electrocatalysts for oxygen electrodes

7/30/2019 Electrocatalysts for Oxygen Electrodes

1/68

LBL-33814

ELECTROCATALYSTS FOR OXYGEN ELECTRODES

Final Report

February 1993

Ernest YeagerCase Center for Electrochemical Sciences

and the Department of ChemistryCase Western Reserve University

Cleveland, Ohio 44106-7078

forExploratory Technology Research Program

Energy &Environment DivisionLawrence Berkeley LaboratoryBerkeley, California 94720

This work was supported by the Assistant Secretary for Conservation and RenewableEnergy, Office of Transportation Technologies, Electric and Hybrid Propulsion Divisionof the U.S. Department of Energy under Contract No. DE-AC03-76SF00098, Subcon-tract No. 4566910 with the Lawrence Berkeley Laboratory.

1


2/68

DISCLAIMERThis report was prepared as an account of work sponsoredby an agency of the United States Government. Neitherthe United States Government nor any agency thereof, norany of their employees, make any warranty, express orimplied, or assumes any legal liability or responsibility forthe accuracy, completeness, or usefulness of anyinformation, apparatus, product, or process disclosed, orrepresents that its use would not infringe privately ownedrights. Reference herein to any specific commercialproduct, process, or service by trade name, trademark,manufacturer, or otherwise does not necessarily constituteor imply its endorsement, recommendation, or favoring bythe United States Government or any agency thereof. Theviews and opinions of authors expressed herein do notnecessarily state or reflect those of the United StatesGovernment or any agency thereof.


3/68

DISCLAIMERPor t i ons o f t h i s docu m en t m ay be i ll eg ib l ein e lec t ro n i c im age p ro du c ts . Im ages a repro du ced f rom th e b es t ava i lab le o r i g ina ld o c u m e n t .

,


4/68

PREFACEThis fhal report includes the important technological and scientific

accomplishments of this project for the period 1 April 1984 to 30 April 1992 and acomprehensive sunu&.ry of the research carried out during the past year (1 May 1991 - 30April 1992). The list of research publications and Ph.D. dissertations obtained under thisproject are also given. Various active catalyst systems for the 0, eduction andgeneration in concentrated alkaline and acid electrolytes as well as for the decompositionof peroxide in alkaline solutions are mentioned in appendicesA and B.

The following personnel were involved in the research on a full or part-time basis.

W. Aldred Research AssistantR. Chen Graduate StudentH. Huang Graduate Student2. hang Graduate StudentS.Gupta* Senior Research AssociateD. Tryk Senior Research AssociateE. Yeager Project Director

*Report prepared by S. Gupta

1


5/68

TABLE OF CONTENTS

I. IMPORTANT TECHNOLOGICALAND SCIENTIFICACC0MPLIS"TS OF THIS PROJECT FOR THEPERIOD 1 APRIL 1984 TO 30 APRIL 1992II. COMPREHENSIVE SUMMARY OF THE RESEARCH

DURING THE PAST YEAR 1 May 1991 - 30 April 1992)1. Transition Metal Macrocycle Catalysts

1.1 In Situ FTIRRAS Studies of FeTsPc on Ag1.2 STM nd AFM Studies of Macrocycles1.3 Solution-Phase Voltammetry of Tetrasulfonated

Phthalocyanines in Organic Solvents1.4 Oxygen Reduction on Ordinary Pyrolytic Graphite andPolycrystalline, Pt and Au Surfaces in Dimethyl Sulfoxidewith and without CoTsPc in Solution1.5 Effects of Adsorbed Tetrasulfonated Phthalocyanines onthe Heterogeneous Redox Behavior of Charged and

Uncharged Species in Aqueous Electrolytes1.6 Effect of Methanol and Other Alcohols on the 02Reduction

Electrocatalytk Activity of Macrocycles on OPG in Acid andAlkaline Solutions

2. Transition Metal Oxides as Catalysts and Supports2.1 02Reduction and Generation on Pyrochlore-Based Gas-Fed

Electrodes in Acid Electrolyte

Page

1

810101114

20

27

3136

362.2 Transition Metal Oxides as Support Materials for02ReductionElectrocatalysts 39

III. RESEARCH RECOMMENDATIONS 44IV. REFERENCES 45

..11


6/68

TABLEOF CONTENTS(Continued)

Page

V. LIST OF RESEARCHPUBLICATIONSUNDERLBL/DOE PROJECT 47VI. PH.D. DISSERTATIONS UNDER LBL/DOE PROJECT 51

Appendix A A-1List of various active catalyst systems prepared and examined for 02reduction and generation in alkaline and acid electrolyteswith porous02-fed (1 atm) electrodes.Appendix B B-1List of various catalysts prepared and examined for the decompositionof peroxide in alkaline electrolytes.

...w


7/68

L IMPORTANT TECHNOLOGICAL AND SCIENTIFIC ACCOMPLISH-MENTS OF THIS PROJECT FOR TEE PERIOD 1 APRIL 1984 TO 30APRIL 1992

transition metal oxides including perovskites and pyrochlores. A new level ofunderstanding of the electrochemical properties of these catalysts and the mechanistic

Extensiveaspects of 02 reduction and generation on these catalysts has been achieved.

This project on electrocatalysts for oxygen electrodes at CWRU was initiated inApril 1984 with the objective to understand the factors controlling the activity of 0 2reduction and generation electrocatalysts and the use of this information to achieve higheractivity combined with long-term stability. Two broad classes of catalysts weredeveloped: (1) transition metal macrocycles in monomeric and polymeric forms, and (2)

studies were also carried out on carbons and graphite as 02 electrocatalysts and substratesfor electrocatalysts in alkaline solution. Some of the important specific accomplishmentsof this research are:

0 Iron phthalocyanine (FePc) and its modified forms such as iron tetrasulfo-phthalocyanine (FeTsPc) and iron tetrapyridinoporphyrazine (FeTPyPz) catalyzethe 4-electron reduction of 02 to O K n alkaline solution when adsorbed onpyrolytic graphite at mono- or submonolayer coverages.Lead-ruthenate pyrochlores in the high-area form are very active bfinctionalcatalysts for 0 2 electrodes in reduction and generation modes in concentratedalkaline solutions. They catalyze the 4-electron reduction ofO2 to OK.Transition metals (e.g., Co y Fe) adsorbed on heat-treated polyacrylonitrile(Pw-based catalysts, particularly Co/PAN/XC-72, show promisingperformance for 0 2 eduction in alkaline and acid electrolytes. These catalystsystems fhction through a peroxide mechanism similar to that involvedwith theheat-treated cobalt porphyrins and have comparable apparent activity.J

1


8/68

0 ]eat-treated ransition metal macrocycles such as CoTMPP dispersed on high-area carbon have high activity for 0 2 eduction in alkaline and acid electrolytes,with good stability in concentrated alkaline solutions.

0 The combined use of FTIR, laser Raman, TEM, Mossbauer, extended X-rayabsorption fine structure (EXAFS), magnetic susceptibility, pyrolysis-gas(chromatograph-mass spectroscopy and linear sweep voltammetry of theadsorbed catalyst has provided new insight into the effects of heat treatment oncatalysts stability.

0 Based on various electrochemical and non-electrochemical studies of the heat-treated macrocycles and related complexes, CWRU proposed a model to explainthe catalytic activityof th e heat-treated transition metal macrocycles. The modelproposed is that of a modified carbon surface on which transition metal ions areadsorbed, principally through interaction with residual nitrogen derived fiom theheat-treated macrocycles.

0 Ex situ and in situ Mossbauer spectroscopywas used successfblly at CWRU todetermine the state of the transition metal in macrocycles both in bulk andadsorbed on high-kea carbon with or without subsequent heat treatment.

0 FePc and its p-oxo derivatives were characterized at Case using FTIR,Mossbauer, STM, cyclic voltammetry, W E nd gas-fed electrodemeasurements. It was found that the activity associated with the p-oxo (1)species (crystallized fiom DMF solution) is comparable to that obtained withPowercat 2000 (a commercially available high-area Pt-based high-performancecatalyst) in alkaline solution.

0 Materials produced by the heat treatment of pyrrole black mixed with transitionmetal salts are promising catalysts for O2 electroreduction in both alkaline and

2i


9/68

acid electrolytes. Their activity is comparable to that of the correspondingtransition metal macrocycles which are heat treated on carbon.The stability of gas-fed 02 electrodes in the reduction as well as generationmodes has been enhanced by using ion conducting polymers in concentrated acidand alkaline electrolytes. Thiswas achieved by incorporating the ionomer in theactive layer of the gas-fed electrodes and also by covering the solution side ofthe active layer with an ionomeric membrane.0 2 eduction to peroxide was examined in aqueous alkaline solution on highlyoriented pyrolytic graphite (HOPG) wth adsorbed 9, 10-phenanthrenequinoneand on ordinary pyrolytic graphite (OPG) with chemically attached 2-aminoanthraquinone. The kinetics of the initial one electron transfer to the quinone onthe surface with the adsorbed quinone are fast, while the reaction of the reducedquinone radical with the solution-phase O2 s the rate determining step. For theoverall 2e- reduction to peroxide on the chemically attached quinone surface, theinitial electron transfer is the rate determining step because the quinone is fartherfiom the surface and in an unfavorable configuration for electron tunneling.02 reduction to superoxide was examined in acetonitrile on the basal planes ofordinary and highly oriented pyrolytic graphite. The reaction has a lowerapparent cathodic transfer coefficient on HOPG than OPG, probably becausepart of the potential drop across the electrode-electrolyte interface occurs in thespace charge region in the HOPG. 02 reduction to superoxide in DMSO wasalso studied on OPG, Au and Pt electrodes with and without CoTsPc in solution.The results indicate that CoTsPc does not catalyze the 02 reduction to 02-naprotic solvents and the reaction appears to occur by an outer-sphere electron-transfer mechanism.

