Effects of electron-deficient β-diketiminate and formazan supporting ligands on copper(I)-mediated dioxygen activation

Article abstract of DOI:10.1021/ic9002466

Copper(I) complexes of a diketiminate featuring CF₃ groups on the backbone and dimethylphenyl substituents (4) and a nitroformazan (5) were synthesized and shown by spectroscopy, X-ray crystallography, cyclic voltammetry, and theory to contain copper(l) sites electron-deficient relative to those supported by previously studied diketiminate complexes comprising alkyl or aryl backbone substituents. Despite their electron-poor nature, oxygenation of LCu(CH3CN) (L = 4 or 5) at room temperature yielded bis(hydroxo)dicopper(II) compounds and at -80 0C yielded bis(w-oxo)dicopper complexes that were identified on the basis of UV-vis and resonance Raman spectroscopy, spectrophotometric titration results (2:1 Cu/02 ratio), electron paramagnetic resonance spectroscopy (silent), and density functional theory calculations. The bis(w-oxo)dicopper complex supported by 5 exhibited unusual spectroscopic properties and decayed via a novel intermediate proposed to be a metallaverdazyl radical complex, findings that highlight the potential for the formazan ligand to exhibit "noninnocent" behavior.

Full text of DOI:10.1021/ic9002466

Inorg. Chem. 2009, 48, 4514-4523
Effects of Electron-Deficient ꢀ-Diketiminate and Formazan Supporting
Ligands on Copper(I)-Mediated Dioxygen Activation
Sungjun Hong,† Lyndal M. R. Hill,† Aalo K. Gupta,† Benjamin D. Naab,† Joe B. Gilroy,‡
Robin G. Hicks,*,‡ Christopher J. Cramer,*,† and William B. Tolman*,†
Department of Chemistry, Center for Metals in Biocatalysis, and Supercomputing Institute,
UniVersity of Minnesota, 207 Pleasant St. SE, Minneapolis, Minnesota 55455, and Department of
Chemistry, UniVersity of Victoria, P.O. Box 3065, Victoria, British Columbia, Canada V8W 3V6
Received February 5, 2009
Copper(I) complexes of a diketiminate featuring CF3 groups on the backbone and dimethylphenyl substituents (4)
and a nitroformazan (5) were synthesized and shown by spectroscopy, X-ray crystallography, cyclic voltammetry,
and theory to contain copper(I) sites electron-deficient relative to those supported by previously studied diketiminate
complexes comprising alkyl or aryl backbone substituents. Despite their electron-poor nature, oxygenation of
LCu(CH3CN) (L ) 4 or 5) at room temperature yielded bis(hydroxo)dicopper(II) compounds and at -80 °C yielded
bis(µ-oxo)dicopper complexes that were identified on the basis of UV-vis and resonance Raman spectroscopy,
spectrophotometric titration results (2:1 Cu/O2 ratio), electron paramagnetic resonance spectroscopy (silent), and
density functional theory calculations. The bis(µ-oxo)dicopper complex supported by 5 exhibited unusual spectroscopic
properties and decayed via a novel intermediate proposed to be a metallaverdazyl radical complex, findings that
highlight the potential for the formazan ligand to exhibit “noninnocent” behavior.
Introduction
topologies, and redox states and to the discovery of structure/
reactivity correlations relevant to biological and other
catalytic processes. Despite this progress, much remains to
be learned, particularly about how the nature of supporting
ligand(s)influencesthestructuresandpropertiesofcopper-oxygen
intermediates. Previously, ligand denticity, steric profile, and
electronic properties were shown to be important determi-
nants, with notable emphasis having been placed on evaluat-
ing how these factors control the relative stability of
interconverting peroxo- and bis(µ-oxo)dicopper isomers.2,3
In prior work, we focused our attention on the O2 reactivity
of copper(I) complexes of strongly electron-donating, monoan-
ionic, and bidentate diketiminate (and anilidoimine)4 ligands,
the steric properties of which can be manipulated readily
through variation of substituents R, R′, and/or R′′ (Figure 1,
top). For most of the diketiminate ligands, bis(µ-oxo)dicopper
The binding and activation of dioxygen by Cu ions is
central to the function of numerous biological and syntheti-
cally useful catalytic systems.1 In efforts to understand the
underlying chemistry, studies of the O2 reactivity of copper(I)
complexes with a variety of supporting ligand types have
been performed.2 These studies have led to the identification
of copper-oxygen intermediates with diverse nuclearities,
* To whom correspondence should be addressed. E-mail: rhicks@uvic.ca
(R.G.H.), cramer@umn.edu (C.J.C.), wtolman@umn.edu (W.B.T.).
†
University of Minnesota.
University of Victoria.
‡
(1) Selected recent reviews: (a) Solomon, E. I.; Chen, P.; Metz, M.; Lee,
S.-K.; Palmer, A. E. Angew. Chem., Int. Ed. 2001, 40, 4570–4590.
(b) Solomon, E.; Sarangi, R.; Woertink, J.; Augustine, A.; Yoon, J.;
Ghosh, S. Acc. Chem. Res. 2007, 40, 581–591. (c) Klinman, J. P.
J. Biol. Chem. 2006, 281, 3013–3016. (d) Arends, I.; Gamez, P.;
Sheldon, R. A. AdV. Inorg. Chem. 2006, 58, 235–279. (e) Battaini,
G.; Granata, A.; Monzani, E.; Gullotti, M.; Casella, L. AdV. Inorg.
Chem. 2006, 58, 185–233. (f) Punniyamurthy, T.; Rout, L. Coord.
Chem. ReV. 2008, 252, 134–154.
(2) Selected recent reviews:(a) Lewis, E. A.; Tolman, W. B. Chem. ReV.
2004, 104, 1047–1076. (b) Mirica, L. M.; Ottenwaelder, X.; Stack,
T. D. P. Chem. ReV. 2004, 104, 1013–1045. (c) Hatcher, L. Q.; Karlin,
K. D. AdV. Inorg. Chem. 2006, 58, 131–184. (d) Hatcher, L.; Karlin,
K. D. J. Biol. Inorg. Chem. 2004, 9, 669–683. (e) Itoh, S. Curr. Opin.
Chem. Biol. 2006, 10, 115–122. (f) Cramer, C. J.; Tolman, W. B.
Acc. Chem. Res. 2007, 40, 601–608. (g) Suzuki, M. Acc. Chem. Res.
2007, 40, 609–617.
(3) For example, see:(a) Cole, A. P.; Mahadevan, V.; Mirica, L. M.;
Ottenwaelder, X.; Stack, T. D. P. Inorg. Chem. 2005, 44, 7345–7364.
(b) Stack, T. D. P. Dalton Trans. 2003, 1881–1889. (c) Lam, B. M. T.;
Halfen, J. A.; Young, V. G., Jr.; Hagadorn, J. R.; Holland, P. L.;
Lledo´s, A.; Cucurull-Sa´nchez, L.; Novoa, J. J.; Alvarez, S.; Tolman,
W. B. Inorg. Chem. 2000, 39, 4059–4072. (d) Hatcher, L. Q.; Vance,
M. A.; Narducci Sarjeant, A. A.; Solomon, E. I.; Karlin, K. D. Inorg.
Chem. 2006, 45, 3004–3013.
