrected electrolysis cell voltage as a function of geometric current density for different anode catalysts (a) 4 mg/cm2 Pt-Ru/C, (b) 2 mg/cm2 Pt/C, (c) 4 mg/cm2 Pt-Ru black, (d) 2 mg/cm2 Pt black, (e) well-defined Pt disk (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC).……………………………………………….......………..66 Dependence of limiting geometric current densities for the oxidation of methanol and water on the CH3OH concentration for various anode catalysts (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC)………………………………………………………………………………………........67 Effect of CH3OH concentration on the individual electrode potential (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC)………..……………………..68 Effect of the IR corrected voltage on electrolysis for Pt/C and Pt black anode catalysts as a function of real electrochemical current density (0, 2, or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC)………….....70 Effect of dry N2 purged cathode on the IR corrected electrolysis cell voltage as a function of geometric current density (0 or 1 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, wet electrolyser catholyte: 0.5 M H2SO4, dry electrolyser catholyte: N2 purge electrolyser, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC)……………..71 Effect of dry N2 purged cathode and a wet cathode on the (a) anode potential and (b) cathode potential, as a function of geometric current density (0 or 1 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, wet electrolyser catholyte: 0.5 M H2SO4, dry electrolyser catholyte: N2 purge, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC)…….72 Effect of CH3OH concentration on the IR corrected electrolysis cell voltage as a function of geometric current density for different temperatures (a) 23 ±2oC (b) 50 ±1oC (c) 75 ±1 o C (0, 2 or 16 M CH 3 OH in 0.5 M H 2 SO 4 electrolyser anolyte, 0.5 M H 2 SO 4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode)……………75 Tafel plot of the IR corrected anodic overpotential as a function of the log of the geometric current density (0, 2, or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 xii  M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt black anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC)…………………………………………………………………………………….......….76 Figure 3.1 Figure 3.2  Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11  Figure 3.12  Figure 3.13  Figure 3.14  Figure 3.15  Figure 3.16 Figure 3.17  Figure 3.18  Figure 3.19  Schematic diagram of the triode electrolyser electrical circuit…………………………....89 Side view of membrane electrode assemblies in different configurations: (a) triode, normal operation, common CE (b) like (a) with auxiliary WEaux away from surface (c) like (a) in inverted operation, and (d) tetrode, normal operation, independent CEs……………………………………………………………………………………….……..90 Schematic drawing of auxiliary TeflonTM cap with apertures………………………….…..94 Front view of the triode or tetrode operation working electrode components…………..94 Apparatus for triode and tetrode electrolysis (a) electrochemical glass cell, (b) auxiliary electrode ……………………………………………………………….……………………....96 Picture of the electrochemical glass cell set-up in triode operation……….……………..97 Potentiostatic working electrode potential experimental settings relative to the standard half-cell electrochemical potentials……………….……………………………………….…98 Schematic diagram of EPOE testing system connections for (a) galvanostatic triode operation, (b) potentiostatic triode operation (c) galvanostatic tetrode operation, and (d) potentiostatic tetrode operation……………………………………………………...…...….99 IR-corrected forward anodic potential sweep (5 mV/s, 0, 2 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, CFP with and without 2 mg/cm2 Pt black WE, Pt flag CE, 23 ±1oC)…………......…….……………………….…101 Triode effect in galvanostatic control (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2, iaux, cell = -50 mA/cm2)……………….…107 Effect of varying Iaux, cell on the electrolyser circuit voltage in galvanostatic control (a) parallel operation (b) reverse operation (4 mg/cm2 Pt-Ru black electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2, iaux, cell = -50 mA/cm2)…...108 Effect of varying Iaux, cell on the electrolyser circuit voltage in galvanostatic control (a) 4 mg/cm2 Pt-Ru black electrolyser ring WEelec, (b) CFP electrolyser ring WEelec (Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2)………………………………....110 Effect of iaux, cell on Uelec, cell and on the triode voltage ratio (R) (a) 4 mg/cm2 Pt-Ru black electrolyser ring WEelec (b) CFP electrolyser ring WEelec (Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2)…………………………………………………………112 Power enhancement ratio (ρelec, cell) and power gain ratio (γelec, cell) as a function of the auxiliary circuit current density (a) 4 mg/cm2 Pt-Ru black electrolyser ring WEelec (b) CFP electrolyser ring WEelec (Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2)……………………………………………………………………………………..113 Benefits of starting in triode operation in galvanostatic control (CFP ring electrolyser WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 50, 75 mA/cm2, iaux, cell = -5 A/cm2)……………………………………………………………………………………........115 Triode effect in potentiostatic control (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)……......……………………………….116 Triode and tetrode effect in potentiostatic control (4 mg/cm2 Pt-Ru black electrolyser ring WEelec, 2 mg/cm2 Pt/C electrolyser CEelec, Pt auxiliary WEaux and CEaux, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE).……………………………………………………………………………………119 Triode effect in potentiostatic control (a) geometric current density (b) counter electrode potential (4 mg/cm2 Pt-Ru black electrolyser ring WEelec, 2 mg/cm2 Pt/C electrolyser CE, Pt auxiliary WEaux, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)……..........……………………………120 Tetrode effect in potentiostatic control (a) geometric current density (b) counter electrode potential (4 mg/cm2 Pt-Ru black electrolyser ring WEelec, 2 mg/cm2 Pt/C xiii  Figure 3.20  Figure 3.21  Figure 3.22  Figure 3.23  Figure 3.24  Figure 3.25  Figure 3.26  Figure 3.27 Figure 3.28 Figure 3.29  Figure 3.30 Figure 3.31  Figure 3.32 Figure 3.33  Figure 3.34  electrolyser CE elec , Pt auxiliary WE aux and CE aux , 2 M CH 3 OH in 0.5 M H 2 SO 4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)...120 Effect of varying Uaux, we on different electrolyser ring WEelec materials in potentiostatic control (a) 0 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte and (b) 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte (2 mg/cm2 Pt black, 4 mg/cm2 Pt-Ru black, and CFP electrolyser WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, Uelec, we = 0.52 V vs. SHE)…................122 Power gain ratio (γelec, cell) as a function of Uaux, we (CFP electrolyser ring WEelec and 4 mg/cm2 Pt-Ru black electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)…………………………………………………………………………125 Durability test in normal, triode and tetrode electrolysis in galvanostatic control (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2, iaux, cell = -50 mA/cm2)………………………………………………………………………..127 Methanol concentration as a function of time as predicted by Faraday’s law, and during normal, triode, and tetrode electrolysis in galvanostatic control (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, ielec, cell = 1.25 mA/cm2, iaux, cell = -50 mA/cm2)….……………………………...…………………...........................................129 Potentiostatic electrolyser circuit control and galvanostatic auxiliary circuit control durability test in tetrode electrolysis (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CEelec, Pt auxiliary CEaux, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 1 V vs. SHE, iaux, cell = -20 mA/cm2)…...131 Methanol concentration as a function of time as predicted by Faraday’s law and as measured during potentiostatic electrolyser circuit control and galvanostatic auxiliary circuit control in tetrode electrolysis (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CEelec, and Pt auxiliary CEaux, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 1 V vs. SHE, iaux, cell = -20 mA/cm2)…………………………………………………………………………………….....133 Effect of varying Uaux, we on the electrolyser current density in tetrode operation (a) Pt auxiliary WEaux 1 cm away from surface (b) Pt auxiliary CEaux 1 cm away from surface (CFP electrolyser ring WEelec, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)…………………………..........135 Nafion® N-117 membrane conductivity and solution conductivity as a function of sulphuric acid concentration………………………………………………………………136 Effect of varying Uaux, we on the electrolyser current density (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 0 M CH3OH in 0.05 or 5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)...137 Resistance measurement over time in normal electrolysis and in triode electrolysis (CFP electrolyser ring WEelec, Pt auxiliary WEaux, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE, Uaux, we = -0.59 V vs. SHE)……………………………………………………………139 Effect of varying Uaux, we on the ielec, cell for different electrolyser ring WEelec gap distances (CFP electrolyser ring WEelec, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)……………...….....................140 Effect of varying Uaux, we on the current density in tetrode operation (a) inverted CEs (CFP ring WEelec, Pt rod WEaux, Pt rod CEelec, Pt/C CEaux) (b) inverted WEs (Pt rod WEelec, CFP ring WEaux, Pt/C ring CEelec, Pt rod CEaux), (2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)...142 Effect of electrolyser ring CFP WEelec carbon SA on ielec, cell (2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE)...144 IR corrected forward anodic potential sweeps (a) 316 SS auxiliary WEaux (b) 316 SS mesh electrolyser ring WEelec (c) Au plated 316 SS auxiliary WEaux (d) Au plated 316 SS mesh electrolyser ring WEelec (e) Pt auxiliary WEaux (5 mV/s, 0, 2, or 16 M CH3OH in 0.5 M H2SO4, Pt flag CE, 23 ±1oC)……….…………….………………………………………147 Effect of varying Uaux, we on (a) Au plated 316 SS electrolyser ring WEelec, Pt auxiliary WEaux, 0 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, (b) Au plated 316 SS electrolyser ring WEelec, Pt auxiliary WEaux, 2 M CH3OH xiv  Figure 3.35 Figure 3.36 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.11 Figure 4.12 Figure 4.13 Figure 4.14 Figure 4.15  Figure 4.16  Figure 4.17 Figure 4.18  Figure 4.19  Figure 4.20  in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, (c) Au plated 316 SS electrolyser ring WEelec, Au plated 316 SS auxiliary WEaux, 0 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, (d) Au plated 316 SS electrolyser ring WEelec, Au plated 316 SS auxiliary WEaux, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte (triode, Pt/C CE, Uelec, we = 0.52 V vs. SHE)…………………………………………………………………………………..149 Power gain ratio (γelec, cell) as a function of Uaux, we (a) 0 M CH3OH, (b) 2 M CH3OH (Au plated 316 SS components, 2 mg/cm2 Pt/C CE, 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, Uelec, we = 0.52 V vs. SHE).……………...152 Schematic diagram of the proposed triode/tetrode mechanism in reverse electrolysis operation………………………………………………...................………………………...154 Schematic diagram of the experimental set-up used for the catalytic reforming of liquid CH3OH using catalyst powders……………………………………………………............167 Schematic diagram of experimental set-up used for the catalytic CH3OH reforming and the electrochemical promotion of catalytic CH3OH reforming using GDEs…………....171 Picture of the experimental set-up used for the catalytic CH3OH reforming and the electrochemical promotion of catalytic CH3OH reforming using GDEs (a) auxiliary equipment (b) glass cell configuration……………………………………………………..172 SEM image of 20 wt% metal Pt-Ru/TiO2 (WD 15.2 mm, 20.0 kV, x2.0 k)……….…….173 EDX spectra for Pt-Ru/TiO2 (a) 10 wt% metal, (b) 20 wt% metal……………………....174 TEM images of 20 wt% Pt-Ru/TiO 2 (a) direct mag. 200000x and (b) direct mag. 600000x (20 wt% Pt-Ru, H 2 reduction method, ball milled 60 min., HV = 100.0 kV)……………………………………………………………………………………….. …...176 X-ray diffraction data for (a) Carbon and 20 wt% Pt-Ru/C, and (b) TiO2 and 20 wt% PtRu/TiO2……………………………………………………………………………...………...177 Survey XPS spectrum of 20 wt% Pt-Ru/TiO2…………………..…………………………181 Pt-Ru/TiO2 BJH desorption pore volume distribution………………………………….....184 Cyclic voltammograms of (a) Pt, (b) Pt-Ru/C on GC, and (c) Pt-Ru/TiO2 on GC (2 M CH3OH in 0.5 M H2SO4, 23 ±2oC, 50 and 75 ±1oC, 50 mV/s, no rotation).……………186 H2 production rate as a function of time (ambient to 100 ±1oC, 0.5 g catalyst, 20 wt% PtRu, Pt:Ru 1:1 a/o, 20 ml of 2 M CH3OH in 0.5 M H2SO4)………………………………..193 H2 production rate as a function of time (ambient to 75 ±1oC, 0.5 g catalyst, 20 wt% PtRu, Pt:Ru 1:1 a/o, 20 ml of 2 M CH3OH in 0.5 M H2SO4)………………………………..194 H2 production as a function of temperature (0.5 g catalyst, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 20 ml of 2 M CH3OH in 0.5 M H2SO4)……………………………………………………..195 CH3OH partial pressure as a function of time (a) Pt-Ru/C, (b) Pt-Ru/TiO2 (ambient to 75 ±1oC, 0.5 g catalyst, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 20 ml of 2 M CH3OH in 0.5 M H2SO4)…………………………………………………………………………………………198 H2 production rate and electrolysis cell voltage as a function of time during unpromoted and promoted catalysis (2.01 cm2, 4 mg/cm2 Pt-Ru/TiO2, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75 ±1oC)…………………………………………………………..202 H2 production rate and electrolysis cell voltage as a function of time during unpromoted and promoted catalysis (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH 3 OH in 0.5 M H 2 SO 4 electrolyser anolyte, 80 ml of 0.5 M H 2 SO 4 electrolyser catholyte, 75 ±1oC)…………….......………………………………………….203 Schematic diagram of the reverse electrolyser electrical circuit and of the proposed EPOC mechanism for the catalytic reforming of methanol………………………………205 H2 production rate and electrolyser current as a function of time during unpromoted and promoted catalysis (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75 ±1oC ).……..………………………………………………………………......207 H2 production rate and electrolyser current as a function of time during unpromoted and promoted catalysis (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75 ±1oC)………………………………………………………………………...208 Absolute Faradaic efficiency as a function of (a) average cell voltage, and (b) average xv  Figure 4.21 Figure 4.22 Figure 5.1 Figure 5.2 Figure B.1 Figure B.2 Figure C.1 Figure D.1 Figure D.2 Figure D.3 Figure D.4 Figure D.5 Figure E.1 Figure E.2 Figure F.1 Figure F.2 Figure F.3 Figure F.4 Figure F.5  Figure F.6 Figure F.7 Figure H.1 Figure H.2  working electrode potential (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75 ±1oC)…………………………………………………………...211 Catalytic H2 production rate as a function of average working electrode potential (2.01 cm2, 4 mg/cm2 Pt-Ru/C, 20 wt% Pt-Ru, Pt:Ru 1:1 a/o, 80 ml of 2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 80 ml of 0.5 M H2SO4 electrolyser catholyte, 75 ±1oC).…212 Electrophilic behaviour of the catalytic reaction rate as a function catalyst WE potential (Modified from Vayenas et al. [18] with permission from Springer)…………………….213 Example of working electrode scale-up design for triode or tetrode operation (a) single auxiliary working electrode, (b) multiple auxiliary working electrodes………………….226 Diagram showing possible electrolysis circuit current reversal under triode or tetrode operation.……………………………………………………………………………………...227 Methanol electro-oxidation parallel pathways………………………………………….....251 Methanol electro-oxidation bi-functional mechanism on Pt-Ru…………………………252 Block diagrams of four H2 generation process schemes (a) PEM methanol electrolyser, (b) PEM water electrolyser, (c) methanol steam reforming, and (d) sodium borohydride hydrolysis.……………………………………………………………………………………..255 CO chemisorption on a transition metal. Molecular orbitals and density of states before (a, b) and after (c, d). Effect of varying Φ and EF on electron back donation (c) and donation (d) (Modified from Vayenas et al. [1] with permission from Springer).………274 Application of Kirchhoff’s first law for fuel cell triode operation………………………….275 Application of Kirchhoff’s first law for triode electrolysis in parallel operation…………276 Application of Kirchhoff’s first law for triode electrolysis in reverse operation…………276 Application of Kirchhoff’s first law for tetrode electrolysis in reverse operation…….....277 Electrolysis in dilute sulphuric acid…………………………………………………………283 Lead-free platinization…………………………………………………………………….....284 Picture of a three-port glass cell for cyclic voltammetry…………………………….……295 Cyclic voltammogram (2.01 cm2, 2 mg/cm2 Pt/C, 5 mV/s, 0.5 M H2SO4, Pt flag CE, 23 ± 2oC)…………………………………………………………………………………...............298 Cyclic voltammograms for a fresh catalyst sample, a used catalyst sample, and an oxidized catalyst sample (2.01 cm2, 2 mg/cm2 Pt/C, 5 mV/s, 0.5 M H2SO4, Pt flag CE, 23 ± 2oC)………………………………………………………………………………………299 Cyclic voltammogram (4 mg/cm2 Pt-Ru, 5 mV/s, 0.5 M H2SO4, Pt flag CE, 23 ± 2oC).302 Typical Nyquist plots for a fresh MEA at varying methanol concentration (initial frequency of 1x105 Hz to final frequency of 0.01 Hz, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte)……………............………………………………………...304 MEA resistance as a function of methanol concentration for various MEA conditions (4 mg/cm2 Pt/Ru anode, 2 mg/cm2 Pt cathode, N-117, 0.5 M H2SO4 electrolyte and catholyte)…………........................................................................................................304 H-cell set-up for OCV measurements……………………………………………………...306 Power consumption as a function of cell voltage for different CH3OH concentrations (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru anode, 2 mg/cm2 Pt cathode, 23 ±2oC)……………...…….324 Rate of CO2 emission and volumetric rate of H2 evolution rate in function of the cell Voltage (2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru or 2 mg/cm2 Pt anode, 2 mg/cm2 Pt cathode, 23 ±2oC)….326  xvi  Nomenclature i) Arabic symbols  Symbol a a A AGeom Ar b B c C Co Cal calf CC CCAP CE CF CH2 COM Cpo D e E Eo Ea EA EoA Eoa Ec Eoc Ee Eff Ep fNafion fPTFE F FCR Fr FWHM i io iaux, cell ielec, cell ik IL I Io  Description  Units  Activity Tafel constant Absorbance Active cell geometric area Pre-exponential factor Tafel slope Levich parameter Charge Concentration Bulk concentration Calibration number Calibration factor Carbon tax Capital cost Energy cost Feed cost Hydrogen cost Operation and maintenance cost Standard heat capacity Diffusion coefficient Charge of an electron Operating cell voltage Standard reversible cell voltage Anodic potential Activation energy Open circuit activation energy Standard anodic potential Cathodic potential Standard cathodic potential Thermodynamic equilibrium (reversible) cell voltage Efficiency Peak potential Fraction of Nafion in solution Fraction of PTFE in solution Faraday constant Fixed charge rate on capital Molar flow rate Full width half max Current density Exchange current density Auxiliary current density in triode or tetrode operation Electrolyser current density in triode or tetrode operation Kinetic current density Limiting current density Total current Exchange current of the metal-solid electrolyte interface  dimensionless V dimensionless cm2 1/s V/dec mA.s1/2/cm2 C M M dimensionless dimensionless $/tonC $-y/GJ $/GJ $/GJ $/GJ $/GJ J/molK m/s2 C V V V kJ/mol eV V V V V GJ produced/GJ feedstock V dimensionless dimensionless C/mol 1/y mmol/min. radians mA/cm2 mA/cm2 mA/cm2 mA/cm2 mA/cm2 mA/cm2 A A xvii  Symbol Iaux Iaux, cell Ioaux, cell Ielec Ielec, cell Ioelec, cell I’elec, cell IFar Ifc Ir l k kA Lcat LCsub m Mi n NH+ Ni NG Ns P Paux, cell PC Pe Pelec, cell Poelec, cell Pi Pp Qb Qcv Qexp QUDP r ro r' re relec, cell roelec, cell rF reH+ R R Re Rs s Sd t T Uaux, cell  Description  Units  Auxiliary circuit current Auxiliary current in triode or tetrode operation Auxiliary current in normal operation Electrolyser circuit current Electrolyser current in triode or tetrode operation Electrolyser current in normal electrolysis Calculated electrolyser current based on measured CH3OH concentration changes in triode or tetrode operation Net Faradaic fuel consuming current Fuel cell current Reversible current Path length of light in solution Shape factor Reaction rate constant Catalyst loading Carbon sublayer loading Mass Mass rate of species i Number of electrons Protonic migration flux Molar rate of species i Number of moles of catalyst on the metal surface Total number of squares Pressure Auxiliary power in triode or tetrode operation Carbon emissions Electric power Electrolyser power in triode or tetrode operation Electrolyser power in normal operation Promotion index Partial pressure Background charge H2 adsorption charge Total charge measured Underpotential deposition charge Electrochemically promoted catalytic reaction rate Open circuit unpromoted catalytic reaction rate before EPOC experiment Open circuit unpromoted catalytic reaction rate after EPOC experiment Electrochemical reaction rate Electrochemically promoted electrochemical reaction rate Unpromoted electrochemical reaction rate Faradaic reaction rate Electrochemical proton reaction rate Universal gas constant Triode or tetrode voltage ratio Reynolds number Ohmic resistance Number of moles of species Standard deviation Time Temperature Auxiliary cell voltage in triode or tetrode operation  A A A A A A A A A A cm dimensionless 1/s mg/cm2 mg/cm2 g g/s electrons mol/m2.s mol/cm2.s moles of metal dimensionless atm W TonC/GJ W W W dimensionless atm C C C C mol/s mol/s mol/s mol/s mol/s mol/s mol/s mol/s J/mol·K dimensionless dimensionless ohm moles dimensionless (s) o C or K V xviii  Symbol Uaux, we Uoaux, cell Uelec, cell Uelec, we Uoelec, cell Ufc Urev, cell UWR UWC V Vi Vr W Wo Y Z Z’ Z”  Description  Units  Auxiliary WE potential in triode or tetrode operation Auxiliary cell voltage in normal operation Electrolyser cell voltage in triode or tetrode operation Electrolyser WE potential in triode or tetrode operation Electrolyser cell voltage in normal operation Fuel cell voltage Reversible cell voltage Working electrode potential with respect to a reference electrode Working electrode potential with respect to a counter electrode Voltage Volumetric rate of species i Volumetric flow rate X-ray wavelength Instrumentation broadening Molar fraction Absolute impedance Real portion of the impedance Imaginary portion of the impedance  V V V V V V V V V V m3/h ml/h  nm radians dimensionless ohm ohm ohm  ii) Greek symbols  Symbol αa αH  β ΔCp ΔG ΔGo ΔGof ΔHo ΔHof ΔHTf ΔHj ΔNmol,g ΔPelec, cell ΔQD ΔQH Δr ΔSo ΔSof ΔΦw  Description Anodic transfer coefficient Enthalpic parameter Symmetry factor Change in heat capacity  Units dimensionless dimensionless dimensionless J/molK  Gibbs free energy of reaction  kJ/mol  Standard Gibbs free energy of reaction  kJ/mol  Standard Gibbs free energy of formation  kJ/mol  Standard enthalpy of reaction Standard enthalpy of formation Change in enthalpy of combustion at temperature T Binding energy or enthalpy of adsorption Number of mole of gaseous species in the products minus in the reactants Change in electrolyser power output in triode or tetrode operation Ru oxide charge gain H2 adsorption charge loss Electrochemically induced change in catalytic reaction rate Standard entropy of reaction Standard entropy of formation Change in the work function of the working electrode  kJ/mol kJ/mol kJ/mol kJ/mol mol W μC μC mol/s or g equivalents/s kJ/molK kJ/molK eV  xix  Symbol  ΔUWR ΔVg ε γelec, cell η ηa ηc ηconc ηE ηEPOC ηfuel ηI ηmax ηmax, system ηoverall ηohm ηs  λ  Λ  τ θB θi  μe  ρe  ρelec, cell ν  Φ Ψ ω  Description  Units  Overpotential between the working electrode catalyst and a reference electrode Change in volume of gaseous components Molar extinction coefficient Electrolyser cell power gain ratio Overpotential Anodic overpotential Cathodic overpotential Concentration overpotential Voltage efficiency EPOC production rate efficiency Fuel efficiency Current efficiency Thermodynamic efficiency System thermodynamic efficiency Overall system efficiency Ohmic overvoltage Charge transfer or activation overpotential Fuel utilization or stoichiometry Rate enhancement factor or Faradaic efficiency Mean size of the crystalline domains Bragg angle Coverage of the promoting ion on the catalyst surface electrochemical potential of an electron at infinite separation from a metal Electrolyser rate enhancement ratio Electrolyser power enhancement ratio Scan rate Work function Outer potential Angular velocity  V l l/mol.cm dimensionless V V V V dimensionless dimensionless dimensionless dimensionless % % dimensionless V V % dimensionless nm radians dimensionless KJ dimensionless dimensionless V/s eV V 1/s  xx  List of Acronyms Acronyms  AC ACS AES APR ATR BET BEV BJH BPG BSE BZY C CE CEaux CEelec CFP CNT CPO CV D DC DI DHE DHRFC DMFC DMR DOE DOS EATR ECSA EDS EDX EIS EMR EP EPOC EPOE ESCA F FC FCV FWHM GB GC GDE GDL GHG  Definition  Alternating current American Chemical Society Atomic emission spectroscopy Aqueous phase reforming Autothermal reforming Brunauer-Emmett-Teller Buttler-Erdey-Gruz-Volmer equation Barrett-Joyner-Halenda Barium-praseodymium-gadolinium oxide Back-scattering electron Barium-zirconium-yttrium oxide Carbon Counter electrode Auxiliary counter electrode Electrolyser counter electrode Carbon fibre paper Carbon nanotube Catalytic partial oxidation Cyclic voltammetry Thermal decomposition Direct current Deionised water Dynamic hydrogen electrode Direct hydrogen redox fuel cell Direct methanol fuel cell Direct methanol reformer Department of energy Density of state Electrochemical autothermal reforming Electrochemical surface area Energy dispersive X-ray spectroscopy Energy dispersive X-ray spectroscopy Electrochemical impedance spectroscopy Electrochemical membrane reactor Electrochemical promotion Electrochemical promotion of catalysis Electrochemical promotion of electrocatalysis Electron spectroscopy for chemical analysis Fuel Fuel cell Fuel cell vehicle Full width half max Global behaviour Glassy carbon Gas diffusion electrode Gas diffusion layer Green house gas xxi  Acronyms  HER HHV HP HPLC ICE ICP ICPMS ID IMFC IR I-R ITO IV LCR LDH LHV MCFC MEA MEPR MPL MS MSE NEMCA OCV OD OES ORR OSR PBI PE+ PEPEM PEMFC PGM PSA PTFE PZC RDE RE RMFC SA SCY SMSI SE SEM SHE SOFC SR SS SSE STP  Definition  Hydrogen evolution reaction Higher heating value High performance High performance liquid chromatography Internal combustion engine Induced couple plasma Induced couple plasma mass spectrometry Inner diameter Indirect methanol fuel cell Current resistance Infra-red Indium tin oxide Inverted volcano Inductance capacitance resistance Layered double hydroxide Lower heating value Molten carbonate fuel cell Membrane electrode assembly Membrane electrochemically promoted reactor Micro-porous layer Mass spectrometer Mercury-mercurous sulphate electrode Non-Faradaic electrochemical modification of catalytic activity Open circuit voltage Outer diameter Optical emission spectrometry Oxygen reduction reaction Oxidative stream reforming Polybenzimidazole Purely electrophilic Purely electrophobic Proton exchange membrane Proton exchange membrane fuel cell Precious group metal Pressure swing adsorption Polytetrafluoro-ethylene Point of zero charge Rotating disk electrode Reference electrode Reformed methanol fuel cell Surface area Strontia-ceria-ytterbia Strong metal-support interactions Secondary electron Scanning electron microscopy Standard hydrogen electrode Solid oxide fuel cell Steam reforming Stainless steel Silver-silver chloride electrode Standard temperature and pressure xxii  Acronyms  TEM TMA TNT TPB UPD UV V WE WEaux WEelec WGS XPS XRD YSZ  Definition  Transmission-electron microscopy Total platinum area Titanium dioxide nanotube Triple phase boundary Under potential deposition Ultra-violet Volcano Working electrode Auxiliary working electrode Electrolyser working electrode Water gas shift X-ray photoelectron spectroscopy X-ray diffraction Yttria-stabilized zirconia  xxiii  Acknowledgements  I wish to thank my supervisor Professor Dr. David Wilkinson for his mentorship, guidance, and support, without which this work would not have been possible. Our discussions helped my progression as a scientific researcher through the strengthening and development of skills that will last a lifetime. I am also very grateful for his encouragement to participate in various learning opportunities throughout the completion of this degree, which contributed to my development at many different levels. I would like to extend my sincere appreciation to my Ph.D. Committee members, Professors Dr. Elöd Gyenge, Dr. Walter Mérida, and Dr. David Dreisinger, for their insightful comments. Additionally, I am grateful to Dr. Ken-Ichiro Ota, from Yokohama National University, for allowing the study of the low temperature contact glow discharge of water during the summer of 2008, Dr. Phillippe Vernoux from the Université Claude Bernard Lyon 1, for hosting the 3rd international training school on the electrochemical promotion of catalysis, Dr. Elena Baranova, from the University of Ottawa, as well as Dr. Constantinos Vayenas, from Patras University, for interesting discussions on electrochemical promotion.  