3


10/68

0 .Another class of compounds which has activity similar to the correspondingmonomeric Pc's, and which is expected to have better stability in concentratedacid and alkaline electrolytes, is the polymeric phthalocyanines. Cobalt and ironphthalocyanine sheet-type polymers were synthesized and purified as described inthe literature. They were separated from the low-molecular-weight fractionsfrom their solutions in DMF. The high-molecular-weight fraction wascharacterized by FTIR, laser Raman, uv-visible and X P S measurements. Thesheet-type configuration of the CoPc oligomer is evident by high-resolutiontransmission electron microscopic (TEM) and atomic force microscopic (AFM)measurements.The orientation of the macrocycle complex on the surface has a strong effect onthe activity as well as the pathway for 02 reduction. In situ SERS, FTIRRASand STM measurements of monomeric TsPc's adsorbed on a substrate at mono-or submonolayer coverages provide evidence that the plane of the adsorbed TsPccomplex is oriented at an angle to the surface. Polymeric sheet-typephthalocyanines, however, in the form of crystallites on a graphite surface showthe sheet-shaped structure of the molecules that are stacked parallel to eachother and parallel to the substrate by usingAFM.The voltammetric peaks for the solution-phase species of CoTsPc and FeTsPc

However, the voltammetricere not obtained so far in aqueous electrolytes.peaks were obtained for these species in organic solvents. A possibleexplanation for this was based on the excess negative charge associated with theligand in aqueous electrolyte as a result of the ionized SO3- groups attached tothe four aromatic rings. The adsorption of such highly charged species on theelectrode surface results in the repulsion of the corresponding solution-phase


11/68

species. This explanation is supported by the effect of adsorbed TsPc's on theredox behavior of negatively charged (ferri and ferrocyanide), uncharged (0-tolidine) and positively charged Fe(ClO& in aqueous acid and alkalinesolutions.Adsorption of FeTsPc and CoTsPc on OPG was investigated as a function ofpH and ionic strengthof the adsorption solution as well as potential. Adsorptionof CoTsPc occurs readily from its freshly prepared aqueous solution and isgenerally independent of pH. For FeTsPc, however, adsorption does stronglydepend on pH. High surface coverage is achieved only from acid solutionsrather than fkom pure water or alkaline solutions. This is explained in terms ofthe form(s)of the complexes existing in the solution phase in air.To obtain krther insight into the redox properties and 0 2 reductionelectrocatalysis on macrocycles, the effect of CN-, which is capable ofcoordinating with the transition metal in the axial position, was examined. Theresults indicate that a strong axial interaction of 0 2 with the transition metal inthe macrocycle is an important factor for 02 electrocatalysis. The blockingeffect of the voltammetric peaks confirms the assignment of the peaks in cyclicvoltammetry curves and the importance of particular redox couple to the 0 2electrocatalysis.The polyvinyl pyridine (PVP)-modified OPG electrode with adsorbed CoTsPcexhibits higher activity for 0 2 reduction than th e OPG electrode with onlyadsorbed CoTsPc in 0.05 M H2S04 solution. This is due to the formation of anadduct between PVP and CoTsPc which is more active and more stable thanCoTsPc. FTIR and uv-visible spectroscopic studies provide evidence for theformation of an adduct between CoTsPc and PIP.

5


12/68

0 The perovskites have not generally been found to be active catalysts for 0 2reduction in gas diffusion 02 cathodes either with or without high-area carbon.Some of the higher-area compounds have considerable activity for peroxidedecomposition and 0 2 generation. When used in conjunction with heat-treatedmacrocycles (even macrocycles without a transition metal center), some of theyerovskites show excellent performance for 0 2 reduction.The effects of alcohols such as methanol, isobutanol and n-amyl alcohol wereexamined to obtain a better understanding about the redox properties and 0 2reduction electrocatalysis on transition metal macrocycles. These alcohols maychange the structure of the interface and orientation of the adsorbed macrocycleson the surface. The effect of methanol and its reaction intermediates on 0 2ireduction are also of interest for the identification of methanol-tolerantlelectrocatalysts for 0 2 reduction in methanol-air fbel cells. For 0 2 reductionon CoTsPc in an alkalime electrolyte, which was pre-adsorbed on an OPG diskelectrode fiom a 0.1 M NaOH solution containing CoTsPc (-1 x lo4 M) inthese alcohols, the half-wave potential (Ex) is more positive (-60 mv). Thismay be due to the increased adsorption of the macrocycle on the surface as aresult of changes in the dielectric property of the solvent. The presence of -1 Mmethanol in 2.5 M H2S04 at 60OC showed no effect on the 0 2 reductionperformance in gas-fed electrodes based on pyrolyzed macrocycles (CoTMPPand CoTAA adsorbed on high-area carbon). This is quite encouraging for airelectrodes in acid electrolyte methanol-air fbel cell.A host of techniques, both electrochemical and non-electrochemical along withspectroscopic techniques, were used to characterize the most promisingelectrocatalysts for 0 2 electrodes. These techniques include:

6


13/68

1.2.3.4.5.6.7.8.9.10.11.12.13.14.15.16.17.18.19.20.

linear sweep voltammetry (LSV) and differential pulse voltammetry (DPV)rotating disk electrode (RDE)and rotating ring-disk electrodeW E )Fourier transform infrared (FTIR) spectroscopyuv-visible spectroscopysurface enhanced Raman scattering (SERS) and resonance Raman (RR)scanning tunneling microscopy (STM) and atomic force microscopy(AFM)high resolution transmission electron microscopy (HRTEM)scanning electron microscopy (SENX-ray absorption fine structure ( X A F S )Mossbauer effect spectroscopy (MES)X-ray ditfkaction(XRD)X-ray photoelectron spectroscopy @PS) and ultraviolet photoelectronsecondary ion mass spectroscopy (SIMS)pyrolysis-gas chromatograph-mass spectroscopy (PGCMS)thermogravimetric analysis (TGA)magnetic susceptibilityelemental analysisBET sudace areaextraction techniquesadsorption techniques

spectroscopyU P S

Pt catalysts supported on lead-ruthenate pyrochlore and with a Nafion 117(DuPont) membrane pressed against the electrolyte side of the gas-fed electrodeshowed promise for bifbnctionalO2 electrodes in acid electrolyte (2.5 M H2S04at 60%).

0 The fabrication of porous gas-fed electrodes was modified to optimize theirstability and performance at high current densities in 85% H3P04 at lOOOC usingcarbon supported Pt catalysts for 0 2 reduction and generation.The RRDE experiments show that the reduction of 02 proceeds by a 4-electronpathway on Ru in alkaline electrolytes. The kinetics and mechanism depend onthe oxidation state of the Ru surface. The reaction mechanisms for the formationof the anodic film and 0 2 reduction on Ru are proposed.


14/68

0 Lithiated NiO powder was examined as a support for Pt-catalyzed gas-fedelectrodes. The performance for 0 2 reduction is inferior to that obtained usingcarbon supports in 5 M KOH at 25%. This oxide has negligible activity for 02reduction at room temperature but has significant activity for 02 reduction attemperatures higher than 150OC. Such a p-type oxide is not expected to beactive for 02 reduction.

0 CoTsPc adsorbed on the perovskite Lac003 and La05Sr(&nO3 powders ascatalyst supports showed higher activity for 0 2 reduction than perovskitesupports without the macrocycle in alkaline solution. The activity is much lessthan that of CoTsPc adsorbed on a high-area carbon under similar conditions.

0 Most of the high-area carbons used as support materials for catalysts for 0 2reduction are generally unstable at the more positive potentials involved in 0 2generation. Efforts were made to find more stable carbon supports. Severaldifferent carbon supports, including both oxidized and non-oxidized Shawiniganblack, LBL carbon, graphitized carbon and mildly fluorinated carbon black, wereexamined at CWRU for use with lead ruthenate pyrochlore. All of these carbonsgave similar performance. This is promising because it shows that highlyoxidation resistant carbons can be used without sacrificing cathode performance.

The details of this research are given in the annual reports prepared by CWRU forthe LBL itswell as in published research papers under this project.

II. COMPREHENSIVESUMMARY OF THERESEARCH DURING THEPi4STYEAR1. ObjectivesThe objectives of the research were:

8


15/68

1. to develop a better understanding of the factors controlling02 reduction andgeneration on various electrocatalysts, including transition metal macrocyclesand oxides;to use this understanding to identifjr and develop much higher activitycatalysts, both monoknction and bfinction;to establish how catalytic activity for a given 02 electrocatalyst depends oncatalyst-support interactions and to identie stable catalyst supports forbfinctional electrodes.

2.

3.

2. IntroductionOver the past three decades a large research effort has been carried out

internationally on a wide range of catalysts for 02 reduction in alkaline and acidelectrolytes and to a lesser extent on 02 generation, particularly in alkaline electrolyes.Much of this prior research by various groups was "semi-Edisonian" and lacked anadequate understanding of the electronic and steric factors controlling th e activity.Consequently, the emphasis on 02 electrocatalysis research at CWRU is to achieve abetter understanding of both 02 reduction and generation in relation to the basicproperties of the catalyst-electrolyte interface. The research also is concerned with thefailure mode of these catalysts.

Research during the past year (1 May to 30 April 1992) at CWRU involved workprincipally on the following aspects of the catalyst systems.

1. Transition Metal Macrocycle Catalvsts1.11.2

In Situ FTIRRAS studies of FeTsPc on AgSTM and AFM studies of macrocycles on surfaces.

9


16/68

1.3

1.4

1.5

1.6

Solution-phase voltammetry of the tetrasulfonated phthalocyaninesin aprotic organic solvents.Oxygen reduction on OPG, Pt and Au surfaces in DMSO with andwithout CoTsPc in solution.Effects of adsorbed tetrasulfonated phthalocyanines on the redoxbehavior of charged and uncharged species in aqueous electrolytes.Effect of methanol and other alcohols on the 02 reductionelectrocatalytic activity of macrocycles in acid and alkalinesolutions.

2. Transition Metal Oxides as Catalyst and Support2.1 0 2 reduction and generation on pyrochlore-based gas-fed elec-

trodes in acid electrolyte.Transition metal oxides as support materials for 0 2 reduction elec-trocatalysts.

2.2

Appendices A and B summarize the catalyst systems which have significant activityfor 0 2 eduction and generation, as well as peroxide decomposition, respectively.

1. Transition Metal Macrocycle Catalysts1.1The iron tetrasulfonated phthalocyanine (FeTsPc) complex adsorbed on an

electrode surface at mono- or submonolayer levels has high activity for the 4-electron reduction of O2 to O H in alkaline solutions. The orientation of thecomplex on the surface has a strong effect on the activity as well as pathway for02 reduction. In-situ surface-enhanced Raman scattering (SERS) measurementsobtained earlier on Ag showed stretching modes involving C=C and C=N bonds of

In Situ FTIRRAS Studies of FeTsPc on Ag.

10


17/68

e-

the macrocycle and provided evidence that the plane of the adsorbed TsPccomplex is oriented at an angle to the surface (1). Selection rules, however, areopen to question for SERS.

