- Email: [email protected]
A selective catalyst for titanium anodes
Journal of Molecular Catalysis, 38 (1986) 81 - 88
81
A SELECTIVE CATALYST FOR TITANIUM ANODES N. V. KRSTAJIC Faculty of Technology
and Metallurgy, University of Belgmde, Belgrade (Yugoslavia)
M. D. SPASOJEVIb Faculty of Agriculture,
University of Kragujevac, fiaEak (Yugoslavia)
and M. M. JAKfb? Faculty of Agriculture,
University of Belgmde, Belgmde (Yugoslavia)
Summary A composite electrocatalytic coating with improved anionic selective features for chlorine evolution has been developed. This coating is intended for both chlorate cells and sea water electrolysis where oxygen evolution is undesirable. An optimized composition (14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd and 34 atm.% Sn) has now been shown to combine high activity for chlorine evolution, good corrosion resistance and low activity for oxygen evolution, both in chlorate cells and sea water electrolysis. The high selectivity stems from high oxygen overpotential as well as from the adsorption behaviour.
Introduction RuOz/TiOz-coated titanium electrodes are now a unique anode material used worldwide for caustic-chlorine, chlorate and sea water electrolysis cells [ 1 - 41. In addition, certain industrial processes need even more selective electrocatalysts to provide maximal efficiency. For instance, both chlorate production and sea water electrolysis have competing electrode reactions as follows : Anode: 2 Cl- -
Cl* + 2e
Cathode: 2 Hz0 + 2e Solution: Clz + Hz0 HOC1 e
Hz + 2 OHHOC1 + Cl- + H+
OCl- + H+
2 HOC1 + OCl- 0304-5102/86/$3.50
(A) (R) (C) (D)
Cl03 + 2Cl- + 2 H+
(E)
@ Elsevier Sequoia/Printed in The Netherlands
82
Anodic loss reactions: 6OCll+ 3Hz0 2H,O--02+4H++4e
2C10,
+ 4Cl- + 6H+ + 3/2 O2 + 6e
(F) (G)
In a chlorate cell, the hypochlorite species undergo a chemical conversion to chlorate, which gives maximal current efficiency [ 51. Reaction (G) plays a minor role, as long as the solution does not become too alkaline or the chloride concentration too low [6]. The anodic chlorate formation (F) and the reduction of the hypochlorite are controlled by the mass transport of the hypochlorite from the bulk solution toward the anode and cathode respectively [ 7, 81. These reactions are suppressed in industrial chlorate cells by addition of dichromate into the solution. Evolution of oxygen by electrolysis of water (G) can not be ignored in dilute NaCl solutions used for sea water electrolysis, because the reversible potentials for chlorine and oxygen discharge are close together. Therefore a selective electrocatalyst is needed to suppress reaction (G) by providing a high oxygen overpotential. In addition, the catalyst must resist corrosion at high anodic current density in solutions containing chloride, chlorine and hypochlorite. Oxide catalysts usually produce oxygen more easily than the corresponding bare metal (similarity factor), Ruthenium is one of the less expensive and more available precious metals, and its mixed oxide with titanium (RuO,/TiOz) is a highly active catalyst for chlorine evolution while being quite durable [ 91. For these reasons it has been widely employed for coating titanium anodes. Chloride ions inhibit the formation of surface oxides on platinum group metals by being adsorbed onto the surface [lo], with the exception of platinum itself which forms surface oxides and thereby exhibits a higher anodic polarization [ 111. Among base metals, tin has a remarkable overpotential [12] but corrodes in solutions containing hypochlorites [13]. A palladium-tin compounds (PdSn,), however, is expected to provide both selectivity (high oxygen overpotential) and a lower corrosion rate. Trassati [14] points out the ability of transition metal ions to form covalent bonds with a metal surface blocking. Phosphate ions are known to adsorb onto palladium, and thus might contribute to its selectivity as a catalyst. Palladium itself could perhaps be the sole electrocatalyst, but RuOz/TiOz is cheaper, more active and more resistant to corrosion. Thus a composite coating of RuOz/TiOz and PdSnz has been optimized as an anionic selective catalyst for chlorate cells and sea water electrolysis. The purpose of the present paper is to describe the catalytic properties, polarization characteristics, corrosion stability and lifetime of this selective catalyst. Experimental The procedure for applying a catalytic coating onto a titanium substrate has been described in detail elsewhere [ 151. The difference between