(4) Reynolds, A. M.; Gherman, B. F.; Cramer, C. J.; Tolman, W. B. Inorg.
Chem. 2005, 44, 6989–6997.
4514 Inorganic Chemistry, Vol. 48, No. 10, 2009
10.1021/ic9002466 CCC: $40.75  2009 American Chemical Society
Published on Web 04/07/2009

Effects of Electron-Deficient Ligands on Cu/O2 Chemistry
supported by 4,9 which we proposed would have electronic/
redox properties similar to those of 2 but by virtue of its
decreased steric bulk could enable the irreversible formation
of a bis(µ-oxo)dicopper complex like that known for 3.5b In
parallel, we turned to nitroformazan 5 as a potential
supporting ligand analogous to the diketiminates10,11 but with
unique properties, including lower electron-donating capa-
bilities and accessibility of a verdazyl radical form.12,13
Herein we report the synthesis and characterization of
copper(I) complexes of electron-poor ligands 4 and 5 (relative
to diketiminates like 3), along with the results of their low-
temperature oxygenation reactions. Combined spectroscopic
and theoretical studies support the formation of bis(µ-
oxo)dicopper complexes in both cases, implicating the
intermediacy of 1:1 Cu/O2 adducts as transient intermediates.
As a result of the unique characteristics of 5, its bis(µ-
oxo)dicopper complex exhibits unusual properties and decays
via the intermediacy of a novel ligand-centered radical
species.
Figure 1. Diketiminate and formazan (5) ligands.
complexes were observed,5 whereas the highly hindered
variants having R′′ ) iPr and R′ ) Me or tBu enabled
characterization of
a side-on monocopper-dioxygen
adduct.2f,5a,6 Detailed spectroscopic and theoretical studies
showed that this adduct is best described as a copper(III)
peroxide as a result of the powerful electron donation of the
diketiminate ligand.6,7 Calculations also showed, however,
that an end-on (diketiminate)copper(II) superoxide isomer
is only ∼5 kcal mol-1 less stable than the side-on product6b
and that decreasing the electron-donating power of the
diketiminate could decrease this energy gap and potentially
enable isolation of the end-on copper(II) superoxide form.8
In an initial experimental test of this hypothesis, copper(I)
complexes of 1 and 2 featuring CF3 replacements for the
backbone methyl groups were prepared and reacted with O2.
While (1)Cu(NCMe) yielded a side-on copper(III) peroxide,
no intermediate was observed for (2)Cu(NCMe), consistent
with calculations showing that the oxygenation is endergonic
in this case.8 It was therefore concluded that, in order for a
bidentate ligand to support end-on copper(II) superoxide
formation, a balance must be achieved such that the ligand
electronics favor this isomer over the side-on copper(III)
peroxide variant, but which also results in favorable ther-
modynamics for O2 binding.8
Experimental Section
General Considerations. All solvents and reagents were ob-
tained from commercial sources and used as received unless
otherwise noted. The solvents tetrahydrofuran (THF), pentane, and
Et2O were dried over sodium/benzophenone and distilled under
vacuum or passed through solvent purification columns (Glass
Contour, Laguna, CA). Acetonitrile (CH3CN) was dried over CaH2
and distilled under vacuum prior to use. [Cu4Mes4] (Mes )
mesityl)14 and the protonated forms of 49 and 510 were prepared
according to published procedures. All metal complexes were
prepared and stored in a Vacuum Atmospheres inert-atmosphere
glovebox under a dry N2 atmosphere or were manipulated using
standard inert-atmosphere vacuum and Schlenk techniques. NMR
spectra were recorded on either Varian VI-300 or VI-500 spec-
trometers at room temperature. Variable low-temperature NMR
spectra were recorded on a Varian VI-300 spectrometer. Chemical
shifts (δ) for 1H (300 MHz) and 13C (75 MHz) NMR spectra were
referenced to residual solvent signal. UV-vis spectra were recorded
on an HP8453 (190-1100 nm) diode-array spectrophotometer.
Low-temperature spectra were acquired through the use of a
Unisoku low-temperature UV-vis cell holder. When necessary,
UV-vis spectra were corrected for drifting baselines due to minimal
frosting of the UV cells caused by the low-temperature device. This
was achieved by subtracting the average of a region with no
absorbance (i.e., baseline, typically 950-1000 nm) from the entire
spectrum. X-band electron paramagnetic resonance (EPR) spectra
were recorded on a Bruker E-500 spectrometer with an Oxford
Instruments EPR-10 liquid-helium cryostat (4-20 K, 9.61 GHz).
Resonance Raman (rR) spectra were recorded on an Acton 506
spectrometer using a Princeton Instruments LN/CCD-11100-PB/
The research described herein was undertaken to further
explore bidentate ligand electronic and structural effects on
the CuI/O2 reactivity. We hypothesized that while not
observed, a 1:1 Cu/O2 adduct from oxygenation of (2)Cu(NC-
Me) might be present at a low equilibrium concentration.
To test this idea, we sought to trap a similar 1:1 Cu/O2 adduct
(5) (a) Spencer, D. J. E.; Aboelella, N. W.; Reynolds, A. M.; Holland,
P. L.; Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 2108–2809. (b)
Spencer, D. J. E.; Reynolds, A. M.; Holland, P. L.; Jazdzewski, B. A.;
Duboc-Toia, C.; Pape, L. L.; Yokota, S.; Tachi, Y.; Itoh, S.; Tolman,
W. B. Inorg. Chem. 2002, 41, 6307–6321.
(6) (a) Aboelella, N. W.; Lewis, E. A.; Reynolds, A. M.; Brennessel,
W. W.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2002, 124,
10660–10661. (b) Aboelella, N. W.; Kryatov, S. V.; Gherman, B. F.;
Brennessel, W. W.; Victor, G.; Young, J.; Sarangi, R.; Rybak-
Akimova, E. V.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Cramer,
C. J.; Tolman, W. B. J. Am. Chem. Soc. 2004, 126, 16896–16911.
(7) (a) Sarangi, R.; Aboelella, N.; Fujisawa, K.; Tolman, W. B.; Hedman,
B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2006, 128,
8286–8296. (b) Gherman, B. F.; Tolman, W. B.; Cramer, C. J.
J. Comput. Chem. 2006, 27, 1950–1961.
(9) Carey, D. T.; Cope-Eatough, E. K.; Vilaplana-Mafe´, E.; Mair, F. S.;
Pritchard, R. G.; Warren, J. E.; Woods, R. J. Dalton Trans. 2003,
1083–1093.
(10) Gilroy, J. B.; Otieno, P. O.; Ferguson, M. J.; McDonald, R.; Hicks,
R. G. Inorg. Chem. 2008, 47, 1279–1286.
(11) Gilroy, J. B.; Patrick, B. O.; McDonald, R.; Hicks, R. G. Inorg. Chem.