During the past year in situ infrared measurements, FTIRRAS, wereconducted to obtain structural information for mono- or submonolayers ofadsorbed FeTsPc on Ag. Enhanced sensitivity was achieved by using a Doveprism (CaF2) s a window, instead of a flat window. Dove prisms provide higherincidence angles. First, cyclic voltammetry of Ag electrode in 0.1 M HClO4containing lo4 M FeTsPc was obtained using the optical cell with the electrodepulled back from the window. The FeTsPc solution was then removed and the cellwas filled with a fiesh solution of HC104, without macrocycle, leaving theadsorbed macrocycle on the surface. The Ag electrode was moved close to theoptical window at +0.1 V vs. SCE. The potential was switched between +O . l Vand -0.4 V vs. SCE. At each potential 200 scans were made. The above processwas repeated until a satisfactory signal-to-noise ratio was achieved (10,000 scans).Spectra at different potentials were obtained which have several derivative peakscorresponding to C=C or C-N vibration modes. In view of the IR selection rules,this provides further evidence that the TsPc is oriented at an angle to the surfaceand is not completely parallel to the surface.1.2 STM and AFM Studies of Macrocycles

Considerable information was obtained on the orientation of Fe- andCoTsPc on Ag and other substrates using surface-enhancedRaman and resonantRaman, uv-vis reflectance, and FT infrared reflectance absorption spectroscopieswith adsorbed mono- or submonolayer coverages. During the last year, research

11


18/68

was focused on the structural features of the adsorbed phthalocyanines using STM(ex situ and in situ) and AFM ex situ ).1.2.1 STM Studies

The CoTsPc was preadsorbed on HOPG surface by dipping the HOPGdisk at open circuit into an air-saturated aqueous solution of CoTsPc (-1 x lW5M) parallel to the liquid surface) for -20 min. The disk was then removed,washed with pure water and dried in a vacuum oven. I3 situ STM measurementswith a Pt-Ir tip showed that the plane of the complex is tilted from theperpendicular to the surface. The complexes are parallel to each other but theadjacent complexes are slipped parallel to each other so as to provide forinclination of the central transition metal ion with the pyrrole or bridging nitrogenof the phthalocyanine. However, the ordered layer was distorted by scanning withth e STM tip. It is probably caused by the interaction force between the tip and theadsorbed molecules which is reported to be about 10-6N for a few nA tunnelingcurrent in vacuum or air.

By operating the STM in a liquid environment, the interaction forcebetween the STM tip and the adorbed molecules is expected to be reduced by 10or 100times because of the difference in the dielectric coefficient of the liquid andair. The orientation of the adsorbed layer of CoTsPc on a HOPG surface was,therefore, studiedwith in situ S T M n 0.05 M H2S04 under potential control. Pttips covered with glass were used. The tip .potential was controlled to maintain aconstant tunneling current. With a modified microcell and Pt wires as referenceaid counter electrodes, cyclic voltammetry showed the submonolayer adsorptionof CoTsPc on the HOPG surface. The in situ STM images showed an orientationoff the adsorbed complex on HOPG similar to that observed initially with ex situ

12


19/68

STM. When the potential was set to the 02 reduction region, the in situ STMimages were distorted and reappeared again when the potential was set to thedouble layer region.1.2.2 AFM Studies

The molecular orientation of pre-adsorbed CoTsPc on a HOPG surfacewas also studied with the Nanoscope 11 AFM in air. The preparation of thepreadsorbed layer of CoTsPc on HOPG was described above. The adsorbed layerwas characterized by cyclic voltammetry. Si,N4 microcantilevers with integratedpyramidel tips were used. The cantilever armwas 100 microns long with a forceconstant of 0.58 N/m. The AFM tips were checked on freshly cleaned HOPGsurface. If the tip gives near atomic resolution on anHOPG surface, it can be usedfor the studies of molecular orientation of CoTsPc on the surface.

A set of AFM topographies of CoTsPc submonolayer adsorbed on HOPGsurface was obtained at a constant force of 2 x lo-* N. A large scan area (300 x300 nm) showed ordered molecular structures of CoTsPc. With highermagnification (30 x 30 nm), the images showed an orientation similar to thatobtained by in situ S T M mages. Rotation of the sample resulted in the expectedchanges of the molecular orientation.

The sheet-type polymeric phthalocyanines are of special interest forunderstanding the 4-electron reduction of 0 2 to OH- on FePc and its modifiedforms such as FeTsPc and FeTPyPz in alkaline solutions. The sheet-like polymerstructures were iirst prepared and proposed by Drinkard and Bailer based onelemental analysis and titrimetric equivalent weights (2). The concentration of theperipheral carboxylic acid groups was determined by neutralizing the acid groupsby standardized NaOH and back titration with standardized HC1. The equivalent

13


20/68

weights were calculated for the various possible oligomers. Achar et al. later 'reported a method (3) to prepare tetramer phthalocyanines as in Figure 1.However, prior to the present study, there was no direct evidence for the existenceof a sheet-type molecular arrangement, except from elemental analysis andtitrimetric equivalent weight which are not very definitive.

CoPc oligomers which are coated on an OPG surface, were studied bytu34 n this laboratory. The oligomer was coated on the OPG surface by placing50 pL of M solution of the polymer in DMSO on the graphite substrate anddrying it in a vacuum oven overnight. Crystallites with particle sizes of 30-60 nmwere observed by AFM. The profiles of the AFM images showed particles with astep height - 5 4 suggesting that the molecules are stacked parallel to each otherand parallel to the graphite surface. Zooming the scanning probe on the top layerof the crystal, one can observe a sheet-shaped molecular structure (Fig. 2). A two-diimensional Fourier transform (2 DFT) of the image given in Figure 2 generates aspectrum with four bright spots with nearly equal distance fiom the origin. A clearmolecular structure is reconstructed by an inverse Fourier transform using thefrequencies represented by the four bright spots (Fig. 3) . The distance between theturo neighboring metal centers is 11 A and th e diameter of the cavity formedbetween four monomer lobes is 8.4 4which agree well with the theoretical valuesof'10.7 A and 8.3 4 respectively, using a molecular modelling program.1.3 Solution-Phase Voltammetry of Tetrasulfonated Phthalocyanines inOrganic Solvents

Earlier studies at CWRU showed that the cyclic voltammetry of Co- andFeTsPc on OPG disk electrodes in aqueous acid or alkaline electrolytescorresponds to the redox processes of the adsorbed macrocycle species and not the

1 4


21/68

\ /w nooc coow

Fig. 1. Polymeric phthalocyanine (fused type, fragment)

15


22/68

Fig. 2. AFM image of the CoPc oligomer which was obtainedby zooming the scanning probe on top of the crystal.

Fig. 3. A molecular structure of CoPc oligomer that wasgenerated by computer using an inverse Fouriertransform from Fig. 2.

16


23/68

dissolved species in solution (4). Iron pyridino-porphyrazinr: (FeTPyPz), aphthalocyanine similar to FeTsPc, also does not show voltammetric peaks inaqueous acid electrolyte associated with the outer-sphere heterogeneous electrontransfer (5). FeTPyPz has appreciable solubility in acid solutions, and under thiscondition most of the pyridino groups are expected to be protonated. Both thesolution-phase and adsorbed FeTPyPz would be positively charged and repulseeach other, thus suppressing the outer-sphere electron transfer kinetics. Theinteraction of the adsorbed macrocycle with the substrate is expected to cause asubstantial difference in the redox behavior of the dissolved and the adsorbedspecies.

To separate the redox peaks associated with the non-adsorbed species fromthose for the adsorbed species on the electrode surface, voltammetric studies ofthese macrocycles were carried out in different solvents. Voltammetricmeasurements of CoTsPc in CH3CN and (CH,),SO containing tetraethylammoniumperchlorate as the supporting electrolyte were obtained on OPG, Auand Pt disk electrodes. Pure water was added to Cl33CN in the ratio 1H20/5CH,CNY and with CoTsPc at a concentration of -1 x lo4 M.macrocycle is sparingly soluble in pure CH3CN.

TheThe acetonitrile (Aldrich,

anhydrous, 99+%, water


24/68

no significant differences in the peak potentials for the three electrodes in the sameorganic solvent (Table I). The currents associated with the voltammetric peaks inaprotic organic solvents are low (of the order of a few pNcm2 at a sweep rate of150 mV/s and concentration of -1 x lo4 M). One explanation for this behaviormiay be that the macrocycle is highly associated in the organic solvents. This canlead to various complications which result in smaller and broader redox peaks.The heights of the peak depend on convection in the solution and are a linearhnction of the square root of the scan rate. These tests indicate that the redoxpeaks in the organic solvent correspond to the dissolved species of CoTsPc.Further, the peak potentials wth CoTsPc and CoPc in (CH3)$0 are almostsimilar. This is probably due to the fact that the S03H groups are not ionized inthe organic solvent. Recently we have observed the outer-sphere electron transferreaction for TsPc's on a tin-doped indium oxide (ITO) electrode in 0.05 M H2SO4.These macrocycles are not adsorbed on this electrode surface in 0.05 M H2SO4.The pH dependence of this reaction on IT0 is under investigation.

An explanation for the absence of voltammetric peaks of thesemacrocycles in aqueous solution is based on the excess negative charge associatedwith the TsPc ligand in solution because of the ionized SO3' groups attached tothe four aromatic rings. Another explanation involves the adsorption of themacrocycle edge-on on the electrode surface, where the distance between theelectrode sucface and the solution-phase macrocycle molecule at the point ofcllosest approach may be -15A. This large distance may impede the electrontunneling between the solution-phase macrocycle and the adsorbed species on theelectrode. Resonance tunneling might be involved, in which case tunneling mayoccur over a large distance.

18


25/68

TABLE ISolution-phase Voltammetric Redox Potentials ofCoTsPc(4 lo4 M) in 5:l Mixture*of CHsCN

and H 2 0 (deaerated) on Various ElectrodesElectrolyte:Scan Rate: 150 mVlsPeak Potentials (V)vs SCE

0.1M Tetraethyl ammonium perchlorate

Electrode Peak 1 Peak 2 Peak 3Pt - -0.40' +0.42'- -020a +0.53aAu - -0.38' +0.42'- -0.20a +0.54a

OPG -1.24' -0.38' +0.43'-l.loa -0.2za +0.54a

~

"CoTsPc is not completely soluble in pure CH3CNaanodic peak potentialCcathodic peak potentialPeak 1corresponds to the reduction of the TsPc ligandPeak 2 corresponds to th e reduction of Co(II)o Co (I)Peak 3 corresponds to the reduction of Co (III) to Co(II)

19


26/68

1 4 Oxygen Reduction on OPG, Pt and Au Surfaces in Dimethyl Sulfoxidewith and without CoTsPc in Solution.Oxygen is reduced by a 2-electron pathway to peroxide on OPG and

polycrystalline Au surfaces in alkaline aqueous solution. However, clean Pt andA.u (100) catalyze the 4-electron reduction of 02 to O K nder similar conditions.To obtain a better understanding of the mechanistic aspects of 02 reduction, thereduction of 02 on OPG, Au and Pt surfaces was examined in an aprotic non-aqueous solvent (DMSO) containingTEAP as the supporting electrolyte, with andwithout the presence of CoTsPc in the solution. CoTsPc is a good catalyst for the2-electron reduction of 0 2 to peroxide in acid and alkaline aqueous solutions.During the past year, 0 2 reduction was studied on OPG, t and Au surfaces inDMSO using linear sweep voltammetry and the rotating disk electrode (RDE)technique at 25OC.

