Molecular Cobalt Catalysts for O2 Reduction: Low-Overpotential

Communication pubs.acs.org/JACS

Cite This: J. Am. Chem. Soc. 2017, 139, 16458-16461

Molecular Cobalt Catalysts for O2 Reduction: Low-Overpotential Production of H2O2 and Comparison with Iron-Based Catalysts Yu-Heng Wang,† Michael L. Pegis,‡ James M. Mayer,‡ and Shannon S. Stahl*,† †

Department of Chemistry, University of Wisconsin−Madison, 1101 University Avenue, Madison, Wisconsin 53706, United States Department of Chemistry, Yale University, New Haven, Connecticut 06520, United States

‡

Downloaded via TUFTS UNIV on July 1, 2018 at 23:37:46 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

cobalt complexes that catalyze reduction of O2 to H2O2 with high selectivity and overpotentials as low as 90 mV. The latter behavior arises from a shallow dependence of the catalytic turnover frequencies (TOFs) on the CoIII/II reduction potentials. This correlation enables direct comparison of the rates and thermodynamic efficiencies of O2 reduction catalyzed by the present Co-based catalysts and a series of reported Fe-porphyrin catalysts.5 Molecular cobalt complexes bearing tetradentate N2O2-based ligands, such as salen, salophen, bis-ketiminates, and related derivatives are common catalysts for aerobic oxidation reactions.6 However, unlike Co-porphyrin and related N4-ligated mononuclear Co complexes,7 N2O2-ligated cobalt complexes have received comparatively little attention as ORR (electro)catalysts.8,9 The 1,2-phenylene-linked bis-ketiminate-CoII complex 1, first reported by Jäger,10 served as a starting point for the present study (Figure 1). A cyclic voltammogram (CV) reveals a

ABSTRACT: A series of mononuclear pseudomacrocyclic cobalt complexes have been investigated as catalysts for O2 reduction. Each of these complexes, with CoIII/II reduction potentials that span nearly 400 mV, mediate highly selective two-electron reduction of O2 to H2O2 (93− 99%) using decamethylferrocene (Fc*) as the reductant and acetic acid as the proton source. Kinetic studies reveal that the rate exhibits a first-order dependence on [Co] and [AcOH], but no dependence on [O2] or [Fc*]. A linear correlation is observed between log(TOF) vs E1/2(CoIII/II) for the different cobalt complexes (TOF = turnover frequency). The thermodynamic potential for O2 reduction to H2O2 was estimated by measuring the H+/H2 open-circuit potential under the reaction conditions. This value provides the basis for direct assessment of the thermodynamic efficiency of the different catalysts and shows that H2O2 is formed with overpotentials as low as 90 mV. These results are compared with a recently reported series of Fe-porphyrin complexes, which catalyze four-electron reduction of O2 to H2O. The data show that the TOFs of the Co complexes exhibit a shallower dependence on E1/2(MIII/II) than the Fe complexes. This behavior, which underlies the low overpotential, is rationalized on the basis of the catalytic rate law.

T

he catalytic reduction of molecular oxygen (O2) is crucial to diverse processes, ranging from biological respiration and fuel cells to the selective oxidation of organic molecules.1 Metalloporphyrins and other molecular transition-metal complexes have been studied as catalysts for the oxygen reduction reaction (ORR),2 typically forming H2O or H2O2 as the reaction product (eqs 1 and 2). O2 + 4H+ + 4e− → 2H 2O O2 + 2H+ + 2e− → 2H 2O2

E° = 1.23 V vs NHE

(1)

E° = 0.68 V vs NHE

(2)

Figure 1. (a) Structure of the cobalt catalyst 1. (b) CVs of 0.50 mM 1 in MeOH with 0.1 M [NBu4][ClO4], and 25 mM AcOH under 1 atm O2 (red) or 1 atm N2 (black). Working electrode: 3.0 mm glassy carbon, reference electrode: Ag/Ag+, counter electrode: Pt wire. Scan rate: 10 mV/s.