Many thanks to the National Sciences and Engineering Research Council of Canada, the Marie Curie Action Project and the European Union, the Japan Society for the Promotion of Science, the Pacific Century, Scott Paper, the University of British Columbia, my supervisor for scholarships and research assistantships, as well as my family for financial aid in times of need.  I would like to address a special thank you to Dr. Kevin Smith for allowing me to use the mass spectrometer in the Catalysis Laboratory, and to Dr. Farnaz Sotoodeh for providing assistance with it. I would like to acknowledge the technical staff at the Bio-imaging laboratories, at the Materials Engineering Department’s Electron Microscope Laboratory, at the Chemistry Department’s Structural Chemistry Facility, at the National Research Council for Fuel Cell Innovations, and at Cansci Glass Inc. I wish to thank the staff in the Chemical and Biological Engineering Department admistration office, store and workshop for their valuable help. I am very grateful to my friends and research colleagues at the Chemical and Biological Engineering Department, at the Clean Energy Research Center, as well as at xxiv  the National Research Council for Fuel Cell Innovations for their aid in this endeavour, and for an enjoyable and supportive work environment. I would also like to acknowledge my friends at the West Coast Symphony Orchestra and all my musically inclined friends for helping me maintain harmony and balance in my life.  Last but not least, I wish to express my deepest gratitude to my father, Michel Cloutier, for his unconditional love, continuous encouragement and support over the years, and to my beloved mother, Colette C. Cloutier and grandmother, Jeanne H. Crégheur, whose passing affected me beyond words. Warm acknowledgements to the loved ones who helped me while completing this thesis, especially Rick Harkness, and to all the people who encouraged me in difficult times. Without you, I could not have done this.  xxv  Dedication  To whom I own all things.  To my father, and to the memory of my mother (12/15/2005) and grandmother (12/12/2010).  You gave me the opportunity to study science and music, you taught me to be strong, and you inspired me to achieve academic excellence. This work is the fruit of the years of support and encouragement you provided me, and for which I will never be able to thank you enough.  I complete this thesis in your honour.  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  À ceux et celles à qui je dois tout.  À mon père, ainsi qu'à la mémoire de ma mère (12/15/2005) et de ma grand-mère (12/12/2010).  Vous m'avez incité à poursuivre mes études en sciences et en musique, vous m'avez appris à être forte et persévérante, et vous avez été mon inspiration dans la poursuite de l'excellence et de la réussite. Ce travail est le fruit de votre soutien et de vos encouragements. Je vous en suis infiniment reconnaissante.  Je vous rends hommage en complétant cette thèse.  xxvi  Chapter 1: Introduction  1.1 Background  1.1.1 Alternative Fuels  The increased global energy demand and environmental challenges we are facing drive efforts in clean energy research. A peak in conventional non-renewable oil production before 2030 appears imminent [1]. The burning of fossil fuels generates green house gases (GHG), which are considered to be the major contribution to climate change. The escalating need to reduce our fossil fuel dependency for increased energy security and to reduce GHG emissions for a cleaner environment, elevate pure hydrogen (H2) as a promising energy fuel for the future. Switching to H2 as an energy carrier for energy production could alleviate many environmental concerns related to the combustion of fossil fuels [2].  Fuel cells (FCs) are electrochemical devices capable of continuously converting the chemical energy of a fuel into electricity. Hydrogen and fuel cells could provide an efficient sustainable alternative to fossil-fuel based technologies. Hydrogen can be stored or readily used in various applications. It can serve as a chemical reagent in hydrogenation processes and in the production of fertilizers, it can be directly combusted in an internal combustion engine (ICE), or used to feed fuel cells. Fuel cell systems can achieve 2 to 3 times greater overall energy efficiencies than conventional gasoline ICE when fuel production is not taken into account [3]. There is no doubt that fuel processing plays an important role in fuel cell development. The strong interest in H2 has led to the search for various methods to deliver it in a convenient way. However, the issues of H2 production, distribution, compression, safety, and public acceptance, limit the large scale adoption of H2 as an energy carrier. The scientific development of practical fuelling technologies is required to close these existing gaps and facilitate the introduction of a H2 economy. Although H2 is the most abundant element on earth, it is not available in its elemental form in sufficient amounts, thus, it needs to be extracted from other sources. It can be produced from primary fossil energy sources which, even if combined with efficient carbon sequestration, emit various pollutants. 1  In addition, the anticipated decrease in oil reserves will result in crude oil price increases. Therefore, ways to produce H2 from non-fossil sources need to be developed. Until then, H2 applications will mostly depend on hydrocarbon fuel processing for H2 generation.  Alternative fuels, such as liquid alcohols, could serve as a bridge between gasoline and gaseous H2. These alternative fuels possess high energy densities and can be easily stored and distributed through the existing gasoline infrastructure. It has also been argued that alcohols are the next liquid fuels to use after the depletion of petroleum resources [4]. Methanol (CH3OH) is a readily-available electroactive alcohol, which can be economically mass-produced catalytically from non-renewable resources, such as natural gas, and coal, and from renewable resources, such as gasified biomass. It may also be electrocatalytically generated from CO or CO2 [5, 6]. Table 1.1 summarizes some important properties of CH3OH.  Table 1.1: Methanol properties. Properties  Methanol  Energy density [kWh/kg]/[kWh/l]  6.4/4.6  Boiling point [oC]  64.7  Freezing point [oC]  -97  Flash point [oC]  11  Toxicity – Oral [mg/kg]  14.3  In addition, methanol possesses a higher hydrogen to carbon ratio than gasoline (octane, C8H18), i.e., 4 vs. 2.25, respectively. Although CH3OH can be used directly in a fuel cell or combustion process to generate electricity, it will be explained that its electrolysis could be an efficient and economical H2 production method to ease H2 transportation and storage issues with minimum infrastructure changes.  2  1.1.2 Technology Gap  There exist four categories of H2 production technologies: biological, chemical, electrochemical and thermal [7].  Most H2 is currently produced by the well-established catalytic steam reforming (SR)  process of hydrocarbon based fuels, which requires temperatures ranging from 250 to 1545oC, depending on the fuel [8, 9]. Reforming can be carried out externally, integrated with a fuel cell, or done directly inside a fuel cell such as a molten-carbonate fuel cell (MCFC, 600–700oC) or a solid oxide fuel cell (SOFC, 600–1000oC).  On-board gasoline SR was originally believed to be the best way of  generating H2 for transportation, but it was determined that it did not offer clear advantages over other available technologies, such as gasoline ICE and battery hybrids [10]. The practical issues of durability, size and weight, resistance to vibration, cold-start, transient response, and H2 purity concerns, complicates the application of SR for H2 generation in transportation.  Combining SR with low  temperature H2 proton exchange membrane (PEMFC) technology is also restricted by the low tolerance of the H2 PEMFC anode catalyst to the residual CO present in the reformed fuel feed.  1.1.3 Proposed solution  As described later in this thesis, there may be various applications for a low temperature H2 production processes. One of them is the production of H2 for direct combustion or for a PEMFC inside a vehicle. An approach to achieving the ultimate performance targets, including start-up time and energy, may be to conduct on-board reforming at temperatures lower than that of conventional steam reforming (SR) processes. Electrocatalysis can be used to extract H2 from CH3OH at low temperatures (20-80oC) [1112]. Methanol catalytic reforming can also be conducted in the liquid phase at lower temperatures (72– 200oC) [13–16].  It will be seen that, using new electrochemical approaches, the electrocatalytic or  catalytic CH3OH reforming process may result in reduced energy consumption over comparable existing technologies. To the author’s knowledge, no studies have verified if these low temperature alcohol electrocatalytic or catalytic reforming processes could meet the on-board fuel processing targets and if their adoption could facilitate the implementation of a H2 economy. This thesis focuses on the catalytic and electrocatalytic oxidation of CH3OH at low temperatures, i.e., below 200oC.  3  1.1.4 A sustainable closed loop system  Low temperature CH3OH oxidation technologies may meet the short-to-medium term demand for H2, until it can be effectively produced without releasing significant CO2. Figure 1.1 (a) shows an example of how conventional methanol synthesis processes can be combined with a direct methanol reformer (DMR) and a PEMFC. For example, without going into the details of the reaction mechanism, syngas can be produced by methane steam reforming (SR):  CH4 + H2O ↔ CO + 3H2  (1.1)  The water-gas shift (WGS) reaction is used to adjust the CO to H2 ratio for the synthesis of CH3OH.  CO + H2O ↔ CO2 + H2  (1.2)  Methanol is synthesized from the highly selective conversion of the synthesis gas generated.  CO + 2H2 ↔ CH3OH  (1.3)  However, as CO is used in the CH3OH synthesis, the WGS reaction reverses, producing more CO. Therefore, the overall reactions, which produces CH3OH from syngas can be summarised in the following overall reaction:  CO2 + CO + 5H2 → 2CH3OH + H2O  (1.4)  Biomass gasification can also be used to produce CH3OH. For example, wood biomass can be gasified to syngas, which can in turn be synthesized to CH3OH.  2C16H23O11 + 19H2O + O2 → 42H2 + 21CO + 11CO2 → 21CH3OH + 11CO2  (1.5)  4  In this case the process is carbon neutral as the CO2 generated is used to produce biomass by photosynthesis. Ideally, the system would exclusively depend on renewable resources: the methanol could be originating from biomass, while the electricity required for direct current (DC) voltage supply could originate from solar, wind, hydro, tidal, or geothermal energy. Figure 1.1 (b) demonstrates how, in the future, the system could be made sustainable when paired with solar energy and an electrochemical synthesis process. The CH3OH would be synthesized from CO2, generated by the DMR, i.e., methanol electrolyser, and H2O, generated by a PEMFC.  The current efficiency for the electro-synthesis of  CH3OH using RuO2/TiO2 nanotube composite electrodes was reported to be about 60% [17]. It may appear to be simpler to produce H2 and O2 directly from water using solar energy, however, the scheme presented in Fig. 1.1 could also use CO2 from external sources (dashed line), such as fossil fuel combustion, and contribute to the reduction of GHG from other existing processes.  Electricity  Stream reforming, or gasification  Natural gas or biomass  CH3OH  DMR  PEM FC  H2  CO2 H2O (a)  Steam  Heat  Solar  CO2  (b)  Electrosynthesis  O2  O2  H2O, CO2  CH3OH  CO2, H2O  Electricity  DMR  CO2  H2  PEM FC  H2O  Figure 1.1: (a) Schematic diagram of conventional methanol synthesis combined with a DMR and a PEMFC (b) Schematic diagram of a future sustainable system comprising a DMR and PEMFC.  5  In this thesis, it will be seen that electrochemical promotion can be applied to improve the current or voltage efficiency of electrochemical processes. Over the long term, this research may benefit the creation of a sustainable energy system for the production of H2 from renewable methanol originating from biomass.  Over the short term, the findings of this research are more likely to impact the  development of transitional low temperature CH3OH based H2 generation technologies, utilizing energy sources currently available.  1.2 Thesis overview  1.2.1 Justification  Ideally, the CH3OH and water catalytic or electrocatalytic H2 production process should possess a higher efficiency than other currently available technologies.  The non-Faradaic enhancement of catalytic  activity (NEMCA), also referred to as the electrochemical promotion (EP) of catalysis (EPOC), was identified as a novel approach to improve the overall performance of catalytic chemical reactions. This phenomenon has been shown to enhance the rate of a variety of catalytic reactions [18]. A new EP approach, known as triode operation, has been shown to affect the overpotential of electrodes in fuel cells [19]. This approach is analogous to EPOC, except that an electrochemical reaction is promoted instead of a catalytic one. We denote it as the electrochemical promotion of electrocatalysis (EPOE). In this thesis, both of these approaches, i.e., EPOC of the catalytic reforming of CH3OH, and EPOE of the electrocatalytic reforming of CH3OH and water, were investigated at low temperature. The application of EP, through the synergy of solid-state electrochemistry and catalysis, could be a key to efficient fuel processing technologies for H2 production. As EP is a fairly new topic, many cutting-edge research opportunities still remain.  This thesis provides valuable information on the effect of EP on the  electrocatalytic and catalytic reforming of CH3OH and water at low temperature, i.e., ambient to 75oC, and gives insights for further improvements. Applying this new electrochemical engineering technique might eventually lead to the development of a ground-breaking H2 production technology, which could have a tremendous market impact on the energy and transportation sectors, as well as on the environment.  6  1.2.2 Research objectives  The main objective of this research was to improve the state-of-the-art methanol and water electrochemical and chemical reforming processes by rendering them more thermodynamically and kinetically attractive. Investigating the thermodynamic and kinetic characteristics of H2 production via catalytic and electrocatalytic methanol liquid phase reforming involved the following goals: 1) Evaluate the influence of different operating conditions, such as CH3OH concentration (0 to 16 M), and low temperatures (25 to 75oC), as well as the performance of different electro-catalysts (Pt, PtRu) on the electrochemical reforming of CH3OH in acidic media; 2) Determine the effect of EPOE on the electrolysis of CH3OH and water using different triode and tetrode configurations and materials in the galvanostatic and potentiostatic modes; 3) Synthesize, characterize, and evaluate the catalytic and electrocatalytic capabilities of Pt-Ru/TiO2; 4) Determine the effect of EPOC on the low temperature catalytic reforming of CH3OH in acidic media.  This applied research project comprises several innovative contributions, which, to the author’s knowledge, have not been reported in the literature to date. For example, there have been no scientific papers on:  1) The passive electrochemical reforming or electrolysis of CH3OH and water in the liquid phase; 2) The use of the triode and tetrode configuration in the electrolysis mode; 3) The application of triode and tetrode operation in the potentiostatic control; 4) The operation of a tetrode in the mixed galvanostatic and potentiostatic control; 5) The low temperature catalytic reforming of CH3OH and water using Pt-Ru/C; 6) The application of EP on the low temperature catalytic reforming of CH3OH and water using protonconducting electrolytes.  This thesis provides a better understanding of the chemical and electrochemical CH3OH reforming processes at low temperature, as well as valuable information on possible ways to improve their effectiveness, particularly through electrochemical promotion.  7  1.2.3 Layout  This study was achieved by drawing from knowledge in electrochemical, chemical, and material engineering. Six key topics were identified: catalytic reforming, electrochemical membrane reactors, H2 pumping, direct fuel cells, proton-conducting electrolytes, and electrochemical promotion. Only the most essential information is discussed in the literature review, which completes Chapter 1 and starts in Section 1.3.  Chapter 2 constitutes a baseline study of the electrochemical reforming of methanol and water using a proton exchange membrane (PEM) electrolyser and extends the existing knowledge of this system, which is presently limited as will be seen in Section 1.3.3. A theoretical thermodynamic evaluation demonstrating that the electrochemical reforming systems could become thermodynamically favourable under certain operating conditions is presented.  The investigation of the kinetic aspect of alcohol  electrochemical reforming systems was valuable in determining if the electrochemical reaction occurs at a rate of practical interest. The characteristics of the alcohol electrochemical reformer were determined under various conditions and using different membrane electrode assembly (MEA) compositions. The content of this Chapter is based on a publication and is reprinted from the International Journal of H2 Energy, 35, C. R. Cloutier, and D. P. Wilkinson, “Electrolytic Production of Hydrogen from Aqueous Acidic Methanol Solutions”, 3967-3984, Copyright (2010), with permission from Elsevier.  Chapter 3 describes the electrochemical promotion of electrocatalysis in triode and tetrode operation in the galvanostatic and potentiostatic mode. This new application was developed based on fuel cell triode operation knowledge presented in Section 1.3.5.  The effects of electrochemical promotion on the  electrolysis of methanol and water are evaluated under various conditions and different design configurations.  The results obtained are used to compare the energetics of triode and tetrode  electrolysis versus conventional electrolysis.  It will be demonstrated that the application of  electrochemical promotion on the electrochemical reforming of alcohols can lead to the elimination of the need for anodic noble metal catalysts. Part of the work presented in this Chapter was published and reprinted from the ECS Transactions, 25 (23), C. R. Cloutier, and D. P. Wilkinson, "Triode Operation of a  8  Proton Exchange Membrane (PEM) Electrolyser", 47-57, Copyright (2010), with permission from the Electrochemical Society. Another more complete article was also recently submitted for publication.  Chapter 4 contains a baseline low temperature chemical reforming study.  The electrochemical  promotion of low temperature liquid phase catalytic reforming of methanol and water requires that the chosen catalyst, in addition to being a chemical catalyst, is also an electrocatalyst for the methanol and water electrochemical reforming reaction. To achieve this requirement, Pt-Ru/TiO2 was synthesized, characterized and tested for its catalytic reforming capabilities. Also in this Chapter, the effect of EP on the catalytic reforming of CH3OH and water was evaluated using the commercial Pt-Ru/C CH3OH electro-oxidation catalyst and the synthesized Pt-Ru/TiO2. The results obtained are used to compare the energetics and viability of the electrochemically enhanced low temperature CH3OH and water reforming systems versus conventional low temperature catalytic reforming, CH3OH and water PEM electrolysis. The content of Chapter 4 is the basis for a journal submission currently in preparation.  Chapter 5 summarizes the research outcomes of Chapters 2 to 4. Possible applications are described and the significance and impact of the work is discussed. Finally, recommendations for future work are given. Appendices A to J contain detailed information which is supporting but not essential to the thesis. Appendix A contains a list of publications, presentations, and posters which resulted from this research. The mechanism for the electro-oxidation of methanol on Pt and Pt-Ru is shown in Appendix B. Appendix C contains an efficiency and economic comparison of H2 production methods. Some EPOC rules and triode equation derivations are provided in Appendix D. Appendix E contains experimental procedures used to carry out this work while Appendix F contains background information on the various electrochemical techniques employed.  Appendices G and H contain thermodynamic and  electrochemical data, and sample calculations.  9  1.3 Literature Review  1.3.1 Heterogeneous catalysis  In heterogeneous catalysis, the reactant(s) phase differs from that of the catalyst. Most of the time, the reactants are in the liquid or gaseous phase, and the catalyst is in the solid phase. The main stages of heterogeneous catalysis includes (1) the diffusion of the reactant(s) to the catalyst surface (2) the adsorption of the reactants on the catalyst active site through the formation of chemical bonds, (3) after chemical reaction, desorption of the products formed from the catalyst surface, and (4) product(s) diffusion away from the catalyst surface. Heterogeneous catalysis is often the most important step in the synthesis of petrochemicals, pharmaceuticals, chemicals, and in environmental applications.  A catalyst is a substance which is not consumed in a chemical reaction, but accelerates its rate. In heterogeneous catalysis, the diffusion and adsorption of reactants to the catalyst surface is crucial for the chemical reaction to occur. Catalysts can reduce the activation energy required for a chemical reaction to take place.  Most metal catalysts used in heterogeneous catalysis are transition metals.  Their  electronic structure impacts the interactions between the reactants and the catalyst as electronically unsaturated metal atoms allow electrons to be accepted by available d orbitals.  