2008, 47, 1287–1294.
(12) Koivisto, B. D.; Hicks, R. G. Coord. Chem. ReV. 2005, 249, 2612–
2630.
(13) Gilroy, J. B.; Ferguson, M. J.; McDonald, R.; Patrick, B. O.; Hicks,
R. G. Chem. Commun. 2007, 126.
(8) Hill, L. M. R.; Gherman, B. F.; Aboelella, N. W.; Cramer, C. J.;
Tolman, W. B. Dalton Trans. 2006, 4944–4953.
(14) Tsuda, T.; Yazawa, T.; Watanabe, K.; Fujii, T.; Saegusa, T. J. Org.
Chem. 1981, 46, 192–194.
Inorganic Chemistry, Vol. 48, No. 10, 2009 4515

Hong et al.
UVAR detector and a ST-1385 controller interfaced with Winspec
software. The spectra were obtained at -196 °C using a backscat-
tering geometry. Excitation at 457.9 and 514.5 nm was provided
by a Spectra Physics BeamLok 2065-7S argon laser. Raman shifts
were externally referenced to liquid indene. Elemental analyses were
performed by Robertson Microlit Laboratory. Electrospray ioniza-
tion mass spectrometry spectra were recorded on a Bruker BioTOF
II instrument. IR spectra were obtained using a ThermoNicolet
Avatar 370 FT-IR equipped with an attenuated total reflectance
attachment, using a CaF2 solution cell (International Crystal
Laboratories). Cyclic voltammograms were recorded using platinum
working and auxiliary electrodes, a silver wire/AgNO3 (10 mM in
CH3CN) reference electrode, and a BAS Epsilon potentiostat
connected to a 22 mL cell in an inert-atmosphere glovebox.
Experiments were performed using analyte concentrations of 1 mM
in THF with 0.3 M tetrabutylammonium hexafluorophosphate,
TBAPF6 (sample volumes of ∼5 mL), at room temperature. The
ferrocene/ferrocenium (Fc/Fc+) couple was recorded for reference.
Theory. All molecular structures were fully optimized using the
generalized gradient approximation (GGA) density functional
(mPW) that combines the exchange functional of Perdew,15 as
modified by Adamo and Barone,16 with the correlation functional
of Perdew and Wang.17 Atomic orbital basis functions were taken
for copper from the Stuttgart effective core potential and basis set,18
including two f functions having exponents 5.208 and 1.315, and
for all other atoms from the 6-31G(d) basis set.19 Resolution of
the identity density fitting was employed in all cases, and restricted
singlet Kohn-Sham wave functions were determined in every
instance to be stable relative to unrestricted alternatives [restricted
to unrestricted singlet instabilities have been demonstrated in other
bis(µ-oxo) species20]. Analytical vibrational frequencies were
computed for all stationary points in order to confirm their nature
as minima or transition-state structures and for comparison to IR
or rR spectra, as appropriate.21 For select molecules, the first 90
vertical electronic excitation energies were computed at the time-
dependent (TD-)DFT level22 using the hybrid GGA B3LYP23-26
functional and a polarized valence double-ꢀ basis set27,28 on all
atoms. Electronic structure calculations were accomplished with
the Gaussian 0329 and Turbomole27 program suites.
Celite to remove any insoluble residue. The solvent was removed
from the filtrate in vacuo to give a brown residue. The product
was isolated in analytically pure form by allowing a concentrated
solution of the residue in CH3CN to stand for 3 days at -20 °C.
The resulting orange crystals were separated from the mother liquor,
washed with cold pentane (∼5 mL), and dried in vacuo (93 mg,
1
75%). H NMR (300 MHz, benzene-d6): δ 6.99 (d, J ) 7.5 Hz,
4H), 6.91 (t, J ) 7.5 Hz, 2H), 6.11 (s, 1H), 2.11 (s, 12H), 0.28 (s,
3H). 13C{1H} NMR (75.0 MHz, benzene-d6): δ 152.40, 148.56,
143.57, 130.33, 124.39, 123.23, 119.43, 83.77, 19.19, 0.27. UV-vis
[λmax, nm (ꢀ, M-1 cm-1) in THF]: 280 (18 900), 381 (16 600), 500
(460). Anal. Calcd for C23H22CuF6N3: C, 53.33; H, 4.28; N, 8.11.
Found: C, 53.56; H, 4.08; N, 8.08.
(5)Cu(CH3CN). This compound was synthesized following the
same procedure as described above, except using H(5) as the starting
material and crystallizing the product by slow diffusion of pentane
1
into a concentrated THF solution at -20 °C (99 mg, 80%). H
NMR (300 MHz, benzene-d6): δ 7.15 (m, 6H), 3.40 (septet, J )
6.9 Hz, 4H), 1.24 (d, J ) 6.9 Hz, 24H), 0.17 (s, 3H). 13C{1H}
NMR (75.0 MHz, benzene-d6): δ 151.46, 141.24, 127.85, 124.29,
29.06, 25.02, 23.71, 0.27. UV-vis [λmax, nm (ꢀ, M-1 cm-1) in THF]:
287 (11 700), 352 (9480), 410 (10 500), 650 (380). Anal. Calcd
for C27H37CuN6O2: C, 59.92; H, 6.89; N, 15.53. Found: C, 59.75;
H, 6.91; N, 15.74.
LCu(CO) (L ) 3, 4, or 5). General Procedure. The complex
LCu(CH3CN) (L ) 3, 4, or 5; 37 µmol) was dissolved in benzene-
d6 (0.8 mL) in a screw-capped NMR tube. Carbon monoxide (CO)
was gently bubbled through the solution for 20 min at ambient
temperature, during which time the color changed to gold for L )
3 and 4 and dark red for L ) 5. The CO adducts were immediately
1
characterized by H and 13C NMR spectroscopy. For IR charac-
terization, the complex LCu(CH3CN) (37 µmol) was dissolved in
THF (2 mL) in a 10 mL Schlenk tube and stirred with CO bubbling
at ambient pressure for 2 h, during which time the solvents
evaporated. Approximately half of the dried solid (golden to dark
red) was dissolved in THF (0.4 mL), and the IR spectrum was
1
recorded immediately. (3)Cu(CO). H NMR (300 MHz, benzene-
d6): δ 7.05 (d, J ) 7.5 Hz, 4H), 6.94 (t, J ) 7.5 Hz, 2H), 4.92 (s,
1H), 2.21 (s, 12H), 1.62 (s, 6H). 13C{1H} NMR (75.0 MHz,
benzene-d6): δ 164.25, 152.56, 130.40, 128.98, 124.24, 95.59, 22.49,
12.29. FT-IR (THF): 2071 cm-1 (νCO). (4)Cu(CO). 1H NMR (300
MHz, benzene-d6): δ 6.96-6.86 (m, 6H), 6.16 (s, 1H), 2.13 (s,
12H). 13C{1H} NMR (75.0 MHz, benzene-d6): δ 153.65, 149.30,
129.92, 125.48, 122.89, 119.09, 85.67, 19.12. FT-IR (THF): 2100
cm-1 (νCO). (5)Cu(CO). 1H NMR (300 MHz, benzene-d6): δ
7.24-7.08 (m, 6H), 3.24 (sept, J ) 6.9 Hz, 4H), 1.20 (s, 24H).