83
the usual RuOz/TiOz coating and this one is the addition of tin and palladium chlorides at various ratios. The materials, RuCls*3H,O, PdC$ (Johnson and Matthey), TiC& and SnC14 (Merck) were of reagent grade purity and were applied as an isopropanol solution containing 10 mg ml-’ of the metals, mixed in different ratios. Usually five coatings were painted onto the titanium substrate to obtain a loading of 10 g mV2at 450 “C. Figure 1 shows the electrochemical apparatus used to measure the electrode potentials. The electrolytic cell, a stirrer, thermostat and pH probe were immersed in the solution in a glass vessel. For chlorate production, the vessel contained 2 dm3 of 300 g NaCl dme3 with added sodium dichromate or Na2HP04 + NaH2P04 or both, all of reagent grade purity. The cell was small, and essentially acted as a generator of available chlorine, while the large retention volume of- the reactor served for its conversion to chlorate. For sea water electrolysis, the reactor contained 50 dm3 of electrolyte to slow down concentration changes. In both cases, the temperature and pH were automatically maintained at the desired set points. Rapid flow was generated through the cell by the gas lift of the cathodic hydrogen. This high recirculation ensured that the change in concentration through the cell was very small (about one part in 800). The production rate and gas composition were followed by means of an Orsat apparatus [16]. To assess cathodic losses, the rate of Hz evolution was measured for a caustic cell which was operated in parallel at the same current density. Available chlorine content was determined by potentiometric titration with sodium arsenite [ 171.
Fig. 1. The set-up for simultaneous investigations of electrode activity and Faradaic yields in the chlorate cell and sea water electrolysis: C: cell, HV: holding volume, M: stirrer, T: temperature control thermostat, P: automatic titrator-p’tf-state, 0: Orsat apparatus for gas analysis, A: multi-range miiliammeter, Re: rectifier, R: reference SCE, V: VTVM.
Results and discussion Crystal structures were determined for various mole ratios of Ru02, Ti02 and PdSnz by X-ray diffraction analysis. Figure 2 shows a typical curve
84
Ti
Fig. 2. X-ray diffraction radiograph for the composite electrocatalyst with optimal phase structure and catalytic activity (Ru: 14 atm.%, Ti: 35 atm.%, Pd: 17 atm.% and Sn: 34 atm.%).
for satisfactory coating. The peaks correspond to reflection from the titanium substrate, from rutile solid state solutions of RuOz and TiOz and from PdSn, [ 131. The rutile peaks are between those of pure RuOz and pure TiOz. Ruthenium dioxide always exists in a rutile solid solution with TiO? and thus forms a phase separate from the other constituents. Palladium and tin may form a number of phases, i.e. PdO, PdSn, PdSn, and SnO, as the Sn/Pd ratio increases. Only when the atomic ratio is between 1 and 2 .do PdSn and PdSnz appear. Corrosion tests have shown that whenever the ratio is greater than 2, Sn02 appears and the coating stability decreases.
Coating composition and activity Polarization characteristics were measured as a function of coating composition. Figure 3 shows the effect of various ratios of Pd/Sn at constant Ru and Ti contents. Any increases in Sn above 34/17 causes a pronounced decrease in catalytic activity, while a decrease below this level gives no improvement. Figure 4 summarizes the effect of composition. The potential at 1 kA m-* is plotted against the ratio of (Ru + Pd) in the coating. Both the
Fig. 3. Polarization characteristics of the composite catalytic coating at constant Ru/Ti atomic fraction (14 atm.% Ru, 35 atm.% Ti) as a function of Pd/Sn atomic ratio for chlorine evolution in brine (NaCl 300 g dmp3, pH 2.0, 80 “C); A: 3 atm.% Pd, 48 atm.% Sn; n : 7 atm.% Pd, 44 atm.% Sn; 0: 12 atm.% Pd, 39 atm.% Sn; 0: 17 atm.% Pd, 34 atm.% Sn; 0: 27 atm.% Pd, 24 atm.% Sn.
