The interplay of electronic structure and reactivity in transition metal complexes is an area of considerable research effort.[1, 2] The cooperative effect of redox-active ligands and metal sites in enzymatic systems,[3] and more recently in synthetic systems,[4] adds significant flexibility to catalyst function. Depending on the relative energies of the redoxactive orbitals, metal complexes with proradical ligands can exist in a limiting description as a metal-ligand radical (Mn+(L•)) or a high valent metal complex (M(n+1)+(L-)). In certain cases, subtle changes to the system through variation of the ligand field, or temperature is sufficient to shift the oxidation locus.[5, 6] Recent work in this area has focused on bis(salicylidene)diamine complexes 1-3 (Scheme 1).[6-11] The one-electron oxidized Ni derivatives exist in the ligand radical form NiII(L•-) in solution and the solid state, however the addition of exogenous ligands to NiII(L•-) in solution results in a shift in the oxidation locus to the NiIII(L2-) form.[7-10] The oxidized Cu derivative of 1 exists as the high valent metal complex in the solid state. In solution this complex exhibits temperature-dependent valence tautomerism between the ligand radical and high valent metal forms, demonstrating the nearly isoenergetic nature of the two species.[6]
Scheme 1.

Nickel bis-phenoxide complexes
Oxidation studies to date have centered on symmetric bis(salicylidene)diamine complexes, resulting in full delocalization of the radical over the ligand framework. Interestingly, recent work has shown that the oxidized Pd analogue of 1 exhibits partial radical localization on one of the two phenolates as this metal ion limits coupling between the redox-active ligands.[12] Intrigued by this example, we have synthesized the Ni analogue of a Salalen ligand 4,[13] a non-symmetric variant of 1.[14] We hypothesized that the reduced amino-phenolate would undergo oxidation at lower potentials as compared to the imino-phenolate, resulting in a localized ligand radical complex in the one-electron oxidized form. Preferential redox tuning of phenolate ligands has been previously demonstrated in a functional model of galactose oxidase.[15]
Compound 4 exhibits two reversible redox couples by cyclic voltammetry (E1/21 = 0.13 V and E1/22 = 0.56 V vs Fc+/Fc; Fc: ferrocene; See Figure S1 in the Supporting Information). The first oxidation for 4 occurs at a potential 0.3 V lower than that for 1, and compares well with the electrochemistry of a Cu tetrahydrosalen complex,[16] and is attributed to oxidation of the more electron rich amino-phenolate. Treatment of 4 in CH2Cl2 with one equivalent of the oxidants AgSbF6 (E1/2 = +0.65 V vs Fc+/Fc), thianthrenyl radical [thianthrene]+• SbF6- (E1/2 = +0.89 V vs Fc+/Fc), or (NH4)2Ce(NO3)6 results in an immediate colour change from red-brown to green, signifying formation of 4+. AgSbF6 as the chemical oxidant provides crystals of 4+ SbF6- suitable for X-ray analysis.[13]
The structures of 4 (See Figure S2 in the Supporting Information) and 4+ possess a slightly distorted square-planar geometry about the Ni centre, similarly to 1 and 1+. While the coordination sphere of 1 contracts symmetrically upon oxidation to 1+,[10] the Ni coordination sphere of 4+ is non-symmetric (Table 1), with considerable lengthening of the amino-phenolate Ni-O1 bond by 0.04 Å in comparison to 4. The lengthening of the Ni-O1 bond is consistent with a decrease in electron donating ability of a phenoxyl ligand relative to that of phenolate,[17, 18] indicating localized oxidation of the amino-phenolate. A similar non-symmetric coordination environment was observed in the solid state for the Pd analogue of 1+.[12] DFT calculations[19] of 4 and 4+ reproduce the observed coordination sphere asymmetry in the oxidized form (Table 1). Interestingly, an elongation of the C-O bond lengths is observed upon oxidation of 4 to 4+, instead of the contraction expected for a phenoxyl radical with semi-quinone character.[17, 20]
Table 1.