(4)Cu(CH3CN). In an inert atmosphere, to a solution of the
ligand precursor H(4) (100 mg, 0.24 mmol) in CH3CN (5 mL) was
added [Cu4Mes4] (44 mg, 0.06 mmol). After the reaction mixture
was vigorously stirred overnight, it was filtered through a plug of
(15) Perdew, J. P. Unified Theory of Exchange and Correlation Beyond
the Local Density Approximation. In Electronic Structure of Solids
′91; Ziesche, P., Eschrig, H., Eds.; Akademie Verlag: Berlin, 1991;
pp 11-20.
(16) Adamo, C.; Barone, V. J. Chem. Phys. 1998, 108, 664–675.
(17) Perdew, J.; Wang, Y. Phys. ReV. B 1992, 45, 13244–13249.
(18) Dolg, M.; Wedig, U.; Stoll, H.; Preuss, H. J. Chem. Phys. 1987, 86,
866–872.
(19) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(20) Gherman, B. F.; Cramer, C. J. Coord. Chem. ReV. 2009, 253, 723–
753.
(21) Cramer, C. J. Essentials of Computational Chemistry: Theories and
Models, 2nd ed.; John Wiley & Sons: Chichester, U.K., 2004.
(22) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454–
464.
(23) Becke, A. D. Phys. ReV. A 1988, 38, 3098–3100.
(24) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(25) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789.
(26) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623–11627.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision
D.02; Gaussian, Inc.: Wallingford, CT, 2004.
(27) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel, C. Chem. Phys.
Lett. 1989, 162, 165–169.
(28) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571–
2577.
(30) Battino, R. Oxygen and Ozone; Battino, R., Ed.; Pergamon Press: New
York, 1981; Vol. 7.
4516 Inorganic Chemistry, Vol. 48, No. 10, 2009

Effects of Electron-Deficient Ligands on Cu/O2 Chemistry
prepared by transferring a sample of the reaction mixture either to
a copper cup or to a precooled NMR tube (-80 °C) followed by
immersion in liquid N2.
(c) EPR Spectroscopy. A 10 mL Schlenk flask was charged
with a THF solution of LCu(CH3CN) (L ) 4 or 5) (2 mM) and
was cooled to -80 °C in an acetone/dry ice bath. Dry O2 was
bubbled through the solution (30 min for L ) 4 and 10 min for L
) 5), and the resulting solution was transferred via a precooled
syringe to a precooled (-80 °C) EPR tube, which was immersed
in liquid N2 for subsequent analysis by EPR spectroscopy. Also, a
quartz cuvette was charged with a THF solution of (5)Cu(CH3CN)
(2 mM), cooled to -30 °C, and oxygenated by bubbling dry O2
with monitoring of the feature at 742 nm by UV-vis spectroscopy.
Aliquots of the solution were rapidly transferred at various time
points via a syringe to a precooled (-80 °C) EPR tube, which was
immersed in liquid N2 for subsequent analysis by EPR spectroscopy.
(d) Spectrophotometric O2 Titrations. A 1 mL sample of a
stock solution of LCu(CH3CN) (L ) 4 or 5) in THF (1.0 mM for
4 and 0.5 mM for 5) was placed in a 0.2-cm-path-length UV-vis
cuvette and cooled to -80 °C, and the headspace was evacuated.
Spectra were taken before and after cooling to ensure that no sample
degradation had occurred. An O2-saturated THF solution (10 mM)
was prepared by bubbling dry O2 gas through argon-saturated THF
in a 10 mL Schlenk tube at 25 °C for 15 min.29 Using a graduated
gastight syringe, portions of the 10 mM O2-saturated THF solution
(10, 20, 30, 40, 50, 75, 100, 150, and 200 µL) were injected into
the cuvette, where it was left to equilibrate with stirring. The
progress of oxygenation was followed by monitoring of the
absorption band at 470 nm (4) or 524 nm (5) in the UV-vis
spectrum until no further increase in absorbance was observed. The
Cu/O2 stoichiometry was calculated from two (L ) 4) or three (L
) 5) replicate runs.
Figure 2. Representation of the X-ray structure of (4)Cu(CH3CN) with all
non-H atoms shown as 50% thermal ellipsoids. Selected interatomic
distances (Å) and angles (deg): Cu1-N3, 1.871(2); Cu1-N1, 1.902(2);
Cu1-N2, 1.998(2); N3-Cu1-N1, 148.46(9); N3-Cu1-N2, 111.57(9);
N1-Cu1-N2, 99.01(9).
13C{1H} NMR (75.0 MHz, benzene-d6): δ 151.62, 151.34, 141.30,
129.92, 124.52, 29.09, 25.40, 23.43. FT-IR (THF): 2092 cm-1 (νCO).
L2Cu2(OH)2 (L ) 4 or 5). Solutions of LCu(CH3CN) (L ) 4
or 5) in THF (∼50 mg in 10 mL) were oxygenated by bubbling of
O2 at ambient temperature for ∼1 min. For L ) 4, the resulting
red-brown solution was layered with ∼10 mL pentane and allowed
to stand at -20 °C. Red single crystals formed after several days
and were either mounted for analysis by X-ray crystallography or
collected by decanting the mother liquor, washing with cold pentane,
and drying in vacuo (19 mg, 40%). Anal. Calcd for
C42H40Cu2F12N4O2: C, 51.06; H, 4.08; N, 5.67. Found: C, 51.76;
H, 4.30; N, 5.27. Red single crystals of the product of the reaction
of (5)Cu(CH3CN) were obtained by slow evaporation of the red-
brown reaction solution and were isolated similarly (21 mg, 44%).
Anal. Calcd for C50H70Cu2N10O6: C, 58.06; H, 6.82; N, 13.54.
Found: C, 57.67; H, 6.91; N, 13.28.
Results and Discussion
Low-Temperature Oxygenations of LCu(CH3CN) (L ) 4
or 5). (a) UV-Vis Spectroscopy. An anaerobically prepared THF
solution of (4)Cu(CH3CN) (0.2 mM) or (5)Cu(CH3CN) (0.1 mM)
in a septum-sealed quartz cuvette was cooled to -80 °C, and a dry
stream of O2 was bubbled through the solution for ∼20 min or
∼100 s, respectively. The spectroscopic changes are shown in
Figure 5. The final spectrum was not perturbed by removal of excess
O2 from the solution by an argon purge. For L ) 5, after bubbling
argon for 30 min through the solution to remove the remaining
unreacted O2, the solution was warmed above -50 °C, resulting in
the disappearance of the band at 525 nm and the growth of a new
sharp feature at 742 nm, which subsequently decayed (Figure 8).
