The volume of activation ∆V el q for the Co(bpy) 3 3+/2+ electrode reaction in aqueous NaCl (0.2 mol L -1 ) is -8.6 ( 0.4 cm 3 mol -1 at 25.0 °C, as expected on theoretical grounds and by analogy with Co(en) 3 3+/2+ and Co(phen) 3 3+/2+ , and neither the rate constant k el at various pressures nor ∆V el q correlate with the corresponding mean diffusion coefficients D for the couple and the diffusional activation volume ∆V diff q , respectively. In organic solvents, however, ∆V el q is strongly positiVe (9.1 ( 0.3, 10.2 ( 0.7, and 12.2 ( 0.9 cm 3 mol -1 for CH 3 CN, acetone, and propylene carbonate, respectively, with 0.2 mol L -1 [(C 4 H 9 ) 4 N]ClO 4 at 25 °C) and correlates with ∆V diff q , while k el correlates with D. These results support the proposition of Murray et al.

Volumes of reaction ΔV Ag/AgCl (vs Ag/AgCl/4.0 mol L-1 KCl) and of activation ΔV el ⧧ for the electrode reactions of the aqueous Co(azacapten)3+/2+, Ru(en)3 3+/2+, and Co(tacn)2 3+/2+ couples have been measured by high-pressure cyclic and AC voltammetry. For the low-spin/low-spin Co(azacapten)3+/2+ couple, ΔV el ⧧ = −3.3 ± 0.4 cm3 mol-1, whereas high-pressure NMR measurements gave a volume of activation ΔV ex ⧧ for the self-exchange reaction of −6.5 ± 0.5 cm3 mol-1, in accordance with the “fifty-percent rule” (J. Am. Chem. Soc. 1997, 119, 7137) and with the prediction of an adaptation of the Marcus theory of intermolecular electron-transfer kinetics (Can. J. Chem. 1996, 74, 631). For the Ru(en)3 3+/2+ self-exchange reaction, ΔV ex ⧧ was estimated indirectly as −15.1 ± 1.7 cm3 mol-1 from the Co(phen)3 3+/Ru(en)3 2+ cross reaction (ΔV 12 ⧧ = −12.9 ± 0.5 cm3 mol-1), for which the rate constant k 12 was consistent with the Marcus cross relation. For the Fe(H2O)6 3+/Ru(en)3 2+ cross reaction (ΔV 12 ⧧ = −18.3 ± 1.2 cm3 mol-1), k 12 was slower than predicted from the Marcus cross relation, and consequently the estimated ΔV ex ⧧ for Ru(en)3 3+/2+ (−18.9 ± 2 cm3 mol-1) may be less reliable. For the Ru(en)3 3+/2+ electrode reaction, ΔV el ⧧ = −7.5 ± 0.4 cm3 mol-1, again in accordance with the fifty-percent rule and, conversely, authenticating the estimated ΔV ex ⧧. The ΔV ex ⧧ estimates for Ru(en)3 3+/2+, however, are some 10 cm3 mol-1 more negative than can be accommodated by the adapted Marcus theory. For the low-spin/high-spin couple Co(tacn)2 3+/2+, ΔV el ⧧ (−5.9 ± 0.9 cm3 mol-1) is intermediate between values expected for CoIII/II clathrochelates and low-spin/high-spin tris(bidentate) chelates, although ΔV Ag/AgCl places this couple within the latter group.

The electronic spectroscopic data for zinc, cobalt, and iron perchlorinated phthalocyanines in several oxidation states are discussed.The magnetic properties of the iron(II) and cobalt(II) derivatives from ambient temperature to 5 K are reported. The electrochemical behavior of these three species is reported in solution and as surfaces on highly oriented pyrolytic graphite. The pH dependence of the surface data is analyzed in detail. The first clearly defined reduction process corresponds with M'[Cll6Pc(-2)]"/M'[C1|6Pc(-3)]12" for M = Co and Fe. The iron and cobalt species' MnPc(-2)/[M'Pc(-2)]~r edox process are broad or ill-defined on the surface and absent from solution. The FeIII[Cl16Pc(-2)]+/Fe"[Cl16Pc(-2)] redox process has atypical pH dependence. The data are explained in terms of the acceptor nature of the chlorine substitution. Oxygen reduction data, catalyzed by these species, are also reported.to Union Carbide, Parma, for a gift of highly oriented pyrolytic graphite.

For the aqueous couples Os(CN)(6)(3-/4-), Mo(CN)(8)(3-/4-), and W(CN)(8)(3-/4-), volumes of reaction DeltaV(Ag/AgCl) relative to Ag/AgCl/4.0 mol L(-)(1) KCl and volumes of activation DeltaV(el)() for the electrode reactions are reported. Values of DeltaV(Ag/AgCl) are consistent with a very small increase in the metal-carbon bond length on reduction in each case, the main component of DeltaV(Ag/AgCl) (other than that of the reference electrode) being electrostrictive solvation change. For media in which the cation is 0.5 mol L(-)(1) K(+) or (for Mo) Na(+), DeltaV(el)() is strongly positive (+9.4 +/- 0.7, +7.3 +/- 0.7, and +10.8 +/- 0.4 cm(3) mol(-)(1) for Os(CN)(6)(3-/4-), Mo(CN)(8)(3-/4-), and W(CN)(8)(3-/4-), respectively, at 25 degrees C), whereas the theoretical expectation for a mechanism involving only the cyanometalate anions is -3 +/- 1 cm(3) mol(-1). For Mo(CN)(8)(3-/4-) in Et(4)NCl, however, DeltaV(el)() is -4.2 +/- 0.2 cm(3) mol(-)(1). In all cases, DeltaV(el)() is just one-half of the corresponding parameter DeltaV(ex)() for the homogeneous (bimolecular) self-exchange reaction of the same couple, giving strong confirmation of the "fifty-percent rule" (Fu, Y.; Swaddle, T. W. J. Am. Chem. Soc. 1997, 119, 7137). These and related results are interpreted in terms of a mechanism for both electrode and homogeneous electron-transfer reactions of cyanometalates in which the counterion mediates the electron-transfer process. For alkali metal cations, partial deaquation to permit this mediation results in positive DeltaV(el)() values, whereas for tetraalkylammonium counterions, there are no aqua ligands to be removed and DeltaV(el)() is "normal".

Ac voltammetry at pressures up to 200 MPa shows that volumes of activation for heterogeneous electron transfer in three aqueous couples {[Fe(CN)6]3-/4-, [C~(sep)]~+'~+ and [C0(en)~]3+'2+) are numerically about 50% of those for homogeneous (bimolecular) electron transfer in these same couples, with the same algebraic sign, as predicted by an extension of Marcus theory. Marcus' has proposed that the free energy AG,1* of activation for electron transfer in a couple ML,(z+ I)+/=+ at an electrode should be approximately one-half that ( AG,,*) for homogeneous (bimolecular) electron transfer in the same couple in solution,? if an adiabatic outer-sphere mechanism is operative [eqn. (111 AGel* ;Ace,* (1) Since free energies of activation AG,* are measurable (indeed, have meaning) only in terms of the corresponding rate constants ki [eqn. (2)],