nearly reversible CoIII/CoII redox couple under anaerobic conditions in MeOH [Figure 1, black CV; E1/2(CoIII/II) = 0.31 V vs Fc*+/0, Fc* = decamethylferrocene]. The CoIII/II redox potential does not change significantly under 1 atm O2 [E1/2(CoIII/II) ≈ 0.28 V], although the CV trace is less reversible than under N2, probably arising from binding of O2 to the Co complex (see below). Efforts to study O2 reduction by cyclic voltammetry were unsuccessful due to the slow reaction rate;11 however, use of decamethylferrocene (Fc*) as a chemical reductant allowed the O2 reduction reaction in eq 3 to be monitored by UV−visible

Diverse complexes have been identified that exhibit high rates and/or high selectively for H2O or H2O2, but an unmet challenge is the identification of molecular catalysts that operate with low overpotential (i.e., at potentials close to the thermodynamic potentials associated with eqs 1 and 2). We recently reported a TEMPO/NOx electrocatalyst system that exhibits an overpotential for O2 reduction to water of ∼300 mV,3 contrasting other molecular catalysts that typically exhibit overpotentials of 500−1000 mV for this reaction.4 Here, we describe a series of © 2017 American Chemical Society

Received: August 25, 2017 Published: October 17, 2017 16458

DOI: 10.1021/jacs.7b09089 J. Am. Chem. Soc. 2017, 139, 16458−16461

Communication

Journal of the American Chemical Society

show that H2O2 is formed with high selectivity (93−99%) in all cases (Chart 1, Figure S5). The rates of ORR mediated by complexes 1−8 were probed under a uniform set conditions, buffered with AcOH/NBu4OAc (25 mM each). TOFs for the conversion of O2 to H2O2 were determined from the initial rates of Fc*+ formation by UV− visible spectroscopy (Figure S10−11).11 The data revealed that complexes with lower CoIII/II redox potentials exhibit higher ORR rates, and a nearly linear free energy relationship (LFER) is evident between log(TOF) and E1/2(CoIII/II) (Figure 3).

Figure 2. UV−visible spectral changes in the two-electron reduction of O2 (10 mM) by Fc* (0.5 mM) with 1 (1.25 × 10−2 mM) in presence of 25 mM each of AcOH and [NBu4][OAc] in O2-saturated MeOH at 298 K. Inset: Absorbance changes at 780 nm are due to formation of Fc*+.

spectroscopy (Figure 2). The increase in absorbance at 780 nm arises from the formation of Fc*+ during the course of the reaction, and the TOF is 0.027 s −1 with respect to O2 consumption.11 Iodometric titration of the final reaction mixture shows the reaction in eq 3 proceeds with nearly complete selectivity for H2O2 (96% yield).7j,11,12 Initial-rate kinetic studies reveal a rate law with a first-order dependence on [Co] and [AcOH], but zero-order with respect to pO2 and [Fc*] (eq 4 and Figure S6).11 To probe the lack of an O2 dependence in the rate law, the CoII complex 1 was added to O2-saturated MeOH and the dissolved O2 was monitored using a Clark electrode. Addition of 1 to the solution resulted in consumption of 1.1 ± 0.1 equiv of O2 (Figure S7−S9).11 This result, which supports a Co−O2 adduct13 as the catalyst resting state, accounts for the zero-order pO2-dependence of the rate. cat.[LCo]

O2 + 2Fc* + 2AcOH ⎯⎯⎯⎯⎯⎯⎯⎯⎯→ H 2O2 + 2Fc*+ + 2AcO−

(3)

rate = kcat.[Co(N2O2 )][AcOH]

(4)

Figure 3. Linear free energy relationship between the log(TOF) for Cocatalyzed ORR and E1/2(CoIII/II) for complexes 1−8. Reaction conditions: 0.9 mM Fc*, 0.02−25 μM Co, 25 mM each of AcOH and [Bu4N][AcO] in O2-saturated MeOH conditions, containing 1 mM urea·H2O2.