The same intermediate spectrum was also generated by directly
oxygenating (5)Cu(CH3CN) above -50 °C.
(b) rR Spectroscopy. An argon-filled Schlenk flask or NMR
tube was charged with 1 mL of a THF solution of LCu(CH3CN)
(L ) 4 or 5; 40 mM for L ) 4 and 10 mM for L ) 5) by syringe
and cooled to -80 °C by submersion in an acetone/dry ice bath
while maintaining an argon purge. Dry O2 was bubbled through
the solution (30 min for L ) 4 and 10 min for L ) 5). Samples
were frozen either in a copper cup attached to a liquid-N2-cooled
coldfinger (L ) 4) or by immersion in the NMR tube in liquid N2
(L ) 5). For reactions with 18O2, a 10 mL Schlenk tube charged
with 1 mL of a THF solution of LCu(CH3CN) (L ) 4 or 5) was
frozen in liquid N2 and ∼10 mL of 18O2 was vacuum-transferred
to the flask. The resulting mixture was warmed to -80 °C by
submersion in an acetone/dry ice bath and stirred for 3 h to ensure
complete reaction. Samples for analysis by rR spectroscopy were
Copper(I) Complexes. The complexes LCu(CH3CN) (L
) 4 or 5) were prepared by the reaction of [Cu4Mes4] with
the protonated forms of 4 or 5 in CH3CN and were
characterized by NMR and UV-vis spectroscopy, elemental
analysis, cyclic voltammetry, and X-ray crystallography
(Figures 2and S1 in the Supporting Information). Both
compounds exhibit three-coordinate CuI ions and gross
structural features similar to other known copper(I) com-
plexes of diketiminate ligands, although the structure for
(5)Cu(CH3CN) suffers from disorder problems that preclude
more detailed analysis of interatomic distances and angles.
In (4)Cu(CH3CN), the CuI ion adopts a geometry close to
T-shaped, with divergent Cu-N(diketiminate) distances of
1.902(2) and 1.998(2) Å. The C(backbone)-N-C(aryl)
angles in (4)Cu(CH3CN) (122-125°) are wider than those
in the copper complex of the analogue 3 (116-120°),31,32
indicative of steric pressure from the CF3 groups that places
the aryl substituents closer to the bound metal ion. The effect
has important implications in reactivity studies (see below),
although even wider angles are observed for systems with
tert-butyl backbone substituents (128-129°).6,33
(31) Dai, X.; Warren, T. H. Chem. Commun. 2001, 1998–1999.
(32) York, J. T.; Young, V. G.; Tolman, W. B. Inorg. Chem. 2006, 45,
4191–4198.
(33) Vela, J.; Vaddadi, S.; Cundari, T. R.; Smith, J. M.; Gregory, E. A.;
Lachicotte, R. J.; Flaschenriem, C. J.; Holland, P. L. Organometallics
2004, 23, 5226–5239.
Inorganic Chemistry, Vol. 48, No. 10, 2009 4517

Hong et al.
Table 1. Properties of Copper(I) Complexes of Diketiminate and
acidic and less readily oxidized copper(I) center. A similar
effect is evident from the ν(CO) values for the complexes
of 3 and 4. Like 2, 4 has two backbone CF3 groups, and it
exhibits a similar ν(CO) value for its copper(I) carbonyl
complex, which implies analogous electron-donating proper-
ties for these two ligands, which only differ with respect to
their aryl substituents (Me vs iPr). Gas-phase theoretical
calculations of the CO stretching frequencies are in good
agreement with the measured values (see Table 1), indicating
the degree to which they reflect ligand effects on the
molecular electronic structure.
However, in apparent contradiction to the conclusions
based on the ν(CO) values, the E1/2 value for (4)Cu(CH3CN)
is 277 mV lower than that for the complex with 2. We
speculate that this difference results from two possible effects
of the differing aryl substituents of 2 and 4. One is the greater
hydrophobicity of the diisopropylphenyl groups in 2 com-
pared to the dimethylphenyl groups in 4, which for 2 would
result in relative destabilization of the more highly charged
copper(II) state and a greater E1/2. Another effect is primarily
steric; the copper(I) complex of more hindered 2 likely
remains three-coordinate upon oxidation, whereas the smaller
steric profile for 4 could enable binding of two (or more)
solvent molecules upon oxidation of its copper(I) complex,
resulting in enhanced stability for the copper(II) state and a
smaller E1/2. Because of these hydrophobic and steric
influences, we view the ν(CO) values as more reliable
indicators of the relative electron density at the metal center,
a conclusion similar to that reached in previous studies of
other diketiminate-copper complexes.5b Thus, on the basis
of the ν(CO) data, nitroformazan 5 and diketiminates 2 and
4 have similar electron-donating characteristics.
Oxygenation of Copper(I) Complexes. (a) Room Tem-
perature Reactions: Formation of Bis(hydroxo)dicopper
Complexes. Treatment of LCu(CH3CN) (L ) 4 or 5) in THF
with O2 at ambient temperature rapidly (<1 min) yielded
red-brown solutions, from which L2Cu2(µ-OH)2 could be
isolated as crystalline solids in modest yields (40-44%). The
products were identified by elemental analysis and X-ray
crystallography (Figures 3 and 4). The structure of
(4)2Cu2(OH)2 features a separation of 3.0195(6) Å between
CuII ions that adopt distorted square-planar geometries
characterized by a τ4 value of 0.31, where τ4 ) 1.00 is
associated with perfect tetrahedral and τ4 ) 0 with perfect
square-planar coordination geometries.34 Steric repulsions
between the aryl groups are apparently minimized by twisting
of the ligand planes with respect to each other, reflected by
a dihedral angle between the N atoms (N1-Cu1-N2/
N1′-Cu1′-N2′) of 56.31° (Figure 3b). Similar twisting has
been observed previously in a bis(µ-chloro)dicopper(II)
complex supported by a diketiminate ligand with the same
dimethylphenyl groups as in 3.5a Similar to the structure of
(4)Cu(CH3CN), the CF3 groups cause canting of the aryl
rings toward the coordinated Cu ion, as reflected by
C(backbone)-N-C(aryl) angles of almost 123°.
Formazan Ligands
ν(CO)
b
c
ligand E1/2 (mV)a exptl (cm-1
)
calcd (cm-1
)
ref
1
2
110
411
2083
2097
2074
8 (exptl), this
work (calcd)
8 (exptl), this
work (calcd)
this work
2082
3
4
5
d
134
103
2071
2100
2092
2070
2088
2084
this work
this work
a Measured for LCu(CH3CN) in THF with TBAPF6 (0.3 M) vs Fc/Fc+.
b Measured for the copper(I) carbonyl complex in THF. c Computed values
for copper(I) carbonyl complexes at the mPW level after correction by 19
cm-1, which is the difference between mPW and experiment for CO gas.
d Value reported previously but under different conditions than those used
in this work (CH3CN solvent), and this fact along with previously noted
complications with reproducibility for this complex makes comparison of
the ν(CO) values more useful for drawing comparisons among the
complexes listed.5b
In order to probe the effects of 4 and 5 on the redox
properties of their copper(I) compounds, we performed
electrochemistry experiments and prepared CO adducts for
characterization by FT-IR spectroscopy. Cyclic voltammo-
grams (Figure S2 in the Supporting Information) of THF
solutions of LCu(CH3CN) (L ) 4 or 5) with TBAPF6 (0.3
M) exhibited pseudoreversible waves with E1/2 values of
+134 mV (4) and +103 mV (5) versus Fc/Fc+ (∆Ep,c
)
118 and 112 mV, respectively, at a scan rate of 20 mV s-1).