Experimental and Calculated (in Parentheses) Coordination Sphere Metrical Parameters for the Complexes in Å.
| Ni-O1 | Ni-O2 | Ni-N1 | Ni-N2 | |
|---|---|---|---|---|
| 4 | 1.846 (1.876) | 1.841 (1.861) | 1.845 (1.945) | 1.845 (1.877) |
| 4+ | 1.883 (1.892) | 1.823 (1.836) | 1.872 (1.951) | 1.824 (1.854) |
The EPR spectum of 4+ in CH2Cl2 at 77 K exhibits a broad (S = 1/2) signal with a giso = 2.018 (Figure 2a), indicating that the unpaired electron is ligand-based. The measured g value is lower than that for 1+ (giso = 2.045)[10] indicating a smaller contribution of metal d-orbitals (mostly dyz) to the SOMO of 4+, and greater ligand radical character. This result corresponds well with the lower calculated Ni spin density (Figure 2 Inset) for 4+ (5 %), in comparison to that for 1+ (10 %).[21] These results suggest that the electronic coupling between the two redox-active phenolates, mediated, presumably through a hole transfer superexchange mechanism involving the Ni dyz orbital,[2, 10, 12] is substantially reduced in 4+ as compared to 1+ (vide infra).
Figure 2.

X-band EPR spectra of 4+ recorded in frozen CH2Cl2 at 77 K (experimental spectra: solid lines; simulations: dashed lines): a) 1 mM 4+, b) 1 mM 4+ + 20 equiv pyridine. Inset: Calculated spin density plot for 4+, showing localization of the unpaired electron on the more electron rich amino-phenolate.
Addition of 20 equivalents of pyridine to 4+ at 233 K and subsequent cooling to 77 K results in an anisotropic EPR pattern (Figure 2b) that is consistent with formation of a NiIII species [4(py)2]+ (gx = gy = 2.230, gz = 2.032, Az = 19 × 10-4 cm-1, gav = 2.16). This result is consistent with modulation of the ligand field upon axial binding of pyridine, and a consequent shift in the locus of oxidation to form a NiIII complex.[8-10]
Ni K-edge X-ray absorption spectroscopy (XAS) was used to further probe the metal oxidation state and structure of frozen solutions of 4, 4+, and [4(py)2]+ (Figure S3 in the Supporting Information). The Ni K-edge 1s→3d transition, or pre-edge, is a successful indicator of Ni oxidation state,[22] and the similar energies of the pre-edge feature for 4 (8332.0 eV) and 4+ (8331.9 eV) are consistent with a NiII oxidation state for both complexes. The shift of the pre-edge to higher energy for [4(py)2]+ (8332.4 eV) signifies oxidation to NiIII, matching the EPR results (vide supra).
The UV-Vis-NIR spectra of 4 is typical of a low spin d8 square planar metal complex, and changes substantially upon oxidation to 4+ (Figure 3). The spectrum of 4+ exhibits a new intense band at 25000 cm-1 (9300 M-1 cm-1), and low energy bands at 13000 cm-1 (3000 M-1 cm-1), 11300 cm-1 (shoulder; 2700 M-1 cm-1), and 6500 cm-1 (2000 M-1 cm-1). The absence of low energy transitions (<12000 cm-1) in 4 and [4(py)2]+ (See Figure S4 in the Supporting Information) indicates that the low energy bands are associated with the ligand radical. The lowest energy NIR band is predicted by time-dependent density functional theory (TD-DFT) to be a phenolate to phenoxyl intervalence charge transfer (IVCT) transition (Figure 3 Inset), and is of much weaker intensity in comparison to the NIR band for the Class III[23] delocalized mixed-valence complex 1+ (4700 cm-1, 21500 M-1 cm-1).[10] This attenuation in NIR band intensity reflects the limited electronic coupling between the two redox-active phenolates in 4+. The intense band at 25000 cm-1 is assigned to a phenoxyl radical π → π* transition.[24] This band is obscured by other intense LMCT transitions in the absorption spectrum of 1+.[9, 11]
Figure 3.
Electronic absorption spectra of 0.08 mM solutions of 4 (solid line) and 4+ (dashed line) in CH2Cl2 at 298 K. Calculated transitions shown as vertical lines. Inset: TD-DFT assignment (β-HOMO→ β-LUMO) of the lowest energy NIR transition (5500 cm-1) for 4+.[13]
The resonance Raman (rR) spectra of 4 and 4+ (Figure 4) exhibit key differences upon oxidation, consistent with localization of the ligand radical on the Raman timescale. rR Modes resonant with both phenoxyl π → π* and phenolate-Ni(II) LMCT transitions are observable in the spectrum of 4+. The features at 1501 and 1581 cm-1 are attributed to characteristic phenoxyl radical C-O stretching, ν7a, and Cortho-Cmeta stretching, ν8a, modes respectively.[25] The rR intensity ratio, I(ν8a)/I(ν7a) ≥ 1, is often used as a spectral marker for metal-coordinated phenoxyl radicals with p-methoxy substituents.[26] However, the t-butyl substituents employed in this work may minimize the semi-quinoid character of phenoxyl radicals, and in addition the elongated Ni-O1 bond length could further influence the enhancement of the ν8a mode.[26, 27] A combination of these factors presumably leads to the reduced intensity of ν8a relative to ν7a observed in 4+; a similar intensity pattern is reported for a NiII-phenoxyl radical complex.[28]
Figure 4.