O2 eduction on these electrode materials in DMSO showed a quasi-reversible one-electron reduction to superoxide. The cathodic peak potentials inthe voltammograms were -0.85 V, -0.89 V and -0.90 V vs. SCE for OPG, andpolycrystalline Au and Pt respectively. This indicates that the rate of electrontransfer in aprotic media depends on the electrode material, in agreement withearlier results (6-8). The separation of the cathodic and anodic peak potentials forClPG (AV= 0.16 V) is less than those for Au (AV=0.22 V) and Pt (AV=0.20 V),respectively. The addition of CoTsPc (-5 x M) to the solution reduces th epeak heights to some extent. There is an increase in the separation of the peakpotentials with all of the electrodes, except for OPG in which case there is a slightdecrease in the separation of the cathodic and anodic peak potentials. The results

20


27/68

show no significant changes in the 0 2 eduction reaction in DMSO which containCoTsPc.

The curves for 0 2 reduction on rotating disk electrode in DMSOcontaining 0.1 M TEAP as the supporting electrolyte are given in Figures 4-6.The half-wave potential (Ex) is -85 mV more positive for OPG than for Au and-25 mV more positive for Au than for Pt. The plots of l i t i n g current densities(iL) vs. square root of rotation rates ( show that iL is approximately equal to thediffusion limiting current density iD (Ncm2) (Fig. 7). The observed B values fiomthe Levich plots are 6.1 x A cmm2rpm-' for Pt respectively. These values agree fairly well with the calculated valueof7.1x A cm-2 rpm-' for n =1 and fiom the other data (7,9,10) (see Fig. 7).The i" vs. f plots show that the slope increases at less negative potentials on all

for OPG, 6.3 x for Au and 6.4x

of the surfaces, indicating the relative influence of the back reaction. The mass-transport-corrected Tafel plots for the forward (positive to negative) potentialsweep for OPG, Au and Pt are shown in Figure 8. The Tafel slopes are -144mV/decade for OPG, -231 mV/decade for Au and -227 mV/decade for Pt and thetransfer coefficients are 0.41 for OPG and 0.26 for Au or Pt, respectively. Thedifferences between these parameters for OPG and Au or Pt electrodes may be dueto differences in the true surface area, fhctional groups on the OPG surface, andthe point of zero charge (PZC) of these materials.

The RDE studies also show no major changes in the 0 2 reduction reactionin DMSO containing CoTsPc. The Au and Pt electrodes show some inhibition forthe electrode reaction (0 2 +e' + 2') in the presence of CoTsPc. The Ex valuesfor Au and Pt are -75 mV and -40 mV more negative, respectively, with CoTsPcin solution. However, the OPG electrode shows negligible effect under similar

21


28/68

-0.8

-0.7

-0.6

-0.5na 0.4EW- -0.3

-0.2

-0.1

0.0

. 3600 rprnG500 rpm-0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4

E (v)vs.SCE

Fig. 4. Disk currents for 02 reduction on OPG in OrsaturatedDMSO (0.1 M TEAP).Sweep rate =20 mV/s.Electrode area =0.2 cm2.Temperature =.25OC.

22


29/68

-1.8

-1.6

-1.4

-1.2

-1o

- -0.8-0.6

-0.4

-0.2

0.0-0.2 -0.4 -0.6 -0.8 -1 o -1.2 -1.4 -1.6E (v) vs. SCE

Fig. 5. Disk currents fo r 0 2 reduction on polycrystalline Au in02-saturated DMSO (0.1 MTEAP).Sweep rate =20 mV/s.Electrode area =0.46 cm ?Temperature =25 OC. 23


30/68

n-au

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

I I I I I I I I I I I I I

I-0.2 -0.4 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6

E ( v ) vs .SCEFig. 6. Disk currents for 0 2 eduction on polycrystalline Pt in02-satu rated DMSO (0.1 MTEAP).

Sweep rate =20mV/s.Electrode area =0.125 cm2.Temperature =25*C.

24


31/68

48

40

32w0x 24E$ 161

cv

2J-8

010 20 30 40 50 60 70

Fig. 7. Levich p lots for 0 2 eduction in DMSO (0.1 M TEAP) onOPG (x), Au (0) nd Pt (A) based on Figs. 4, 5and 6respectively. Current values were measured at -1.4 Vfor OPG -1.5 V for Au and -1.6 V vs. SCE for Pt. Solidline calculated from the literature values of D o 2 , v andC o p or n =1. Temperature =25OCDo 2= 2.82 x 10-5 m2 d ( 9 )VC02= 2.1 x 10-6mol cm-3(7)= 1.79 x 10-2 cm2 s-1(10)

25


32/68

LLI0v,u j>>W'nW

-0.75

-0.85

-0.95

-1.05

-1 I 5

-1.25

-1.35% b Q

.-1.4!50.0 0.5 1o 1.5 2.0 2.5 3.0log i rnb cm-2

Fig. 8. Mass-transport-corrected Tafel plots for 0 2 eduction in02-saturated DMSO (0.1 M TEAP) on OPG (o),Au(A)and Pt (x) for fowvard scan. Data taken from figs. 4, 5an d 6, respectively, for 2500 rpm. Temperature =25OC.

26


33/68

conditions (-10 mV more negative value of EyJ with CoTsPc in solution. Theseresults confirm that the rate of the electron transfer for 0 2 eduction in an aproticmedia depends on the electrode material. The addition of CoTsPc to the DMSOsolution with 0.1M TEAP as the supporting electrolyte causes only a small changein the rate constant (see TablesII and III).

The standard rate constants for the reaction 0 2 +e' 0 2- on varioussurfaces, with and without the CoTsPc, are calculated from voltammetric studies(1 1) and are recorded in Tables 11and 111along with other parameters. The valuesof the standard heterogeneous electron transfer rate constants (ko) calculated fromvoltammetry and rotating disk electrode techniques agree fairly well.

1.5. Effects of Adsorbed Tetrasulfonated Phthalocyanines on the RedoxBehavior of Charged and Uncharged Species in Aqueous ElectrolytesCobalt and iron tetrasulfonated phthalocyanines adsorbed on OPG disk

electrode at mono- or submonolayer coverages, were found to be goodelectrocatalysts for 0 2 eduction in aqueous alkaline and acid electrolytes. Earlierstudies at CWRU showed that the solution-phase cyclic voltammetry of Co andFeTsPc in aqueous acid or alkaline electrolytes on OPG corresponds to the redoxprocesses of the adsorbed macrocycle species and not the dissolved species insolution (4). However, the solution-phase redox peaks for these macrocycles onOPG, Au and Pt electrodes were recently observed in organic solvents such as(CH&SO with tetraethyl ammonium perchlorate (TAEP) as the supportingelectrolyte. An explanation for the absence of the solution-phase peaks of themacrocycle was based on the excess negative charge associated with the TsPcligand in aqueous solution as a result of the ionized SO3' groups attached to thefour aromatic rings. Another explanation may be related to the edge-on adsorption

27


34/68

TABLEIIStandard Electrode Reaction Rate Constantsko for

0 2 Reduction on OPG, Au and Pt in Dimethyl Sulfoxide(0.1 M TEAP) with CoTsPc in the Solution (-1 x lo4 M) at 25OC

Electrode ScanRateMaterials (mV s-1)

OPG 150(a=0.38) 100

50-

Au(a 0.21)

Pt(a=0.25)

2010

150loo502010

15010050

, 2010

mP(mV)16015013010090

275250205160140

3 10285230185158

YGb

0.1900.2200.3000.5800.810

0.0650.0650.1100.1900.250

0.0650.0650.0800.1400.200

ko (cmd) Average

0.00290.00280.00270.00330.0033

0.00120.0010.00120.00130.00120.001 10.00090.00080.00090.0009

ko(cm 5-1)0.003

0.0012

0.0009

aDetermined by interpolation from values in Table 1 of reference 11bSweep direction, positive to negativeaTransfer coefficientYA knction of the anodic - cathodic peak potential separation and is tabulated in Ref. 11

28


35/68

TABLE IIIStandard Electrode Reaction Rate Constants kofor 0 2Reduction on OPG,Au and Pt in Dimethyl Sulfoxide

(0.1 M TEAP )withoutCoTsPc in the Solution (-1 x lo 4 M) at 25OC

ElectrodeMaterialsOPG

(a=0.41)

Au(a=0.26)

Pt(a=0.26)

Scan Rate(mV -1)150100502010

150100502010

150100502010

AEP(mV)18516514011095

260225185145125

280240200150130

\ya,b

0.1300.1800.2500.4450.680

0.0650.0850.1400.2350.330

0.0650.0700.1 180.2200.300

kox 10-2 Average(cm 5-1) kox 100.19 0.230.220.220.240.26

0.11 0.130.120.140.150.15

0.11 0.120.100.120.140.14

aDetermined by interpolation from values in Table 1of reference 11bSweep direction, positive to negativeTransfer coefficientyA bnction of the anodic - cathodic peak potential separation (see Table 11)


36/68

of the macrocycle on the electrode surface. In this case the distance between theelectrode surface and the solution-phase macrocycle molecules at the point ofclosest approach may be -15A. This large distance may impede electron tunnelingbetween the solution-phase macrocycle and the adsorbed species on the electrode.Resonance tunneling might be involved, in which case tunneling may occur over alarger distance.

To examine further the heterogeneous redox processes of the TsPc's inaqueous electrolytes, the effects of preadsorbed tetrasulfonated phthalocyanines onOPG or in solution (-1 x lo4 M) was investigated using linear sweepvoltammetry. This has involved the redox behavior of negatively chargedK3Fe(CW6 and K&(CN)6, uncharged o-tolidine and positively chargedF c ( C ~ O ~ ) ~n aqueous acid and alkaline solutions.phthalocyanines were used, including CoTsPc, FeTsPc, CuTsPc and H2TsPc.

Various tetrasulfonated

In general, the redox behavior for K3Fe(CN), and K,$?e(CN)6or their 1:1mixture (1 x 10-3M) on OPG electrodes with preadsorbed macrocycle becomesless reversible (more separation of the voltammetric cathodic and anodic peakpotentials) in acid electrolyte (0.05 M H2S04). There is also a decrease in themagnitude of the cathodic and anodic peaks of the redox species of the pre-adsorbed macrocycle as compared to those for the same electrode withoutmacrocycle present. The irreversibility is further increased and the magnitude ofthe peak heights decreased if the macrocycle (-1 x lo4 M) is in th e solution. Asi~milar edox behavior is obtained for the ferri- and ferro-cyanide in the alkalinesolution (0.1 M NaOH) for various macrocycles pre-adsorbed on the OPG oradded in the solution except for the CoTsPc. When it is added to the alkalinesolution containing ferri- or ferro-cyanide or their mixture, the redox behavior

30


37/68

became more reversible and the peak heights increased. This is probably due tothe fact that CoTsPc acts as a mediator for electron transfer because the redoxpotential of CoTsPc lies in the same potential range as that of the fem- andferrocyanide species in alkaline solution.