These electrochemical results, as well as ν(CO) values for
their CO adducts, are compared to data for related com-
pounds obtained under similar conditions in Table 1. As
noted in previous work,8 replacement of the backbone methyl
group in 1 by a CF3 group to yield 2 results in a 14 cm-1
increase in the ν(CO) value for LCuCO and a ∼300 mV
increase in the E1/2 value for LCu(CH3CN), consistent with
2 being a poorer electron donor that yields a more Lewis
Figure 3. (a) Representation of the X-ray structure of (4)2Cu2(OH)2 with
all non-H atoms shown as 50% thermal ellipsoids. (b) View down the
Cu-Cu vector; C and F atoms are not shown for clarity. Selected interatomic
distances (Å) and angles (deg): Cu1-O1′, 1.9105(18); Cu1-O1, 1.9121(18);
Cu1-N2, 1.940(2); Cu1-N1, 1.943(2); Cu1-Cu1′, 3.0195(6); O1′-Cu1-O1,
75.44(9); O1′-Cu1-N2, 159.07(9); O1-Cu1-N2, 97.82(8); O1′-Cu1-N1,
97.93(8); O1-Cu1-N1, 157.40(9); N2-Cu1-N1, 95.28(8).
(34) Yang, L.; Powell, D.; Houser, R. Dalton Trans. 2007, 955–964.
4518 Inorganic Chemistry, Vol. 48, No. 10, 2009

Effects of Electron-Deficient Ligands on Cu/O2 Chemistry
M-1 cm-1; 5) that are suggestive of charge-transfer (CT)
transitions associated with bis(µ-oxo)dicopper cores.2a,b,36
Spectrophotometric titration results confirmed the requisite
2:1 Cu/O2 stoichiometry (insets to plots in Figure 5).
However, the UV-vis bands are shifted to longer wave-
lengths relative to typical [Cu2(µ-O)2]2+ core electronic
transitions that occur in the ranges 300-330 nm (ꢀ ∼
10 000-20 000 M-1 cm-1) and 400-450 nm (ꢀ ∼ 10 000-
30 000 M-1 cm-1). The 525 nm band for the complex
supported by 5 is a particularly notable outlier at energy
significantly different from that of typical bis(µ-oxo)dicopper
core transitions. A low-energy feature at 550 nm with an
extinction much lower than what we observe (1300 M-1
cm-1) was reported for a bis(µ-oxo)dicopper complex
supported by a guanidine ligand and assigned as a guanidine
f [Cu2(µ-O)2]2+ CT transition.37
To provide further insights, we performed rR spectroscopy
experiments on samples prepared using 16O2 or 18O2 with
excitation wavelengths of 457.9 nm (4) or 514.5 nm (5). The
spectrum for the oxygenated product of (4)Cu(CH3CN)
(Figure 6a) contains two peaks at 598 and 610 cm-1 that
are replaced by a single peak at 566 cm-1 upon 18O labeling.
This pattern has been seen previously for bis(µ-oxo)dicopper
complexes36,38 and is usually attributed to a symmetric Cu2O2
core vibration that appears as a Fermi doublet with 16O2 but
converts to a single peak with 18O2. For the case of the
reaction with (5)Cu(CH3CN) (Figure 6b), a peak at 581 cm-1
appears to shift to 558 cm-1 upon use of 18O2, although an
overlapping additional peak at 581 cm-1 complicates the
analysis. As shown in the inset to Figure 6b, frequency-
doubled overtone peaks are observed at 1160 cm-1 (16O2)
and 1112 cm-1 (18O2). The combined data for both cases 4
and 5 are consistent with the assignment of the 470 and 525
nm transitions, respectively, as CT transitions involving
bis(µ-oxo)dicopper cores. Also consistent with these formu-
lations, the intermediates are EPR-silent (X-band, 4 K).
Figure 4. (a) Representation of the X-ray structure of (5)2Cu2(OH)2 with
all non-H atoms shown as 50% thermal ellipsoids. (b) Alternate view by
rotating part (a) by 90°. Selected interatomic distances (Å) and angles (deg):
Cu1-O3, 1.9179(13); Cu1-O3′, 1.9294(13); Cu1-N1, 1.9519(15); Cu1-N3,
1.9646(15); Cu1-Cu1′, 3.0278(5); O3-Cu1′, 1.9294(13); O3-Cu1-O3′,
76.19(6); O3-Cu1-N1, 172.51(6); O3′-Cu1-N1, 96.95(6); O3-Cu1-N3,
98.50(6); O3′-Cu1-N3, 173.05(6); N1-Cu1-N3, 88.09(6); N3-Cu1-Cu1′,
136.57(5); Cu1-O3-Cu1′, 103.81(6).
Complex (5)2Cu2(OH)2 also features a planar bis(hydroxo)-
dicopper(II) core, with a Cu-Cu distance of 3.0278(5) Å
(Figure 4). Interligand interactions are minimized in this case
by canting of the ligands away from the plane of the core
by ∼27° (Figure 4b). Also, the lack of substituents on the N
atom adjacent to the ligated N atom results in small
N-N-C(aryl) angles of ∼112° and a decreased steric profile
for the formazan, notwithstanding the presence of the
isopropyl substituents on the aryl rings.5b The formazan-
copper metallocycle appears as a boatlike conformation
because of displacement of the central C atom and the Cu
atom away from the plane defined by the formazan N atoms.
Similar conformational properties have been seen for related
formazan complexes.11,13,35
Computational
Assessment
of
Oxygenation
Reactions. DFT calculations at the mPW level of theory were
carried out to gain additional insight into structural and
spectroscopic characteristics of the bis(µ-oxo)dicopper com-
plexes of 4 and 5 (the full calculated geometries of all species
are available as Supporting Information; the structure of 5
is shown in Figure S5). For 4, two different geometric
isomers are predicted to be minima; the lower energy isomer
belongs to the D2 point group and has a core structure
resembling that of the bis(hydroxo)dicopper complex shown
in Figure 3, including an N-Cu-Cu-N torsion angle of
47.7°. The other structure belongs to the C2h point group
(b) Low-Temperature Reactions: Identification of
Intermediates. Oxygenation of LCu(CH3CN) (L ) 4 or 5)
at -80 °C in THF resulted in the growth of new features in
the UV-vis spectra that were unperturbed upon removal of
excess O2 but which bleached upon warming (Figure 5). New
features below 400 nm (4) (not shown) or 500 nm (5) are
likely to arise in large part from ligand-based π-π*
transitions, and a weak band at 780 nm (ε ) 200 M-1 cm-1)
was also observed for the system supported by 4. Most
importantly, oxygenation resulted in intense new bands at
470 nm (ꢀ ∼ 15 000 M-1 cm-1; 4) or 525 nm (ꢀ ∼ 23 000
(36) Henson, M. J.; Mukherjee, P.; Root, D. E.; Stack, T. D. P.; Solomon,
E. I. J. Am. Chem. Soc. 1999, 121, 10332–10345.
(37) Herres-Pawlis, S.; Verma, P.; Haase, R.; Kang, P.; Lyons, C. T.;
Wasinger, E. C.; Flo¨rke, U.; Henkel, G.; Stack, T. D. P. J. Am. Chem.