These data may be analyzed further to assess the overpotential for the reactions using reported methodology for overpotential determination under nonaqueous conditions.5,14 The effective overpotential (ηeff) is defined as the difference between the thermodynamic potential for O2 reduction to hydrogen peroxide (EO2/H2O2) under the reaction conditions and the E1/2(CoIII/II) values (Figure 4). The standard potential for O2/H2O2 in

A series of related N2O2-ligated cobalt complexes 2−8 (Chart 1) were prepared to compare their ORR reactivity with that of 1. Cyclic voltammograms of complexes 1−8 show (quasi-) reversible redox couples, with E1/2(CoIII/II) values in the range −0.085−0.31 V vs Fc*+/0 (Chart 1, Figure S4). Analysis of their reactivity with O2 in the presence of Fc* and AcOH (cf. eq 3) Figure 4. Redox potential scale showing E1/2(CoIII/II) for 1−8 and thermodynamic reduction potentials for H+/H2, O2/H2O2 and O2/H2O vs Fc*+/0 under the catalytic reaction conditions.

Chart 1. Molecular N2O2-Ligated Cobalt Complexes Tested for O2 Reduction

methanol may be estimated from (i) the standard aqueous cell potential for O2 + H2 → H2O2, (ii) the measurement of the open circuit potential (OCP) for H+/H2 (EH+/H2) under the reaction conditions (Figure S12)15 and (iii) the free energy to transfer H2O2 from H2O to MeOH.16 The solvation energy of H2O2 in MeOH is expected to be nearly identical that in H2O and, therefore, the transfer free energy is negligible.17 The thermochemical cycle associated with i−iii (see Supporting Information, section VIII) corresponds to the standard O2/H2O2 potential in methanol with acetic acid/acetate buffer (EO2/H2O2). This value is obtained by adding 0.68 V to the EH+/H2 measured by OCP (−0.31 V vs Fc*), which corresponds to 0.37 V versus Fc*+/0. This value is adjusted to 0.46 V to account for the 16459


Communication


Both classes of ORR reactions in Figure 5 exhibit high selectivity (Co: H2O2; Fe: H2O) across the entire range of redox potentials. This sharp distinction in selectivity contradicts a recent suggestion that H2O2/H2O selectivity is directly correlated to ORR overpotential, irrespective of catalyst identity,23 and it highlights to the need for further insights into the factors that contribute to product selectivity for mononuclear complexes such as these. Such efforts are the focus of ongoing studies. The different slopes for the Co- and Fe-catalyzed reactions in Figure 5 may be rationalized by the rate laws for the Co- and Febased reactions and the simplified catalytic cycle in Scheme 1.

nonstandard-state background concentration of 1 mM H2O2 using the Nernst equation (cf. eq 5).18 EO2 /H2O2 = E°O2 /H2O2 −

− 2 2.303RT ⎛ [H 2O2 ][AcO ] ⎞ ⎟ log⎜ 2 nF ⎝ pO2 [AcOH] ⎠

(5)

The thermodynamic analysis yields ηeff values that range from 150 to 550 mV. Further analysis of the O2/H2O2 ORR catalyzed by 1 showed the rate is not altered in the presence of higher [H2O2], ranging from 1 to 100 mM.11 With a background [H2O2] of 100 mM, the O2/H2O2 ηeff is only 90 mV with 1. Direct comparison of this result with those of other catalysts is difficult. Most prior studies employed non-buffered conditions and lack the OCP measurements needed to estimate overpotentials. Nonetheless, a 90 mV overpotential is much lower than is typically associated with ORR.3,4,19,20 One catalyst system for which direct comparison is possible is a series of Fe-tetraphenylporphyrin [Fe(por)] derivatives, which catalyze selective conversion of O2 to water.5 The Fe(por)catalyzed O2 reduction catalysts follow a rate law with a firstorder dependence on [Fe(por)], [H+] and [O2] (eq 6).21 Overpotential analysis, similar to that described above, was performed for these catalysts. The logarithm of the ORR rates measured for the Fe and Co catalysts are plotted relative to their respective ηeff values for formation of H2O2 (red, Co) and H2O (blue, Fe) in Figure 5.