Resonance Raman (rR) spectra of 4 (solid line) and 4+ (dashed line) in CH2Cl2 at 213 K (λex = 413 nm). Solvent = •.
The phenolate/phenoxyl radical ν8a mode in 1+ [9] is red-shifted by 22 cm-1 (1625 cm-1 to 1605 cm-1) in comparison to the phenolate ν8a mode in 1,[29] consistent with delocalization of the ligand radical for 1+ on the rR timescale. The feature at 1615 cm-1 in 4+, assigned to the phenolate ν8a mode, is only red-shifted by 4 cm-1 in comparison to the corresponding feature in 4.[29] These rR results further support a localized ligand radical description for 4+.
In summary, we have characterized the electronic structure of a non-symmetric one-electron oxidized NiII bis-phenoxide complex 4+. While the symmetric derivative 1+ is a Class III mixed valence species, 4+ is best described as Class II mixed valence species due to localization of the ligand radical on the more electron rich amino-phenolate.
Experimental Section
Synthesis of 4+ SbF6- : 4 (0.097 g, 0.16 mmol) was dissolved in CH2Cl2 (3 mL) and solid AgSbF6 (0.055 g, 0.16 mmol) was added. A bright green suspension formed immediately. After 1 hr the mixture was filtered through celite and the solvent was removed in vacuo to afford a green solid. The material was recrystallized from CH2Cl2/ pentane to afford 4+ SbF6- as green block-like crystals. Magnetic susceptibility; (Evan’s Method) μeff = 1.6 BM. Anal. Calcd (found) for C36H54N2O2NiSbF6: C, 51.40 (51.56); H, 6.47 (6.18); N, 3.33 (3.28).
X-ray details for 4 and 4+ SbF6- are available in the supporting information. CCDC-770780 and CCDC-770781 contain supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. Synthetic procedures, X-ray data, computations, electrochemistry, UV-Vis data, X-ray absorption spectroscopy, and Raman analysis are available as supplementary information.
Supplementary Material
Figure 1.

Molecular structure of 4+SbF6- (50 % probability ellipsoids). Selected interatomic distances (Å) and angles [°]: Ni1-N1 1.872(5), Ni1-N2 1.824(5), Ni1-O1 1.883(4), Ni1-O2 1.823(3), C1-O1 1.365(6), C20-O2 1.329(6), N1-C7 1.472(2), N2-C14 1.283(7), N1-Ni1-N2 86.7(2), O1-Ni1-O2 84.2(1), N1-Ni1-O1 93.3(2), N2-Ni1-O2 95.3(2).
Acknowledgments
This work was supported by NIH grant GM-50730 (TDPS), the CSUC College of Natural Science (ECW), an NSERC post-doctoral scholarship (TS), and a Grant-in-Aid for Scientific Research No. 17750055 (YS). Dr. Allen Oliver is thanked for X-ray analysis. SSRL operations are funded by the DOE, Office of Basic Energy Services. Prof. Fumito Tani, Kushu University is acknowledged for Raman analysis. Mr. Mithi Adhikari is thanked for the solution magnetic measurement of 4+.
Footnotes
Supporting information for this article is available on the WWW under http://www.chemeurj.org/ or from the author.
Contributor Information
Prof. Yuichi Shimazaki, Email: yshima@mx.ibaraki.ac.jp.
Prof. T. Daniel P. Stack, Email: stack@stanford.edu.