Soc. 2009, ASAP; doi: 10.1021/ja807809x.
(38) Holland, P. L.; Cramer, C. J.; Wilkinson, E. C.; Mahapatra, S.; Rodgers,
K. R.; Itoh, S.; Taki, M.; Fukuzumi, S.; Que, L., Jr.; Tolman, W. B.
J. Am. Chem. Soc. 2000, 122, 792–802.
(39) Theory also offers support for the possible experimental occurrence
of Fermi resonances in these cases, insofar as there are many normal
modes predicted to have fundamental frequencies near half of the
breathing-mode frequencies, so that overtones of the former might be
expected to couple with the latter.
(35) Gilroy, J. B.; Ferguson, M. J.; McDonald, R.; Hicks, R. G. Inorg.
Chim. Acta 2008, 361, 3388–3393.
Inorganic Chemistry, Vol. 48, No. 10, 2009 4519

Hong et al.
Figure 5. UV-vis changes accompanying oxygenation of (left) (4)Cu(CH3CN) and (right) (5)Cu(CH3CN) at -80 °C in THF, using starting concentrations
of 0.2 or 0.1 mM, respectively. The initial and final spectra are in blue and red, respectively. The insets show the results of spectrophotometric titrations of
the complexes with O2 as linear fits to (left) data for two replicate runs or (right) average data from three replicate runs.
Figure 6. rR spectra of the frozen solutions resulting from oxygenations of (a) (4)Cu(CH3CN) (40 mM) and (b) (5)Cu(CH3CN) (10 mM) in THF at -80
°C, using 16O2 (solid line) or 18O2 (dashed line). The inset to plot (b) shows overtone bands. Excitation wavelengths were 457.9 and 514.5 nm, respectively.
The asterisk indicates a solvent peak.
and has all of the rhomb (Cu2O2) and diketiminate heavy
atoms in a common plane. Because this structure is predicted
to be 3.4 kcal mol-1 higher in free energy than the D2
minimum at 298 K, it makes little contribution to the
experimental equilibrium, but it is interesting that it is
predicted to be stable. The only local minimum structure
found for 5 belongs to the C2 point group and has a core
structure resembling that of the bis(hydroxo)dicopper com-
plex shown in Figure 4. In terms of metric data, the Cu-Cu
distances in the bis(µ-oxo)dicopper complexes of 4 and 5
are predicted to be 2.860 and 2.783 Å, respectively, and the
Cu-O distances are predicted to be 1.832 Å in 4 and 1.829
and 1.833 Å in 5 (where they are not required to be identical
by symmetry). The substantially shorter Cu-Cu distance in
5 compared to 4 is attributable to the lack of substitution at
the 2 and 4 positions of the formazan ring compared to the
diketiminate. Thus, in 4, the CuNCipso angle, which describes
the degree to which the aryl rings project forward from the
diketiminate, is predicted to be 115.5°; in 5, on the other
hand, the corresponding angle is predicted to be 120.6°. The
effect of this difference, which is experimentally shown
(Figures 2-4 and S1 in the Supporting Information), is
multiplied 2-fold upon dimerization, reducing unfavorable
steric interactions associated with shorter Cu-Cu distances.
As a point of comparison that offers insight into the
sensitivity of the Cu2O2 rhomb structure to electron-donating
or -withdrawing groups on the ligands, the Cu-Cu and
Cu-O distances in the bis(µ-oxo)dicopper complex of 3,
where the CF3 groups in 4 are replaced by CH3 groups, are
predicted at the mPW1 level to be 2.836 and 1.831 Å,
respectively. These values are quite similar to those for the
bis(µ-oxo)dicopper complex of 4, so the electron-withdraw-
ing character of the CF3 groups does not seem to have a
particularly strong influence on the core rhomb structure. This
observation is consistent with the experimental and predicted
rR data. For the bis(µ-oxo)dicopper complex of 3, the core
breathing frequency is observed at 608 cm-1 (∆18O2 ) 27
cm-1).5b As shown in Figure 6, the corresponding frequency
with 4 is at about 604 cm-1 if one assumes that the pair of
peaks at 598 and 610 cm-1 result from a Fermi resonance.
Theory provides some insights because predicted spectra are
not complicated by the Fermi resonance phenomenon. For
3, the breathing vibration is predicted at the mPW level to
occur at 582 cm-1 with an 18O2 isotope shift of 24 cm-1,
while for 4 the analogous values are 577 and 25 cm-1. Thus,
theory is consistent with an experimental assignment of the
fundamental core frequency for the bis(µ-oxo)dicopper
complex of 4 being only 4 or 5 cm-1 smaller than that of 3
4520 Inorganic Chemistry, Vol. 48, No. 10, 2009

Effects of Electron-Deficient Ligands on Cu/O2 Chemistry
and showing an almost identical isotope shift.39 The analo-
gous frequencies in the bis(µ-oxo)dicopper complex of 5 are
more difficult to assign because the totally symmetric
breathing mode couples quite strongly with the aryl bending
and formazan modes. One such frequency is predicted to be
565 cm-1 with an 18O2 isotope shift of 20 cm-1, and these
values are roughly consistent with the mPW/experiment
offsets for 3 and 4, but the analysis in this case must be
considered more tentative in the absence of a more sophis-
ticated model that would include predicting rR intensities.
Having computed structures and energies for the bis(µ-
oxo)dicopper and monocopper carbonyl complexes, we may
also compute the energy changes for the reaction
2LCuCO + O2 f [LCu]2(O)2 + 2CO
(1)
Figure 7. Predicted donor (below) and acceptor (above) Kohn-Sham
molecular orbitals for the longest-wavelength transitions observed for the
bis(µ-oxo)dicopper complexes of 4 (a) and 5 (b). The 0.02 au isodensity
surfaces are visualized, and H atoms are removed for clarity. Cu atoms are
peach, F atoms are aqua, O atoms are red, N atoms are blue, and C atoms
are gray; the molecules are oriented with the Cu-Cu axis horizontal.