85
0.0
Lo.0
8oD
ot%iRu+~
Fig. 4. The effect of precious metals (Pd + Ru) content in electrocatalytic coating on titanium anode potential for chlorine evolution in brine at constant current density (NaCI) 300 g dmA3, 80 “C, pH 2.0 and j = 1.0 kA md2); 0: 35 atm.% Ti, 17 atm.% Pd, 48 atm.% (Ru+Sn);~:35atm.%Ti,14atm.%Ru,51atm.%(Pd+Sn);~:100atm.%(Ru+Ti).
standard RuOz/TiOz (cf. [15]) and the composite coating need about 30 atm.% noble metals. Below that level, the activity decreases faster when the coating contains tin. SnOz as a separate phase reduces the conductivity, the activity and the durability of the coating, while at higher palladium levels, the stable and active but expensive PdSn is formed. Baking temperature effects At the optimal coating composition (atm.% : Ru = 14. Ti = 35, Pd = 1’7, Sn = 34), heating between 300 and 600 “C gives the separate well-distinguished phases of rutile RuOz/TiOz and PdSn,. Below 300 ‘C, the diffraction peaks are small, pointing to incomplete conversion to these phases and hence to the presence of residual chloride ions. Above 700 “C, the peaks are split and the RuO, and TiOz become separate rutile phases, while PdSn, remains unaltered (cf. [3, IS]). Figure 5 shows the effect of temperature on the anode potential during chlorine evolution at 3 kA rnm2,both for the composite coating and the regular Ru02/Ti02 coating. The optimal temperature range is 450 - 550 “C. Selectivity The selectivity was evaluated at both chlorate-producing and sea water electrolysis conditions. Although chlorine evolution is the main anodic reaction, at pH values above 7 and especially at low chloride concentrations,
86
t(“cl
1d.cp+,cl&Old6?)
Fig. 5. The effect of temperature for thermal processing of titanium anodes on their potential for chlorine evolution at constant current density (j = 3 kA mp2) and for optimal catalyst compositions; 0: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn; 0: 40 mol,% RuO2,60 mol.% Ti02. Electrolysis conditions: NaCl 300 g dmp3; pH 2,0; 80 OC; (final processing time for catalytic coating, 60 min). Fig. 6. Oxygen content in the cell gas recalculated for maximal hydrogen current yields in diluted brine (NaCl 30 g dmV3) electrolysis plotted as a function of available chlorine concentration in the bulk at constant current density 0’ = 1.0 kA mm2) and low temperature (25 “C, pH 7 - 9); 0: 40 mol.% Ru02, 60 mol.% TiOz, 0: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn.