Most hydrogen is produced by the heterogeneous catalytic reforming of fossil fuels [20]. Three main vapour phase production paths exist for the generation of H2 from methanol: thermal decomposition (D), steam reforming (SR), and catalytic partial oxidation (CPO). Autothermal reforming (ATR), also referred to as oxidative stream reforming (OSR), combines the SR with the CPO reactions. Of these processes, only SR is discussed, as it produces the highest H2 concentration. Methanol can be converted to H2 at lower temperatures than other hydrocarbons, making it possible to catalytically reform it by aqueous phase reforming (APR) or liquid phase reforming.  10  1.3.1.1 Steam reforming  The catalytic SR of CH3OH is an endothermic reaction, which is conducted at temperatures ranging from 250 to 350oC [6]. Heat is needed to attain a reasonable CH3OH conversion. The SR process starts with the splitting of CH3OH into CO and H2, followed by the exothermic water gas shift (WGS) reaction (Eq. 1.9), as per the following chemical reactions, which do not describe the details of the reaction mechanism:  Step 1:  CH3OH ↔ CO + 2H2  (1.6)  Step 2 (WGS):  H2O ↔ 1/2O2 + H2  (1.7)  CO + 1/2O2 ↔ CO2  (1.8)  Overall Step 2:  CO + H2O ↔ CO2 + H2  (1.9)  Overall:  CH3OH + H2O ↔ CO2 + 3H2  (1.10)  The enthalpy of the overall reaction at standard conditions is ΔHo = 49.2 kJ/mol. It will be later explained that the overall CH3OH chemical reforming reaction is the same as the overall electrochemical reaction for the electrolysis of CH3OH in a proton exchange membrane (PEM) electrolyser. Catalysts commonly used for the steam reforming of CH3OH are composed of copper supported on zinc oxides, such as CuO/ZnO and CuO/ZnO/Al2O3, in which Al2O3 can be added for thermal stability [21]. Different Cu-Al catalysts of varying Cu contents have been investigated for the production of H2 from CH3OH.  A  maximum H2 production efficiency of 78 mol% was obtained for the SR reaction at 250oC using a catalyst containing 27.8 wt% Cu and calcined at 700oC [22]. Since the WGS reaction (Eq. 1.9) is reversible, CO can be generated. The activity of layered double hydroxide (LDH) catalysts containing various metal combinations was evaluated for the SR of CH3OH at 150–400 °C and atmospheric pressure [23].  Again, the most favourable LDH for the SR of CH3OH was Cu-Al based.  A  WO3/CeO2/YSZ nanocomposite catalyst was used for the reforming of CH3OH, with and without H2O, at temperatures ranging from 100 to 300oC [24]. In both cases, the only product was CO2, which formed at  11  room temperature when H2O was present.  The SR CO2 emissions need to be suppressed by  concentrating and sequestering CO2, which results in energy losses.  1.3.1.2 Liquid phase reforming  The terms aqueous phase reforming (APR) are used to designate the reforming of aqueous solutions composed of direct biomass products, such as methanol, ethylene glycol, glycerol, sugars and sugar alcohols (sorbitol, etc.). Biomass is a renewable carbon-containing fuel, which can be obtained from a variety of sources such as animal, agricultural, and municipal wastes.  In this process, a feed of  oxygenated hydrocarbons, having a limited volatility and a C:O ratio of 1:1, is catalytically converted with liquid H2O over a heterogeneous catalyst at temperatures around 230oC to produce H2 and CO2 [25]. This single-step catalytic process is achievable at conditions where the WGS reaction is favourable, making it possible to generate H2 with low amounts of CO. The APR process consumes less energy than the conventional SR process, as it does not require the vaporization of the oxygenated hydrocarbon and water feeds. Platinum metal has a high activity for the dehydrogenation of CH3OH, so typically, Pt/Al2O3 is used for APR, while other promising materials include Pt and Ni-Sn based catalysts [26-28]. Bimetallic catalysts were shown to result in higher activities for the APR reaction than monometallic catalysts [25]. It was noted that the oxygenated hydrocarbons have various aqueous phase reforming reaction pathways, resulting in catalyst selectivity challenges. The process selectivity depends on many factors, including the catalytic metal, the support, the solution pH, the feed and the operating conditions [25]. The CO species adsorbed at the catalyst surface result in low catalytic activity and are partially removed by the WGS to form CO2. The development of heterogeneous catalysts comprising metals, metal alloys, support and WGS reaction promoters, which are stable under APR conditions, is still under way.  There are only a limited number of studies focusing on the aqueous phase catalytic reforming of CH3OH, which do not originate from biomass sources. Hence, these processes are not referred too as APR in the literature. This liquid phase reforming reaction was first studied on a Ru/C catalyst with external heating at 100-200oC [13]. Compared to Pt-based catalysts, commercial Cu/ZnO reforming catalysts were found to have no activity under liquid phase reaction conditions, although their activity is high for 12  the steam reforming reaction. Silica-supported Pt-Ru catalysts were also investigated at 77 to 84oC at ambient pressure, and CH3OH dehydrogenation was determined to be the rate limiting step of the reaction [14]. The formation of CO2 over Pt-Ru/SiO2 did not proceed via HCHO decomposition, and partly-dehydrogenated CH3OH (CH2OH*) was determined to be the initial reaction intermediate, from which H2 and CO2 were formed through HCOOCH3 and HCOOH as successive reaction intermediates. The catalytic activity and selectivity towards CO2 increased with basic oxide catalyst supports and decreased with acidic catalyst supports [29]. Platinum was most active for liquid phase H2 production reforming when supported on TiO2. In addition, Pt-Ru/SiO2 resulted in an even greater activity than Pt/SiO2. The addition of Ru to Pt/SiO2 accelerated the reaction of the formaldehyde (H2CO) intermediate to form methyl formate (HCOOCH3), as the product formation rate was not restricted in the presence of CO. The dehydrogenation of CH3OH was reported as the rate determining reaction. Ir-Re/SiO2 catalysts were also studied at 105oC and resulted in an activity comparable to Pt-Ru/SiO2 [15]. The highest activity for the catalytic reforming of liquid CH3OH was obtained with Pt-Ru/TiO2 between 77 and 84oC [16]. It proceeded through the partially dehydrogenated HCOOCH3 and HCOOH intermediates through a mechanism similar to that obtained using Pt-Ru/SiO2.  1.3.1.3 Hydrogen spillover in heterogeneous catalysis  Spillover is defined as a phenomenon in which species activated on one phase are transported, usually across a surface, to another phase where they may then participate directly or indirectly in catalysis [30]. The second surface should not normally adsorb or form the active species which are adsorbed or formed on the first surface under the same conditions. The species spilled on the second surface may (1) diffuse on the surface, (2) diffuse or react in the bulk, (3) react on or with the surface, (4) create sites capable of adsorption or catalysis [30]. Spillover can extend from the initial accepting surface to adjacent surfaces in direct or indirect contact with the source of spillover, or the surface that initially accepted the spillover species. This means that spillover species can be transferred over long distances, i.e., over millimetre and centimetre distances [30]. Spillover constitutes an important mechanistic step which can occur during adsorption in heterogeneous catalysis and was found to be involved in many surface reactions and transport phenomena.  13  H2 spillover from a metal to an oxide or carbon surface was extensively studied because most heterogeneous catalysts comprise metal particles supported on high surface area carbon or oxides, and because H2 is involved in many catalytic reactions. It is the fastest spillover process and may occur at room temperature [30]. H2 spillover has been demonstrated on several supported metals, which are known to adsorb H2 dissociatively.  A model assuming the dynamic equilibrium of two coexisting  activated H2 spillover species (H atoms and H+ ions) was proposed to explain the synergy between the components of bifunctional catalysts [31].  It was concluded that the nature of the activated hydrogen  species depended on the physico-chemical properties of the catalyst [32]. For the promotion of acidic catalysis, the spillover species is proposed to be part of the active site as H+ [28]. The principles of H2 spillover have been applied to the design of highly selective and active catalysts from physical mixtures [33].  The spillover of H2 over macroscopic distances was exploited in the design of a dual-bed reactor comprising two separate zones of metal and bimetallic catalysts. Using this design, activities 2.7 times greater than that of the noble metal alone were achieved, demonstrating that a catalytic reaction may occur at different reactive sites as the activating species moved via surface diffusion [34]. Indirect catalyzed hydrogenation was shown to occur via a mechanism in which H2 is activated on a metal catalyst, desorbed in an activated form, which may then react with a second reactant not in contact with the catalyst [35]. The deactivation kinetics of H2 spillover in the gas phase was found to be a first-order reaction, which is accelerated by glass surfaces [36]. The effect of H2 spillover through the gas phase on the hydrogenation of methane was studied on graphite and activated carbon using various independent methods [37]. Studies of the H2 spillover to the carbon support of metal catalysts demonstrated that the uptake and spillover of H2 was greater for Ru containing catalysts than for Pt/C catalysts [38].  1.3.1.4 Catalyst local heating  Catalysts can be thermally heated locally by using heating elements or by the passage of an electric current. The selective oxidation of CH3OH was investigated at temperatures ranging from 100 to 350oC on indium tin oxide (ITO). The study was carried out by (1) thermally heating the catalyst with a ceramic heating element, (2) by heating the catalyst though ohmic heating due to the passage of an electric 14  current, and (3) by thermal heating followed by ohmic heating [39]. It was found that the electrically activated and thermally heated catalyst was much more active than the non-electrically activated but thermally heated catalyst. The electrically activated catalyst was active even after the current was cut-off. This phenomenon was thought to be analogous to the EPOC effect, which is described in more details in Section 1.3.4. However, it was determined that the mechanism was different in this case, as reversing the catalyst electrical leads did not affect the reaction kinetics or selectivity, and the catalyst was still active after the current was cut-off, and the reaction took place on ITO deposited on a porous insulating surface and not on a metal deposited on a solid ion conducting electrolyte. An electrically heatable device using electrically conductive non-metallic materials [40] and a process for the electrically activated transformation of chemical and material compositions have been proposed [41].  Electrically activated  catalysis was also studied for the steam reforming of methane at 600oC [42]. A conversion of 76% H2 was obtained upon passing a current through the catalyst, while a conversion of 66% was obtained upon thermal heating only. More detailed investigations are required to elucidate the reasons behind the catalyst activity enhancement observed.  1.3.2 Proton exchange membrane fuel cell  The electrochemical operating principles of FCs are similar to those of batteries as neither requires combustion. However, fuel cells can effectively convert chemical energy into electrical energy as long as a fuel is supplied, so they do not require recharging like batteries. They are quiet and flexible devices, which convert a fuel to electricity and heat through a spontaneous electrochemical reaction. H2 fuel cells were initially developed for the space program for providing electricity and drinking water to astronauts. As they can produce electrical energy efficiently and can generate less emissions than combustion engines, various types of FCs are being considered for many applications. Most FC research is focused on the development of direct H2 and direct liquid fuel cells. Only low temperature proton exchange (or polymer electrolyte) membrane (PEM) fuel cells are discussed in this thesis.  15  1.3.2.1 Membrane electrode assembly  The membrane electrode assembly (MEA) forms the core of a PEMFC. It is usually inserted between flow field plates, which allow current flow and reactant and product distribution. The MEA comprises a PEM, which separates an anodic electrode from a cathodic electrode. The most common PEM is a perfluorinated ion exchange membrane from DuPont, which is referred to as Nafion®. The mobile ion of this sulphonated fluoro-polymer is H+. This acidic polymer is an excellent gas separator and electrical insulator, and has good chemical resistance and mechanical strength.  However, it dehydrates at  temperatures exceeding 100oC and loses its proton conductivity. Therefore, Nafion® membranes must be kept humidified to maintain their high proton conductivity. The presence of cationic contaminants also affects the mechanical properties and the protonic conductivity of Nafion® [43, 44].  Due to these  shortcomings, work has been done on the development of other proton conducting electrolyte membranes for PEMFC applications, such as ceramics [45-50], composite [51-54], other polymers [5557], and metal coated polymers [58, 59].  The anode and cathode of fuel cells are usually composed of electroactive Pt group metal (PGM) based catalysts. As platinum has a high cost and a low abundance, precious metal nanoparticles are often dispersed or supported on larger carbon particles to reduce the amount of Pt needed while increasing the electrochemically active surface area (ECSA) of the catalyst. ECSA losses deteriorate the PEMFC performance. Improving the catalyst activity, utilization and stability are important topics in PEMFC research. The main approaches taken to reduce or replace Pt while maintaining PEMFC performance and efficiency are to reduce the noble metal loading or use less expensive noble metals [60], or explore using non-Pt electrocatalysts [61, 62]. Changing the carbon catalyst support is also looked at as it may affect the catalyst activity and lead to the partial reduction of the Pt loading [63].  A triple phase boundary (TPB) between the electrolyte, reactants and catalyst is required for an electrochemical reaction to occur.  Different methods can be used to create this three-dimensional  electrochemical reaction zone. The most common is the application of an ink comprised of a polymer electrolyte solution and electrocatalytic particles. Thin films of catalyst inks usually contain an electrolyte ionomer binder (typically Nafion®), and supported or unsupported nanocatalyst Pt particles. A three 16  phase contact between the H2, protons and solid catalyst is obtained by applying this electronically connected ink coating onto a porous conducting carbon substrate or a membrane. The three main methods consist of applying an ink onto (1) a Teflon film which is transferred by decal to the membrane through a hot pressing process [64] (2) a porous conducting carbon substrate (gas diffusion layer, GDL) to form a gas diffusion electrode (GDE) [65], or (3) onto each side of the membrane to form a catalystcoated membrane (CCM) [66].  1.3.2.2 Overvoltage  In fuel cells, the operating voltage, E (V), is always greater than the standard reversible cell voltage, Eo (V):  Eo = −  ΔG o  (1.11)  nF  where ΔGo (kJ/mol) is the change in Gibbs free energy of reaction, n is the number of electrons participating in the reaction, and F is Faraday’s constant (96485 C/mol). When not operating at standard conditions, the thermodynamic equilibrium (reversible) cell voltage at non-standard temperature and pressure, Ee (V), can be estimated with the Nernst equation:  ⎛ Π a Rer , Jd , j 2.303RT ⎜ j o Ee = E − Log ⎜ s nF ⎜ Π aOxO ,,jj ⎝ j s  ⎞ ⎟ ⎟ ⎟ ⎠  (1.12)  where R is the ideal gas constant (8.314 J/molK), T is the temperature (K), and a is the activity (dimensionless). The excess amount of electrical energy required for the electrocatalytic process to occur is known as overvoltage, η (V). It is the difference between the operating cell voltage, E, and the equilibrium cell voltage, Ee:  17  η = E − Ee  (1.13)  The overvoltage represents the sum of the cell efficiency losses which are mainly due to the activation overvoltage or charge transfer overvoltage, ηs, the concentration overvoltage, ηconc, and the ohmic overvoltage, ηohm.  E = E e − η s − η conc − η ohm  (1.14)  The activation overvoltage and the concentration overvoltage in Eq. 1.14 can be separated in their anodic and cathodic components. If so, the absolute values in Eq. 1.14 are then removed, and the anodic component is subtracted as it is a positive value, while the cathodic component is added, as it is a negative value. The activation overvoltage is related to the activity of the electrode surface which affects the charge transfer process for a particular electrochemical reaction. The concentration overvoltage is linked to the change in the mass transport of reactants to the electrode surface as the electrochemical reaction proceeds. The ohmic resistance is caused by the PEM/electrocatalyst interface resistance, as well as the bulk resistance and the ionic conductor’s resistance to the transport of electrons. Other losses include fuel crossover and internal currents [67]. The power output and thermodynamic efficiency of electrochemical systems depend on the minimization of overpotential losses at the anode and/or cathode. Significant overpotentials will result in  inefficient electrocatalysis and restrict the  commercialization of electrochemical devices as they are often not sufficiently cost-effective. For this reason, research efforts are concentrated towards the minimization of overvoltage losses, which is crucial to the development of effective electrocatalytic processes.  1.3.2.3 Hydrogen proton exchange membrane fuel cell  The H2 PEMFC uses H2 as the fuel and O2 as the oxidant. The oxidation occurs at the anode and the reduction occurs at the cathode. The electrons pass through an external electrical circuit while the protons diffuse through the membrane. The overall reaction is the sum of the anodic and cathodic reactions. The PEMFC electrochemical reactions are as follows: 18  Anode half-cell reaction:  H2 → 2H+ + 2e-  Eoa = 0 V vs. SHE  (1.15)  Eoc = 1.23 V vs. SHE  (1.16)  Eo = 1.23 V  (1.17)  Cathode half-cell reaction:  1/2O2 + 2H+ + 2e- → H2O  Overall reaction:  H2 + 1/2O2 → H2O  Note that throughout this thesis, the convention of reversing the Eoa values reported for the half-cell reactions written as electro-oxidation was adopted. The subscript “a” designate the standard potential of the half-cell reaction occurring on the anode, while the subscript “c” designate the standard potential of the half-cell reaction occurring on the cathode. The standard reversible cell voltage, Eo, based on thermodynamics is 1.23 V. The rate limiting process is the cathode oxygen reduction reaction (ORR), which suffers from kinetic limitations. For this reason, a larger quantity of catalyst is needed for the slow ORR cathodic reaction compared than for the fast H2 anodic reaction. The H2 PEMFC has significant water management issues, which need to be continuously addressed to reach stable operation, and for this reason, their operation is usually limited to temperatures around 60-90oC.  Hydrogen purity requirements  The PEMFC electrodes usually contain a Pt catalyst, which promotes the reactions occurring at the electrodes. The electrocatalytic properties of Pt are greatly affected by contaminants, which adsorb to the surface of the catalyst and prevent H2 adsorption onto the catalyst surface [68]. At temperatures below 150oC, CO strongly adsorbs onto Pt and poisons the H2 reaction. This CO poisoning effect occurs at concentrations of CO as low as 5-100 ppm (0.0005-0.01%) in the H2 inlet stream, depending on the catalyst loading [69]. A tolerance to CO levels of 100 ppm or more would greatly simplify the PEMFC 19  fuelling system when H2 is produced by catalytic reforming processes. Typically, the H2 fuel stream needs to contain less than 10 ppm of CO in order to be fed directly to a PEMFC.  Different options exist to overcome electrocatalyst CO poisoning in the PEMFC [70]. These include increasing the cell operating temperature, which weakens the CO bond to the Pt active sites and reduces the catalyst’s sensitivity to CO poisoning. It is also possible to remove CO from the reformate using advanced fuelling system design or membranes for CO separation. Preferential oxidation reactors or H2 separation membranes containing Pd-based alloys or proton-conducting ceramics can be used to remove CO. For example, Pt-alloy catalysts can preferentially oxidize CO and selectively remove it [71]. Other methods are the introduction of oxidant in the H2 fuel feed stream, referred to as O2 bleeding, or the use of cyclic operation to clean the surface from adsorbed species.  1.3.2.4 Direct methanol fuel cell  The direct methanol fuel cell (DMFC) is the most advanced type of direct liquid fuel cell. In a DMFC, a diluted aqueous solution of CH3OH is fuelled directly to the anode of a PEM fuel cell, while air or O2 is supplied to the cathode, as per the following electrochemical reactions:  Anode half-cell reaction:  CH3OH + H2O → CO2 + 6H+ + 6e-  Eoa = -0.016 V vs. SHE  (1.18)  Eoc = 1.23 V vs. SHE  (1.19)  Eo = 1.21 V  (1.20)  Cathode half-cell reaction:  3/2 O2 + 6H+ + 6e- → 3H2O  Overall reaction:  CH3OH + 3/2O2 → 2H2O + CO2  20  The DMFC has a standard reversible cell voltage of 1.21 V. The MEA arrangement for the DMFC is similar to that for the H2 PEMFC. When liquid CH3OH mixtures are used as the fuel, the membrane can maintain high humidification levels, but cross-over of the liquid CH3OH can be a problem. While the cathode reaction of the PEMFC suffers from poor kinetics, in the DMFC, the anode and the cathode reactions both result in kinetic losses. Thus, the two main issues which restrict the performance of DMFCs are the poor anodic CH3OH oxidation kinetics, and CH3OH crossover.  Anodic methanol oxidation  The slow CH3OH oxidation kinetics at the electrode/electrolyte interface in acidic media limits the electrochemical reaction rate. The high activation overpotentials observed are mostly caused by the complex anodic electrochemical reaction mechanism [4]. Possible CH3OH electro-oxidation pathways are illustrated in Appendix B. The direct electrochemical oxidation of CH3OH on Pt occurs in multiple stages, which include various adsorbed intermediates. Strongly adsorbed CO remains un-oxidized at the surface and blocks the electrode active sites from adsorbing reaction intermediates which are formed during the oxidation process [4, 72]. The poisoning species which deactivates the electrode surface is CO. It is the rate limiting component for the electro-oxidation of CH3OH.  Typical methods used to prevent CO poisoning in DMFCs include modifying the anode electrocatalyst structure/composition (e.g., Pt-Ru) to enhance the reaction rate, adding oxygen in the fuel (O2 bleeding), using cyclic operation, and increasing the operating temperature. It was found that adding a second component to Pt prevented the formation of strongly absorbed CO by accelerating its oxidation kinetic at lower potentials. To date, the preferred anode material for the electrochemical oxidation of CH3OH is an optimized Pt-Ru binary electrocatalyst [73]. The presence of Ru results in the oxidation of CO via the formation of adsorbed hydroxide ions at low potential and the lowering of the oxidation potential of CO [72]. The bi-functional mechanism for CH3OH electro-oxidation on Pt-Ru is described in Appendix B. Binary metal-oxide Pt catalysts, which are poor electrical conductors, were also studied to promote the electrochemical oxidation of CH3OH [74]. It was reported that Pt-WO3 electrodes are more active and resistant to poisoning than Pt or Pt-Ru alloy catalysts [75]. In this work, Pt and Pt-Ru supported and unsupported catalysts will be employed. 21  Methanol crossover  Methanol is fully miscible, i.e., a highly soluble molecule which readily mix with water.  Nafion®  membranes are permeable to CH3OH. In DMFCs, water and CH3OH molecules are transferred through the membrane by diffusion due to the driving force of concentration gradients between the anode and the cathode.  They are also transferred through the PEM by electro-osmotic drag, as induced by the  movement of protons, and by hydraulic pressure gradients which is usually small compared to the two other crossover mechanisms.  In a flowing-electrolyte DMFC, diffusion is replaced by convection-  diffusion [76]. Water diffusion can be neglected in a well-humidified membrane.  However, CH3OH  diffusion cannot be neglected because a CH3OH concentration gradient develops between the anode and the cathode. The diffusive flux and electro-osmotic drag of CH3OH are both directed from the anode to the cathode.  While anodic gaseous CO2 results from the electro-oxidation of CH3OH, cathodic  gaseous CO2 is usually attributed to the combustion of the crossed-over CH3OH with O2. However, it was shown that anodic permeation CO2 to the cathode by diffusion and convection due to electroosmosis also occurs in DMFC, and is not negligible [77]. The anodic CO2 permeation was determined to be 20% of the total cathodic flux in a DMFC at room temperature.  