In comparing the effect of adsorbed CoTsPc on th e redox behavior ofM each) in 0.1 M HClO4 solution,3Fe(CN)6, 0-tolidine and Fe(C104)3 1 x

it was observed that the separation of the voltammetric anodic and cathodic peakpotentials become greater with K$e(CN)6, almost no change with the o-tolidineand less with Fe(C104)3 (Figures 9-11). This double layer effect is morepronounced when CoTsPc (-1 x 104 M) is in solution. The results provideevidence that the charge associated with the TsPc ligand in aqueous solution isresponsible for th e absence of the solution-phase voltammetric peaks fortetrasulfonated phthalocyanines in aqueous electrolytes. Iron pyridino-porphyrazine (FeTPyPz), a phthalocyanine similar to FeTsPc, also does not showvoltammetric peaks in aqueous electrolytes (5 ) . In contrast to FeTsPc, however,the FeTPyPz does not have large charge on the ligand.1.6 Effect of Methanol and Other Alcohols on the 0 2 Reduction

Electrocatalytic Activity of Macrocycles on OPG in Acid and AlkalineSolutionsTransition metal macrocycles on high-area carbons were found to be good

electrocatalysts for 0 2 eduction in alkdiie solutions. Earlier studies showed theeffect of coordinating agents on the redox properties and 0 2 reductionelectrocatalytic activity of transition metal macrocycles in alkaline solution (12). Itis shown that CN- ions coordinate with specific valency states of the transitionmetals such as Fe and Co in phthalocyanines and porphyrins, and have a substantialpoisoning effect on02 reduction.

31. .


38/68

Fig.9. Cyclic voltammetryof K3Fe(CN)6 (-1 x 10-3 M) in 0.1 M HC104(deaerated) inthe presenceand absenceof CoTsPc adsorbedonthe electrode surfaceor insolutionat25OC.1- Voltammetric curvefor K3Fe(CN)6in0.1 M HC104 onOPG.

NoCoTsPconthesurface or insolution.2- Voltammetric curve for K3Fe(CN)6 in 0.1 M HC104 onCoTsPc preadsorbed onOPG.3 - Voltammetric curvefor K3Fe(CN)6 in0.1 M HC104 onOPGwith -1 x 10-3M CoTsPc insolution.Area ofthe electrode=0.2 cm2Scan rate=100mV/s 32


39/68

FOR CURVES 2 , 3 A NDFOR CURVE 1

SCE

I 'L2

Fig. 10. Cyclic voltammetry of tolidine (-1 x 10-3 M) on OPG in 0.1 MHC104 (deaerated) in the presence and absence of CoTsPcadsorbed on the electrode surface or insolution at 25OC.1- Voltammetric curve for tolidine in0.1 M HC104 on OPG. No

CoTsPc on the surface or insolution.2 - Voltammetric curve for tolidine in 0.1 M HC104 on CoTsPc

preadsorbed on OPG.3 Voltammetric curve for tolidine in0.1 M HC104 on QPG with-1 x 10-4M CoTsPc insolution.Area of the electrode =0.2 cm2Scan rate=100mV/s.

33


40/68

TI' -2I

Fig. 11. Cyclic voltammetry of Fe(C104)3(-1 x 10-3M) on OPG in0.1 MHC104 (deaerated) in the presence and absence of CoTsPcadsorbed on the electrode surface or insolution 25OC.1- Voltammetric curve for Fe(ClO4)gin 0.1 M HCI04 on OPG.No CoTsPc on the surface or insolution.2 - Voltammetric curve for Fe(CI04)3 in 0.1 M HC104 onCoTsPc preadsorbed on OPG.3 - Voltammetric curve for Fe(C104)3 in 0.1 M HClOq on OPGwith -1 x 10-4M CoTsPc insolution.Area of the electrode =0.2 cm 2scan rate=100 mV/s.

34


41/68

.To better understand the redox properties and 0 2 reduction electrocatalysison transition metal macrocycles, the effect of alcohols such as methanol,isobutanol and n-amyl alcohol on the electrochemical properties and 02 reductionbehavior of CoTsPc were examined. These alcohols may change the structure ofthe interface and orientation of the adsorbed macrocycle on the surface. Theeffects of methanol and its reaction intermediates on 02 reduction are also ofinterest for the identification of methanol-tolerant electrocatalysts for O2 reductionin methanol-air he1 cells.

The addition of isobutyl alcohol and n-amyl alcohol in alkaline solution (0.1M NaOH) containing CoTsPc (-2 x M) on OPG electrodes showed adecrease in the magnitude of the redox peak Co(II)/Co(I) with completesuppression of this peak at -1 M and 0.4 M concentration of the alcohols,respectively. When this electrode was washed and dipped in a fresh alkalinesolution without the macrocycle, the redox peak reappeared at about the samepotential, but decreased in magnitude (-33%) as compared to that with no alcoholin solution. No sigruficant suppression of the redox peaks was found with theaddition of methanol (52M) in solution. The half-wave potential for 0 2 eductionon CoTsPc, (preadsorbed on an OPG disk electrode fi-om alkaline solution (-1 xlo4 M)) in the presence of these alcohols is -60 mV more positive than 02reduction in the absence of these alcohols. This may be due to the increase in theadsorption of the macrocycle on the electrode surface as a result of changes in thedielectric property of the solvent containing alcohols.

02-reduction polarization measurements were also carried out in 2.5 MH2SO4 at 6OoC using porous gas-fed electrodes containing pyrolyzed macrocycles(CoTMPP and CoTAA adsorbed on high area-carbon). The presence of -1 M

35


42/68

methanol in the solution hardly affected the 0 2 reduction performance (Fig. 12).Tlhis is quite encouraging for air electrodes in acid electrolyte methanol-air fbelcell.2. Transition Metal Oxides as Catalyst and Support

2.1 0 2 Reduction and Generation on Pyrochlore-Based Gas-FedElectrodes in Acid Electrolyte

Metal oxides with the pyrochlore structure, particularly the stoichiometriclead ruthenate, Pb2R~207-~,ere found to be very active catalysts for both O2reduction and generation in alkaline solutions with carbon supports or in a self-supporting mode (i.e., without high-area carbon as catalysts support). The lead-ruthenate pyrochlores were examined both as an electrocatalyst and as a supportmaterials to Pt for 0 2 eduction and generation in acid electrolyte (2.5 M H2S04at 6OoC) (Fig. 13). A Nafion 117 (DuPont) membrane was pressed on theel'ectrolyte side of the electrode to prevent the dissolution of Pb and Ru fiom thepyrochlore in the acid solution. The performance for 0 2 reduction with thepyrochlore in gas-fed electrodes was found to be significantly lower in acid than inalkaline solution (i.e., over 200 mv more polarization at -10 mA/cm2). Previouswork (13) showed that the reaction order for O K as -0.5 over the pH range 12to1 14, but the mechanism changes in going from pH 14to 1. It was found that the0 2 generation reaction occurred almost as readiiy in acid as in alkaline solution butwith a higher Tafel slope (-80 mV/decade in acid vs. -40 mV/decade in alkalinesolution). The higher Tdel slope is not encouraging because the polarization athigher current densities becomes larger than that for the lower slope.

The 0 2 eduction performance was much better with Pt supported on apyrochlore as compared with a pyrochlore alone. Approximately 10 wt% Pt was

36


43/68

0.0 I I I I I l l 1 1 I 1 I 1 I I I I I I I 1 1 1 1 1 I 1 I 1 1 1 1 I I I I I I l l +

0.2

WU

0.4uj>W

1o

0, Reduction:2M H,SO, 6OoC, 1atm0,10wt% CoTMPP (55OoCHT) on RB carbon

after addition of -1 M CH,O

before addition of CH,OH

0.01a I 1 fi l r l r l fi L fi a f i l l l l fi I I fi a 1 1 1 1 I 1 I I I I l l 1 fi 1 I 1 1 1

0.1 1 10 100CurrentDensity, mA cmm2

lo00

Fig. 12. Steady-state polarization curves (IR-free) for 0 2reduction with porous gas-fed electrodes.

37


44/68

1.E

1.4

1.2

1o

0.8

0.6

0.4

0.2

. 0, eduction and Generation:2.5 M H,SO, /Nafion 117,6OoC,1atm0,

10wt% Pt on Pb2Ru02 7-y(without carbon support)

r

'00\0

0bn

Current Density, mA cm-*Fig. 13. Steady-state polarization curves (IR-free) for 02reduction (1) and generation (2) with porous gas-fed electrodes.

38


45/68

deposited using the Prototech method (14). The 0 2 generation performance wasalso significantly improved (-40 mV/decade Tafel slope) with deposited Pt (Fig.13). Thus there is some promise for the bifinctional 02 electrode in acidelectrolyte usingR upported on lead ruthenate pyrochlore.

Measurements of 0 2 reduction on gas-fed electrode were also carried outin concentrated KOH -150C) using Li-doped NiO as a catalyst support for Ptand Au catalysts. The Pt was deposited on the Li-NiO using the Prototechmethod, which involves the use of the Pt sulfite complex (15). The Au catalystwas prepared according to a Prototech patent, using chloroauric acid (16). ThePt/Li-NiO catalyst gave much better performance than Au/Li-NiO.

2.2 Transition Metal Oxides as Support Materials for 0 2 ReductionElectrocatalys ts

Transition metal oxides are of interest as alternative supports to carbon.Some of these oxides are quite stable in alkaline solution and have reasonableelectronic conductivity. The effect of adsorbed CoTsPc on lithiated NiO for 02reduction was investigated using the rotating disk electrode (RDE) technique andgas-fed electrodes. In the RDE experiments, mosaic single-crystal NiO andpolycrystalline NiO grown on a metallic Ni disk were used. A significantenhancement of the 02 reduction kinetic current (-15 fold) on CoTsPc wasobserved in 0.1 M KOH t -0.4 V vs. Hg/HgO, OH- at room temperature. Thepresence of CoTsPc in solution phase was necessary to obtain this effect. It wasfound that CoTsPc (negatively charged) was weakly bound to the electrodesurface when it was adsorbed from an air-saturated aqueous solution (-5 xM) with the electrode at open circuit and at low rotation rate (-100 rpm). Apositively charged macrocycle, cobalt tetra kis-(4-t1imethyl ammonium phenyl)

39


46/68

porphyrin also showed weak adsorption under similar conditions. This indicatestbat the excess negative charge (17, 18) on the WiO surface has practically noeflect on the adsorption of the macrocycle.

Another reason for the weak adsorption may be because the NiO surface iscovered by strongly bonded O K groups which prevent Ni+2 fiom binding otherligands. Efforts were made to bind to the O K group through hydrogen bondingusing small organic molecules with various functional groups. Efforts were alsomade to interpret the effect of adsorption in terms of electrophoretic properties ofthe oxide including the isoelectric point.