oxygen evolution becomes significant due to the discharge of hypochlorite (reaction (F)), and to some extent to the discharge of water (reaction (G)). The optimized composite catalyst was compared to the regular Ru02/ Ti02 rutile coating at identical experimental conditions (Fig. 6). The oxygen readings were corrected to the values they would have had at 100% efficiency. As the hypochlorite increases to 0.01 molar, there is a rapid drop in percent oxygen, especially for the composite coating. Above this concentration, the oxygen increases and eventually rises linearly with the hypochlorite concentration, with both coatings having the same rate of increase. This behaviour indicates that there are two oxygen evolution processes. One is the discharge of hypochlorite, its rate being controlled by mass transport [6] rather than by the nature of the catalyst. The other is the discharge of water, and its rate is controlled by the nature of the catalyst. The presence of hypochlorite blocks those sites where water is discharged, decreasing its contribution to A for the composite coating and A’ for the regular coating at maximal coverage. The rapid decrease in oxygen in the low hypochlorite region resembles a typical adsorption isotherm in the reverse position, thus supporting the explanation proposed above. The same selectivity is shown for chlorate cells in the presence of phosphate buffers [19]. At high hypochlorite concentrations characteristic of alkaline brine solutions, the rate of oxygen evolution controlled by mass transport can be very high (Fig. 7). In phosphate-free brine, oxygen evolution is directly proportional to hypochlorite concentration, as expected for mass transfer rate control. In the presence
87
Iu
Fig. 7. Dependence of oxygen gas content in the chlorate cell on the available chlorine concentration at constant current density (j = 3.0 kA mm2) in presence and absence of phosphate buffer for the composite catalytic coating (14 atm.% Ru, 35 atm.% Ti, 17 atm. % Pd, 34 atm.% Sn); 0: sodium dichromate buffer (3 g dmm3), 0: equimolar primary and secondary sodium phosphate buffer system (3 g dmM3). Fig. 8. The time function of available chlorine current efficiency (7) and the anode potential for two catalytic coatings at constant current density (j = 1.0 kA me2) in diluted brine (NaCl 30 g dme3, pH 6.5 - 7.6, NaClO 1.0 X 10m2 mol dmV3) at low temperature (25 “C); 0 and A: 40 mol.% Ru02, 60 mol.% Ti02; l and A: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn; (circles = current efficiency, triangles = anode potential).
of phosphate however, the composite electrode has a lower rate of oxygen evolution, apparently due to adsorption of phosphate ions on the active sites. Regular RuOz/Ti02 does not exhibit this property [ 191. Durability The efficiencies and anode potentials of the two catalysts were recorded against time for sea water electrolysis conditions (Fig. 8). During nearly a year in dilute brine (NaCl = 30 g 1-l) however, the composite coating gave about 6% higher efficiency. The activity and stability of these coatings were nearly equal. Concluding remarks In conclusion, the experiments with the composite coating and the regular RuOz/TiOz have unambiguously shown that these catalysts have almost equal durability and activity, whilst the composite coating suppresses oxygen evolution due to the increased oxygen overvoltage on tin and the adsorption behaviour of palladium.
88
References
10 11 12 13 14 15 16 17 18 19
0. de Nora, Chem. Zng. Tech., 42 (1.970) 222. 0. de Nora, Chem. Zng. Tech., 43 (1971) 182. S. D. Argade and F. B. Leitz, J. Electrochem. Sot., 124 (1977) 12C. C. J. Hurke and J. Renner, J. Electrochem. Sot., 125 (1978) 455C. M. M. JakHi& J. Electrochem. Sot., 121 (1974) 70. N. Ibl and D. Landolt, J. Electrochem. Sot., 115 (1968) 713. L. Hammar and G. Wranglen, Electrochim. Acta, 9 (1964) 1. F. Hine and M. Yasuda, J. Electrochem. SOL, 118 (1971) 170. H. V. K. Udupa, R. Thangappan, B. R. Yadav and P. Subbiah, Chem. Age India, 23 (1975)545. A. T. Kuhn and P. M. Wright, J. Electroanal. Chem., 41 (1973) 329. F. Foerster, 2. Elektrochem., 22 (1916) 85. T. A. Chertykovtseva, D. M. Shub and V. I. Veselovskii, Elektrokhim., 14 (1978) 275. M. Spasojevie, Ph. D. Thesis, Faculty of Technology and Metallurgy, University of Beograd, Belgrade, 1983. S. Trasatti, J. Electroanal. Chem., 39 (1972) 163. M. Spasojevie, N. Krstajib and M. JakliiC, J. Res. Inst. Cat., Hokkaido Univ., 31 (1983) 77. M. M. JakSX, A. R. DespiE, I. M. Csonka and B. 2. NikoliE, J. Electrochem. SOC., 116 (1969) 1316. E. Miiller, Die Elektrometrische (Potentiometrische) Massanalyse, 7th edn., T. Steinkopf, Dresden, 1944. Yu. E. Rogonskaya, V. I. Bistrov and D. M. Shub, Zh. Neorg. Khim., 22 (1977) 201. M. SpasojeviE, N. Krstajid and M. JakgiE, Surf. Technol., 21 (1984) 19.