Scheme 1. General Catalytic Cycle for the O2 Reduction Catalyzed by Co(N2O2) and Fe(por) Catalysts

Both Co- and Fe-catalyzed ORR are proposed to be initiated by O2 binding to the MII species, followed by a series of proton- and electron-transfer steps to afford H2O2 or H2O as the final product. The rate law for Fe(por)-catalyzed ORR in eq 6 arises from pre-equilibrium O2 binding to FeII, followed by protonation of the FeIII(O2•) adduct.21 Facile and favorable O2 binding to CoII results in a CoIII(O2•) adduct as the catalyst resting state and a rate law with zero-order dependence on pO2.11 Several protontransfer steps could account for the [AcOH] dependence evident in the catalytic rate law in eq 4; however, previous studies provide support for protonation of a CoIII(O2•) adduct as the ratelimiting step.7f,g,k,o Both O2 binding and protonation steps should be favored for complexes with lower E1/2(MIII/II). The steeper slope with Fe, relative to Co, complexes may be rationalized by the additive effect of two electronically sensitive steps (O2 binding and protonation) for Fe, but only one step (protonation) for Co. More quantitatively, the slope of the Cocatalyzed ORR correlation in Figure 5 (mCo = 6.1 (logTOF)/V) is similar to the slope observed for Fe(por) catalysts (mFe) when the overpotential is adjusted solely by changing the pKa of the acid source with a single Fe(por) catalyst (5.1 (logTOF)/V).24 In summary, this study shows that Co(N2O2) complexes serve as effective homogeneous ORR catalysts and form H2O2 as the near-exclusive product. Determination of the thermodynamic potential for EO2/H2O2 by measuring the open circuit potential for H+/H2 under the nonaqueous reaction conditions reveals that H2O2 formation proceeds with quite low overpotential (ηeff). The relatively slow ORR rates initially observed with these catalysts is a consequence of their low overpotential, and analysis of the linear free energy relationships shows that Co(N2O2) catalysts are considerably faster than analogous Fe(por) catalyst would be at similarly low overpotentials. Overall, the overpotential analysis used here and in the recent study of Fe(por)catalyzed ORR5 provide an important foundation for more

Figure 5. Correlations between log(TOF) versus ηeff for the O2 reduction to H2O2 or H2O, catalyzed by Co(N2O2) (red), and Fe(porphyrin) (blue). The Fe(porphyrin) data are taken from Table 1 in ref 5.22 ORR reactions with Co(N2O2) complexes were conducted in the presence of urea·H2O2: 1 mM (•), 10 mM (■), and 100 mM (▲), respectively (see also Table S1).

rate = kcat.[Fe(por)][H+][O2 ]

(6)

The trendline for the Co complexes exhibits a significantly shallower slope than that for the Fe complexes. Thus, while Co catalysts are slower than Fe catalysts at high overpotentials, Co catalysts are faster at low overpotentials. This distinction between Co and Fe is even more noteworthy in light of the production of H2O2 with Co, since formation of H2O2 from O2 has a 550 mV lower driving force relative to the formation of water (E° = 0.68 versus 1.23 V; eqs 1 and 2). 16460