The isoelectric point for lithiated NiO was determined to be at pH 9.7 . , andrelported in the literature to be at pH 10.4 for undoped NiO (17). At pH>9.7 thesurface of the Li-NiO is negatively charged and this would inhibit the adsorption ofthe negatively charged (COTSPC)~s experimentally found. On the other hand,adsorption was also weak at pH 1.4. Further, a macrocycle with peripheral tetra-allrylammonium groups also adsorbed only weakly on Li-NiO in alkaline solution.Thus it appears likely that adsorption is weak due to the difficulty of displacingstrongly bound surface OH, O K or H20. Other approaches to assist in theadsorption process are being considered, including the use of ion-conductingpollymer layers.

02-reduction on CoTsPc preadsorbed on Lac003 powder obtained fromBasic Volume Ltd., England, in alkaline solution was also investigated. Thetypical surface area and particle size of this perovskite are 10-30m2/g and 0.1-0.05pnn, respectively. Thin-porous-coatings (TPC) on the electrode was used to assessthe activity. Cyclic voltammetry of the CoTsPc on Lac003 showed a smallrounded peak corresponding to the redox process Co(II)/Co(I), indicating

40


47/68

adsorption of CoTsPc on the LaCo03 substrate. The half-wave potential for 0 2reduction was -100 mV more positive for CoTsPc adsorbed on Lac003 than forsubstrate alone (Fig. 14). The limiting current approached the same value at morenegative potential (-0.9 V vs. SCE) with and without CoTsPc on LaCoO3,showing that the same number of electrons are involved in the reduction process.The positive shift of the potential for 0 2 eduction with CoTsPc on Lac003 maybe due to firther decomposition of the peroxide by CoTsPc inside the poresbecause of the slow di&sion of peroxide ions. The 0 2 eduction currents with theperovskites as substrate are about twenty times less than that with high-areacarbon as the substrate. This difference is attributed to the low surface area andlow electronic conductivity with perovskite substrate.obtained for 0 2 eduction on CoTsPc adsorbed on Lao$3-0 n O 3 .

Similar results were

Efforts were also made to prepare high area, conductive metal oxides basedon nickel. Such materials should be more stable than carbon as supports for 0 2reduction or for bibctional electrocatalysts, especially at more positive potentialsduring 0 2 generation.

Among the various methods used to prepare the high-area ransition metaloxides, a new method was recently developed at Battelle Northwest Laboratories(19). This method, which has been referred to as the glycine-nitrate method,involves a rapid combustion-like reaction between glycine and transition metalnitrates, reaching 1000-14OO0Cor just a few seconds. This method was used toprepare the perovskite LaNi03, which is a metallic conductor. The conventionalprocedure, which involves thermal decomposition of the nitrates, typically yieldspowders with BET surface areas less than 5 m2/g Using the glycine-nitratemethod, which involves a quick combustion step (15 s) and 45OoC heat treatment


48/68

-1 .o-0.8

-0.6naE- 0.4-

-0.2

0.0

8 I 8 I 8 I I 1 8 I 8

0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2E (v )vs . SCEFig. 14. Disk currents for 0 2 eduction on a thin porous coatinglayer of Lac003 (1) and CoTsPc adsorbed on Lac003

(2) on OPG surface in Orsaturated 0.1 M NaOH.Fluon (-10%) was used for the fabrication of the TPCelectrode. Curves obtained point by point in thenegative direction at room temperature.Disk area (geometrical) =0.45 ern22500 rpm

42


49/68

in air, single-phase LaNiO3 was obtained with a BET surface area of 15.3 m2/g asshown by x-ray diffraction. This method appears to show promise for perovskite-type oxides in general, according to the developers (20).

Efforts were also continued to prepare higher area forms of lithiated nickeloxide, Nil+LiXO. The efforts were hampered by the need for higher heattreatment temperatures to enable the lithium to diffuse into th e NiO particles.Undoped NiO can easily be prepared in high area form (2100m2/g). A modifiedthermal decomposition in vaccum was used with Ni(Oq2 and CH3CO,Li.2H,Oas the starting materials followed by heat treatment in air at 350% The BETsurface area was quite encouraging;-23 m2/g. The resistivity at room temperatureand 34% theoretical denisty was -20 Ocm, which is also quite encouraging. (Theresistivity decreases at higher tempeature). Since the X-ray difiaction patternshowed some evidence of particle size broadening effects, this material will becharacterized shortly using EXAFS at Stanford in conjunction with ProfessorDaniel Scherson of CRWU.

The other support materials such as WC and Sic, in the powder form, werealso used as substrates for cobalt tetramethoxy phenyl porphyrin (CoTMPP) as thecatalyst. Porous gas-fed electrode measurements for 0 2 reduction were made onthese materials in 5.5 M KOH at 25OC with and without CoTMpP on the supportand using NH4HCO3 as the pore former. The catalytic activity of CoTMPP for02reduction was inferior to the activity of CoTMPP supported on a high-area carbonwith or without heat treatment. This may be due to the larger particle size andlower surface area of these substrates as well as the poor conductivity as comparedto the high-area carbons.


50/68

III. RESEARCH RECOMM ENDATIONSThe following list summarizes research topics that should be considered:1. Studies of factors controlling stability of precious metals and macrocycle

catalysts including:- oxidation of carbon supports and ways of impeding oxidation by

surface treatments such as selective fluorination and the use of certaincarbons.- changes in the morphology and characteristics of the catalyst active

layer.- sensitivties of oxygen catalysts to the transport of components of theanolyte into the catholyte.- use of ionomers to stabilize catalyst layers such as adsorbed

macrocycles andUPD submonolayers2. Studies of factors controlling activity ofcatalysts for oxygen reduction:

- identification of intermediates, particularly adsorbed species- measurements of redox properties of reactants, intermediates and

44

products both in solution-phase and as adsorbed species3 . Establish effective ways to bind transition metal to surfaces


51/68

Iv.1.

2.3.4.

5 .

6.7.

8.9.

10.11.12.13.

14.15.16.17.

REFERENCESB. Simic-Glavaski, S. Zecevic and E. Yeager, J. Am. Chem. SOC.,107. 5725(1985).W. C. Drinkard and J. C. Bdar, Jr., J. Am, Chem. SOC.,81,4795 (1959).B. N. Achar, G. M. Fohlen and J. A. Paker, J. Polym. Sci., 201785 (1985).E. Yeager, "Oxygen Electrodes for Energy Conversion and Storage," AnnualReport (1 October 1979 - 30 September 1980), Contract No. DEACO2-77-ET25502. Prepared by CWRU for Diamond Shamrock Corporation, November1981.

E. Yeager, "Electrocatalysts for Oxygen Electrodes," Final Report, LBL No.3 1278, prepared by CWRU for LBL, October 1991.M.E. Peover and B. S. White, Electrochim Acta,11, 1061 (1966).D. T. Sawyer, G. Chiericato, Jr., C. T. Angelis, E. J. Nanni, Jr. and T. Tsuchiya,Anal. Chem., 54, 1720 (1982).M. S.Hossain, Case Western Reserve University, Ph.D. Dissertation (1986).E. L. Johnson, K. H.Pool and R. E. Hamm, Anal. Chem., 3 . 6 1 (1968).I.M. Kolhoff and T. B Reddy, J. Electrochem. SOC., 08,980 (1961).R. S.Nicholson, Anal. Chem., 3 1351 (1965).S.Gupta. C. Fierro and E. Yeager, J. Electroanal. Chem., 306.239 (1991).E. Yeager, "Electrocatalysts for 0 2 Electrodes," Annual Report, LBL No. 26054,prepared by CWRU for LBL, August 1988.H. G. Petrow and R. J. Allen, U.S. Patent 3,992,33 1, November 16, 1976.H. G. Petrow and R. J. Allen, U.S. Patent 4,044,192, August 23, 1977.H. G. Petrow, U.S. Patent 3,776,776, December 4, 1973.G.A. Parks. Chem. Rev.. 65. 177 (1965).

45


52/68

18. G.. . Parks and P. L. deBruyn,J. Phys. Chem.,66.967 (1962).19. L. A. Chick, L. R Pederson, G. D. Maupin, J. L. Bates, L. E. Thomas and G. J.

Exarhos, Mat. Let.,10.6(1990.20. J. L. Bates, personal communication.

46


53/68

V. LIST OF RESEARCH PUBLICATIONS UNDER LBL/DOE PROJECT

1. "The Electrochemical Properties of Graphite and Carbon,," E. Yeager, J. A. Molla and S .Gupta, Proceedings of the Workshop on The Electrochemistry of Carbon, Cleveland,Ohio, August 17-19, 1983, S. Sarangapani, J. R Akridge and B. Schumm, eds., TheElectrochemical Society, Pennington, NJ, 1984,pp. 123-157,

2. "Effect of Surface Treatment of Glassy Carbon on 02Reduction in Alkaline Solution,,"Z.W. Zhang, D. A. Tryk and E. Yeager, Proceedings of the Workshop on theElectrochemistry of Carbon, Cleveland, Ohio, August 17-19, 1983, S. Sarangapani, J. R.Akridge and B. Schumm, eds., The Electrochemical Society, Pennington, NJ, 1984, pp.158-178.

3. "Carbon in Bifinctional Air Electrodes in Alkaline Solution," D. Tryk, W. Aldred and E.Yeager, Proceedings of the Workshop on the Electrochemistry of Carbon, Cleveland,Ohio, August 17-19, 1983, S. Sarangapani, J. R. Akridge and B. Schumm, eds., TheElectrochemical Society, Pennington, NJ, 1984, pp. 192-220.

4. "Mechanisms for the Hydrogen and Oxygen Electrochemical Generation Reactions,"Ernest Yeager and Donald Tryk, Hvdrogen Energy Progress, Vol. V, T. N. Veziroglu andJ. B. Taylor, eds., International Association for Hydrogen Energy, Pergamon Press, NewYork, 1984, pp. 827-844.

5. "In-situ Mossbauer Spectroscopy on an Operating Fuel Cell," D. Scherson, C. Fierro, E.Yeager, M. E. Kordesch, J. Eldridge, R. W. Hoffman and A. Barnes, J. Electroanal.Chem., 169.287-302 (1984).

6 . "Electrocatalysts for 0 2 Reduction,," E. Yeager, Electrochim. Acta, 29, 1527-1537(1984).

7. "Mechanistic Aspects of Oxygen Electrochemistry," E. Yeager, D. Scherson, and B.Simic-Glavaski, Proceedings of the Symposium on The Chemistry and Physics ofElectrocatalysis, San Francisco, California, May 8-13, 1983, J.D.E. McIntyre, M. J.Weaver, and E. Yeager, eds., The Electrochemical Society, Pennington, NJ, 1984, Vol.84-12, pp. 247-270.