Methanol crossover from the anode to the cathode results not only in fuel waste, but also reduces the voltage and Faradaic efficiency of DMFCs [78, 79]. This crossover phenomenon lowers the DMFC performance at higher fuel concentrations [80].  For this reason, there is an optimum CH3OH  concentration to use at specific DMFC operating conditions [81]. With Nafion®, the crossover typically decreases with decreasing operating temperatures and concentrations, increasing cathodic pressure, increasing membrane thickness and equivalent weight, as well as increasing current density.  Typical measures to minimize CH3OH crossover and its effects include the optimization of the protonconducting membrane structure and composition, the use of oxygen reduction electrocatalysts insensitive to CH3OH at the cathode, and the improvement of electrochemical oxidation catalyst utilization and optimization of the cell operating conditions.  Modifying the CH3OH membrane  permeability without decreasing its ionic and/or electrical conductivity is not a simple process [79]. High  22  temperature (120-200oC) acid-doped polybenzimidazole (PBI) membranes have high conductivities [82] and improve DMFC cell performance by decreasing CH3OH crossover [83].  1.3.2.5 Indirect methanol fuel cell  The indirect methanol fuel cell (IMFC), also referred to as the reformed methanol fuel cell (RMFC), is the combination of a H2 PEMFC with an upstream catalytic CH3OH SR. This approach has the benefits of both the high electrochemical activity of H2, as well as the convenience of liquid CH3OH storage and distribution. However, as mentioned earlier, the SR product stream needs to be purified before H2 can be fed to the PEMFC. Any remaining un-recoverable H2 is combusted with air. As the IMFC system typically operates at higher temperatures than the DMFC, heat losses have to be properly managed. Due to the energetic inefficiencies associated with the catalytic CH3OH SR, the IMFC tends to be larger, heavier, and have longer start-up times than the DMFC [84]. For this reason, most research is directed towards the DMFC instead of the more complex IMFC system.  Feeding liquid CH3OH directly to the anode of a DMFC eliminates the need for a fuel processor, but, as discussed earlier, the full potential of the DMFC is hindered by the slow CH3OH electro-oxidation kinetics, and CH3OH crossover. To reduce the effect of crossover, the operation of DMFCs is limited to low CH3OH concentrations, which renders their performance more suitable for low power density applications. As the H2 PEMFC can attain higher power densities than the DMFC, the DMFC’s low power density would require improvements to be competitive with the IMFC. The IMFC process could become a more interesting option if a low temperature CH3OH fuel reforming process could be developed.  The drive cycle dynamic response and efficiency of an IMFC’s CH3OH SR has been investigated [85, 86]. A study from the Los Alamos National Laboratory compared the trade-offs of using a PEMFC, DMFC or an IMFC for portable power applications [87]. Ishihara et al. compared the energy of an IMFC with that of a DMFC and determined that, at a current density of 600 mA/cm2, the DMFC needed to operate at a minimum voltage of 0.5 V to have a greater energy efficiency than the IMFC [88]. Brown compared seven fuels, including CH3OH, gasoline and H2, as possible H2 sources for PEMFC for use in fuel cell 23  vehicles (FCVs) [89]. Lattner et al. compared CH3OH based fuel processors using commercial catalysts for PEMFC systems [90]. A high-level efficiency and economic comparison of some H2 production methods is provided in Appendix C.  1.3.3 Proton exchange membrane electrolysis  The principle of electrolysis has been known since the early 19th century. It consists in the separation of chemically-bonded elements and substances by the passage of an electric current. The electrochemical decomposition of water is the most practical electrolytic process for the production of pure H2 and O2. Existing H2 electrolytic production methods include alkaline electrolysis, PEM electrolysis, ceramic oxide electrolysis, and photoelectrolysis. PEM water electrolysis was developed over the last decades [91]. It is an acidic process which possesses many benefits over traditional alkaline electrolysis: it results in higher energy efficiencies and power densities, and it is a simple system that can operate at low temperature and high pressure [92, 93]. It has been used in the industrial gas markets for H2 production, in the military and aerospace markets for O2 production, and is now close to entering new emerging energy markets, such as H2 vehicle refuelling [94]. It represents an attractive technology for portable applications due to its compactness and its ability to effectively produce pure H2 on demand.  1.3.3.1 Proton exchange membrane water electrolysis  PEM water electrolysers are the reverse of H2 PEMFCs. Water is split into oxygen and protons at the anode. The protons travel through the proton conducting membrane and combine with electrons at the cathode to form H2 via the hydrogen evolution reaction (HER). The water electrolysis electrochemical reactions are as follows:  Anodic half-cell reaction:  H2O → 1/2O2 + 2H+ + 2e-  Eoa = -1.23 V vs. SHE  (1.21)  24  Cathodic half-cell reaction:  2H+ + 2e- → H2  Eoc = 0 V vs. SHE  (1.22)  Eo = -1.23 V  (1.23)  Overall reaction:  H2O → 1/2O2 + H2  Water electrolysis is an endothermic reaction with a positive Gibbs free energy, which results in a negative standard equilibrium cell voltage of -1.23 V. It is a power sink which requires electrical energy input to extract H2 from water. The amount of potential/current applied controls the electrocatalytic reaction rate as per Faraday’s law. Commercial PEM water electrolysers operate between 60-80oC. Although there exists significant opportunities for the development of H2O electrolysis systems, they remain economically limited by the high overall cost required to split H2O. While the H2O electrolyser operating costs may be more affordable where electricity prices are low, the cost of electricity greatly fluctuates with geographical location. When combined with renewable sources, PEM H2O electrolysis becomes a possible pathway to zero pollution fuelling [95]. To render the electrochemical generation of H2 as economic as the production of hydrocarbon fuels, significant research advancements are needed to reduce the high energetic cost of H2O splitting, increase its operating efficiency, and reduce its capital cost.  The core structure of PEM water electrolysers is similar to that of H2 PEMFC and they are subject to similar overpotential losses. Fortunately, some of the challenges encountered in PEMFCs, such as cooling and water management, do not apply to PEM electrolysers as there is a constant contact with H2O.  The anode, where the electro-oxidation of water to oxygen occurs, has a high activation  overpotential and represents the largest portion of the PEM water electrolyser losses. Like in the case of the PEMFC, various PEM water electrolysis research approaches are taken to address the concerns related to catalytic activity, electronic conductivity and stability. Non-noble metal catalysts corrode and Pt forms a conducting oxide film when in contact with the acidic Nafion® membranes. Pt has a low  25  overpotential at low current densities, but its resistance must be reduced at high current densities. The performance of the membrane and electrocatalysts is crucial to maintain high electrolyser efficiencies at high current densities. Since the development of dimensionally-stable anodes (DSA), the anode of PEM electrolysers usually comprises noble metal oxides (Ir, Ru) combined with oxides of non-noble metal (Sn, Ti) as electrocatalysts. These anodes usually possess large active surface areas. However, it was reported that there may be a poor contact between the catalyst layer and the PEM electrolyte because of high open porosity, and that the thin-film electrocatalyst layer technique, also used in PEMFCs, may be a preferred option [96]. Fujishima et al. photo-electrochemically produced H2 using n-type semiconductor TiO2 at the electrolyser anode [97]. Composite membranes based on Nafion® and TiO2 or Nafion® and SiO2 were also developed to improve the membrane conductivity for high temperature PEM electrolysis [98, 99].  Some PEM water electrolysers are commercially available. Hydrogenics Corporation’s  HyLYZERTM consists of 7 PEM cells which can generate up to 1 Nm3/h of 99.99% H2 at a stack efficiency of 4.9 kWh/Nm3 and ambient temperature (5 to 35oC) [100]. Its maximum power consumption is 7.2 kW, when all equipment is included. The H-Tec EL 30 electrolyser has a maximum H2 production rate between 0.3 m3/h (13 cells) and 3.6 m3/h (144 cells) at ambient temperatures ranging from 4 to 50 oC [101]. Their rated power range is from 1.8 to 20 kW, but no information is given about the EL 30 bar electrolysers efficiencies. It is not clear if these commercial electrolysers operate at lower currents to attain greater efficiencies. While reduction in the electrical energy required for the electrolysis may be obtained by reducing the overpotential losses, it is not possible to operate the electrolysis system at voltage more positive than the H2O electrolysis reversible cell voltage at standard conditions. Further energy cost reductions might be achieved through depolarized electrolysis methods as described below.  1.3.3.2 Depolarized proton exchange membrane water electrolysis  Increasing the electrolysis efficiency and reducing the reversible cell voltage can be accomplished by substituting the O2 evolution reaction (Eq. 1.16) with another anodic reaction taking place at a lower potential than that of O2 evolution. This technique, called anodic depolarization, consists of depolarizing the H2O electrolysis anodic electro-oxidation reaction by oxidizing a fuel, which forms oxidized species along with H+ ions and electrons [102]. The electro-oxidation of the chosen depolarization fuel must have more favourable thermodynamics than the O2 evolution reaction, i.e., a smaller reversible standard 26  cell voltage, Eo. In addition, Eo and the operating electrolyser cell voltage in normal operation, Uoelec, cell, would both be further reduced, if the chosen anodic electro-oxidation involved the transfer of more than two electrons. Therefore, depolarized electrolysis may result in reduced energy consumption, hence cost reductions, over conventional water electrolysis. Depolarization fuels studied in the past include coal [103], glucose (C6H12O6) [104], sulphur dioxide (SO2) [105, 106], methanol (CH3OH) [107, 108], and ferrous ions (Fe2+) [109]. Only the case of methanol depolarization is discussed here.  Methanol depolarized PEM electrolysis  Methanol is a particularly attractive water electrolysis depolarizer for the generation of H2. The DMR or CH3OH PEM electrolyser is analogous to a DMFC as the anodic reaction is the electrochemical oxidation of CH3OH in both cases. Consequently, most issues taking place at the electrodes during CH3OH depolarized water PEM electrolysis are encountered at the DMFC anode. The electrochemical reactions are as follows:  Anode half-cell reaction:  CH3OH + H2O → CO2 + 6H+ + 6e-  Eoa = -0.016 V vs. SHE  (1.24)  Eoc = 0 V vs. SHE  (1.25)  Eo = -0.016 V  (1.26)  Cathode half-cell reaction:  6H+ + 6e- → 3H2  Overall reaction:  CH3OH + H2O → CO2 + 3H2  The overall reaction for CH3OH catalytic reforming process (Eq. 1.10) is the same as for the electrochemical CH3OH reforming process (Eq. 1.26). The standard reversible cell voltage required to  27  drive the CH3OH depolarized PEM electrolysis (Eq. 1.26), is much less (minimum 0.016 V) than for water PEM electrolysis (minimum of 1.23 V). Like in the PEM water electrolysis case, the amount of electrical energy required mainly depends on the anodic activation overvoltage. This is because the H2 evolution reaction is very rapid, and the CH3OH oxidation is rate limiting. The electrolyzer electrode where the electro-oxidation of CH3OH occurs causes most of the overpotential in this electrochemical system. As in the case of the DMFC anode, the anodic activation overvoltage of the CH3OH depolarized water electrolyser results from the poor alcohol electro-oxidation kinetics of the complex reaction mechanism involving adsorbed intermediates. In CH3OH electrolysis, CH3OH crossing over to the cathode will not be oxidized, but will result in a fuel loss, which should be minimized. It appears that no scientific papers were published on the CH3OH depolarization of PEM water electrolysis in acidic media until 2007. However, there was some prior activity on this topic in the patent literature. Some electrochemical reforming membrane reactors for the electrochemical reforming of CH3OH investigated are described in Table 1.2.  Table 1.2: Relevant methanol electrochemical reforming membrane reactors.  Reactant  Product  Solid Electrolyte  Porous Electrode  Temperature  Reference  Electron Donor  Electron Acceptor  CH3OH (l)  H+  H2, CO2  Aqueous NaOH or KOH [6-12 M]  Pt/C  23-60  [110]  CH3OH  H+  H2, CO2  -  -  -  [111]  CH3OH (g)  H+  H2, CO2  Nafion®  Pt, Pt/Ru anode, Pd-Ag foil cathode  50-100  [112]  [oC]  The electrochemical production of H2 from CH3OH in the presence of a base, such as NaOH or KOH, was carried out at Pt/C electrodes [110]. The inclusion of a base permitted the generation of H2 without emission of greenhouse gas and without the need to separate H2 from the other gaseous products. However, the resulting carbonate ion by-product needs to be dealt with. An electrochemical reformer and fuel cell system was developed to reform organic fuels into H2 and CO2 [111]. It uses electricity  28  and/or thermal energy to supply the necessary reaction energy to convert the fuel to H2. There is no mention of the operating temperature, and of the type of electrode or electrolyte used.  An  electrochemical process to reform a mixture of CH3OH and water at temperatures between 50 and 100oC was proposed [112]. As shown in Fig. 1.2, the methanol electrochemical reforming membrane cell operating in the gas phase resembles a DMFC.  DC Power Source e-  eCO2 H+  H2  CH3OH H2 O e-  e-  Nafion®  Pt, Pt/Ru Anode  Pd Cathode Pd-Ag alloy  Figure 1.2: Schematic diagram of a methanol electrochemical reforming membrane cell (Modified from patent DE 197007384 [112]).  A polymeric proton-conducting membrane and Pt or Pt/Ru catalysts were used. A Pd-Ag foil was hotpressed on the cathode polymer side of the membrane to prevent water and CH3OH crossover. A cell voltage between 300 and 500 mV was sufficient to generate H2.  There are three Japanese patents pertaining to the electrolysis of alcohols: two of them relate to the electrolysis of alcohols in conjunction with a PEMFC, which is similar to an entirely electrochemical version of an IMFC [113. 114], and the other describes an organic electrolytic synthesis in the presence of a base [115]. The first American patent on the generation of H2 production by electrolysis of organic solutions dates from 2001 [116]. It was later followed by a patent on the electrolytic production of H2 in 29  the presence of a base [110]. Narayanan et al. last re-amended their original 2001 US patent in 2006 [117].  An example of a PEM CH3OH electrolyser in transient operation is shown in Fig. 1.3. The anode chamber is continuously fed with un-acidified CH3OH aqueous solution, which is recycled using a pump. The cathode chamber is purged with argon. The anode catalyst is Pt or Pt-Ru/C and, typically, the cathode catalyst is Pt/C, but the performance of a Pt-WC/C catalyst has also been investigated. The electrolysis of CH3OH resulted in an absolute voltage of 0.4 V to 1 V at the same current density depending on the cathode catalyst used [7, 118]. The current collectors and diffusion layers were made of carbon paper.  As Nafion® 117 was the electrolyte, the CH3OH electrolyser was operational at  temperatures up to 120oC.  Proton Conductor Electrolyte  CO2  Ar, H2 Anode  H+  Fluid Pump  Cathode  Ar  CH3OH + H2 O  DC Power Source Anode Reaction: + CH3OH + H2O → CO2 + 6H + 6e  Cathode Reaction: 6H+ + 6e- → 3H2  Figure 1.3: Schematic diagram of an alcohol electrolyser (Modified from Hu et al. [7] with permission from Elsevier).  At low current densities, the voltage reached a steady-state approximatively an hour after electrolysis started. At higher current densities, a rapid drop in negative voltage was observed. The drop happened because the applied current could not be sustained with only the oxidation of CH3OH and drifted to a 30  value where O2 evolution could also occur [118]. Hu et al. reported that the electrolysis voltage did not depend on the CH3OH concentration at concentrations greater than 2 M and at temperatures below 60oC [7]. The explanation given was that at CH3OH concentrations exceeding 2 M, CH3OH crossover to the cathode resulted in a mixed potential which reduced the activity of the electrocatalyst. Shen et al. reported that the electrolyser voltage was about the same at CH3OH concentrations between 2 and 4 M, and that the voltage was greater outside this concentration range [11]. Again, this was explained by mass transport limitations at low CH3OH concentrations, and by CH3OH crossover to the cathode, which supposedly created a mixed potential, at higher CH3OH concentrations.  Yet, Take et al. confirmed that the voltage did not depend on the CH3OH concentration, but only on the current density [118]. However, the limiting current density at which the voltage rapidly decreases varies with CH3OH concentration. It attained a maximum value when the methanol-water ratio was equal to 1 [118]. In the same study, it was also established that the oxidation of crossover methanol in the cathode compartment does not occur. The CO2 in the cathode exhaust is not produced at the cathode because no O2 is supplied at the cathode. In fact, any cathodic CO2 permeated from the anode in a similar manner as to that previously discussed for DMFCs. The CO2 permeation rate diminished with increasing anode CH3OH concentrations and increased with increasing current densities. Also like for the DMFC, the water and CH3OH would also cross over to the cathode at permeation rate which decreased as the current density increased. Take et al. verified that the CH3OH crossover rate escalated with increasing CH3OH concentrations, while the water permeation rate was not affected [118]. The water and CH3OH at the cathode can be separated from the H2 produced by molecular sieves. However, since it was found that the permeated water-methanol solution concentration was almost the same as the solution supplied to the anode, the permeated solution was recycled back to the anode. The flow rate of the H2 produced increased proportionally to the current density and came close to the theoretical H2 production rates [118]. The electrolysis voltage became more positive with increasing temperature as the mass transport and the kinetics are more favourable and the activation and concentration polarizations are reduced. For example, the voltage at 80oC was reduced by half compared to the voltage at 20oC [11].  31  Applications of depolarized methanol PEM electrolysis  The CH3OH PEM electrolysis H2 production has advantages over conventional CH3OH steam reforming as it may result in reduced energy consumption.  First, the single-reactor CH3OH electrochemical  reforming process has the potential to be conducted in the liquid phase at a lower operating temperature and at higher overall system efficiencies compared to a conventional high-temperature multi-reactor SR system. Secondly, since the production of H2 from depolarized electrolysis should result in lower power consumption than H2 production from water electrolysis. Therefore, its large scale commercialization might not be as significantly energetically restrained as that of water electrolysis. Narayanan et al. estimated that H2 production from the electrolysis of CH3OH would cost about 50% less compared to that of water, even when the cost of CH3OH is taken into account [117].  An efficiency, energy and  economical comparison of various H2 production methods is presented in Appendix C.  Although CO2 is produced whether CH3OH is thermally or electrochemically reformed, it is more localized and concentrated in the electrochemical reforming process than in the chemical reforming process.  For this reason, electrochemical reforming may result in more effective CO2 capture for  sequestration and in lower CO2 disposal cost in many applications.  The cost of CO2 capture and  disposal for various coal-fired generation technologies ranges from 1.1 to 4.9 cent/kWhe and 0.6 to 6.7 cent/kWhe, respectively [119]. In some cases, it may be desirable to combine the CO2 produced with the H2 generated at high temperature and pressure to obtain methane, as per the Sabatier reaction:  CO2 + 4H2 → CH4 + 2H2O  (1.27)  Alternatively, other depolarized water PEM electrolysis processes using compounds which do not produce CO2, such as the ferrous ions (Fe2+), could be evaluated for H2 production.  Methanol electrolysis could effectively be used as part of systems cogenerating chemical energy and electrical energy, which may include a H2 PEMFC, as long as the amount of energy required to generate H2 does not exceed the energy resulting from using the H2 produced. A system composed of an efficient electrochemical CH3OH electrolyser providing H2 to a PEMFC could be more advantageous than a 32  DMFC because it might be possible to use higher concentration fuel mixtures in a system combining a CH3OH electrolyser with a H2 PEMFC than in a DMFC. Without taking the storage and infrastructure issues into account, the overall performance of a system composed of a CH3OH electrolyser and a H2 PEMFC is expected to be less than a system in which gaseous H2 tanks directly feeds a PEMFC, but greater than that for a IMFC. The system combining the principles of H2 pumping and autothermal reforming referred to as electrochemical autothermal reforming (EATR) [120, 121] could be modified to include an electrochemical reformer in order to simultaneously produce H2 and useful heat.  The  resulting higher temperatures would render the thermodynamic and kinetic aspects of the electrochemical CH3OH oxidation more attractive.  The prompt start-up and shut-down time of PEM electrolysers and their ease of scalability render them appropriate to generate pure H2 for portable and stationary applications. Having similar scalability limits as PEMFCs, multiple PEM electrolysis cells can be put into stacks. Currently, some of the largest PEM electrolysis stacks commercially available have a rated power of 14 to 20 kW [101]. The combination of an electrochemical alcohol reformer and a fuel cell system could potentially be made compact, light and simple enough for use in small low power applications, which usually do not justify the cost of a separate fuel processor. It could be developed into a portable fuelling option for high and low power output PEMFCs, as well as for micro-fuel cell power sources, and used to power laptops, cell phones and other small portable devices. Under thermodynamically favourable operating conditions, the direct electrochemical CH3OH reformer could be used as a stand-alone H2 generation unit for H2 combustion engine vehicles or FCVs requiring fuelling at local electrochemical stations. Distributed H2 production at a H2 refuelling station constitutes an attractive option to supply H2 to FC vehicles, allowing the supply to match the demand as more H2 vehicles are driven. As it has clear advantages over high temperature SR, the low temperature CH3OH electrochemical reformer may meet the DOE on-board fuel processing targets for automotive applications. It may represent an attractive on-board H2 storage alternative to high pressure H2 storage in FCVs, as the H2 could be produced at the PEMFC’s demand. A CH3OH fuel processor and fuel cell system could be optimized to give maximum power density as an off-grid generator system for back-up and remote power applications for the telecommunications, utilities and military sectors. As it requires less energy to electrolyser CH3OH than water, and depending on the source of CH3OH, CH3OH electrolysis may be a preferred option to produce H2 during off-peak hours so 33  electricity can be provided by H2 PEMFCs during high grid loads or blackouts. Based on these examples, it can be seen that an efficient CH3OH electrochemical reformer has a significant market potential.  1.3.4 Electrochemical promotion of heterogeneous catalysis  Catalysts typically consist of an active metal or metal oxide phase, a support, and one or more promoters. The promoters affect the properties of the supports or of the active phase [122]. The metal-support interactions between catalytic nanoparticles and their support is a well studied heterogeneous catalysis concept. A special type of heterogeneous catalysis, in which the catalyst properties are modified by the application of an electric field, was discovered in 1980, and was reported as a new electrochemically induced catalytic effect in 1988 [123, 124]. It was found that solid electrolytes or active catalyst supports may be used to modify the catalytic properties of metals and metal oxides. Applying polarization controls the surface work function by creating promoters and modifying their quantity on the catalyst surface, which in turns affects the adsorption, amount, and surface diffusion properties of reaction intermediates [125].  