81
A SELECTIVE CATALYST FOR TITANIUM ANODES N. V. KRSTAJIC Faculty of Technology
and Metallurgy, University of Belgmde, Belgrade (Yugoslavia)
M. D. SPASOJEVIb Faculty of Agriculture,
University of Kragujevac, fiaEak (Yugoslavia)
and M. M. JAKfb? Faculty of Agriculture,
University of Belgmde, Belgmde (Yugoslavia)
Summary A composite electrocatalytic coating with improved anionic selective features for chlorine evolution has been developed. This coating is intended for both chlorate cells and sea water electrolysis where oxygen evolution is undesirable. An optimized composition (14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd and 34 atm.% Sn) has now been shown to combine high activity for chlorine evolution, good corrosion resistance and low activity for oxygen evolution, both in chlorate cells and sea water electrolysis. The high selectivity stems from high oxygen overpotential as well as from the adsorption behaviour.
Introduction RuOz/TiOz-coated titanium electrodes are now a unique anode material used worldwide for caustic-chlorine, chlorate and sea water electrolysis cells [ 1 - 41. In addition, certain industrial processes need even more selective electrocatalysts to provide maximal efficiency. For instance, both chlorate production and sea water electrolysis have competing electrode reactions as follows : Anode: 2 Cl- -
Cl* + 2e
Cathode: 2 Hz0 + 2e Solution: Clz + Hz0 HOC1 e
Hz + 2 OHHOC1 + Cl- + H+
OCl- + H+
2 HOC1 + OCl- 0304-5102/86/$3.50
(A) (R) (C) (D)
Cl03 + 2Cl- + 2 H+
(E)
@ Elsevier Sequoia/Printed in The Netherlands
82
Anodic loss reactions: 6OCll+ 3Hz0 2H,O--02+4H++4e
2C10,
+ 4Cl- + 6H+ + 3/2 O2 + 6e
(F) (G)
In a chlorate cell, the hypochlorite species undergo a chemical conversion to chlorate, which gives maximal current efficiency [ 51. Reaction (G) plays a minor role, as long as the solution does not become too alkaline or the chloride concentration too low [6]. The anodic chlorate formation (F) and the reduction of the hypochlorite are controlled by the mass transport of the hypochlorite from the bulk solution toward the anode and cathode respectively [ 7, 81. These reactions are suppressed in industrial chlorate cells by addition of dichromate into the solution. Evolution of oxygen by electrolysis of water (G) can not be ignored in dilute NaCl solutions used for sea water electrolysis, because the reversible potentials for chlorine and oxygen discharge are close together. Therefore a selective electrocatalyst is needed to suppress reaction (G) by providing a high oxygen overpotential. In addition, the catalyst must resist corrosion at high anodic current density in solutions containing chloride, chlorine and hypochlorite. Oxide catalysts usually produce oxygen more easily than the corresponding bare metal (similarity factor), Ruthenium is one of the less expensive and more available precious metals, and its mixed oxide with titanium (RuO,/TiOz) is a highly active catalyst for chlorine evolution while being quite durable [ 91. For these reasons it has been widely employed for coating titanium anodes. Chloride ions inhibit the formation of surface oxides on platinum group metals by being adsorbed onto the surface [lo], with the exception of platinum itself which forms surface oxides and thereby exhibits a higher anodic polarization [ 111. Among base metals, tin has a remarkable overpotential [12] but corrodes in solutions containing hypochlorites [13]. A palladium-tin compounds (PdSn,), however, is expected to provide both selectivity (high oxygen overpotential) and a lower corrosion rate. Trassati [14] points out the ability of transition metal ions to form covalent bonds with a metal surface blocking. Phosphate ions are known to adsorb onto palladium, and thus might contribute to its selectivity as a catalyst. Palladium itself could perhaps be the sole electrocatalyst, but RuOz/TiOz is cheaper, more active and more resistant to corrosion. Thus a composite coating of RuOz/TiOz and PdSnz has been optimized as an anionic selective catalyst for chlorate cells and sea water electrolysis. The purpose of the present paper is to describe the catalytic properties, polarization characteristics, corrosion stability and lifetime of this selective catalyst. Experimental The procedure for applying a catalytic coating onto a titanium substrate has been described in detail elsewhere [ 151. The difference between