8. "In-situ Mossbauer Spectroscopy and Electrochemical Studies of the Thermal Stability ofIron Phthalocyanine Dispersed in High Surface Area Carbon,," D. A. Scherson, C. A.Fierro, D. Tryk, S . L. Gupta and E. Yeager, J. Electroanal. Chem., 184.419-426 (1985).

9. "Sauerstoffreduktionselektroden ir Elektrolyse - und BrennstofEellen," R. Holze and E.Yeager, Technische Elektrolysen, Dechema Monograph, Vol. 98.No. 1993-2030, VerlagChemie, Weinheim, 1985, pp 499-509.

47


54/68

10.

11.

12.

13.

14.

15.

16.

17.

18 .

19.

"In-situ Spectroscopic Studies of Oxygen Electrocatalysts Involving Transition MetalMacrocycles," E. Yeager, D. Scherson, and C. Fierro, in Catalyst CharacterizationScience: Surface and Solid State Chemistry, M.L. Deviney and J.L. Gland, eds., ACSSymposium Series288, American Chemical Society, Washington, DC, 1985, pp. 537-549."Spectroscopic and Electrochemical Studies of Transition-Metal TetrasulfonatedlPhthalocyanines. 3. Raman Scattering fiom Electrochemically Adsorbed TetrasulfonatedPhthalocyanines on Silver Electrodes," B. Simic- Glavaski, S . Zecevic and E. Yeager, J.Am. Chem. SOC., 07.5625-5635 (1985)."Spectroscopic and Electrochemical Studies of Transition-Metal TetrasulfonatedPhthalocyanines." 6. The adsorption of Iron Tetrasulfonated Phthalocyanine on SingleCrystal Silver Electrodes, R. Adzic, B. SimioGlavaski and E. Yeager, J. Electroanal.Chem.,194 155-163 (1985)."Spectroscopic and Electrochemical Studies of Transition Metal TetrasulfonatedPhthalocyanines. 5. Voltammetric Studies of Adsorbed Tetrasulfonated Phthalocyanines(MTiPc) in Aqueous Solutions," S . Zecevic, B. Simic-Glavaski, E. Yeager, kB.P. Leverand P.C. Minor, J. Electroanal. Chem.,196(1985) 339-358."Catalytic Activity of Manganese Dioxide for Hydrogen Peroxide Decomposition inAlkaline Solutions," Y. Zalyoksnis, D. Tryk and E. Yeager, Proceedings of the 2ndInternational Battery Material Association, Vol. 2, Grm,1985, pp. 467-476."Transition Metal Macrocycles Supported on High Area Carbon: Pyrolysis-MassSpectrometry Studies," D. Scherson, A. A. Tanaka, S . L. Gupta, D. Tryk, C. Fierro, R.Holze, E. Yeager and R. P. Lattimer, Electrochim. Acta,31.1247-1258 (1986)."Thermal Decomposition of Surfactants in PTFE-Dispersions," R. Holze, A. T. Riga andE. Yeager, J. Materials Science Letters, 5,819-822 (1986)."Dioxygen Electrocatalysis:Yeagtx, J. Molecular Catalysis, 38,5-25 (1 986).Mechanisms in Relation to Catalyst Structure," Ernest

"Heat-Treated Pyrrole Black-Based Electrocatalysts for 0 2 Reduction,," S. Gupta, D.Tryk, M. Daroux, W. Aldred and E. Yeager, in "Proceedings of the Symposium on LoadLevelllig and Energy Conservation in Industrial Processes," D. Chin, Editor, TheElectrochemical Society, Pennington, NJ, 86-IO, 207-218 (1986)."E1ed:rocatalyticAspects of Iron Phthalocyanine and Its pox0 Derivatives Dispersed onHigh Surface Area Carbon,," A. A. Tanaka, C. Fierro, D. Scherson and E. B. Yeager, J.Phys. Chem., 2,3799-3807 (1987).

48


55/68

20. 'lSpectroscopic and Electrochemical Studies of Transition-Metal TetrasulfonatedPhthalocya-nines. Part VIII. Spectroelectrochemical Studies of Stacked-Ring SiliconPhthalocyanines," B. Simic-Glavaski, A. A. Tanaka, M. E. Kenney and E. Yeager, J.Electroanal Chem., 229.285-296 (1987).

21. "TheUse of Electrochemical and Non-Electrochemical Methods for the Determination ofPeroxide Decomposition Rate Constants," R. E. Carbonio, D. Tryk and E. Yeager,Proceedings of the Symposium on Electrode Materials and Processes for EnergyConversion and Storage, S. Srinivasan, S. Wagner and H. Wroblowa, eds., TheElectrochemical Society, Pennington, NJ, Vol. 87-12, pp. 238-255 (1987).

22. "Electrochemistry of Iron Oxides: An in-situ Mossbauer Study," C. Fierro, R. E.Carbonio, D. Scherson and E. Yeager, J. Phys. Chem.,91 579-6581, (1987).

23. "Perovskite-Type Oxides: Oxygen Electrocatalysis and Bulk Structure," R. E. Carbonio,C. Fierro, D. Tryk, D. Scherson and E. Yeager, J. Power Sources, 2 387-398 (1988).

24. "In-situ Mossbauer Effect of a Spectroscopy Model Iron Perovskite Electrocatalyst," C.Fierro, R. E. Carbonio, D. Scherson and E. Yeager, Electrochim. Acta, 33, 941-945(1988).

25. "Electrochemistry of Carbon Relative to Batteries and Fuel Cells," E. Yeager and D. Tryk,Proceedings of the 33rd International Power Sources Symposium, Cherry Hill, PA, June13-16,1988, The Electrochemical Society, Peimington, NJ, 1988, pp. 114-128.

26. "Heat-Treated Polyacrylonitrile-Based Catalysts For Oxygen Electroreduction,,"S.Gupta,D. Tryk, I. Bae, W. Aldred and E. Yeager, J. Applied Electrochem., 19 19-27 (1989).

27. "Oxygen Reduction on Adsorbed Irdn Tetrapyridinoporphyrazine," A. A. Tanaka, C.Fierro, D. A. Scherson and E. Yeager, Materials Chemistry and Physics, 22, 431-456(1989).

28. "Examination of the IonomerElectrode Interface using the Ferric/Femous Redox Couple,"D. Chu,D. Tryk, D. Gervasio and E. Yeager, J. Electroanal. Chem., 272.277-284 (1989).29. "The Electrochemistry of Graphite and Modified Graphite Surfaces: The Reduction of

02,"M. S.Hossain, D. Tryk and E. Yeager, Electrochim. Acta, 34, 1733-1737 (1989).30. "Electroreduction of Oxygen on Pyrolytic Graphite Electrode with Adsorbed Layer ofCobalt TetrasuKonated Phthalocyanine in Acetronitrile and Dimethylfomamide," H.

Huang, D. Tryk and E. Yeager, Chem. J. Chinese Univ. 5,164-170 (1989).31. "Infrared Reflectance Absorption Spectroscopy (IRRAS). Study of the Thermal Stability

of Perfluorinated Sulphonic Acid Ionomers on Pt,"D. Chu, D. Gervasio, M. Razaq and E.Yeager, J. Appl. Electrochem.20, 157-162 (1990).

49


56/68

32.

33.

34.

35./

36.

37.

38.

"Infrared Spectroscopic Determination of pH Changes in Diffusionally Decoupled Thin-Layer Electrochemical Cells," I. T. Bae, D. Scherson and E. Yeager, J. Anal. Chem.,a.,45 -491 (1 990)."Eleatrocatalysts for Oxygen Electrode in Fuel Cells and Water Electrolyzers for SpaceApplications," J. Prakash, D. Tryk and E. Yeager, J. Power Sources, 2 413-422 (1990)."ESCA and Electrochemical Studies of Chelate-Modified Carbon Materials for Cathode-glen Reduction," D. Ohms, S.Gupta, D. A. Tryk, E. Yeager and K. Wiesener, Z.Physilk. Chem., 272,451-459 (1990)."The Effects of Cyanide on the Electrochemical Properties of Transition MetalMacrocycles for Oxygen Reduction in Alkaline Solutions," S.Gupta, C. Fierro and E.Yeager, J. Electroanal. Chem., 306,239-250 (1991)."In-Situ Infrared Spectroscopic Studies of Redox Active Self-Assembled Monolayers onGold Electrode Surfaces," I. T. Bae, H. Huang, E. Yeager and D. A. Scherson, Langmuir,-,15!58-1562 (1991)."Oxygen Reduction on Poly(4-~inylpyridine)-Modified Ordinary Pyrolytic GraphiteElectrodes with Adsorbed Coablt Tetra-sulphonated Phthalocyanine in Acid Solutions," Z.Y. Zeng, S.L. Gupta, H. Huang and E. Yeager, J. Appl. Electrochem., 3,973-981(1991)."Studies of the Adsorption of Tetrasulfonated Phthalocyanines on Graphite Substrate," S.Gupta, H. Huang and E. Yeager, Electrochim. Acta, 36,2165-2169 (1991).

50


57/68

VI.

1.

2.

3 .

4.

5 .

6.

7.

Ph.D. DISSERTATIONS UNDER LBL/DOE PROJECT

C. Fierro,

S. Hossain,

A. Tanaka,

D. Chu,

I. Bae,

J. Prakash,

H. Huang,

"Transition Metal Macrocycles as Catalysts for the ElectrochemicalReduction of 02," Case Western Reserve University, May, 1985."Oxygen Reduction on Carbon and Modified Carbon Sufaces,"Case Western Reserve University, May, 1986."Electrochemical Properties of Transition Metal Macrocycles forOxygen Reduction," Case Western Reserve University, May, 1987"Studies of the Solid Polymer Electrolyte-Electrode Interface,"Case Western Reserve University, May, 1989."In-situ Infrared Spectroscopic Studies of ElectrocatalyticSystems," Case Western Reserve University, August, 1989."Oxygen Electrocatalysis on Ruthenium Metal and its Pyrochlores,"Case Western Reserve University, June, 1990."The Role of Adsorbed Species in Various ElectrocatalyticProcesses: Electrochemical and In-situ Infrared SpectroscopicStudies," Case Western Reserve University, May, January 1992.

51


58/68

Appendix AThis gives the list of various active catalyst systems prepared and examined for

0 2 reductionand generation in alkaline and acid electrolyteswith porous 02-fed (1 atm)electrodes.

A- 1


59/68

TABLE A-I0 2 Reduction Polarization for Various Catalyst Systems in4M NaOH at 6OoC with Porous 02-Fed (1 atm) Electrodes

Potential in mVvs Hg/HgO, O H

S . No. Catalyst Systems Red. Pot. (mv)at -100 mA em-*

1.2.3.4.5 .6.7.8.

9.

10.

11.12.

13.