The multidisciplinary principles of this phenomenon combine features of heterogeneous catalysis and electrocatalysis as the working electrode is used to catalyze a chemical reaction and an electrochemical reaction simultaneously. Two types of processes occur at electrodes: Faradaic and non-Faradaic [126]. In Faradaic processes, the electrons transferred across the metal-solution interface cause an oxidation or reduction reaction to occur at a rate proportional to the current. Under unfavourable thermodynamic and kinetic conditions, charge transfer reactions may not occur. However, non-Faradaic processes like adsorption or desorption may occur.  The structure of the electrode-solution interface may then be  altered by changing the current or potential, or the solution composition [126]. In the electrical promotion of catalysis, most of the yield is obtained through the heterogeneous catalytic reaction, and is nonFaradaic.  The superimposed electrochemical reaction used to control and improve the catalytic  properties of the electrode surface result in the Faradaic part of the yield [125].  34  1.3.4.1 Theory  The non-Faradaic electrochemical modification of catalytic activity (NEMCA), also called the electrochemical promotion of catalysis (EPOC), is ascribed to the modification of the catalytic activity of metals on ionic conducting substrates. It is obtained upon variation of the potential on the working electrode, which is used as an electrode and as a catalyst for the heterogeneous catalytic reaction under study [127]. The application of a current (± 5-50 mA/cm2) or voltage (± 1-2 V) between the catalyst working electrode WE and a counter electrode CE, also deposited on the solid electrolyte, polarizes the catalyst and changes its work function. This property correlates the electrocatalytic and catalytic reaction rates with the properties of the catalyst involved. The catalyst work function is directly proportional to the catalyst electrode overpotential, which is the extent of polarization at the metal catalyst [128]. The work function, Φ (eV), is the minimum energy required to remove an electron from the interior of a solid to a position just outside. It is defined as:  − Φ = μ e − eΨ  (1.28)  where μe is the electrochemical potential of an electron at infinite separation from a metal, i.e., at the Fermi level (KJ), e is the charge of an electron (C), and Ψ is the outer potential (V), which represents the work required to bring a charge from infinity to a point just outside a charged phase and is equal to zero if there is no net charge on the metal surface. It was determine that the changes in the catalyst surface work function closely follow the changes in potential difference under galvanostatic or potentiostatic transients for a large number of systems [129]. The change in the work function of the catalyst surface,  ΔΦW (eV), can be controlled in situ by an applied potential, UWR (V) as per the following relationship:  eΔU WR = ΔΦ W  (1.29)  where ΔUWR (V) is the overpotential between the working electrode catalyst and a reference electrode [130]. The catalyst potential is represented by UWR and is defined as the catalyst WE potential with  35  respect to a reference electrode (RE) while UWC represents the cell voltage and is defined as the catalyst-working electrode potential with respect to a counter CE.  Vayenas et al. qualitatively explained this effect by the scheme represented in Fig. 1.4 for a metalelectrode deposited on an oxide-conducting and proton-conducting solid electrolyte. For a system where the solid electrolyte mobile ions are H+, the reaction at the electrode surface would be:  H+ + M + e- → Mδ- - Hδ+  (1.30)  The migration of ions to or from the ion conducting solid electrolyte from or to the catalyst surface is called backspillover, in analogy with the term spillover, which refers to migration from a metal to a support [129]. This backspillover of ions forms neutral spillover dipoles which establish an effective chemical double layer over the entire catalyst surface which is exposed to the reactants. Hence, EPOC is catalysis in the presence of an electrochemically controllable double layer at the catalyst/reactant interface [18]. The charged double layers are accumulations of charges on a surface, which are either formed due to diffusion effects, reactions between electrons in the electrodes and ions in the electrolyte, or results from applied currents or voltages. The classical metal-solid electrolyte double layer and the effective double layer resulting from the current or potential controlled ion migration are indicated in Fig. 1.4 (a) and (b) for an oxide conductor and for a proton conductor, respectively.  All EPOC studies to date have been carried out using gaseous reactants. The electrocatalytic reaction takes place at the catalyst-solid electrolyte-gas three phase boundaries (TPB) while the catalytic reaction with no charge transfer takes place at the metal-gas interface [128]. Here, the term catalytic refers to a chemical reaction and the term electrocatalytic refers to an electrochemical reaction which includes charge transfer.  The electrochemically-induced protonic current supplies promoting H+ ions to the  catalytically-active metal-gas interface but do not directly affect the catalytic reaction rate [131]. It is the effective chemical double layer which interacts electrochemically with covalent chemisorbed reactants and reaction intermediates by lowering their energy and modifying their binding strength.  36  O  Reactants (ex. H2O, CH3OH)  Metal-Reactant Interface (catalytic reaction)  Adsorbed Species C (ex. CO) Working catalyst electrode  Effective Double Layer  δ-  O  Spillover ions  +  +  δ  δ  O  δδ-  δ-  O  +  +  δ  δ  O  W +  δ  O  G/P  δ-  Metal (ex. Pt)  δ-  O  +  _ ++ 2O  Double Triple Phase Layer Boundary (TPB) (electrocatalytic reaction)  ++ 2O  δ  ++ 2O  UWR  Oxide conducting solid electrolyte (ex. YSZ)  C A  (a)  R  1-2 V  UWC  c O  Reactants (ex. H2O, CH3OH) Adsorbed Species (ex. CO) Working catalyst electrode  H  Spillover ions  Metal-Reactant Interface (catalytic reaction) Effective Double Layer  C  δ+  -  -  δ H  δ  H  δ+  δ+ -  -  δ  δ  Double Layer  H  δ+  W H  G/P  δ+  Metal (ex. Pt) -  δ  _ H  H  δ  _  _ +  -  +  H  H  δ+  Triple Phase Boundary (TPB) (electrocatalytic reaction)  UWR  +  Proton conducting solid electrolyte (ex. Nafion)  C  (b)  A  1-2 V  R  UWC  Figure 1.4: Schematic diagram of a metal catalyst-electrode deposited on (a) an oxide-conducting solid electrolyte, and (b) a proton-conducting solid electrolyte (Modified from Vayenas et al. [18] with permission from Springer).  37  A change in the strength of the chemisorptive bonds can be induced upon polarization of the metal-solid electrolyte interface. The corresponding decrease in overpotential can enhance the catalytic reaction rate at the gas-metal interface in a very pronounced and reversible manner [18].  This induced  exponential catalytic rate non-Faradaic enhancement is analogous to the Buttler-Erdey-Gruz-Volmer (BEV)  equation  for  high  electrode  overpotentials,  highlighting  the  resemblances  between  electrochemistry and heterogeneous catalysis [127].  Some of the rules of EPOC are listed in Appendix D. Supplying H+ with a proton conductor (positive voltage application) is equivalent to removing O2- with an oxide conductor (negative voltage application) as they both decrease the catalyst work function. An electron donor is defined as a compound that gives electrons during its oxidation, while an electron acceptor is defined a compound which accepts electrons during its reduction. As per the first rule of EPOC, adding electropositive promoters, such as H+, weaken the chemisorptive bond of electron donor (negatively charged) adsorbates and strengthen the chemisorptive bond of electron acceptor (positively charged) adsorbates. This first principle has been experimentally confirmed and rationalized on an electrostatic and quantum mechanic basis as per Eq. 1.26. The binding energy or enthalpy of adsorption, ∆Hj, is related linearly to the change in work function,  ΔΦ W , as follows:  Δ ΔH j ≈ α H J ⋅ ΔΦ W  (1.31)  where αH,j is a parameter which is positive for electropositive (electron donor) adsorbates, and negative for electronegative (electron acceptor) adsorbates [18]. Although not a general fundamental equation, Vayenas et al. are not aware of any exceptions to the physical meaning which Eq. 1.26 conveys, i.e., that ∆Hj, and hence, the coverage of an electron acceptor/donor adsorbate decreases/increases with increasing work function, and thus, decreasing Fermi level, EF. More details on this can be found in Appendix D. It is known that CH3OH will be an electron donor (O bonding) on surfaces with high work functions and will be an electron acceptor (CO bonding) on surfaces with low work functions [18]. It was also established that the strength of the CO metal bond was reduced by the addition of electronegative modifiers (e.g., O2-) as it decreases the activation energy of CO desorption [18]. 38  Varying the potential and work function of the catalyst affects the heterogeneous reaction catalytic rate and the activation energy of the reaction. There is a linear variation in the activation energy, EA (kJ/mol):  E A = E Ao + α H ⋅ ΔΦ W  (1.32)  where αH is the enthalpic parameter, a constant, which is usually negative for electrophobic reactions (donor reactions), and  E Ao is the open circuit activation energy value (eV). Hence, the activation energy  of an electron acceptor (positively charged) decreases linearly with decreasing work function [18]. Reactions for which the activation energy decreases with increases in the catalyst work function are termed electrophobic, and when the opposite occurs, the reactions are termed electrophilic.  1.3.4.2 Parameters  The magnitude of EPOC is described by some important parameters: the rate enhancement factor, the rate enhancement ratio, and the promotion index. Normally, the electrochemical reaction rate, re (mol/s), is given by Faraday’s law:  re =  i ⋅ AGeom I = , nF nF  (1.33)  where i is the applied current density (A/cm2), A is the active geometric surface area (cm2), I is the applied current (A), n is the number of electrons transferred, and F is the Faraday constant (C/mol). Following the established EPOC terminology conventions, in the non-Faradaic enhancement case, this equation becomes:  Δr = Λ ⋅  I , nF  (1.34)  39  where ∆r is the non-Faradaic electrochemically induced change in the catalytic reaction rate (mol/s) [132], and is equal to the difference between the promoted catalytic reaction rate, r and the unpromoted catalytic reaction rate, ro. Here, n represents the number of electrons promoting the electrochemical reaction at the triple phase boundary. Hence, it is equal to 2 in the case of an oxide conductor and equal to 1 in the case of a proton conductor, as per the following electrochemical reactions:  For an oxide conductor:  ½ O2 + 2e- → O2-  (1.35)  For a protonic conductor:  ½ H2 + 1e- → H+  (1.36)  Rearranging equation 1.29, and knowing that for a proton conductor n = 1, the dimensionless rate enhancement factor, also referred to as the Faradaic efficiency, Λ, is defined as:  Λ=  nFΔr , I  (1.37)  For catalytic oxidation reactions, Λ > 1 implies an electrophobic (donor reaction) behaviour for which the rate increases with increasing catalyst work function, while Λ < -1 implies an electrophilic (acceptor reaction) behaviour for which the rate increases with decreasing catalyst work function. For a pure electrocatalyst, Λ is always unity [18].  There are four types of global catalytic reaction r vs. Φ behaviours over the entire experimentally accessible Φ range: purely electrophobic (∂r/∂Φ>0), purely electrophilic (∂r/∂Φ<0), volcano type (∂r/ ∂Φ>0 followed by ∂r/∂Φ<0), and inverted volcano-type (∂r/∂Φ<0 followed by ∂r/∂Φ>0).  All purely  electrophobic reactions are positive order in electron donor and zero or negative order in electron acceptor. All purely electrophilic reactions are positive order in electron acceptor and zero or negative order in electron donor. Volcano-type reactions are always in positive order for one reactant and purely negative order in the other. Inverted volcano-type reactions are positive in both reactants [18]. The EPOC global behaviour rules are listed in Appendix D.  40  A reaction is electrochemically promoted when ΙΛΙ > 1 and it is electrocatalyzed when ΙΛΙ ≤ 1. The exchange current of the metal-solid electrolyte interface, Io (A), can be extracted from Tafel plots and used to estimate the expected magnitude of the absolute value of the rate enhancement factor as per:  Λ ≈  nFr0 I0  (1.38)  where ro is the open circuit unpromoted catalytic reaction rate (mol/s) [133]. This expression defines the applicability limit of EPOC: when the absolute value of the rate enhancement factor well exceeds unity, EP is observed and the kinetic efficiency is enhanced. For a reaction to be electrochemically-promoted, the open circuit catalytic rate, ro, must be greater than Io/nF and the catalytic reaction must be faster than the electrocatalytic one [127]. The fact that Io increases exponentially with temperature in conjunction with the fact that Λ is inversely proportional to Io explains why most EPOC studies are restricted to lower temperatures [134].  The dimensionless rate enhancement ratio, ρ, is defined as the ratio of the  promoted to unpromoted catalytic rate:  ρ=  r , r0  (1.39)  where r is the electrochemically promoted catalytic reaction rate (mol/s).  1.3.4.3 Electrochemical promotion of methanol oxidation  The EPOC effect has been demonstrated for a variety of catalytic reactions, including the CH3OH oxidation on oxide-conducting ceramics. Table 1.3 summarizes the rate enhancement factors and rate enhancement ratios reported for CH3OH oxidation EPOC studies. All studies listed in Table 1.3 were conducted using gaseous reactants. The most pronounced rate enhancements were typically observed at the lowest temperature examined [128]. It is important to remark that most EPOC catalytic oxidation reaction studies were conducted via oxygen pumping using solid oxide ion conductors.  41  Table 1.3: Vapour phase methanol oxidation EPOC studies with oxide conductors. Reactant Electron Electron Donor Acceptor  Product  Electrolyte  Catalyst  T [oC]  Λ  ρ  Global Behaviour (GB)*  Ref.  CH3OH  O2  H2CO, CO2  YSZ  Pt  250320  100  -  PE-  [136]  CH3OH  O2  H2CO, CO2  YSZ  Ag  500  -95.5  2  PE+  [137]  CH3OH  O2  H2CO, CO2  YSZ  Pt  250500  -  2.5  -  [132]  CH3OH  O2  H2CO, CO, CH4  YSZ  Ag  550750  -25  6  PE+  [138]  CH3OH  O2  H2CO, CO2  YSZ  Pt  300500  1x104  4, 15  IV  [18]  CH3OH  O2  H2CO, CO2  YSZ  Pt  400500  -10  3  PE+  [18]  *GB: global behaviour, PE-: purely electrophobic, PE+: purely electrophilic, IV: inverted volcano, V: volcano. Note: Although EPOC was mostly studied on oxidation reactions, it has been applied to many hydrogenation and dehydrogenation reactions.  1.3.4.4 Electrochemical promotion using protonic electrolytes  The use of the electrochemical promotion of heterogeneous catalyst surfaces has been investigated on various reactions with proton-conducting electrolytes.  The EPOC effect was observed at high  temperatures using proton-conducting ceramics, and also at low temperatures using Nafion® as a protonic polymer electrolyte.  Table 1.4 summarizes the findings of EPOC studies using protonic  conductors.  All studies in Table 1.4 were conducted using gaseous reactants.  No EPOC studies on the  electrochemical reforming of CH3OH using proton-conducting electrolytes have been reported in the literature. By comparing the rate enhancement factors and rate enhancement ratios reported in Table 1.3 and Table 1.4, it can be generally concluded that, to date, it seems that there are less reactions having a high rate enhancement factor with proton conductors than with solid oxide ion conductors. Table 1.5 lists the few EPOC studies which have been carried out in alkaline aqueous media.  42  Table 1.4: Vapour/gaseous phase EPOC studies using protonic conducting electrolytes. Reactant Electron Electron Donor Acceptor O2 C2H4  Product  Electrolyte  Catalyst  T [oC]  Λ  ρ  GB*  Ref.  CO2  Ba3Ca1.18 Nb1.82O9-a  Pt  250 350  -1340  4  PE+  [139]  H2  N2  NH3  CaIn0.1Zr0. 9O3-a-a  Fe  440  6  ∞  PE+  [13, 140]  C2H4  O2  CO2  CaIn0.1Zr0. 9O3-a  Pt  385 470  -3x104  5  PE+  [18]  N2, H2  CaIn0.1Zr0. 9O3-a  Fe  530 600  150  3.6  PE-  [18]  NH3  H2  N2  NH3  SrCe0.95Yb 0.05O3-a (SCY)  Pd  550 750  ~ 1-2  -  -  [141]  CH4  -  C2H6, C2H4  SCY  Ag  750  -  8  PE-  [18, 128]  H2  CO2  H2O, CO  SrZr0.90Y0.1 0O3-a  Cu  550 750  <1  -  -  [142]  C2H4  O2  CO2  Gd-doped BaPrO3 (BPG) and Y-doped BaZrO3 (BZY)  Pt  400 600  ~1  1.3  PE+  [143]  H2  C2H4  C2H6  CsHSO4  Ni  150 170  300  2  PE+  [18, 144]  C2H4  O2  CO2  Nafion® 117  Pd/C  -  6  1230  -  [145]  H2  O2  H2O  Nafion® 117  Pt  25  20  6  V  [18, 146]  C4H10, Pd 70 -28 40 PE+ [18, 147, Nafion® 2-C4H8 117 148] *GB: global behaviour, PE-: purely electrophobic, PE+: purely electrophilic, IV: inverted volcano, V: volcano. 1-C4H8  -  43  Table 1.5: EPOC studies in aqueous systems. Reactant Electron Electron Donor Acceptor  Product  Electrolyte  Catalyst  T [oC]  Λ  ρ  GB*  Ref.  H2  O2  H2O  0.1 M KOH  Pt  25-50  20  6  PE-  [18]  H2  O2  H2O  0.01-0.5 M KOH / LiOH  Pt  -  7.2  -  PE-  [135]  *GB: global behaviour, PE-: purely electrophobic, PE+: purely electrophilic, IV: inverted volcano, V: volcano. It is important to note that, for H2 oxidation using Pt as a catalyst in aqueous media, non-Faradaic rate changes were only obtained in alkaline aqueous solutions. Only Faradaic changes were observed in acidic solutions as the catalytic rates were much higher than those measured in the alkaline aqueous solutions [135]. This was attributed to the larger electrocatalytic activity measured in acidic aqueous solutions. In aqueous media, the concentration of electrolyte affects the magnitude of EPOC [135]. It was also reported that the low temperatures of aqueous electrochemistry may have limited the number of reactions where non-Faradaic enhancements may have been obtained, as the open circuit catalytic activity, ro, must be measurable to obtain an EPOC effect [132].  No EPOC studies on the  electrochemical reforming of liquid CH3OH in acidic media have been reported in the literature.  1.3.4.5 Applications for the electrochemical promotion of catalysis  Currently, there is a strong industrial interest in developing commercial EP applications with most efforts focused on lab-scale EPOC research. The acceptance and understanding of EPOC is still limited [125]. The use of EPOC in commercial applications depends on technical and economical aspects: material cost minimization, ease of electrical connection, efficient and compact reactor design [149]. Although applied research efforts are needed to develop industrial EPOC applications, fundamental research is still required to improve commercialization possibilities.  For example, low temperature EPOC with  aqueous electrolytes has a tremendous potential for many applications but has not been thoroughly studied in the literature.  Suitable chemical processes for industrial EPOC applications are slow  processes, which can be activated through the use of EPOC, and processes requiring low investments, 44  as it would reduce the market entry risks for the first commercial EPOC application. EPOC has been recognized as a mean to facilitate the commercialization of fuel cells [125, 149]. H2 production and utilization was identified as a potential niche market where EP could find an open field opportunity for industrialization [149].  A compact and flexible EPOC reactor design will minimize heat losses, maximize efficiencies, and facilitate the practical utilization of EPOC in industrial settings. To reduce application implementation costs, an efficient EPOC reactor design based on established technologies which have simple electrical connections, and use inexpensive materials would need to be developed.  Durability, lifetime, and  scalability are other aspects which will need to be addressed to develop a practical EPOC reactor [149]. To date, most EPOC reactor designs have been based on heterogeneous catalysis reactors and fuel cells. Simplified EPOC reactors may be obtained through single chamber reactors and bipolar systems. Scalable liquid electrolyte EPOC reactors designs may be obtained based on scrubber technologies [127].  Most EPOC studies have been conducted in electrochemical membrane reactors (EMR) using solid electrolytes [150]. Two main types of catalytic-electrocatalytic reactors have been used: double-chamber reactors and single-chamber reactors. In the double-chamber reactor, the catalyst WE is exposed to the catalytic reactants and products, while the CE and RE are exposed to a reference gas in a separate compartment. This design possesses an accurate RE but is difficult to scale-up. The EPOC effect has also been studied in continuous flow single-chamber reactors containing three electrodes (WE, CE and RE), which are all exposed to the reactants and products in order to promote the rate of catalytic reactions. The CE and RE are both made of catalytically-inert materials so that no catalytic reaction occurs on their surface. The single-chamber reactor requires the use of a quasi-reference electrode, which may cause inaccuracies in measuring the catalyst potential, but is easier to scale-up.  It was shown that direct electrical contact to the catalyst electrode is not necessary to induce EP in a bipolar monolithic reactor [151]. EPOC can be induced without external potential application by using the potential difference developed between the catalyst WE and a catalytically-inert CE. This single pellet wireless configuration was used to demonstrate the effect of EPOC on the oxidation of CH3OH at 45  250oC with an oxide ion-conducting electrolyte [132]. A change in the CH3OH consumption rate was obtained by using the RE as a CE, and short-circuiting it with the WE. A potential difference developed between the CE and the WE, and resulted in a current flow between two electrodes [136]. The wireless EPOC for the oxidation of CH3OH in a mixed-reactant single-chamber alkaline fuel cell has been patented [152, 153]. The design of EPOC reactors based on this wireless EPOC concept may lead to simpler designs.  Yiokari et al. evaluated the effect of EPOC on the synthesis of NH3 via the Haber-Bosh process, using proton conducting ceramic catalyst pellets [154]. This study was the first to use an industrial catalyst, to be carried out at high pressure (50 atm), and it was the first attempt at an EPOC scale-up. St-Pierre et al. employed EPOC in an electrolytic cell to purify a reformate stream H2 feeding a PEMFC [155]. EPOC can efficiently be carried out on thin sputtered metal films which possess stability and endurance. Balomenou et al. conducted development work on a high temperature membrane electrochemically promoted reactor (MEPR), which is a hybrid between a monolithic honeycomb catalytic reactor and a planar flat or ribbed plate SOFC for the EPOC of the catalytic oxidation of hydrocarbons [156, 157]. In this scalable reactor design, two external electrical connections are needed to dynamically control the applied current/potential required to induce EPOC to the catalyst film plates and only one gas stream containing all reactants and products was used.  The benefits of EPOC were also applied to the treatment of automotive exhaust in catalytic converters to reduce CO emissions [158-161]. EPOC studies were extended to non-redox systems and Salazar et al. electrochemically enhanced the isomerization reaction of olefin by over a thousand fold in a PEMFC using Pd/C catalyst at the cathode [162]. Sapountzi et al. have shown that EPOC could efficiently purify the H2 stream from the CO formed during the H2 production from hydrocarbons or alcohols, by electrochemically enhancing the water-gas-shift reaction or by improving the selectivity towards CO oxidation [163]. EPOC was used in the development of a catalytic system for the after-treatment of diesel exhaust. The unit combustion, tested on a commercial diesel engine, was found to be satisfactory for soot but not for NOx, and its Faradaic efficiency was estimated to be 66 [149]. More recently, a method using EPOC in a tubular reactor at temperatures from 150 to 600oC was developed to improve the reaction kinetics of biofuel production from biomass [164]. 46  1.3.5 Electrochemical promotion of electrocatalysis  Recently, a new engineering solution analogous to EPOC was developed to enhance electrochemical reactions instead of chemical ones.  Hence we refer to it as the electrochemical promotion of  electrocatalysis (EPOE). This approach, involving a third electrode, is referred to in the literature as triode operation. Balomenou et al. have shown that it can enhance the power output of fuel cells [19, 165]. A dual-chamber triode fuel cell arrangement was designed to study the application of EP in fuel cells using gaseous reactants. It comprised three electrodes: the fuel cell anode WE, the fuel cell cathode CE, and an auxiliary CE. The fuel cell anode WE was also the auxiliary WE. The electrodes were in electrolytic contact and form two electrical circuits as shown on Fig. 1.5. The fuel cell circuit consisted of the fuel cell anode WE and the fuel cell cathode CE. The auxiliary circuit consisted of the auxiliary WE, which was also the anode WE of the fuel cell circuit, and the auxiliary CE.  Polymer Electrolyte Membrane Fuel Cell Ring Common WE CEfc WEaux = WEfc Common Current Collector  Fuel Cell Current Collector Auxiliary CEaux  Iaux Uaux G/P  Ifc Ufc Fuel Cell Circuit Auxiliary Circuit  Figure 1.5: Schematic diagram of a triode fuel cell electrical circuit.  