Communication


Anson, F. C. Inorg. Chem. 1995, 34, 2771−2780. (j) Fukuzumi, S.; Okamoto, K.; Gros, C. P.; Guilard, R. J. Am. Chem. Soc. 2004, 126, 10441−10449. (k) Fukuzumi, S. Chem. Lett. 2008, 37, 808−813. (l) Kadish, K. M.; Shen, J.; Frémond, L.; Chen, P.; Ojaimi, M. E.; Chkounda, M.; Gros, C. P.; Barbe, J.-M.; Ohkubo, K.; Fukuzumi, S.; Guilard, R. Inorg. Chem. 2008, 47, 6726−6737. (m) Zagal, J. H.; Griveau, S.; Silva, J. F.; Nyokong, T.; Bedioui, F. Coord. Chem. Rev. 2010, 254, 2755−2791. (n) Honda, T.; Kojima, T.; Fukuzumi, S. J. Am. Chem. Soc. 2012, 134, 4196−4206. (o) Mase, K.; Ohkubo, K.; Fukuzumi, S. J. Am. Chem. Soc. 2013, 135, 2800−2808. (p) Mase, K.; Ohkubo, K.; Fukuzumi, S. Inorg. Chem. 2015, 54, 1808−1815. (8) N2O2-based ligands have been employed with molecular vanadium ORR catalysis: (a) Yamamoto, K.; Oyaizu, K.; Tsuchida, K. J. Am. Chem. Soc. 1996, 118, 12665−12672. (b) Tsuchida, E.; Oyaizu, K.; Dewi, E. L.; Imai, T.; Anson, F. C. Inorg. Chem. 1999, 38, 3704−3708. (c) Liu, Z.; Anson, F. C. Inorg. Chem. 2000, 39, 274−280. (d) Liu, Z.; Anson, F. C. Inorg. Chem. 2001, 40, 1329−1333. (9) For examples of electrocatalytic O2 reduction by immobilized Co(N2O2)-type catalysts, see the following: (a) Choi, Y.-K.; Park, J.-K.; Jeon, S. Electroanalysis 1999, 11, 134−138. (b) Okada, T.; Katou, K.; Hirose, T.; Yuasa, M.; Sekine, I. J. Electrochem. Soc. 1999, 146, 2562− 2568. (c) Shamsipur, M.; Najafi, M.; Hosseini, M.-R. M.; Sharghi, H. Electroanalysis 2007, 19, 1661−1667. (10) (a) Jäger, E.-G.; Knaudt, J.; Rudolph, M.; Rost, M. Chem. Ber. 1996, 129, 1041−1047. (b) Seidel, R. W.; Goddard, R.; Breidung, J.; Jäger, E.-G. Z. Anorg. Allg. Chem. 2014, 640, 1946−1952. (11) See Supporting Information for details. (12) For protocol of iodometric titration, see: (a) Mair, R. D.; Graupner, A. J. Anal. Chem. 1964, 36, 194−204. (b) Fukuzumi, S.; Kuroda, S.; Tanaka, T. J. Am. Chem. Soc. 1985, 107, 3020−3027. (13) Characterization of the O2 adduct of complex 5 in MeOH was reported: Anson, C. W.; Ghosh, S.; Hammes-Schiffer, S.; Stahl, S. S. J. Am. Chem. Soc. 2016, 138, 4186−4193. (14) Appel, A. M.; Helm, M. L. ACS Catal. 2014, 4, 630−633. (15) Roberts, J. A. S.; Bullock, R. M. Inorg. Chem. 2013, 52, 3823− 3835. (16) Pegis, M. L.; Roberts, J. A. S.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M. Inorg. Chem. 2015, 54, 11883−11888. (17) For relevant considerations, see: (a) Ohgaki, K.; Sano, F.; Katayama, T. J. Chem. Eng. Data 1976, 21, 55−58. (b) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378−6396. (18) Complexes 1−8 were found to decompose in the presence of a standard-state concentration of H2O2 (1 M), so the rate measurements in Figure 3 were conducted in the presence of 1 mM H2O2·urea to establish a stable reference state. This 103-fold difference in [H2O2] leads to a 90 mV Nernstian shift in the thermodynamic O2/H2O2 potential. Control experiments showed urea present in H2O2·urea does not influence the reaction (also see Supporting Information, section IX). (19) Onset potentials for the ORR to H2O with heterogeneous Pt catalysts exhibit overpotentials of ∼300 mV: Gasteiger, H. A.; Kocha, S. S.; Sompalli, B.; Wagner, F. T. Appl. Catal., B 2005, 56, 9−35. (20) O2 reduction to H2O2 has received less attention, but heterogeneous catalysts for this process can have lower overpotentials (50−450 mV): Siahrostami, S.; Verdaguer-Casadevall, A.; Karamad, M.; Deiana, D.; Malacrida, P.; Wickman, B.; Escudero-Escribano, M.; Paoli, E. A.; Frydendal, R.; Hansen, T. W.; Chorkendorff, I.; Stephens, I. E. L.; Rossmeisl, J. Nat. Mater. 2013, 12, 1137−1143. (21) Wasylenko, D. J.; Rodríguez, C.; Pegis, M. L.; Mayer, J. M. J. Am. Chem. Soc. 2014, 136, 12544−12547. (22) Reaction conditions: 0.3 mM Fe(porphyrin) derivatives were added in DMF solution containing 0.1 M electrolyte [Bu4N][PF6] in the presence of 20 mM [DMF-H]OTf under 1 atm O2. Scan rate: 100 mV/s. See ref 5 for details. (23) Passard, G.; Ullman, A. M.; Brodsky, C. N.; Nocera, D. G. J. Am. Chem. Soc. 2016, 138, 2925−2928. (24) Pegis, M. L.; Wise, C. F.; Koronkiewicz, B.; Mayer, J. M. J. Am. Chem. Soc. 2017, 139, 11000−11003.

thorough comparison of ORR catalysts, such as those composed of different metal ions and/or ligand classes.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b09089. Experimental procedures, NMR spectroscopic data, reaction time courses, cyclic voltammograms, iodometric titrations (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Michael L. Pegis: 0000-0001-6686-1717 James M. Mayer: 0000-0002-3943-5250 Shannon S. Stahl: 0000-0002-9000-7665 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Drs. Colin W. Anson and James B. Gerken for helpful discussions. This research was supported as part of the Center for Molecular Electrocatalysis, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences.