83
the usual RuOz/TiOz coating and this one is the addition of tin and palladium chlorides at various ratios. The materials, RuCls*3H,O, PdC$ (Johnson and Matthey), TiC& and SnC14 (Merck) were of reagent grade purity and were applied as an isopropanol solution containing 10 mg ml-’ of the metals, mixed in different ratios. Usually five coatings were painted onto the titanium substrate to obtain a loading of 10 g mV2at 450 “C. Figure 1 shows the electrochemical apparatus used to measure the electrode potentials. The electrolytic cell, a stirrer, thermostat and pH probe were immersed in the solution in a glass vessel. For chlorate production, the vessel contained 2 dm3 of 300 g NaCl dme3 with added sodium dichromate or Na2HP04 + NaH2P04 or both, all of reagent grade purity. The cell was small, and essentially acted as a generator of available chlorine, while the large retention volume of- the reactor served for its conversion to chlorate. For sea water electrolysis, the reactor contained 50 dm3 of electrolyte to slow down concentration changes. In both cases, the temperature and pH were automatically maintained at the desired set points. Rapid flow was generated through the cell by the gas lift of the cathodic hydrogen. This high recirculation ensured that the change in concentration through the cell was very small (about one part in 800). The production rate and gas composition were followed by means of an Orsat apparatus [16]. To assess cathodic losses, the rate of Hz evolution was measured for a caustic cell which was operated in parallel at the same current density. Available chlorine content was determined by potentiometric titration with sodium arsenite [ 171.
Fig. 1. The set-up for simultaneous investigations of electrode activity and Faradaic yields in the chlorate cell and sea water electrolysis: C: cell, HV: holding volume, M: stirrer, T: temperature control thermostat, P: automatic titrator-p’tf-state, 0: Orsat apparatus for gas analysis, A: multi-range miiliammeter, Re: rectifier, R: reference SCE, V: VTVM.
Results and discussion Crystal structures were determined for various mole ratios of Ru02, Ti02 and PdSnz by X-ray diffraction analysis. Figure 2 shows a typical curve
84
Ti
Fig. 2. X-ray diffraction radiograph for the composite electrocatalyst with optimal phase structure and catalytic activity (Ru: 14 atm.%, Ti: 35 atm.%, Pd: 17 atm.% and Sn: 34 atm.%).
for satisfactory coating. The peaks correspond to reflection from the titanium substrate, from rutile solid state solutions of RuOz and TiOz and from PdSn, [ 131. The rutile peaks are between those of pure RuOz and pure TiOz. Ruthenium dioxide always exists in a rutile solid solution with TiO? and thus forms a phase separate from the other constituents. Palladium and tin may form a number of phases, i.e. PdO, PdSn, PdSn, and SnO, as the Sn/Pd ratio increases. Only when the atomic ratio is between 1 and 2 .do PdSn and PdSnz appear. Corrosion tests have shown that whenever the ratio is greater than 2, Sn02 appears and the coating stability decreases.
Coating composition and activity Polarization characteristics were measured as a function of coating composition. Figure 3 shows the effect of various ratios of Pd/Sn at constant Ru and Ti contents. Any increases in Sn above 34/17 causes a pronounced decrease in catalytic activity, while a decrease below this level gives no improvement. Figure 4 summarizes the effect of composition. The potential at 1 kA m-* is plotted against the ratio of (Ru + Pd) in the coating. Both the
Fig. 3. Polarization characteristics of the composite catalytic coating at constant Ru/Ti atomic fraction (14 atm.% Ru, 35 atm.% Ti) as a function of Pd/Sn atomic ratio for chlorine evolution in brine (NaCl 300 g dmp3, pH 2.0, 80 “C); A: 3 atm.% Pd, 48 atm.% Sn; n : 7 atm.% Pd, 44 atm.% Sn; 0: 12 atm.% Pd, 39 atm.% Sn; 0: 17 atm.% Pd, 34 atm.% Sn; 0: 27 atm.% Pd, 24 atm.% Sn.