References
- 1.a) Chaudhuri P, Wieghardt K. Prog Inorg Chem. 2001;50:151. doi: 10.1021/ic010552y. [DOI] [PubMed] [Google Scholar]; b) Thomas F. Eur J Inorg Chem. 2007:2379. [Google Scholar]; c) Pierpont CG. Coord Chem Rev. 2001;216:99. [Google Scholar]
- 2.Ray K, Petrenko T, Wieghardt K, Neese F. Dalton Trans. 2007:1552. doi: 10.1039/b700096k. [DOI] [PubMed] [Google Scholar]
- 3.Stubbe J, van der Donk WA. Chem Rev. 1998;98:705. doi: 10.1021/cr980059c. [DOI] [PubMed] [Google Scholar]; a) Whittaker JW. Chem Rev. 2003;103:2347. doi: 10.1021/cr020425z. [DOI] [PubMed] [Google Scholar]
- 4.a) Bouwkamp MW, Bowman AC, Lobkovsky E, Chirik PJ. J Am Chem Soc. 2006;128:13340. doi: 10.1021/ja064711u. [DOI] [PubMed] [Google Scholar]; b) Chirik PJ, Wieghardt K. Science. 2010;327:794. doi: 10.1126/science.1183281. [DOI] [PubMed] [Google Scholar]; c) Konigsmann M, Donati N, Stein D, Schonberg H, Harmer J, Sreekanth A, Grutzmacher H. Angew Chem Int Ed. 2007;46:3567. doi: 10.1002/anie.200605170. [DOI] [PubMed] [Google Scholar]; d) Nguyen AI, Blackmore KJ, Carter SM, Zarkesh RA, Heyduk AF. J Am Chem Soc. 2009;131:3307. doi: 10.1021/ja808542j. [DOI] [PubMed] [Google Scholar]; e) Ringenberg MR, Kokatam SL, Heiden ZM, Rauchfuss TB. J Am Chem Soc. 2008;130:788. doi: 10.1021/ja076801k. [DOI] [PubMed] [Google Scholar]; f) Dzik WI, Reek JNH, de Bruin B. Chem Eur J. 2008;14:7594. doi: 10.1002/chem.200800262. [DOI] [PubMed] [Google Scholar]; g) Hetterscheid DGH, Kaiser J, Reijerse E, Peters TPJ, Thewissen S, Blok ANJ, Smits JMM, de Gelder R, de Bruin B. J Am Chem Soc. 2005;127:1895. doi: 10.1021/ja0439470. [DOI] [PubMed] [Google Scholar]
- 5.a) Dolphin D, Niem T, Felton RH, Fujita I. J Am Chem Soc. 1975;97:5288. doi: 10.1021/ja00851a050. [DOI] [PubMed] [Google Scholar]; b) Ohtsu H, Tanaka K. Angew Chem Int Ed. 2004;43:6301. doi: 10.1002/anie.200460023. [DOI] [PubMed] [Google Scholar]; c) Puschmann FF, Harmer J, Stein D, Ruegger H, de Bruin B, Grutzmacher H. Angew Chem Int Ed. 2010;49:385. doi: 10.1002/anie.200903201. [DOI] [PubMed] [Google Scholar]
- 6.Storr T, Verma P, Pratt RC, Wasinger EC, Shimazaki Y, Stack TDP. J Am Chem Soc. 2008;130:15448. doi: 10.1021/ja804339m. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rotthaus O, Jarjayes O, Thomas F, Philouze C, Del Valle CP, Saint-Aman E, Pierre JL. Chem Eur J. 2006;12:2293. doi: 10.1002/chem.200500915. [DOI] [PubMed] [Google Scholar]
- 8.Rotthaus O, Thomas F, Jarjayes O, Philouze C, Saint-Aman E, Pierre JL. Chem Eur J. 2006;12:6953. doi: 10.1002/chem.200600258. [DOI] [PubMed] [Google Scholar]
- 9.Shimazaki Y, Tani F, Fukui K, Naruta Y, Yamauchi O. J Am Chem Soc. 2003;125:10512. doi: 10.1021/ja035806o. [DOI] [PubMed] [Google Scholar]
- 10.Storr T, Wasinger EC, Pratt RC, Stack TDP. Angew Chem Int Ed. 2007;46:5198. doi: 10.1002/anie.200701194. [DOI] [PubMed] [Google Scholar]
- 11.Rotthaus O, Jarjayes O, Del Valle CP, Philouze C, Thomas F. Chem Comm. 2007:4462. doi: 10.1039/b710027b. [DOI] [PubMed] [Google Scholar]
- 12.Shimazaki Y, Stack TDP, Storr T. Inorg Chem. 2009;48:8383. doi: 10.1021/ic901003q. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.See supporting information for details.