Transition Metal Macrocycles

10% CoTMPPDRB (45OoCHT)10% CoTPPfllRB (450OCHT)20% FeTMPPDRB (80OoCHT)15%CoTAA/P-33 (650OCHT)5% CoTMPP/SASB (450OC HT)8.9% CoTMPP/SASB (80OoCHT)20% CoTMPPBP (80OoCHT)4.8% CoTMPP/XC-72 (45OOC HT)

I2TMPP+Metal Oxides3% H2TMPP/SB (450OCHT)+100! Ru and Co as4.4% H2TMPP/XC-72+2.5% Fe and Co ashydroxides

hydroxides (450OC HT)Naturally Occurring Macrocycles

10%Hemine/DU3 (80OoC HT)10% ChlorophyllDRB+2.5% Co as hydroxide

(450OC 3")40% PAN /XC-72 +lo% CO(OAC)~8OOOCHT)

-47-81-38-5 1-70-65-70-75

-42

-85

-73

-92-75

A-2


60/68

TABLEA-I(continued)

0 2 Reduction Polarization for Various Catalyst Systems in4M NaOH at 6OoCwith Porous 02-Fed (1 atm) Electrodes

Potential in mV vs Hg/HgO, OK

S. No. Catalyst Systems Red. Pot. (mv) at-100 m~ cm-2

14.15.16.

17.18.19.

20.

21.

Macrocycles+PerovskitesPB+2.5% CO(OAC)~7OOOC HT)PB+2.5% Fe as acetate (85OoCHT)3% CoTMPP/SB (450OC HT) +3% CoTMPP/SB (450OC HT)+L Q . ~ S ~ O . ~ C O O ~3% CoTMPP/SB (45OoCHT)+La-Fe Perovskite

L%.5sr0.5c00.&%. 13

5% CoOEP/XC-72 (400OC HT)+L%.8sr0.2~~0.dRuo.03

15% COTAA/P-33 (65OOC HT)+L%.5sr0.5c00.8Rb.203

3% H2 TMPP/SB (450OCHT) +L%.8sr0.2c00.9Rb. 13

-90-88-5 8-67-68-45

-48

-56

A-3


61/68

TABLEA-I(continued)

0 2 Reduction Polarization for Various Catalyst Systems in4M NaOH at 60C with Porous 02-Fed (1 atm) Electrodes

Potential in mV vs Hg/HgO, O K

S. NO^. Catalyst Systems Red. Pot. (mV)at -100 mA cm-2

Polymers+Perovskites22. 40% PANBC-72+5% CO(OAC)~800OC HT)+

L%.8sr0.2c00.9Ru0. 13 -6423. 40% PAN/SASB+5% CO(OAC)~(SOO"CFT) +

LaF%. 1Nb.9O3 -7224. 40% PAN/SASB +5% CO(OAC)~800OC HT)+

Platinum25. 10%Pt/XC-72 (Prototech electrode)26. 5% Pt/DRB (Electromedia electrode)

DRB ==De-ashed RB carbon (Calgon)SB=Shawinigan black (Gulf)SASB=Steam activated SB (Electromedia)P-33 (IEast Germany)XC-72, (Calgon)BP =I3lack Pearls (Cabot)CoTMPP=Cobalt tetramethoxy phenyl porphyrinH2TMPP=Metal-fiee tetramethoxyphenyl porphyrinLBL =Lawrence Berkeley Laboratory carbon

-68

-53-5 1

TAA=tetraazaannuleneTPP=tetraphenyl porphyrinPAN =PolyacrylonitrilePAA=PolyacrylamidePB =Pyrrole blackRAI =Anion exchange membraneCo-OEP =Cobalt octaethyl porphyrinDMDAAC =Dimethyldiallyl ammoniumchloride

A-4


62/68

TABLE A-II

0 2 Reduction and Generation Polarization Datafor Various Catalysts in 5.5M KOH at 22OC

with Porous 02-Fed (1 atm) Electrodes

Potentials in mVvs.Hg/HgO,OK

S. No. Catalyst Systems* Red. Pot. (mv) at OHi. Pot. (mv) -100 mA cme2 +IO m~ cm-2

1.2.3.

4.5.

6.

7.8.9.10.11.12.

13.

Pb2Ru206.5/SB air oxidized)Pb2(Ru1.67Pb0.33)06.5/SBair oxidized)Pb2(Ru1.67Pb~.33)06.5/SBair oxidized)/RAI

membrane on E.S.pb2(Ru167Pb0.33)06.5/COTMPP/SBpb2(Ru1 67pb0.33)06. .5/c0TMpp/sB/RAI

membrane on E.S.

DMDAAChJafionPb2(Ru1.6Pb0.33)/SB (air oxidized)/

Pb2Ru206.5( Self supported)Pb2Rul.51r0.507-y/SB (air oxidized)Pb1.5C00.5Ru206.5/SB (air oxidized)3% CoTMPP/SB (45OOC HT)3% Co"MPP/SB (450OCHT)+Pb2kl206.53% H2TMPP/SB (450OCHT)+10% CO,RUas hydroxides

6.5% Fe +Ni ashydroxides3% H2TMPP/SB ( 4 5 W HT)+10%CO,

-75-16

-23-32

-3 8

-15+10-3 1-60-67-35

-60

-75

474477

442447

449

462459483472-

-

-

*For notations, seeTable A-I.

A-5


63/68

TABLEA-II(continued)

0 2 Reduction and Generation Polarization Datafor Various Catalysts in 5.5M KOH at 22OCwith Porous 02-Fed (1 atm) Electrodes

Potentials in mV vs. Hg/HgO, O K

S. No. Catalyst Systems* Red. Pot. (mV) t-100 mA cm-2 Oxi. Pot. (mv) a+IO m~ cm-2

14.

15 .

16.

17.

18.19.20.

21.22.

23.

24.25.

3% H2TMPP/SB (4500C HT)+Fe, Ni, Co3% H2TMPP/SB (45OOC HT) +perovskite3% H2TMPP/SB (4500 CHT) +perovskites:

%.5sr0.5c00.5~.503 4- Ni0.5R%.5La033% CoTMPP/SB (45OOC HT) +perovskite

perovskites

L%.8sr0.2c00.9R%.103

L%.8sr0.2c00.9R~.13410%PAN+5% CO(OAC)~XC-72 (800OC HT)5% CoOEP/XC-72 (400% HT)5% CoOEP/SASB (NoHT)+perovskite1daF%.15x0.8503/DRB (No HT)

LaF%,15N6.850320% PAA+40% PAN+5% CO(OAC)~XC-72

(8000C HT)3% CoTMPP/SB (450OC HT) +pyrochlore1.5% CoTMPP/SB (450OC HT)1.5% CoTMPP/SB (450OC HT)+pyrochlore

pb15c00.5Ru206.5

Bi1.5C00.5RLlI*7-y

-87

-65-75

-85-75-85

-75-65

-70

-58-70

-66

606---598I--612-I495-574

*for notations, see Table A-I.

A-6


64/68

TABLE A-II(continued)

0 2 Reduction and Generation Polarization Datafor Various Catalysts in 5.5M KO H at 22OCwith Porous 02-Fed (1 atm) Electrodes

Potentials in mV vs. Hg/HgO, O K

S. No. Catalyst Systems Red. Pot. (mv) at Oxi. Pot. (mv)-100 mA cm-2 +IO m~ cm-

26. Pb2R~206.5/SBair oxidized)/RAI membrane onthe E.S. -20

27. Pb2RU206.5+4Hco3 (self supported) +2028. 1PhPt (Prototech)/Pb2Ru206-5-tN&Hco3 -1029. 10%PtLBLcarbon (Prototech method) -65

*for notations, see Table A-I.

A-7

464-


65/68

TABLE A-ID0 2 Reduction Polarization Data for Various Catalysts in 85%

H~PO.Qt 100C with Porous 02-Fed (1 atm) ElectrodesPotential in mV vs RHE

S.No, Catalysts* Red. Pot. (mV) t-100 mA cm-2

1.2.3.4.5.6 .7.8.9.10.11.12.13.14.

15.16.

:LO% FeTMPPDRE3 (80OOC HT)4.75% CoTMPP/SASB (450OC HT)+Fluon+Nafion5% FeTMJ?P/XC-72 (80OoCHT)+Fluon+Nafion10% CoTMPPDRB (SOO0CHT)9.3% CoTMPP/SASB (80OoCHT)+Fluon+Nafion110%HemidDRE3 (800C HT)li0% WConsel (Powercat 2000)110% Pt/XC-72 (Prototech)YO% Pt/XC-72 +SASB (CWRU Fabrication)1.0% Pt/XC-72 (Prototech) (CWRU Fabrication)5% WDRB (EMC) (CWRU Fabrication)0.5 mgWcm2 (commercial, Prototech)0.5mg Pt/cm2 (commercial, Prototech) impregnatedPowercat 2000 (CWRU Fabrication)0.5 mg Pt/cm2 (commercial, Prototech) impregnated

115% C O T M - 3 3 65OOC HT)

with Nafion

with DOW compound

680580570470670600650640770710770650730

700760

730

*For notaitions, see Table A-I.

A-8


66/68

TABLE A-IV

O2 Reduction Data for Catalysts in2.5 M H2S04 and Nafion Membrane at 60OC

Potential in m V sRBE

S. No. Catalysts BET Surface Red. Pot. (mv)Area (m2g-l) at -100 mA cm-2

1.2.3.

4.5.6.7.

8.9.10.

11.12.

Macrocycles (w/o Nafion)a15% C o T M - 33 (63OC HT)20% CoTAAlBP (630OC HT)10% CoTMPPDIB (8OOOC HT)

Pyrochlores with carbonbPb2RU2O6.5 -tSBPb2RU206.5 +SBpb2Ru206.5 +SBPb2Ru206.5+ SB

Pt doped pyrochloresb

pb2Ru1 8%.2O6.5pb2Ru 1.SPb.O6.5pb2Ru1.Ph.O6.5

Pt on supportsbPrototechPt10%Pt/Pb2b206.5

-1 000-1400-1200

115650

3.5

6848

3.1

-11

680700600

475475448383

405278204

806754

aat25OCbPyrochlore with different BET surface area

A-9


67/68

AppendixBThis gives a list ofvarious catalysts prepared and examinedforthe decomposition

of peroxide in alkaline electrolytes.

B-1


68/68

TABLEB-IPeroxide Decomposition RateConstants or SelectedCatalystsin Alkaline Solutiona

a. 0.1 M NaOH.b. The gasometric method was used. The sample was dispersed using ultrasonic agitation in 45 cm3NaOHsolutioxi for 1min.and then stirred with a magnetic stir bar for 5 min. Then 5 cm3 f 2 MH202 solutionw as adtied to the suspension,and he volume of evolved0 2was monitored as a functionof time. The InCHO-was plotted as a function of time, and the initial slopewas used to evdutate the rate constants.

electrocatalysts for oxygen electrodes

Documents