The net Faradaic fuel-consuming current, Ifar, is the difference between the fuel cell current, Ifc, and the auxiliary circuit current, Iaux [14]. When the auxiliary circuit is not used, the fuel cell operates normally. When a current is supplied in the auxiliary circuit, the fuel cell anode and cathode are forced to operate under a controlled potential [14, 166]. It was reported that the triode improvement resulted from using 47  the auxiliary circuit potential to keep the fuel cell anode (or cathode) at a corrosion-type potential inaccessible during normal fuel cell operation.  At the new WE operating potential resulting from triode operation, the surface coverage of the promoting species and the adsorption of adsorbates is modified. In some cases, the reaction at the WE surface is enhanced in triode operation as it is no longer limited by the supply and adsorption of the fuel. Triode operation is advantageous under high anodic and cathodic overpotentials [14]. The application of EPOE resulted in gains similar to that obtained in classical EPOC experiments. The only results available in the literature pertain to triode operation of an SOFC and a PEMFC using gaseous reactants. Balamenou et al. obtained power enhancement ratios up 8, i.e., efficiencies over 700% for a SOFC operating on dry H2, ethane or methane at 400-750oC in triode operation [14]. In this case, the low anodic potential created in triode operation decreased the surface coverage of O2-, increased the adsorption of H2, CO, or CH4, and enhanced the rate of the anodic oxidation. In SOFC’s, the solid electrolyte ohmic resistance contribution typically exceeds that of the electrodes, and as the use of the auxiliary circuit diminishes the ohmic losses between the fuel cell anode and cathode, triode operation can enhance the fuel cell performance [14]. Triode operation was also shown to improve the efficiency of a PEMFC running on pure H2 or H2 poisoned with CO, using an anode feed of humidified H2/CO/He or H2/CO2/CO/N2 at 30oC [165]. Using 4 mg/cm2 of a Pt-Ru/C anode catalyst and 4 mg/cm2 of a Pt black cathode catalyst, the application of a current of 0.09-0.5 mA to the auxiliary circuit enhanced the power enhancement ratio of a CO-poisoned PEMFC by a factor of 5 [165]. The enhancement was attributed to the electrolytic H+ supplied by the auxiliary circuit, which decreased the coverage of adsorbed CO and enhanced the membrane conductivity due to additional H+ pumping. Vayenas et al. [166] developed the triode gas phase fuel cell and battery concept as it is an advantageous alternative method to conduct electrochemical reactions in non-liquid systems. It appears that no EPOE studies have been carried out using liquid phase reactants and that triode operation has not been evaluated in the electrolysis mode to date. In this thesis, the EPOE effect was evaluated in triode and tetrode operation using liquid phase reactants and in the electrolysis mode for the first time.  48  Chapter 2: Electrocatalysis Baseline Study  2. 1 Synopsis  As discussed in Section 1.3.3, all published work relevant to CH3OH electrolysis or DMR was carried out in a recycle operation mode or in a flowing mode, in which an un-acidified CH3OH aqueous solution is pumped to the anode, while the H2 produced at the cathode was continuously purged with Ar. In this Chapter, the electrochemical production of H2 from the PEM electrolysis of liquid CH3OH in acidic aqueous media was investigated in the static mode (non-flowing, stirred). Experiments were carried out in a two-compartment glass cell with a MEA composed of a Nafion® 117 membrane and gas diffusion electrodes (GDE). This glass cell configuration allowed for the separate measurements of the anodic and cathodic potential contributions to the overall cell voltage. Most tests were conducted with 0.5 M H2SO4 in the anode and cathode compartment, but the effect of having a dry N2 purge cathode was also investigated. Methanol electrolysis was studied at concentrations ranging from 0 to 16 M, where 0 M corresponds to water electrolysis. The characteristics of the CH3OH - H2O electrolysis were described and compared to that of the H2O electrolysis under the same acidic conditions.  The influence of  supported (carbon) and unsupported (black) conventional fuel cell catalysts (Pt and Pt-Ru), operating temperatures (23, 50 and 75oC) and operating modes (dry and wet cathode) on the electrocatalytic reforming of CH3OH and water were evaluated. A theoretical thermodynamic analysis of the system was conducted and the limiting current densities, kinetic parameters, including the Tafel slopes and current exchange density, and apparent activation energies were determined.  Additional electrochemical  sample calculations can be found in Appendix H. The work presented in this section was published and is reprinted from the International Journal of H2 Energy, 35, C. R. Cloutier, and D. P. Wilkinson, “Electrolytic Production of Hydrogen from Aqueous Acidic Methanol Solutions”, 3967-3984, Copyright (2010), with permission from Elsevier.  49  2.2 Experimental  2.2.1 Materials  Certified electronic grade CH3OH, and American Chemical Society (ACS) certified plus H2SO4, both from Fisher Scientific, were used with deionised (DI) H2O having a resistivity of 18 MΩcm to make up the solutions. The H2O associated with H2SO4 was taken into account when making the CH3OH solutions. Behmann et al. [167] stated that the addition of CH3OH to H2SO4 and H2O mixtures containing less than 18 M H2SO4 had no influence on the dissociation behaviour of H2SO4 in the ternary mixture, and had dissociation behaviour is similar to that of the H2SO4 and H2O binary system. Hence the dissociation behaviour of H2SO4 will be the same for all the aqueous acidic CH3OH solution concentration used in this study. Ultra high-purity H2 and N2 gases from Praxair were used in some tests. Potassium dichromate salt (K2Cr2O4) reagent plus ≥ 99.5 % from Sigma-Aldrich was used for the spectrophotometric studies with a Shimadzu UV Vis spectrophotometer (UV mini 1240). A supporting electrolyte concentration of 0.5 M H2SO4 was used in the anode and cathode cell compartments in most tests, unless other wise indicated. A volume of 80 ml of solution was used in the anode and cathode cell compartments in most tests, unless otherwise indicated.  Nafion® 117 (Ion Power Inc.), was selected as the solid polymer proton exchange membrane. Conditioning was conducted by first boiling the membrane in a solution of 3 wt% H2O2 (certified, Fisher Scientific), for 30 minutes, then boiling it in DI H2O for 30 minutes and finally, boiling it in 0.5 M H2SO4 for 30 minutes. It was rinsed with DI between each boiling step and then cut to shape and stored in DI H2O until used. TorayTM carbon paper, TGPH-060, with 20 wt% Teflon® (PTFE) wet-proofing from BASF Fuel Cell Inc., was used as the gas diffusion layer (GDL). A micro-porous layer (MPL) consisting of a 1 mg/cm2 coating composed of carbon black and 20 wt% PTFE was sprayed on top of the cathode GDL surface, to form a double-layer gas diffusion layer. No MPL was used on the anode side.  Conventional CH3OH fuel cell noble metal electrocatalysts were employed: 20 wt% high performance (HP) Pt supported on Vulcan XC-72 and 20 wt% HP Pt:Ru alloy (1:1 atomic ration (a/o)) supported on  50  Vulcan XC-72, both from E-TEK, Pt black (99.9+%, fuel cell grade) and Ru black (99.9%) both from Sigma Aldrich, and Pt:Ru alloy (1:1 a/o) from Alfa Aesar. The anode Pt-Ru catalyst loading was 4 mg/cm2, while the cathode Pt loading was only 2 mg/cm2, as the cathode half-cell reaction is simple and straightforward compared to the anodic half-cell reaction.  The catalyst inks were composed of the  supported catalyst powder, 30 wt% of a proton conductive ionomer in solution (Nafion®, 5 wt% perffluorosulfonic acid-PTFE copolymer, Alfa Aesar), H2O and isopropanol (IPA, 2-propanol certified ACS plus from Fisher Scientific). Catalyst inks were prepared and sprayed with an Accuspray® air gun (Model 07HS, ISAAC Series HVLP Spray), on one side of the GDL, forming a gas diffusion electrode (GDE). More details of the ink preparation and spraying procedure are given in Appendix E. A planar Pt disk (0.25 mm, 99.99%+ Pt foil, Goodfellow Cambridge Limited) of a 2.01 cm2 area was used as an anode catalyst in some experiments for comparison purposes. The current collectors were made from an annealed 0.25 mm thick Nb foil, 99.9% pure from Sigma-Aldrich.  2.2.2 Equipment  All experiments were conducted in a dual-chamber borosilicate glass (Pyrex) cell composed of two chambers as shown in Fig. 2.1 and 2.2, which was built by CanSci Glass Products Ltd. Various ports permitted the inclusion of reference electrodes, thermocouples, and gas spargers or bubblers in the anode or cathode compartments. Slight internal bevels were provided in the upper part of the anode and cathode compartments to facilitate the gas exit. The CO2 gas forming at the anode and H2 gas forming at the cathode were vented to atmosphere in all experiments.  The MEA holder contained different layers: the membrane electrode assembly, the current collectors, two silicon sealing rings and a PTFE spacer ring, as shown in Fig. 2.1 (b). The MEA holder was machined in polyetherimide (ULTEM 1000) by Core Tools Ltd. and is shown in Fig. 2.1 (c). Its central opening had a circular active geometric area of 2 cm2. The silicon rings were made in house using a procedure described in Appendix E. A 25 mm diameter membrane sample was sandwiched between the anodic and cathodic 16 mm diameter GDE samples. Two niobium current collectors (25 mm OD, 16 mm ID), having a thin current collecting metal cross through their circular openings were placed on each side of the MEA. Silicone seal rings (25 mm OD, 16 mm ID, 1 mm thickness) made in house were 51  placed on each side of the current collectors. Finally, a PTFE spacer ring (54273 AMG06, 25 mm OD, 16 mm ID, 1 mm thickness) was placed as the last layer on the cathode side to reduce friction when turning the cap to close the MEA holder.  Multistat and PC Mercury Sulphate Reference electrode  Temperature Indicator  Cathode Current Collector  Thermocouple  Anode Current Collector CO2 Outlet  H2 Outlet  Threaded Caps with Plugs  Heater / Circulator  Heater / Circulator  Anolyte Reservoir  Catholyte Reservoir  Magnetic Stirrer  (a)  Connection to/from Temperature Controlled Water Bath  O-rings  Magnetic Stirrer  MEA Holder Flow  Current Collectors Gas Diffusion Electrodes  Spacer  Seals (b)  Membrane  (c)  Figure 2.1: Schematic diagram of (a) the electrochemical glass cell, (b) the MEA components (c) the MEA holder. 52  The glass cell, MEA holder and O-rings were joined and held together using a screw clamping apparatus. It was designed to press the thick flat ends of the glass cell compartments together and ensured alignment of the glass compartments. This permitted the application of a uniformly distributed force to seal the glass cell flanges to the MEA holder O-rings. All experiments carried out at room temperature were conducted in ambient air (23oC ± 2oC). For experiments at 50 and 75oC ± 1oC, H2O was circulated within the double wall of the glass cell and the temperature was controlled by a Haake DC-30 immersion circulator. Thermocouples were used to monitor the temperature in each of the glass cell compartments.  Reference electrode  Multistat connections  MEA holder  Cathodic compartment  Anodic compartment Magnetic stirrers  Glass cell holder  (a)  Current Collectors  Gas Diffusion Electrode O-ring  (b)  Figure 2.2: Picture of (a) the electrochemical glass cell, (b) the MEA holder.  53  2.2.3 Electrochemical measurements  All electrochemical measurements were carried out at steady-state using a multistat (Solartron Analytical, Model 1470E) connected to a computer, using CoreWare software. A single junction mercury-mercurous sulphate (MSE, Hg/Hg2SO4, Radiometer Analytical) reference electrode located in the anode compartment was used for all experiments at ambient temperature. At 50 and 75oC, a double junction silver/silver chloride (SSE, Ag/AgCl, Radiometer Analytical) reference electrode was used, with 0.5 M H2SO4 in its outer compartment, as it is more stable than the MSE at higher temperatures. This is because the MSE has an outer porous plug tip while the SSE electrode outer tip is a porous glass frit, which limits the ion migration to increase with temperature. The reference electrode was placed in the anode compartment. All voltages are reported versus the standard hydrogen electrode (SHE). All electrolysis experiments were conducted in the galvanostatic, current-controlled mode. More information on the electrochemical technique used to obtain current/potential transients is given in Appendix F. AC impedance measurements were carried out with an impedance/gain phase analyzer (Solartron Analytical, Model 1260A) to obtain the MEA and solution resistance. Measurements were performed with an AC amplitude of 10 mV, in the frequency range of 10-3 to 107 Hz. More details on this electrochemical technique can be found in Appendix F. As the measured potential values were affected by the ohmic drop, they were corrected for the current resistance (IR) before kinetic parameter information was extracted. Using this technique, it was confirmed that the MEA resistance increased with increasing CH3OH concentration, MEA usage and electrode degradation due to carbon oxidation.  2.2.4 Characterization  Cyclic voltammetric techniques were used to determine the true active electrochemical surface area (ECSA) of the catalysts in a conventional three electrode single chamber electrochemical cell, using a platinized Pt counter electrode (CE) (see Appendix E for platinization procedure), an Ag/AgCl double junction reference electrode (RE), and various working electrodes (WE). The platinization procedure employed can be found in Appendix E. The ECSA of the various Pt catalyst working electrodes (WEs) was estimated from the H2 adsorption charge on the cyclic voltammograms [168], while that of Pt-Ru catalysts was estimated by the copper under potential deposition (UPD) method [169, 170]. More details 54  on the CV technique employed can be found in Appendix F. The average total Pt area (TPA) and ECSA values obtained by H2 adsorption or Cu UPD for the different catalysts are summarized in Table 2.1. The values reported represent the average of three tests conducted on three different catalyst samples of the same batch of GDE. The Brunauer, Emmett and Teller (BET) surface area (SA) from the catalyst powder certificate of analysis were provided for comparison purposes only. The BET estimated ECSA exceeded the measured ECSA in all cases. The experimentally determined values were later used in the estimation of kinetic parameters and in Fig. 2.9. For the pure Pt disk, a surface roughness factor of 4 was assumed, giving an ECSA of 8 cm2/g Pt. These techniques were also used to confirm that the catalyst electrochemical area decreases with increasing utilization and carbon oxidation of the catalyst support.  55  Table 2.1: Electrochemical characterization of supported and unsupported catalysts studied. Total metal area (TMA) by Different Methods BET SA*  Typical BET SA Range  Actual Loading  Geometric Area  [m2/g]  [m2/g]  [mg/cm2]  Pt-Ru/C  130  120-140  Pt-Ru/C  130  Pt/C  Electrochemical Surface Area (ECSA) by Different Methods Roughness H2 Cu UPD Factor of 4 adsorption [cm2/g Pt [cm2/g Pt] [cm2/g Pt] and Ru]  Roughness Factor of 4  H2 adsorption  Cu UPD  [cm2]  [cm2 Pt]  [cm2 Pt]  [cm2 Pt and Ru]  4.01  2.01  -  -  562  -  -  140  120-140  3.99  2.01  -  -  562  -  -  140  180  160-200  2.04  2.01  -  3929  -  -  1926  -  Pt/C  180  160-200  2.01  2.01  -  3746  -  -  1863  -  Pt-Ru black  86.65  80-90  3.98  2.01  -  -  69  -  -  17  Pt black  29  25 to 34  2.06  2.01  -  248  -  -  120  -  Pt disk  -  -  -  2.01  8  -  -  8  -  -  Catalyst  * From catalyst powder vendor certificate of analysis.  56  2.3 Results and Discussion  2.3.1 Thermodynamic evaluation of methanol reforming at low temperatures  A theoretical thermodynamic analysis was carried out for a DMR in acidic media and a water PEM electrolyser as a function of temperature. The effect of pressure on the theoretical cell voltage and thermodynamic efficiency was found to be negligible compared to the effect of temperature, and hence is not discussed here. The thermodynamic information for the species involved, including the standard Gibbs free energy of formation (ΔGof), the standard enthalpy of formation (ΔHof), the standard entropy of formation (ΔSof), and the heat capacity (Cpo), are available in the literature [171, 172] and are listed in Appendix G.  These values were used to establish the theoretical cell voltage (Ee) and theoretical  thermodynamic efficiency (ηmax) as a function of temperature (25 to 150oC) assuming a constant pressure (1 atm). Examples of the thermodynamic calculations for the DMR electrochemical reactions in the liquid phase and in the gas phase are provided in Appendix G. The standard Gibbs free energy associated with the electrochemical reaction is given by:  ΔG o = ∑ i si ΔGio, f ( products − reactants)  (2.1)  The standard enthalpy of the electrochemical reaction can be calculated by:  ΔH o = ∑ i s i ΔH io, f  (2.2)  Similarly, assuming that the entropy is independent of pressure, the standard entropy of formation of the electrochemical reaction can be determined by:  ΔS o = ∑ i s i ΔS io, f  (2.3)  57  The standard reversible cell voltage can then be calculated by Eq. 1.6. Various Shomate equations of the form  N  C P = a o + ∑ a nT bn  (2.4)  n =1  based on empirical data, were used to evaluate the heat capacity of the various species as a function of temperature. Over the temperature range of 25 to 50oC, the average change in the Cp values was about 8 % for methanol in the liquid phase, and over the temperature range of 100 to 150oC, the change in the Cp value was about 2 % for water in the gas phase, respectively. As demonstrated in Appendix G, the heat capacities hence obtained were used to evaluate the molar enthalpy and entropy of formation at temperature T, assuming a constant pressure, are given by:  ΔH T = ΔH o + ∫ T298 C p dT  ΔS T = ΔS o + ∫ T298  (2.5)  1 C p dT T  (2.6)  For CH3OH electrolysis, over the temperature range of 25 to 50oC, the temperature correction was about 1 % for the ΔST value in the liquid phase and over the temperature range of 76.7 to 150oC, and the temperature correction was about 2 % for the ΔST value in the gas phase. The reversible cell voltage approximated as a function of temperature by:  ⎛ ΔS Ee = E o + (T − 298) ⋅ ⎜⎜ T ⎝ nF  ⎞ ⎟⎟ ⎠  (2.7)  And the corresponding Gibbs free energy of formation was calculated:  ΔGT = − nFE e  (2.8)  58  Finally, the maximum thermodynamic efficiency, ηmax, can be calculated, at a temperature T and 1 atm, as per:  ⎛ ΔG T ⎜ ΔH T ⎝  η max = ⎜  ⎞ ⎟ ⋅ 100 ⎟ ⎠  (2.9)  These theoretical calculations assume that there is no gas forming before the fuel boiling point is reached and that there is no liquid present after the fuel boiling point is attained. The boiling point of the CH3OH-H2O mixture will change depending on the concentration.  For the DMR thermodynamic  calculations an equimolar mixture of H2O and CH3OH, which has a boiling point of 76.7oC, was assumed. The effect of the electrolyte is also ignored. The CO2 released at the anode of the DMR was assumed to be in the gas phase at all temperatures.  It is important to note that the sign of the thermodynamic efficiency for the electrolysis systems can be positive or negative, depending on the temperature at which the system is evaluated.  For both  electrolysis systems, the enthalpy of formation was positive for the temperature range studied, indicating that both overall reactions are endothermic. The Gibbs free energy of formation for the water electrolysis system is positive (> 0) under the entire temperature range studied, indicating that the electrolysis of water is non-spontaneous. Hence, the thermodynamic efficiency of water electrolysis was positive at all temperature studied. Although, the methanol electrolysis Gibbs free energy of formation was positive at 25oC, indicating that the reaction is non-spontaneous at this temperature, the Gibbs free energy of formation was negative (< 0) at higher temperatures, indicating that the reaction became spontaneous. Under these conditions, the thermodynamic efficiency was a negative value.  As shown in Fig. 2.3, the PEM H2O electrolyser resulted in a positive theoretical thermodynamic efficiency over the entire temperature range studied, while that of the DMR was positive at 25oC and became negative at higher temperatures. The theoretical thermodynamic efficiencies determined for the PEM H2O electrolyser indicate that work is required on the system in order to generate H2 under these conditions. The theoretical thermodynamic efficiencies determined for the DMR indicate that the system  59  requires work input to generate H2 at 25oC, but that work is generated by the system at higher temperatures.  As shown in Fig. 2.3, for both systems, the theoretical cell voltage became less negative as temperature increased. In the case of the DMR, the theoretical cell voltage even became positive as temperature increased. It was determined that the DMR Gibbs free energy of formation and theoretical cell voltage of methanol electrolysis become positive at a temperature of about 41oC.  Hence, according to  thermodynamics at temperatures greater than 50oC, the DMR should be a spontaneous source of electricity, as well as H2. However, favourable reaction thermodynamics do not imply that the reaction is kinetically favourable. The Gibbs free energy of formation and theoretical cell voltage of the PEM H2O electrolyser, on the other hand, do not become positive over the temperature range investigated, i.e., up to 150oC.  The thermodynamic analysis demonstrates which electrochemical reactions can occur  spontaneously at specific conditions and equilibrium, but does not provide any kinetic information about the electrochemical reactions. The actual amount of electrical energy which will be required to generate H2 is equal to the Gibbs free energy of the electrochemical reaction plus the losses in the system.  60  0.2  100 80  Power Source Power Sink  0  40  Thermodynamic efficiency, Methanol Electrolyser Thermodynamic efficiency, water electrolyser  20  Theoretical cell voltage, methanol electrolyser Theoretical cell voltage, water electrolyser  0  Liquid  -40  Boiling Point of Pure Methanol (64.7oC)  -60  -0.8  Boiling Point of Pure Water (100oC) Liquid  -1  Gas  -1.2  -80 -100 0  20  40  60  -0.4  -0.6  Gas  Boiling Point of Equimolar Methanol/Water Mixture (76oC)  -20  -0.2 Theoretical Cell Voltage (V)  Thermodynamic Efficiency (%)  60  80  100  120  140  -1.4 160  o  Temperature ( C)  Figure 2.3: Theoretical thermodynamic efficiency and theoretical cell voltage as a function of temperature (1 atm) for a DMR and PEM water electrolyser.  61  2.3.2 Electrolysis Polarizations  2.3.2.1 Cell voltage stability  Typical responses obtained for different electrolysis current densities are shown in Fig. 2.4. The open circuit voltage (OCV) for the reaction was measured in an H-cell as described in Appendix F. At low current densities, steady-state was rapidly reached and the CH3OH electrolysis voltage stabilized. At high current densities, the CH3OH oxidation current alone was insufficient to sustain the high current density. The electrolysis voltage dropped off until water oxidation and carbon oxidation started to occur.  0.0 OCV  -0.5  Cell Voltage (V)  -1.0 0.02 mA/cm2  -1.5  5 mA/cm2  200 mA/cm2  -2.0 -2.5 -3.0 -3.5 -4.0 -50  0  50  100  150  200  250  300  350  400  450  500  Time (s)  Figure 2.4: Electrolysis cell voltage stability with respect to current density (16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC).  Water electrolysis starts to occur at voltages below -1.23 V vs. SHE, and carbon oxidation, e.g., the anode oxidation catalyst support starts to oxidize at a cell voltage of about -1.8 V vs. SHE. Hence, a limiting current density is reached for the electro-oxidation of CH3OH beyond which no stable cell voltage measurements could be obtained. Although the experiments were done in the static mode, this effect has also been observed for CH3OH electrolysis in the active mode [118]. Voltage measurements were  62  recorded at current densities where steady-state was reached, i.e., when the change in cell voltage was less than 20 mV over five minutes. The region of cell voltage stability varied as a function of current density, CH3OH concentration, temperature and the type of anode electrocatalyst. In the limiting current density region, steady-state could not be reached.  Figure 2.5 shows an example of the electrolysis test repeatability in the different regions. In the stable region, the electrolysis polarization curve is very reproducible.  The experiments conducted in the  forward direction, from small current densities to large current densities, in the backward direction, from large current densities to small current densities, and in a random order of current densities all resulted in very similar voltages at a particular current density. When the test was started backward from a point in the unstable region, the results were still reproducible, but the error for the first data point in the stable region was slightly larger. When the test was started at a current density where the safety limit of -4 V was reached, irreversible damage was observed as the catalyst was oxidized and precious metal was lost from its surface. This was also verified by conducting cyclic voltammetric experiments on fresh, used, and oxidized supported catalyst coupons.  