85
0.0
Lo.0
8oD
ot%iRu+~
Fig. 4. The effect of precious metals (Pd + Ru) content in electrocatalytic coating on titanium anode potential for chlorine evolution in brine at constant current density (NaCI) 300 g dmA3, 80 “C, pH 2.0 and j = 1.0 kA md2); 0: 35 atm.% Ti, 17 atm.% Pd, 48 atm.% (Ru+Sn);~:35atm.%Ti,14atm.%Ru,51atm.%(Pd+Sn);~:100atm.%(Ru+Ti).
standard RuOz/TiOz (cf. [15]) and the composite coating need about 30 atm.% noble metals. Below that level, the activity decreases faster when the coating contains tin. SnOz as a separate phase reduces the conductivity, the activity and the durability of the coating, while at higher palladium levels, the stable and active but expensive PdSn is formed. Baking temperature effects At the optimal coating composition (atm.% : Ru = 14. Ti = 35, Pd = 1’7, Sn = 34), heating between 300 and 600 “C gives the separate well-distinguished phases of rutile RuOz/TiOz and PdSn,. Below 300 ‘C, the diffraction peaks are small, pointing to incomplete conversion to these phases and hence to the presence of residual chloride ions. Above 700 “C, the peaks are split and the RuO, and TiOz become separate rutile phases, while PdSn, remains unaltered (cf. [3, IS]). Figure 5 shows the effect of temperature on the anode potential during chlorine evolution at 3 kA rnm2,both for the composite coating and the regular Ru02/Ti02 coating. The optimal temperature range is 450 - 550 “C. Selectivity The selectivity was evaluated at both chlorate-producing and sea water electrolysis conditions. Although chlorine evolution is the main anodic reaction, at pH values above 7 and especially at low chloride concentrations,
86
t(“cl
1d.cp+,cl&Old6?)
Fig. 5. The effect of temperature for thermal processing of titanium anodes on their potential for chlorine evolution at constant current density (j = 3 kA mp2) and for optimal catalyst compositions; 0: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn; 0: 40 mol,% RuO2,60 mol.% Ti02. Electrolysis conditions: NaCl 300 g dmp3; pH 2,0; 80 OC; (final processing time for catalytic coating, 60 min). Fig. 6. Oxygen content in the cell gas recalculated for maximal hydrogen current yields in diluted brine (NaCl 30 g dmV3) electrolysis plotted as a function of available chlorine concentration in the bulk at constant current density 0’ = 1.0 kA mm2) and low temperature (25 “C, pH 7 - 9); 0: 40 mol.% Ru02, 60 mol.% TiOz, 0: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn.
oxygen evolution becomes significant due to the discharge of hypochlorite (reaction (F)), and to some extent to the discharge of water (reaction (G)). The optimized composite catalyst was compared to the regular Ru02/ Ti02 rutile coating at identical experimental conditions (Fig. 6). The oxygen readings were corrected to the values they would have had at 100% efficiency. As the hypochlorite increases to 0.01 molar, there is a rapid drop in percent oxygen, especially for the composite coating. Above this concentration, the oxygen increases and eventually rises linearly with the hypochlorite concentration, with both coatings having the same rate of increase. This behaviour indicates that there are two oxygen evolution processes. One is the discharge of hypochlorite, its rate being controlled by mass transport [6] rather than by the nature of the catalyst. The other is the discharge of water, and its rate is controlled by the nature of the catalyst. The presence of hypochlorite blocks those sites where water is discharged, decreasing its contribution to A for the composite coating and A’ for the regular coating at maximal coverage. The rapid decrease in oxygen in the low hypochlorite region resembles a typical adsorption isotherm in the reverse position, thus supporting the explanation proposed above. The same selectivity is shown for chlorate cells in the presence of phosphate buffers [19]. At high hypochlorite concentrations characteristic of alkaline brine solutions, the rate of oxygen evolution controlled by mass transport can be very high (Fig. 7). In phosphate-free brine, oxygen evolution is directly proportional to hypochlorite concentration, as expected for mass transfer rate control. In the presence
87
Iu
Fig. 7. Dependence of oxygen gas content in the chlorate cell on the available chlorine concentration at constant current density (j = 3.0 kA mm2) in presence and absence of phosphate buffer for the composite catalytic coating (14 atm.% Ru, 35 atm.% Ti, 17 atm. % Pd, 34 atm.% Sn); 0: sodium dichromate buffer (3 g dmm3), 0: equimolar primary and secondary sodium phosphate buffer system (3 g dmM3). Fig. 8. The time function of available chlorine current efficiency (7) and the anode potential for two catalytic coatings at constant current density (j = 1.0 kA me2) in diluted brine (NaCl 30 g dme3, pH 6.5 - 7.6, NaClO 1.0 X 10m2 mol dmV3) at low temperature (25 “C); 0 and A: 40 mol.% Ru02, 60 mol.% Ti02; l and A: 14 atm.% Ru, 35 atm.% Ti, 17 atm.% Pd, 34 atm.% Sn; (circles = current efficiency, triangles = anode potential).