- 14.a) Nakano K, Nakamura M, Nozaki K. Macromolecules. 2009;42:6972. [Google Scholar]; b) Berkessel A, Brandenburg M, Leitterstorf E, Frey J, Lex J, Schafer M. Adv Synth Catal. 2007;349:2385. [Google Scholar]
- 15.Thomas F, Gellon G, Gautier-Luneau I, Saint-Aman E, Pierre JL. Angew Chem Int Ed. 2002;41:3047. doi: 10.1002/1521-3773(20020816)41:16<3047::AID-ANIE3047>3.0.CO;2-W. [DOI] [PubMed] [Google Scholar]
- 16.Pratt RC, Stack TDP. J Am Chem Soc. 2003;125:8716. doi: 10.1021/ja035837j. [DOI] [PubMed] [Google Scholar]
- 17.a) Benisvy L, Blake AJ, Collison D, Davies ES, Garner CD, McInnes EJL, McMaster J, Whittaker G, Wilson C. Chem Comm. 2001:1824. doi: 10.1039/b105186p. [DOI] [PubMed] [Google Scholar]; b) Sokolowski A, Bothe E, Bill E, Weyhermüller T, Wieghardt K. Chem Comm. 1996:1671. [Google Scholar]
- 18.Benisvy L, Blake AJ, Collison D, Davies ES, Garner CD, McInnes EJL, McMaster J, Whittaker G, Wilson C. Dalton Trans. 2003:1975. [Google Scholar]
- 19.Frisch MJ, et al. Gaussian 03, Revision C.02. Gaussian, Inc.; Wallingford, CT: 2004. [Google Scholar]
- 20.Manner VW, Markle TF, Freudenthal JH, Roth JP, Mayer JM. Chem Comm. 2008:256. doi: 10.1039/b712872j. [DOI] [PubMed] [Google Scholar]
- 21.The slight difference in Ni spin density for 1+ (10%) in comparison to our previous report (14%; see reference 10) is due to the use of the 6-311+g* basis set (Ni, O, N) here in comparison to the use of the TZVP basis set of Ahlrich’s (Schafer A, Horn H, Ahlrichs R. J Chem Phys. 1992;97:2571.) in the previous report.
- 22.Fujita K, Schenker R, Gu WW, Brunold TC, Cramer SP, Riordan CG. Inorg Chem. 2004;43:3324. doi: 10.1021/ic049876n. [DOI] [PubMed] [Google Scholar]
- 23.a) Robin MB, Day P. Adv Inorg Chem Radiochem. 1967;10:247. [Google Scholar]; b) D’Alessandro DM, Keene FR. Chem Soc Rev. 2006;35:424. doi: 10.1039/b514590m. [DOI] [PubMed] [Google Scholar]
- 24.Schnepf R, Sokolowski A, Müller J, Bachler V, Wieghardt K, Hildebrandt P. J Am Chem Soc. 1998;120:2352. [Google Scholar]
- 25.a) Radziszewski J, Gil M, Gorski A, Spanget-Larsen J, Waluk J, Mroz B. J Chem Phys. 2002;116:5912. [Google Scholar]; b) Spanget-Larsen J, Gil M, Gorski A, Blake DM, Waluk J, Radziszewski JG. J Am Chem Soc. 2001;123:11253. doi: 10.1021/ja0113596. [DOI] [PubMed] [Google Scholar]
- 26.Schnepf R, Sokolowski A, Muller J, Bachler V, Wieghardt K, Hildebrandt P. J Am Chem Soc. 1998;120:2352. [Google Scholar]
- 27.Tripathi GNR, Schuler RH. J Phys Chem. 1988;92:5129. [Google Scholar]
- 28.Shimazaki Y, Huth S, Karasawa S, Hirota S, Naruta Y, Yamauchi O. Inorg Chem. 2004;43:7816. doi: 10.1021/ic049040k. [DOI] [PubMed] [Google Scholar]
- 29.Selected raman peaks [cm-1] (λex 413 nm) 4: 1350, 1530, 1619 (ν8a). 4+: 918, 1350, 1370, 1501 (PhO• ν7a), 1530, 1581 (PhO• ν8a), 1615 (ν8a). 1: 1351, 1533, 1627 (ν8a).1+: 1370, 1504 (PhO• ν7a), 1605 (PhO• ν8a).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.