0.0  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 Forward Backward Random Backward, from unstable region Backward, from unstable region Backward, after C oxidation  -2.5 -3.0 -3.5 -4.0 0  5  10  15  20  25  30  35  40  45  50  2  Geometric Current Density (mA/cm )  Figure 2.5: IR corrected electrolysis cell voltage repeatability as a function of geometric current density in the stable and unstable regions (2 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 2 mg/cm2 Pt/C anode and cathode, 23 ±2oC).  63  The data points shown in the figures represent the arithmetic average of values obtained for a minimum of three different samples. The calculated standard deviation, Sd, representing the upper and lower limits, was typically in the range of 10 mV but closer to 200 mV in the limiting current region. In the limiting current density region, voltage values were unstable and were empirically obtained by linear interpolation between the last stable data points and the safety limit, which was set at -4 V.  2.3.2.2 Effect of methanol concentration and anode catalyst  In order to investigate the effect of CH3OH concentration on the polarization curve, the anode compartment CH3OH concentration was varied from 0 to 16 M (i.e., 0, 1, 2, 6, 16 M), with 0 M representing H2O electrolysis. Figure 2.6 shows the dependence of the cell voltage corrected for the ohmic losses on the current density for different CH3OH concentrations and represent typical electrolysis polarization curves. They show open-circuit voltages (OCV) less than the equilibrium cell voltage likely due to fuel cross-over; activation control at lower current densities resulting in a sharp initial decrease in cell voltage as the current density increases; a subsequent linear slow voltage decrease with current density due to ohmic control followed by a rapid decrease in cell voltage due to mass transport limitations. The limiting current densities, iL, are reached in the transport control region when the fuel is used at a rate equal to its maximum transport supply rate. At the limiting current density, voltages where other competing reactions can occur, such as H2O electrolysis and/or carbon oxidation, were reached. As can be seen in Fig. 2.6, the iL value varied with CH3OH concentration. At low current densities (<12 mA/cm2), the CH3OH concentration did not impact the cell voltage significantly, while at higher current densities, lower overpotentials were obtained at higher CH3OH concentrations. As expected, for CH3OH oxidation the required electrolytic voltages were significantly less than for H2O electrolysis at the same current density.  In contrast with active CH3OH electrolysis studies where the anode reactants are  continuously flowing, the static system electrolytic polarizations are more sensitive to CH3OH concentration. Hu et al. [7] reported that the electrolysis performance was not dependent on the CH3OH concentration at concentrations greater than 2 M at 60oC, but that the concentration polarization was obvious at high current densities for concentrations lower than 1 M. In another study, by Shen et al. [11], the voltage losses for CH3OH electrolysis were reported to be almost the same in the 2-4 M range, while the voltage losses were greater at concentrations lower or higher than this range.  At low CH3OH  64  concentrations, high current densities could not be sustained.  At high CH3OH concentrations, less  voltage was required to hold the same current densities and higher current densities could be sustained. Methanol permeation from the anode compartment to the cathode compartment will increase with increasing CH3OH concentrations and will decrease fuel efficiency. However, in the system under study, CH3OH present in the cathode compartment can not be oxidized at the reducing cathode potentials observed during electrolysis. In their study, Hu et al. attributed the poor electrolysis performance at high CH3OH concentration to the creation of a mixed potential at the cathode, which in turn reduced its electrocatalytic activity.  Other phenomena such as adsorption and catalyst site blockage were not  considered but are possible. As the concentration of CH3OH increases, the solution conductivity will decrease, and the mass transfer limitations will decrease, which will affect rate of CH3OH electrolysis.  As the data presented in Fig. 2.6 is for catalysts having different ECSAs, it is not possible to make a direct comparison of the performance of the three-dimensional catalysts studied for the electrolysis of CH3OH or H2O on a geometric current density basis. Nevertheless, Fig. 2.6 (e) clearly demonstrates that the well-defined low electrochemical surface area non-porous Pt disk did not perform as well as the dispersed three-dimensional electrode catalysts. Furthermore, it can be seen that the Pt disk electrode, for which no carbon is present, was able to sustain greater current densities for H2O electrolysis than for CH3OH oxidation. Interestingly, for the 1 M CH3OH case with the well-defined Pt disk at low current densities, two different potential values were obtained for the same current density. In this system, proton conductivity of the MEA was provided by the electrolyte in the cathodic compartment. The bistability was observed in multiple tests which were performed separately at the same conditions. This phenomenon was also observed for a PEM fuel cell running at 37oC using H2 partial pressure less than 1.5 kPa, and 4 mg/cm2 Pt-Ru black anode and a 4 mg/cm2 Pt black cathode [173]. It was attributed to the total resistance of Nafion®, which is split in an ohmic component, due to proton migration in the membrane’s aqueous phase, and a non-ohmic one, due to proton tunnelling in the membrane.  65  0.0  0.0  IR Corrected Cell Voltage (V)  -1.0 -1.5 -2.0 -2.5 -3.0 -3.5  -1.0 -1.5 -2.0 -2.5 -3.0 -3.5  -4.0  -4.0  0  (a)  10  20  30  40  50  60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Geometric Current Density (mA/cm 2)  10  20  30  40  50  60  70 80 90 100 110 120 130 140 150 160 170 180 190 200 Geometric Current Density (mA/cm 2)  0.0  0 M CH3OH 2 M CH3OH 16 M CH3OH  -1.0 -1.5 -2.0 -2.5 -3.0  0 M CH3OH 2 M CH3OH 16 M CH3OH  -0.5 IR Corrected Cell Voltage (V)  IR Corrected Cell Voltage (V)  0  (c)  0.0 -0.5  -3.5  -1.0 -1.5 -2.0 -2.5 -3.0 -3.5  -4.0  -4.0 0  (b)  0 M CH3OH 2 M CH3OH 16 M CH3OH  -0.5 IR Corrected Cell Voltage (V)  0 M CH3OH 1 M CH3OH 2 M CH3OH 6 M CH3OH 16 M CH3OH  -0.5  10  20  30  40  50  60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Geometric Current Density (mA/cm 2)  (d)  0  10  20  30  40  50  60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Geometric Current Density (mA/cm 2)  0.0 0 M CH3OH 1 M CH3OH (low) 1 M CH3OH (high) 2 M CH3OH 6 M CH3OH 16 M CH3OH  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0  (e)  0  10  20  30  40  50  60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Geometric Current Density (mA/cm 2)  Figure 2.6: Effect of methanol concentration on the IR corrected electrolysis cell voltage as a function of geometric current density for different anode catalysts (a) 4 mg/cm2 Pt-Ru/C, (b) 2 mg/cm2 Pt/C, (c) 4 mg/cm2 Pt-Ru black, (d) 2 mg/cm2 Pt black, (e) well-defined Pt disk (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC). 66  Figure 2.7 shows that the iL values varied with the CH3OH concentration and the type of anode catalyst used.  Limiting Geometric Current Density 2 (mA/cm )  160 4 mg/cm 2 Pt-Ru/C 2 mg/cm 2 Pt/C 4 mg/cm 2 Pt-Ru black 2 mg/cm 2 Pt black Well-defined Horizontal Pt Disk  140 120 100 80 60 40 20 0 0  2  4  6  8  10  12  14  16  18  Methanol Concentration (M)  Figure 2.7: Dependence of limiting geometric current densities for the oxidation of methanol and water on the CH3OH concentration for various anode catalysts (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC).  For CH3OH oxidation, the largest limiting current densities were reached with the Pt-Ru/C catalyst. For H2O electrolysis, the largest limiting current density was reached with the well-defined Pt disk. However, the iL values obtained for this catalyst are lower in the case of CH3OH oxidation than for H2O oxidation. Although this well-defined Pt catalyst is not affected by carbon oxidation thus allowing higher current densities to be achieved, its low electrochemical surface area was strongly affected by CO adsorption in the case of CH3OH oxidation. The iL values for the Pt/C catalyst were lower than the one for the unsupported catalyst at low concentrations, while the iL values were similar to the unsupported catalyst ones at high concentration. The iL values for the 4 mg/cm2 Pt-Ru black catalyst and for the 2 mg/cm2 Pt black catalyst were within the same range. It would be expected that the supported catalysts, which possess higher active areas, would result in higher limiting current densities. Yet, this was not observed for Pt/C and Pt black. Take et al. [118] determined limiting current densities for an active CH3OH electrolysis electrolytic cell using 5.4 cm diameter Pt catalyst samples (real surface area of 23 cm2) at the anode and cathode, of 130 mA/cm2 for 1 M CH3OH, 170 mA/cm2 for 2 M CH3OH, 210 mA/cm2 for 6 M  67  CH3OH, and 260 mA/cm2 for 16 M CH3OH.  These reported values are greater than the limiting  geometric current densities shown in Fig. 2.7 for the same CH3OH concentrations using the 2 mg/cm2 supported or unsupported Pt catalyst.  In the passive mode, the transport of the fuel to the electrode is  slower and its access to the electrode is more restrained by the CO2 production. This resulted in lower limiting current densities compared to the active mode.  Figure 2.8 shows the individual IR corrected anode potential and cathode potential, respectively, obtained for stable voltage values before the limiting current density region. The change in cathode potential is insignificant compared to the change in anode potential. As the cathode potential remained more or less the same, it can be deduced that the increase in cell voltage required is mainly due to the anode. It can also be seen that higher CH3OH concentrations reduce the anode overpotential.  Anode and Cathode Potential (V vs. SHE)  1.6  1.2  0 M CH3 OH 0.8  1 M CH3 OH 2 M CH3 OH  Anode Potential  6 M CH3 OH 0.4  Cathode Potential  16 M CH3OH  0.0  -0.4 0  25  50  75  100  125  150  2  Geometric Current Density (mA/cm )  Figure 2.8: Effect of CH3OH concentration on the individual electrode potential (0, 1, 2, 6 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC).  As shown in Fig. 2.8, the electrolysis of H2O in 0.5 M H2SO4 commences at the anode at a potential of about 1.3 V vs. SHE (Eq. 1.21). At anode potentials equal to and greater than this value for the CH3OH oxidation reaction, the current efficiency is not 100 %, as there is a current loss due to the H2 evolution  68  reaction. Beyond this point, the H2O electrolysis current density should be subtracted from the CH3OH electro-oxidation current density. At higher cell voltages, the current density is also increasingly sustained by carbon corrosion. According to thermodynamics, carbon can be oxidized on the anode at a standard potential of -0.21 V vs. SHE:  C + 2H2O → CO2 + 4H+ + 4e-  Eoa = -0.21 V vs. SHE  (2.7)  C + H2O → CO + 2H+ + 2e-  Eoa = -0.52 V vs. SHE  (2.8)  However, the carbon oxidation is kinetically-inhibited and it does not usually begin until a potential of about 1.8 V vs. SHE is reached at the anode [174]. Beyond this potential, the cell current density is increasingly sustained by carbon corrosion. This was confirmed by the observation of a decrease in cell voltage after re-utilization of catalysts which were subjected to anode potentials greater than this value. The degradation of the catalyst carbon support and the consequent loss in electrochemical surface area was also confirmed by cyclic voltammetry (refer to Appendix F for more details). Furthermore, Ru can dissolve at anodic potentials greater than the standard potential of -0.25 V vs. SHE, as per the following electrochemical reactions:  Ru → Ru2+ + 2e-  Eoa = -0.46 V vs. SHE  (2.9)  Ru2+ → Ru3+ + e-  Eoa = -0.25 V vs. SHE  (2.10)  Hence, as the absolute current density increases and the anode potential become more positive, a decrease in the CH3OH oxidation performance due to Ru loss can also be anticipated. However, this is likely kinetically limited, similar to the case of carbon oxidation.  The impact of Ru dissolution and  crossover on the performance of a PEMFC and on the functionality of the anode was studied in the literature [175].  69  Figure 2.9 compares the performance of Pt/C and Pt black catalysts on a real electrochemical current density basis for different CH3OH concentrations. Both catalysts had the same catalyst loading of 2 mg Pt/cm2 and a Nafion® content of 30 wt%, even though the optimized amount of ionomer required for the unsupported catalyst is likely different than for the supported catalyst. 0.0  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 -2.5 Pt/C, 0 M CH3OH Pt/C, 2 M CH3OH Pt/C, 16 M CH3OH Pt Black, 0 M CH3OH Pt Black, 2 M CH3OH Pt Black, 16 M CH3OH  -3.0 -3.5 -4.0 0.00  0.10  0.20  0.30  0.40  0.50  0.60  0.70  0.80  0.90  2  Real Current Density (mA/cm )  Figure 2.9: Effect of the IR corrected voltage on electrolysis for Pt/C and Pt black anode catalysts as a function of real electrochemical current density (0, 2, or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 23 ±2oC).  The Pt black catalyst had a much smaller ECSA than the Pt/C catalyst. It is possible that the nonoptimized Nafion® content partly contributed to a decrease in the surface area of the unsupported catalyst. However, the required applied voltage to oxidize H2O or CH3OH was less for the Pt black catalyst than for the Pt/C catalyst at the same current density. Thus, the Nafion® loading used did not restrain the access to the active surface sites of the black catalyst and did not render it less active than the Pt/C catalyst. Moreover, the Pt black catalyst was able to sustain higher real current densities than the Pt/C catalyst at the same CH3OH concentration. It appears that carbon oxidation started to occur at lower real current densities for the Pt/C electrode than for the Pt black electrode. This suggests that the carbon of the supported catalyst oxidizes before the carbon of the carbon fibre paper on which the both catalysts were sprayed start to oxidize.  70  2.3.2.3 Effect of dry purged cathode  Experiments were also conducted with a dry cathode, which was purged with N2 gas, to confirm the feasibility of this possible design simplification. Figure 2.10 compares the polarization curves for a DMR running with the cathode compartment filled with 0.5 M H2SO4 and for one with a dry cathode compartment purged with N2. The cell voltage became more negative for the DMR operating with a dry purged cathode compartment compared to the DMR with a cathode compartment containing 0.5 M H2SO4.  IR Corrected Cell Voltage (V)  0.0 0 M CH3OH, wet cathode 0 M CH3OH, dry cathode 1 M CH3OH, wet cathode 1 M CH3OH, dry cathode  -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 0  10  20  30  40  50  60  70  80  2  Geometric Current Density (mA/cm )  Figure 2.10: Effect of dry N2 purged cathode on the IR corrected electrolysis cell voltage as a function of geometric current density (0 or 1 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, wet electrolyser catholyte: 0.5 M H2SO4, dry electrolyser catholyte: N2 purge electrolyser, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC).  In the case of CH3OH electrolysis, slightly more voltage is required to drive the electrochemical reaction when the cathode does not contain any electrolytic solution. In both cases, the required voltage for the dry and wet cases were within 10 % at high current densities and similar limiting current densities were attained in the dry and wet cathode case for 0 and 1 M CH3OH. However, the AC impedance (Appendix F) results indicated that the ohmic resistance across the MEA can be up to about 55 % higher in the dry cathode case compared to the wet cathode one, i.e., 0.22 Ω versus 0.1 Ω for 1 M CH3OH. This indicates  71  humidification of the membrane by diffusion of H2O from the liquid anode side to the dry cathode side in this case might be insufficient to maintain the membrane conductivity.  Of course, membrane  humidification could be improved by decreasing the membrane thickness and pressurizing the anode side, for example. Fig. 2.11 shows the corresponding anode and cathode potentials respectively, for the wet and dry cathode cases.  The cathode potential remained largely unaffected by the wet or dry  cathode operation compared to the anode potential. No tests were carried out using humidified N2.  Anode and Cathode Potential (V vs. SHE)  2.5 0 M CH3OH, wet cathode 0 M CH3OH, dry cathode 1 M CH3OH, wet cathode 1 M CH3OH, dry cathode  2.0 1.5 1.0 0.5  (a) Anode Potential  (b) Cathode Potential  0.0 -0.5 -1.0 0  5  10  15  20  25  30  35  40  45  50  2  Geometric Current Density (mA/cm )  Figure 2.11: Effect of dry N2 purged cathode and a wet cathode on the (a) anode potential and (b) cathode potential, as a function of geometric current density (0 or 1 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, wet electrolyser catholyte: 0.5 M H2SO4, dry electrolyser catholyte: N2 purge, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC).  Fig. 2.11 shows that for CH3OH electrolysis the anode potential was significantly larger with the dry purged cathode than with the wet cathode. The CH3OH cross-over and MEA resistance are likely greater in the dry N2 case. It is not clear why the anode potential is larger in the dry cathode case. At high enough current densities in both modes, the anode potential is high enough that H2O electrolysis will also occur in parallel. Using a dried N2 purged cathode compartment seems to represent a possible design simplification for the liquid CH3OH electrochemical reformer.  72  2.3.2.4 Effect of temperature  The effect of temperature was studied for CH3OH and H2O electrolysis using the 4 mg/cm2 Pt-Ru/C catalyst. As can be seen in Fig. 2.12, increasing the operating temperature significantly improved the PtRu/C electrode kinetics of the CH3OH electrolysis. There was a large improvement in the electrolyser performance for all catalyst types with temperature. For example, for the 16 M CH3OH solution and a current density of 150 mA/cm2, the cell voltage was -1.55 V, -1.26 V and -0.87 V corresponding to temperatures of 23, 50 and 75oC, respectively. Hence, there was a reduction in the electrolyser cell voltage of 31 % between 50 and 75oC.  The reduced electrolytic overpotentials resulted from an  improvement in the mass transport and in the electrode kinetics at the higher temperature. Compared to the limiting current density at 23oC, at 50oC, the limiting current density was about 25 mA/cm2 higher in the case of the 2 M CH3OH solution, and 50 mA/cm2 higher in the case of the 16 M CH3OH solution. However, the increase in limiting current densities between 50 and 75oC was not significant, and a similar limiting current was obtained at both temperatures in the case of water and CH3OH electrolysis. This is not in agreement with what has been previously observed for active CH3OH electrolysis systems [7, 11]. For example, Shen et al. [11] using a 2.5 mg/cm2 Pt-Ru/C anode and a 1.5 mg/cm2 Pt/C cathode in a 2 M CH3OH active system shown that over the temperature range of 50 to 80oC, the electrolysis cell voltage at 80oC was reduced to half of the electrolysis cell voltage at 50oC. There is no indication in this paper that measures were taken to prevent CH3OH loss to the vapour phase. However, as a closed-loop CH3OH-H2O anode feeding system, it can be assumed that any CH3OH evaporating would be condensing back in the CH3OH-H2O storage tank, and that the concentration of CH3OH fed to the anode is constant. In our static system set-up, the anolyte is not constantly replenished and any CH3OH evaporating is loss. This may explain why the electrolyser cell voltage did not decrease as significantly as for active systems between 50 and 75oC in our experiments.  As the 16 M CH3OH solution has a boiling point of 76.7oC, and our study uses a non-flowing system, a partial loss of solution to the vapour phase was possible. Using a K2Cr2O7 reduction technique combined with spectophometric measurements, which is described in Appendix E, it was verified that, the CH3OH concentration change for a 16 M CH3OH at 75oC was less than 2 M over the period of time for which polarization measurements were taken. This small concentration change might partly explain why the  73  limiting current density was similar at 50 and 75oC in the case of the 16 M CH3OH solution. Another possible explanation may be that there is a greater crossover rate of CH3OH at 16 M CH3OH, which may affect the kinetics of the H2 production at the cathode. However, in the case of H2O electrolysis, while minor improvements in the anode kinetics were observed with increasing temperature, the limiting current also remained the same over the temperature range of 23 to 75oC studied. Therefore, CH3OH crossover can not explain why similar limiting current densities were obtained for the electrolysis of 2 and 16 M CH3OH at 50 and 75oC. Nevertheless, it may be preferable to operate the CH3OH electrolyser at higher temperatures as less current would be wasted for secondary reactions and performance is better under these conditions.  74  0.0 0 M CH3OH 2 M CH3OH 16 M CH3OH  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0  (a)  0  10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 Geometric Current Density (mA/cm 2)  0.0 0 M CH3OH 2 M CH3OH 16 M CH3OH  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0  (b)  0  10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 Geometric Current Density (mA/cm 2)  0.0  IR Corrected Cell Voltage (V)  -0.5 -1.0 -1.5 -2.0 -2.5 -3.0  0 M CH3OH 2 M CH3OH 16 M CH3OH  -3.5 -4.0  (c)  0  10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 Geometric Current Density (mA/cm 2)  Figure 2.12: Effect of CH3OH concentration on the IR corrected electrolysis cell voltage as a function of geometric current density for different temperatures (a) 23 ±2oC (b) 50 ±1oC (c) 75 ±1oC (0, 2 or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt-Ru/C anode, 2 mg/cm2 Pt/C cathode).  75  2.3.2.5 Anode Tafel kinetics  In the case CH3OH and H2O electrolysis, the anodic reaction kinetics is slow, resulting in a high overpotential. The overvoltage at the cathode is negligible compared to the overvoltage at the anode. The Buttler-Erdey-Grutz-Volmer (BEV) equation can be simplified to the Tafel equation as per a wellknown treatment described elsewhere [126, 176] and in more details in Appendix F. The anodic form of the Tafel equation was used to make Tafel plots (log(i) vs. ηa) from which the Tafel slope and the Tafel constant were extracted to obtain kinetic information on the CH3OH and H2O electrolysis. Fig. 2.13 shows an example of the IR corrected anode overpotential vs. log of the geometric current density for different CH3OH concentrations, demonstrating the linearity of the Tafel lines obtained.  5.2 4.8  0 M CH3OH  IR Corrected Anodic Overpotential (V vs SHE)  4.4  2 M CH3OH  4.0  16 M CH3OH  3.6 3.2 2.8  Limiting current  2.4  Limiting current  2.0 1.6  Limiting current  1.2 0.8 0.4 0.0 -2.0  -1.8  -1.6  -1.4  -1.2  -1.0  -0.8  -0.6  -0.4  -0.2  0.0  0.2  0.4  2  0.6  0.8  1.0  1.2  1.4  1.6  1.8  2.0  2.2  2.4  Log(i) (mA/cm )  Figure 2.13: Tafel plot of the IR corrected anodic overpotential as a function of the log of the geometric current density (0, 2, or 16 M CH3OH in 0.5 M H2SO4 electrolyser anolyte, 0.5 M H2SO4 electrolyser catholyte, 4 mg/cm2 Pt black anode, 2 mg/cm2 Pt/C cathode, 23 ±2oC).  The slope of the anodic Tafel curves for the CH3OH electrolysis changed at about 0.6 V for the Pt and Pt-Ru catalysts, indicating two distinct kinetic control regions. Wu et al. [177] have reported that these two regions represent two different reaction mechanisms. In the first region, where ηa < ~0.6 V vs. SHE, the splitting of the first C-H bond with the first electron charge transfer reaction is the rate-determining  76  step, while in the second region, where ηa > ~0.6 V vs. SHE, the reaction between the adsorbed CO on Pt sites is the rate-determining step [178]. For H2O electrolysis on the other hand, only one apparent slope was obtained over the entire potential range for the different catalysts studied. Hence, it is likely that there is only one reaction mechanism over the entire region. However, Kinoshita [179] reported that two distinct Tafel slopes were obtained for most metal oxide based catalysts studied in 1 N H3PO4 at 25oC. They reported a low Tafel slope value at low current densities and a high Tafel slope value at high current densities.  Tafel parameters reported in the literature for Pt and Pt-Ru catalysts for CH3OH and H2O electrolysis, along with the values obtained in this study, are summarized in Table 2.2 and 2.3, respectively. For the case of the 4 mg/cm2 Pt-Ru anode, the exchange current density values, io, were approximated by using twice the ECSA determined for the 2 mg/cm2 Pt catalyst, as well as the ECSA determined through Cu UPD, for comparison purposes. This is indicated by the superscript 1 and 3, respectively, as indicated in the legend provided at the bottom of Table 2.3.  77  Table 2.2: Literature Tafel kinetic parameters for the electro-oxidation of methanol and water on Pt and Pt-Ru catalysts. Experimental Conditions Electrode Method IR, P.  Catalyst Pt (111)  IR, P.  Pt (110)  IR, P.  Pt (100)  IR, P.  Pt (111)  IR, P.  Pt (110)  M, P.  Pt/C  P.  Pt black  Rs, P.  Pt black  Rs, P.  Pt/C  IR, G.  Pt-Ru/C  IR, G.  Pt-Ru/C  IR, G.  Pt-Ru/C  IR, G.  Pt-Ru/C  IR, G.  Pt-Ru/C  78  IR, G.  Pt-Ru/C  IR, G.  Pt-Ru/C  M, P.  Pt-Ru/C  Rs, P.  Pt-Ru black  Rs, P.  