of phosphate however, the composite electrode has a lower rate of oxygen evolution, apparently due to adsorption of phosphate ions on the active sites. Regular RuOz/Ti02 does not exhibit this property [ 191. Durability The efficiencies and anode potentials of the two catalysts were recorded against time for sea water electrolysis conditions (Fig. 8). During nearly a year in dilute brine (NaCl = 30 g 1-l) however, the composite coating gave about 6% higher efficiency. The activity and stability of these coatings were nearly equal. Concluding remarks In conclusion, the experiments with the composite coating and the regular RuOz/TiOz have unambiguously shown that these catalysts have almost equal durability and activity, whilst the composite coating suppresses oxygen evolution due to the increased oxygen overvoltage on tin and the adsorption behaviour of palladium.
88
References
10 11 12 13 14 15 16 17 18 19
0. de Nora, Chem. Zng. Tech., 42 (1.970) 222. 0. de Nora, Chem. Zng. Tech., 43 (1971) 182. S. D. Argade and F. B. Leitz, J. Electrochem. Sot., 124 (1977) 12C. C. J. Hurke and J. Renner, J. Electrochem. Sot., 125 (1978) 455C. M. M. JakHi& J. Electrochem. Sot., 121 (1974) 70. N. Ibl and D. Landolt, J. Electrochem. Sot., 115 (1968) 713. L. Hammar and G. Wranglen, Electrochim. Acta, 9 (1964) 1. F. Hine and M. Yasuda, J. Electrochem. SOL, 118 (1971) 170. H. V. K. Udupa, R. Thangappan, B. R. Yadav and P. Subbiah, Chem. Age India, 23 (1975)545. A. T. Kuhn and P. M. Wright, J. Electroanal. Chem., 41 (1973) 329. F. Foerster, 2. Elektrochem., 22 (1916) 85. T. A. Chertykovtseva, D. M. Shub and V. I. Veselovskii, Elektrokhim., 14 (1978) 275. M. Spasojevie, Ph. D. Thesis, Faculty of Technology and Metallurgy, University of Beograd, Belgrade, 1983. S. Trasatti, J. Electroanal. Chem., 39 (1972) 163. M. Spasojevie, N. Krstajib and M. JakliiC, J. Res. Inst. Cat., Hokkaido Univ., 31 (1983) 77. M. M. JakSX, A. R. DespiE, I. M. Csonka and B. 2. NikoliE, J. Electrochem. SOC., 116 (1969) 1316. E. Miiller, Die Elektrometrische (Potentiometrische) Massanalyse, 7th edn., T. Steinkopf, Dresden, 1944. Yu. E. Rogonskaya, V. I. Bistrov and D. M. Shub, Zh. Neorg. Khim., 22 (1977) 201. M. SpasojeviE, N. Krstajid and M. JakgiE, Surf. Technol., 21 (1984) 19.