■

REFERENCES

(1) For general references, see: (a) Babcock, G. T.; Wikstrom, M. Nature 1992, 356, 301−309. (b) Stahl, S. S. Science 2005, 309, 1824− 1826. (c) Lee, J.; Jeong, B.; Ocon, J. D. Curr. Appl. Phys. 2013, 13, 309− 321. (d) Katsounaros, I.; Cherevko, S.; Zeradjanin, A. R.; Mayrhofer, K. J. J. Angew. Chem., Int. Ed. 2014, 53, 102−121. (2) (a) Collman, J. P.; Wagenknecht, P. S.; Hutchison, J. E. Angew. Chem., Int. Ed. Engl. 1994, 33, 1537−1554. (b) Kobayashi, N.; Nevin, W. A. Appl. Organomet. Chem. 1996, 10, 579−590. (c) Anson, F. C.; Shi, C.; Steiger, B. Acc. Chem. Res. 1997, 30, 437−444. (d) Rosenthal, J.; Nocera, D. G. Acc. Chem. Res. 2007, 40, 543−553. (e) Scanlon, M. D. ChemCatChem 2013, 5, 1696−1697. (f) Zhang, W.; Lai, W.; Cao, R. Chem. Rev. 2017, 117, 3717−3797. (3) Gerken, J. B.; Stahl, S. S. ACS Cent. Sci. 2015, 1, 234−243. (4) Variations in reaction conditions make it difficult to compare overpotentials associated with molecular ORR catalysts; however, estimated overpotentials for diverse catalyst systems are compiled in Table S2 of the Supporting Information for ref 3. (5) Pegis, M. L.; McKeown, B. A.; Kumar, N.; Lang, K.; Wasylenko, D. J.; Zhang, X. P.; Raugei, S.; Mayer, J. M. ACS Cent. Sci. 2016, 2, 850−856. (6) For leading references, see: (a) Bäckvall, J. E.; Awasthi, A. K.; Renko, Z. D. J. Am. Chem. Soc. 1987, 109, 4750−4752. (b) Bäckvall, J. E.; Piera, J. Angew. Chem., Int. Ed. 2008, 47, 3506−3523. (c) Wendlandt, A. E.; Stahl, S. S. J. Am. Chem. Soc. 2014, 136, 11910−11913. (7) (a) Alt, H.; Binder, H.; Sandstede, G. J. Catal. 1973, 28, 8−19. (b) Jahnke, H.; Schönborn, M.; Zimmermann, G. Organic dyestuffs as catalysts for fuel cells. In Physical and Chemical Applications of Dyestuffs; Schäfer, F. P., Gerischer, H., Willig, F., Meier, H., Jahnke, H., Schönborn, M., Zimmermann, G., Eds.; Springer Berlin Heidelberg: Berlin, Heidelberg, 1976; pp 133−181. (c) Geiger, T.; Anson, F. C. J. Am. Chem. Soc. 1981, 103, 7489−7496. (d) Chan, R. J. H.; Su, Y. O.; Kuwana, T. Inorg. Chem. 1985, 24, 3777−3784. (e) Kobayashi, N.; Nishiyama, Y. J. Phys. Chem. 1985, 89, 1167−1170. (f) Fukuzumi, S.; Mochizuki, S.; Tanaka, T. Chem. Lett. 1989, 18, 27−30. (g) Fukuzumi, S.; Mochizuki, S.; Tanaka, T. Inorg. Chem. 1989, 28, 2459−2465. (h) Fukuzumi, S.; Mochizuki, S.; Tanaka, T. Inorg. Chem. 1990, 29, 653−659. (i) Kang, C.; 16461


Molecular Cobalt Catalysts for O2 Reduction: Low-Overpotential

Recommend Documents