Pt-Ru/C  Description 0.2 cm OD spheres 0.2 cm OD spheres 0.2 cm OD spheres 0.2 cm OD spheres 0.2 cm OD spheres 2 14 ug/cm , ® Nafion film on GC. 5.6 and 4.2 2 mg/cm , on N315. 5.6 and 4.2 2 mg/cm , on N315. 40 wt%, on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 40 wt%, 1:1, 3.2 2 mg/cm , on GC. 2 14 ug/cm , ® Nafion film on GC. 5.6 and 4.2 2 mg/cm , on N315. 30 wt%, 1:1, on GC  [H2SO4]  [CH3OH]  T o  Anode Overpotential  Tafel Slope  Exchange Current Density  ηa [mV vs. SHE]  b [mV/dec]  io [A/cm ]  2  Ref.  [M]  [M]  [ C]  Zone I  Zone II  Zone I  Zone II  Zone I  Zone II  0.1  0.2  20  <422  -  119  -  1.00x10  -2  -  -  [180] [180]  0.1  0.2  20  <422  -  123  -  1.00x10  -2  0.1  0.2  20  <422  -  63  -  1.00x10  -2  -  0.1  0.2  20  <422  -  106  -  6.31x10  -6  -  [181]  0.1  0.2  20  <422  -  123  -  6.31x10  -6  -  [181]  0.5  0.5  60  <550  -  80  -  -  -  [182]  0.5  1  16-20  400-700  400-700  100-140  100-140  -  -  181]  0.5  1  16-20  <500  >600  60  120  4.00x10  0.5  0.1-2  30-80  <650  -  80-94  -  2.00x10  -8 -5  8.95x10  -8  [180]  1.55x10  -6  [183] [184]  -4  [177]  0.5  0.5  40-70  <550  >550  108-128  300-700  5.24x10  0.5  0.5  40  <550  >550  128  525  2.51x10  -4  7.41x10  -4  [177]  0.5  0.5  40  <550  >550  123  647  3.51x10  -5  8.19x10  -4  [177]  2.24x10  -4  [177]  2.27x10  -3  [177] [177] [177]  0.5  0.5  40  <550  >550  116  465  1.48x10  -6  0.5  0.5  70  <550  >550  114  650  1.64x10  -5  1.00x10  -3  3.16x10  -4  0.5  0.5  70  <550  >550  124  777  1.00x10  -5  0.5  0.5  70  <550  >550  108  537  1.93x10  -7  0.5  0.5  60  <550  -  195  -  -  0.5  1  16-20  <500  >600  70  163  2.88x10  0.5  0.1-2  30-80  <650  -  140-192  -  4.50x10  [182]  -7  -12  2.88x10 -  -6  [183] [184]  Experimental Conditions Electrode Method  Catalyst  Rs, IR, V.  Pt-Ru/C  V.  Pt-Ru/C  V.  Pt/C  V.  Pt/C  R.  Pt  F, P.  Pt/C  R  Pt (oxide free)  [H2SO4]  Description 40 wt%, 1:1, 1 2 mg/cm , on TGPH-60. 2 1.5 mg/cm , carbon cloth with GDL, N-117. 2 1.5 mg/cm , carbon cloth with GDL, N-117. 2 1.5 mg/cm , carbon cloth with GDL, N-117. 20 wt%, 50-100 2 ug/cm , on GC. -  [CH3OH]  T o  Anode Overpotential  Tafel Slope  Exchange Current Density  ηa [mV vs. SHE]  b [mV/dec]  io [A/cm ]  2  [M]  [M]  [ C]  Zone I  Zone II  Zone I  Zone II  Zone I  Zone II  1  1  20-80  <750  -  159-169  -  -  -  [185]  0.5  0.1, 1  20-45  -  -  -  -  -  [186]  0.5  0.1, 1  25  -  -  -  -  -  [186]  0.5  0.1, 1  45  -  -  90  -  -  -  [186]  0.5  1  Room T  -  -  125  -  -  -  [187]  0.5  0.1  100  >330  -  83  -  -  0.1 N  -  -  -  -  60 80  70  -  -  [188]  -11  -  [179]  -9  -  [179]  -10  -  [179]  -6  -  [179]  1.00x10  -12  -  [179]  7.60x10  -10  -  [179]  3.70x10  -10  -  [179]  6.40x10  -10  -  [179]  1.30x10  -11  -  [179]  1.80 x10  -8  -  [179]  5.10 x10  -9  -  [179]  -9  -  2.00x10  G  Pt  -  1N  -  25  -  -  65  -  1.00x10  G  Pt black  -  20 wt%  -  70  -  -  65  -  4.00x10  R -  Pt Ru Pt Pt (111) Pt (100) Pt  -  0.05 1N  -  1N 1N 1N 1N  25 25 20  -  20 20 80  -  -  145 80 107 120 114 90  -  Pt-Ru  50 atom % Ru  1N  -  80  -  -  120  -  Ru  -  1N  -  80  -  -  41  P P, IR  RuO2 RuOx  -  Ref.  1 1N  -  25  -  -  30-40 32  -  50-65 42  3.00x10  5-15 x10 7.00 x10  -13  1.00x10  [179] -11  [179]  Legend: all italic values are average values estimated from graphs, F = flow through, G = galvanostatic mode (constant current applied vs. time, monitor V vs. time), IR = IR corrected current density, M = mass specific current density, P = potentiostatic mode (constant potential applied vs. time, monitor I vs. time), R = Rotating disk electrode using the Levich method, RS = real surface area current density, V = voltammetry.  79  Table 2.3: Tafel kinetic parameters from current work for the electro-oxidation of methanol and water on Pt and Pt-Ru catalysts. Experimental Conditions  Anode Overpotential  Tafel Slope  Exchange Current Density  ηa [mV vs. SHE]  b [mV/dec]  io [A/cm ]  2  Electrode  [H2SO4]  [CH3OH]  T  [M]  [M]  [ C]  Zone I  Zone II  Zone I  Zone II  Zone I  Method  Catalyst  Description  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru black  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru black  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru black  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt black  2 mg/cm .  G, IR.  Pt black  2 mg/cm .  G, IR.  Pt black  2 mg/cm .  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt Disk  0.25 mm thick, 99.99% Pt.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  o  2  1  0.5  0  23  <600  >600  125.90  125.90  7.33x10  2  1  0.5  1  23  <600  >600  134.50  557.9  2  1  0.5  2  23  <600  >600  159.70  2  1  0.5  6  23  <600  >600  2  1  0.5  16  23  <600  2  1  0.5  0  23  2  1  0.5  2  2  1  0.5  2  3  2 2  Zone II  -16  7.33x10  -16  9.77x10  -8  1.27x10  -4  649.20  3.46x10  -7  1.50x10  -4  191.6  1407.8  2.89x10  -6  1.60x10  -3  >600  186.40  274.60  8.22 x10  -7  1.03x10  -5  <600  >600  176.60  176.60  1.94x10  -12  1.94x10  23  <600  >600  113.40  447.80  1.42x10  -8  7.53x10  -5  16  23  <600  >600  104.40  284.90  5.40x10  -9  2.19x10  -5  0.5  0  23  <600  >600  244.80  244.80  4.40x10  -9  4.40x10  -9  3  0.5  2  23  <600  >600  144.80  701.60  5.45x10  -6  1.16x10  -2  3  0.5  16  23  <600  >600  156.10  527.90  1.20x10  -5  7.13x10  -3  2 4  0.5  0  23  <600  >600  150.30  150.30  1.19x10  -12  1.19x10  2 4  0.5  2  23  <600  >600  147.20  640.60  3.10x10  -6  5.03x10  -3  2 4  0.5  16  23  <600  >600  142.00  522.30  2.11x10  -6  2.74x10  -3  -77  6.54x10  -7  7.75x10  -7  -15  6.60x10  -5  -12  -12  +  2  0.5  0  23  <600  >600  120.60  120.6  6.54x10  -27  +  2  0.5  1  23  <600  >600  400.30  400.30  7.75x10  +  2  0.5  1  23  <600  >600  49.30  408.70  1.19x10  +  2  0.5  2  23  <600  >600  143.90  325.20  1.29x10  -5  2.37x10  -3  +  2  0.5  6  23  <600  >600  145.50  393.90  3.70x10  -5  8.68x10  -3  +  2  0.5  16  23  <600  >600  155.30  533.20  1.24x10  -4  4.49x10  -2  2  1  0.5  0  50  <600  >600  359.60  359.60  5.83x10  -12  5.83 x10  -12  2  1  0.5  2  50  <600  >600  145.40  140.00  4.52x10  -10  3.08 x10  2  1  0.5  16  50  <600  >600  126.20  216.10  3.10x10  -10  3.34x10  -4  -8  80  Experimental Conditions  Anode Overpotential  Tafel Slope  Exchange Current Density  ηa [mV vs. SHE]  b [mV/dec]  io [A/cm ]  2  Electrode  [H2SO4]  [CH3OH]  T  [M]  [M]  [ C]  Zone I  Zone II  Zone I  Zone II  Zone I  Method  Catalyst  Description  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  G, IR.  Pt-Ru/C  50:50, 4 mg/cm , on TGPH-60, N-117.  o  2  1  0.5  0  75  <600  >600  325.50  325.50  1.43x10  2  1  0.5  2  75  <600  >600  206.10  285.90  2  1  0.5  16  75  <600  >600  215.10  2  3  0.5  0  23  <600  >600  2  3  0.5  2  23  <600  2  3  0.5  16  23  2  3  0.5  0  2  3  0.5  2  3  2  Zone II  -11  1.43x10  -11  2.98x10  -7  1.59x10  -6  222.30  2.93x10  -7  2.98x10  -7  -  -  3.01x10  -24  3.01x10  >600  -  -  5.77x10  -6  8.24x10  -4  <600  >600  -  -  1.33x10  -5  2.22x10  -2  50  <600  >600  -  -  7.90x10  -11  7.90x10  2  50  <600  >600  -  -  6.12x10  -9  4.17x10  -3  0.5  16  50  <600  >600  -  -  4.20x10  -9  4.53x10  -7  3  0.5  0  75  <600  >600  -  -  1.93x10  -10  1.93x10  2  3  0.5  2  75  <600  >600  -  -  4.03x10  -6  2.15x10  -5  2  3  0.5  16  75  <600  >600  -  -  3.98x10  -6  4.03x10  -6  -24  -11  -10  Legend: all italic values are average values estimated from graphs, F = flow through, G = galvanostatic mode (constant current applied vs. time, monitor V vs. time), IR = IR corrected current density, M = mass specific current density, P = potentiostatic mode (constant potential applied vs. time, monitor I vs. time), R = Rotating disk electrode using the Levich method, RS = real surface area current density, V = voltammetry. 1  2  2  3  Based on the electrochemical surface area of a 2 mg/cm Pt/C catalyst determined by H2 adsorption, Geometric surface area assuming a roughness factor of 4, Based on the 2 4 2 electrochemical surface area of a 4 mg/cm Pt-Ru black catalyst determined by Cu UPD, Based on the electrochemical surface area of a 2 mg/cm Pt black catalyst determined by H2 adsorption.  81  A proper comparison with the literature is difficult due to differences in the experimental conditions and whether geometric or real surface area was used. The type of electrolyte, the electrolyte and CH3OH concentration, as well as the catalyst loading used varied among the studies and most Tafel parameters available were mainly obtained using single-compartment cells without membranes. In general, the reported Tafel slope using Pt-Ru catalysts ranged from 70 to 128 mV/dec at low overpotentials, and ranged from 75 to 777 mV/dec at high overpotentials. This is comparable to the 4 mg/cm2 Pt-Ru/C catalyst Tafel slopes obtained at low CH3OH concentrations, but smaller than the ones obtained at high CH3OH concentrations in our work. Amongst different systems for which Tafel slopes were reported, the one used by Aramat et al. [183] seems to be the closest to our system. Tafel data was obtained using a two-compartment cell for unsupported catalyst coated with Nafion® 315 at 16-20oC in 1 M CH3OH and 0.5 M H2SO4. The Tafel slopes were in the range of 100-140 mV/dec for Pt black- Nafion® 315 in the overvoltage range of 0.4 to 0.7 V vs. SHE, 70 mV/dec for Pt-Ru black-N315 at ηa < 0.5 V vs. SHE and 163 mV/dec for the same catalyst at ηa > 0.6 V vs. SHE. The Tafel slopes obtained in our work using a 2 mg/cm2 Pt black GDE anode and N117 membrane, were in agreement with the Pt-N315 reported values for ηa < 0.6 V vs. SHE and CH3OH concentrations equal to or lower than 2 M. However, the Tafel slope values obtained in this study for CH3OH electrolysis with the Pt black catalyst at greater CH3OH concentrations and at ηa > 0.6 V vs. SHE are up to four times higher than the ones reported in the literature. This may be linked to a change in the rate determining step at higher potentials, as it is known that CO stripping from an unsupported Pt-Ru catalyst surface normally occurs at potentials ranging from about 0.5 to 0.7 V vs. SHE at 25oC [189, 190].  Tafel slopes were also evaluated as a function of temperature using the 4 mg/cm2 Pt-Ru/C anode at three different CH3OH concentrations. Zhang et al. [184] reported that at low temperatures, non-linearity is due to CO poisoning, while at high temperature, it is due to concentration polarization. They also noticed that Tafel slope data reported in the literature conflictingly increase or decrease with increasing temperature. This was attributed to different rate determining steps. In our study, the Tafel slopes values obtained for Pt-Ru/C were larger at ambient temperature than at 50oC, and the values obtained for the same catalyst at 75oC were the largest of all. Although it might be possible that the catalyst went through degradation at higher temperature, it is also possible that operation at 50oC is more desirable for  82  the Pt-Ru/C catalyst as it improves its activity for the electrolysis of CH3OH.  From an efficiency  standpoint, faster oxidation kinetics might allow the use of higher CH3OH concentrations. In general, reported Tafel slope values for Pt-Ru/C varied between 108 to 195 mV/dec in the temperature range of 60 to 80oC [177, 182, 184, 185], compared to experimental values obtained in this study which ranged from 126 to 215 mV/dec between 50 to 75oC . For ηa < 0.6 V, the Tafel slopes obtained at 75oC and 2 to 16 M CH3OH averaged 211 mV/dec. This seems to be in reasonable agreement with the 192 mV/dec reported on 30 wt% Pt-Ru/C on glassy carbon at 80oC and low CH3OH concentration in a single compartment cell [184].  The effect of CH3OH concentration on the kinetic parameters seems to be complex and no clear trend could be established for the change in b and io values as a function of increasing CH3OH concentration. However, it can be observed that at ambient conditions and low overpotentials, the lowest Tafel slopes were obtained using the Pt/C anode, while the greatest Tafel slopes were obtained using the Pt-Ru/C anode. It was suggested by Zhang et al. [184] that high b values on Pt-Ru/C are due to the initial dissociative dehydrogenation of CH3OH while low b values for Pt/C are due to the oxidation of CO rather than an adsorption process. Based on the real surface area of a 2 mg/cm2 catalyst determined by H2 adsorption, at 2 M CH3OH and for ηa < 0.6 V, the b value obtained in this study was about 30% larger for Pt-Ru/C than for Pt/C, and the io value was about 96 % smaller for Pt/C than for Pt-Ru/C. Under those conditions, the oxidation of CH3OH may occur faster on Pt-Ru/C than on Pt/C. For the same conditions, but at ηa > 0.6 V, the b value obtained was also about 30 % larger for Pt-Ru/C than for Pt/C. Similarly, the value of io was about still 96 % smaller for Pt/C than for Pt-Ru/C. The io value determined based on the ECSA of the 4 mg/cm2 Pt-Ru/C catalyst obtained by Cu UPD was even larger. Based on this, it is likely that Pt-Ru/C possesses a superior catalytic activity for the electrolysis of CH3OH than Pt/C. This difference was not observed between the Pt-Ru black and Pt black anodes at ambient conditions, which had similar Tafel slopes at all concentrations. For pure Pt, at CH3OH concentrations of 1 M or less, reported Tafel slopes ranged from 63 to 125 mV/dec below 0.6 V, and no Tafel data was reported for Pt above 0.6 V. As previously mentioned in Section 2.3.2.2, bi-stability was observed for the well-defined Pt case at 1 M CH3OH and two Tafel slopes were calculated. In the same potential range, it seems that  83  the Tafel value obtained for the more positive curve (49 mV/dec) is in much better agreement with the literature than the one obtained for the more negative curves (400 mV/dec).  The Tafel slope value of 121 mV/dec obtained for H2O electrolysis using the well-defined Pt disk is in agreement with the reported value range of 65 to 145 mV/dec at 20-25oC, however the corresponding io value is 100 % smaller than the ones reported in the literature. For H2O electrolysis, the Tafel slope was about 60 % larger for the Pt-Ru black catalyst than for the Pt black catalyst, which would indicate that Pt black is more favourable to H2O electrolysis than Pt-Ru black. However, the io value obtained was three orders of magnitude larger for Pt-Ru black than for Pt black.  It was the largest io value obtained  compared to all other catalysts tested for H2O electrolysis. Based on this, it seems like Pt-Ru black is a more favourable catalyst than Pt black for H2O electrolysis. The latter is in agreement with Miles et al. [191], who reported that Pt-Ru black was a better catalyst than Pt black for the O2 evolution reaction in 0.1 M H2SO4 at 80oC.  They attributed this to the formation of an oxide film which decreased the Pt  surface catalytic activity towards O2 evolution. For H2O electrolysis at ambient conditions, the Pt/C catalyst had a b 40 % greater than that for the Pt-Ru/C catalyst, and its io was 4 orders of magnitude smaller. This indicates that, under these conditions, Pt-Ru/C is more active for H2O electrolysis than Pt/C. The Tafel slopes for H2O electrolysis with the Pt-Ru/C at 50 and 75oC, were about 60 % higher than when using the same catalyst at ambient conditions.  The kinetic parameters reported for electrolysis are varied and often contradictory. It is important to note that at IR corrected anodic overpotentials greater than 0.6 V, mass transport effects are not negligible. Hence, the kinetic parameters extracted in this region do not likely represent true kinetic behaviour and have to be considered with caution as mass transport effects could explain some of the variability in reported results.  A low overpotential, which is desired to use energy efficiently and to operate as close  as possible to equilibrium, can be obtained by choosing a catalyst material and system resulting in a b value as small as possible and an io value as large as possible for the desired electrochemical reaction.  84  2.3.2.6 Activation energy  The activation energy was determined for CH3OH electrolysis using the Pt-Ru/C catalyst. It can be determined from the Arrhenius equation:  ln(io ) = ln(io0 ) −  EA RT  (2.11)  0  where io is the absolute exchange current density of the reaction, which represents the maximum io at infinite temperature, i.e., io (T →∞ ) . Using the exchange current densities based on the ECSA estimated from the intercept of the Tafel Plots, Arrhenius plots (log(io) vs. 1/T) were constructed for various CH3OH concentrations in the linear kinetic Tafel region (<0.6 V) and in the region just above 0.6 V, where the effect of mass transfer starts to appear. The apparent activation energy, EA, was determined from the slope (-EA/R) of the fitted linear trends. The values obtained, as well as some values reported in the literature, are included in Table 2.4. In agreement with the literature [184], over the concentration range studied, the larger the CH3OH concentration, the larger was the resulting Ea value. The Ea values obtained with the Pt-Ru/C anode in the kinetic region below 0.6 V are significantly lower than the smallest values reported in the literature, although the Ea values reported in the literature were obtained in single compartment cells and not using MEAs as in the present study. Zhang et al. [184] have reported that small Ea values (20-30 kJ/mol) obtained at low CH3OH concentrations (0.1-0.5 M), indicate that CO surface diffusion is the rate determining step. These researchers obtained larger Ea values (5184 kJ/mol), at higher CH3OH concentrations (1, 2 M) representing a change in the rate determining step. At Ea values of about 60 kJ/mol, the dissociative adsorption of CH3OH is the rate determining step, while at Ea values around 70 kJ/mol, CO oxidation is the rate determining step. In this study, the Ea values obtained in the region below 0.6 V vs. SHE for 2 and 16 M CH3OH are both below 30 kJ/mol. This might signify that, under those conditions, CO surface diffusion is the rate determining step. The Ea values obtained in the region above 0.6 V vs. SHE for 2 and 16 M CH3OH are 50 and 154 kJ/mol, respectively. Although there were no changes in the rate determining step for the electrolysis of CH3OH on Pt-Ru/C as  85  a function of CH3OH concentration in the low potential region, there might be one in the high potential region.  Table 2.4: Activation energy data for the electro-oxidation of methanol on Pt and Pt-Ru catalysts. Concentration  Temperature  Anode Overpotential  Activation Energy  ηa [V vs. SHE]  Ea [kJ/mol]  Anode  Anode  [H2SO4]  [CH3OH]  Range  Catalyst  Description  [M]  [M]  [ C]  Zone I  Zone II  Pt-Ru  50:50  0.5  1  17-50  0.50  -  Pt  -  0.5  1  17-50  0.50  -  -  0.5  0.6  0-65  0.25  -  0.5  0.6  0-65  -  0.5  0.6  -  0.5  -  Pt(100)Ru Pt(100)Ru Pt(100)Ru Pt(110)Ru Pt(110)Ru Pt(110)Ru Pt(111)/R u Pt(111)/R u Pt(111)/R u Pt(111)/R u Pt(111)/R u Pt(111)/R u  Deposited Ru -4 from 2x10 M Ru(NO)(NO3)3 in 0.5 M H2SO4 Deposited Ru -4 from 2x10 M Ru(NO)(NO3)3 in 0.5 M H2SO4 Deposited Ru -4 from 2x10 M Ru(NO)(NO3)3 in 0.5 M H2SO4 -5 5X10 M RuCl3 in 0.1 M HClO4 -5 5X10 M RuCl3 in 0.1 M HClO4 -5 5X10 M RuCl3 in 0.1 M HClO4  o  Zone I 60  Ref.  Zone II -  [186]  30  -  [192]  -  61 ± 11  -  [192]  0.31  -  53 ± 7  -  [192]  0-65  0.37  -  65 ± 4  -  [192]  0.6  0-65  0.25  -  38 ± 4  -  [192]  0.5  0.6  0-65  0.31  -  50 ± 5  -  [192]  0.5  0.6  0-65  0.37  -  47 ± 8  -  [192]  0.5  0.6  0-65  0.25  -  84 ± 10  -  [192]  0.5  0.6  0-65  0.31  -  66 ± 7  -  [192]  0.5  0.6  0-65  0.37  -  56 ± 5  -  [192]  0.5  0.6  0-65  0.25  -  70 ± 7  -  [192]  0.5  0.6  0-65  0.31  -  56 ± 5  -  [192]  0.6  0-65  0.37  -  54 ± 4  -  [192]  30-80  -  -  54-84  -  [184]  30-80  -  -  23-78  -  [184]  0.5  Pt-Ru/C  30 wt%, 1:1  0.5  Pt/C  40 wt%  0.5  Pt-Ru/C  40 wt%, 1:1  Pt-Ru/C  3.2 mg/cm  2  Pt-Ru/C  50:50, 4 mg/cm  Pt-Ru/C  50:50, 4 mg/cm  0.1, 0.5, 1, 2 0.1, 0.5, 1, 2  1  1  20-80  -  -  33.6  -  [185]  0.5  0.5  23-70  -  -  62.5  -  [174]  2  0.5  2  23-75  <0.60  >0.60  12.1  59.72  2  0.5  16  23-75  <0.60  >0.60  27.46  154.12  Current work Current work  86  2.4 Summary  It was thermodynamically and experimentally demonstrated that H2 production from CH3OH electrolysis in aqueous acidic media is significantly more effective than the H2 production from acidic H2O electrolysis. The theoretical thermodynamic evaluation proved that alcohol electrochemical reforming systems could become thermodynamically-favourable under certain operating conditions. It was estimated that in the liquid phase CH3OH electrolysis becomes thermodynamically favourable at about 41oC. However, it was demonstrated that at 50 and 75oC, the kinetics were not favourable and energy was still required to electrochemically generate H2 from the acidic aqueous CH3OH solution with all catalysts. The kinetics of the PEM electrolyser were investigated for the first time under static conditions, i.e., stirred solutions, but no flow-through or recycling systems were investigated. For all catalysts studied, it was possible to operate the CH3OH electrolyser at much greater current densities than the H2O electrolyser.  The  electrolysis has to be carried-out at currents (< iL) and voltages (> -1.8 V) that avoid carbon oxidation or more stable materials comprising no carbon should be used.  In acidic aqueous media and for all  operating conditions, carbon oxidation and catalyst degradation prevented the DMR or PEM electrolyser from sustaining higher currents in the vicinity of the limiting current. Operating the CH3OH electrolyser with a dry cathode, purged with N2 resulted in similar cell voltage requirements for electrolysis, particularly at high current densities.  Very limited kinetic data was found to describe similar  electrochemical systems in the literature. Only two studies were found to report kinetic data on catalyzed membranes or electrodes with membranes in the literature. All data available focusing on the electrooxidation of CH3OH was determined for low concentrations of CH3OH (< 2 M). The Tafel kinetic analysis carried out revealed that Pt-Ru binary catalysts enhance the kinetics of the CH3OH oxidation. The activation energy obtained with the MEA was found to be in the same range as previously reported values for similar systems. A comprehensive baseline study was established for future comparison with other PEM electrolysers.  87  Chapter 3: Electrochemical Promotion of Electrocatalysis (EPOE)  3.1 Synopsis  In this Chapter, the use of electrochemical promotion in triode and tetrode operation was investigated as a method to improve the kinetics of the direct aqueous acidic CH3OH PEM electrolyser or reformer studied in Chapter 2.  To the author’s knowledge, the application of electrochemical promotion of  electrocatalysis concept described in Section 1.3.5 to the electrolysis mode remains an unexplored area which requires more research. This Chapter consists of a comprehensive study of the triode and tetrode electrolysis concept, and addresses aspects of this knowledge gap. Triode and tetrode operation was shown to have an electrochemical promotion effect on the PEM electrolysis of liquid CH3OH and water for H2 production in a non-flowing system at ambient conditions in the galvanostatic mode and in the potentiostatic mode.  A decrease in electrolysis voltage requirements proportional to the current or  potential imposed in the auxiliary circuit was observed only when the auxiliary current or potential was opposite to the electrolyser circuit current or potential. The effect was observed with catalytic and noncatalytic electrolyser ring working electrode and auxiliary working electrode materials. It was shown that the improvement obtained was not caused by intrinsic membrane or solution conductivity enhancements during triode operation and was not due to carbon oxidation of the carbon fibre paper at the electrolyser ring electrode. Some design aspects of the triode and tetrode electrolyser were investigated in order to further elucidate the mechanism causing the enhancements observed. The need for electrode contact with the electrolyte was investigated, as well as the effect of the electrolyser working electrode ring geometry, the orientation of the proton flux lines, and the use of a high-surface area carbon electrolyser working electrode were studied. It appears that triode and tetrode operation enhanced the electrooxidation through the electrochemical pumping of protons and proton spillover.  Triode and tetrode  operation may lead to electrolysis cost reduction through the use of smaller amounts or no precious metal catalyst at the working electrodes while maintaining an adequate performance. Part of the work presented in this section was published and reprinted from the ECS Transactions, 25 (23), C. R. Cloutier,  88  and D. P. Wilkinson, "Triode Operation of a Proton Exchange Membrane (PEM) Electrolyser", 47-57, Copyright (2010), with permission from the Electrochemical Society.  3.2 Overview  Similar to the triode fuel cell, the triode electrolyser is composed of three electrodes: the electrolyser electrode, an auxiliary electrode, and a counter electrode (CE). Three electrodes form two electrical circuits: (a) the electrolyser circuit, which includes the electrolyser ring electrode and the CE, and (b) the auxiliary circuit, which includes the auxiliary electrode and the CE. The electrolyser ring electrode and the auxiliary electrode serve as the working electrodes (WE) of each respective circuit while the CE is the same for both circuits. A typical PEM triode electrolyser arrangement is shown in Fig. 3.1. The electrolyser WEelec and the auxiliary WEaux, are not physically or conductively connected to one another, i.e., electrons are not traveling between the WEs of both circuits. Therefore, the anodic or cathodic current of an