At the mPW level, the computed 298 K free energies of
reaction for this process are 25.4 and 9.9 kcal mol-1 for 4
and 5, respectively. The bis(µ-oxo)dicopper species are not
synthesized from the carbonyl compounds, but these values
provide an indication of the differential stabilities of the two
bis(µ-oxo)dicopper cores. The much greater stability of the
complex with 5 compared to that with 4 may be traced to
the reduced steric clash between the aromatic rings in the
former, as noted above. As an interesting aside, we were
successful in locating computationally a bis(µ-oxo)dicopper
structure of 2 that was predicted to be a local minimum
(belonging like 4 to the D2 point group). However, the free
energy of reaction for eq 1 with this ligand is predicted to
be 45.6 kcal mol-1, indicating how much more unfavorable
the steric interactions between the isopropyl-substituted
phenyl rings are compared to the methyl-substituted ones
and rationalizing the failure to observe this dimerization
experimentally.6
Turning last to the UV-vis spectra of the bis(µ-oxo)di-
copper complexes of 4 and 5, TD-DFT calculations using
the B3LYP functional were analyzed to understand the nature
of the longest-wavelength absorptions in these species. In
the case of the complex with 4, TD B3LYP predicts there
to be only a single excitation having significant oscillator
strength at low energy, and the wavelength of the transition
is predicted to be 540 nm. This prediction is well to the red
of the experimentally observed value, which is typical of
CT excitations with TD B3LYP.40 The transition itself is
computed to be an excitation out of the antibonding π*
orbital formed from the out-of-plane O 2pz orbitals (some-
times called the πv* orbital20) into an antibonding combina-
tion of Cu dxy orbitals and the σu* orbital formed by the
antibonding combination of the O 2py orbitals along the O-O
axis (Figure 7a). This excitation is exactly the same as that
predicted by Estiu´ and Zerner, at the INDO/S level, to occur
at lowest energy for a bis(µ-oxo)dicopper core supported by
three imidazole ligands per Cu atom.41 In the case of the
complex with 4, the donor orbital is predicted also to be
Figure 8. (a) UV-vis spectra of (5)Cu(CH3CN) (dotted line) and the
intermediate observed after oxygenation for 150 s at -10 °C in THF (solid
line). (b) X-band EPR spectrum of the intermediate (4 K).
substantially delocalized onto the diketiminate aryl rings, but
the core contribution is dominated by the O-O πv*.
The most intense low-energy absorption for the bis(µ-
oxo)dicopper complex with 5, on the other hand, is predicted
to occur at 548 nm at the TD B3LYP level and involves
excitation from an occupied orbital comprised of the O 2py
σu* orbital and formazan in-plane ligand orbitals into the
πσ* orbital, which is formed as an antibonding combination
(40) Cramer, C. J. Essentials of Computational Chemistry: Theories and
Models, 2nd ed.; John Wiley & Sons: Chichester, U.K., 2004; pp
501-504.
(41) Estiu´, G. L.; Zerner, M. C. J. Am. Chem. Soc. 1999, 121, 1893–1901.
Inorganic Chemistry, Vol. 48, No. 10, 2009 4521

Hong et al.
Scheme 1
of the Cu dxy orbitals and the O 2px π* orbital (Figure 7b).
This acceptor orbital also has a significant formazan π*
component. The excitation is very similar to that proposed
by Henson et al. based on SCF-Xa-SW calculations for the
lowest-energy excitation in a bis(µ-oxo)dicopper core sup-
ported by two ammonia ligands per copper.36 In the case
considered by Henson et al., however, there was substantial
contribution to the occupied orbital by the Cu dxy orbitals,
while in this case, they are predicted to make negligible
contribution to the occupied orbital. There is thus a more
substantial overall CT character to this excitation that likely
contributes to its intensity. The unique characteristics of the
formazan ligand, with its framework orbitals lying at high
energy in the occupied orbital manifold, are responsible for
this interesting variation.
Decay of (5)2Cu2(µ-O)2. As noted above, the UV-vis
spectroscopic features of the bis(µ-oxo)dicopper complexes
decayed upon warming above -80 °C, yielding L2Cu2(OH)2
complexes that were isolated by performing the oxygenation
at ambient temperature. Similar decompositions have been
noted previously for bis(µ-oxo)dicopper compounds.2,37 For
the system supported by 5, we observed an additional
intermediate species during the decay process, the formation
and decay of which was most conveniently monitored by
performing the oxygenation of (5)Cu(CH3CN) at tempera-
tures above -50 °C. This new intermediate exhibits a narrow
low-energy absorption feature at 742 nm (ꢀ ∼ 1100 M-1
cm-1 per Cu; Figure 8a). While d-d transitions for copper(II)
complexes often appear in this region of the UV-vis
spectrum, such an assignment seems unlikely considering
the band shape and intensity observed here. Kinetics experi-
ments at -10 °C support a first-order rate law for both the
formation (k ) 0.037 s-1) and decay (k ) 0.0024 s-1) of the
742 nm feature (Figure S3 in the Supporting Information).
Measurements of the rates of formation in THF and THF-d8
at -30 °C (Figure S4 in the Supporting Information) revealed
a modest kinetic isotope effect (KIE ) kH/kD) of 1.6,
consistent with attack on the solvent C-H(D) bond during
the conversion of the bis(µ-oxo)dicopper complex to the
intermediate.42 Most intriguing is the 4 K X-band EPR
spectrum for the intermediate comprising an isotropic signal
at g ∼ 2.0 (Figure 8b), for which double integration indicates
a total spin quantity of 77% relative to the bis(µ-oxo)dicopper
precursor.
We hypothesize that the 742 nm absorption feature and
the signal in the EPR spectrum arise from a species
containing a ligand-centered radical. Scheme 1 depicts
possible decomposition pathways to (5)2Cu2(OH)2 via a
proposed formazyl radical ligand intermediate (A or B).12
Precedent for these formulations can be found in a related
boron-formazan complex, which could be reduced to a
“borataverdazyl” radical anion with similar spectroscopic
features (a band at 738 nm in the solid-state electronic
spectrum and an isotropic EPR signal at g ∼ 2.0 in the solid
state).13 Furthermore, the acceptor orbital for (5)2Cu2(O)2
(Figure 7) features significant contributions from formazan
π* orbitals that are analogous to the singly occupied
molecular orbital of verdazyls, a well-established family of
stable radicals.43 Formulation A or B would result from an
internal redox isomerization from an L2CuIIICuII species,
which would be the first product generated upon H-atom
abstraction by the bis(µ-oxo)dicopper(III) complex (consis-
tent with the observed KIE). The redox isomerizations
concievably could involve electron transfer from the CuII
site to the nitroformazan ligand or from the ligand to the
CuIII site to yield A or B, respectively. The former process
to yield A would appear most likely, in view of the absence
of any copper-based hyperfine splitting in the EPR signal
and the fact that the borataverdazyl with analogous spectro-
scopic properties is generated by reduction rather than
oxidation.13 On the other hand, species B cannot be ruled
out if strong antiferromagnetic coupling between the cop-
per(II) centers and little if any coupling between those centers
and the verdazyl radical are assumed.
Conclusions
Compared to copper(I) complexes of diketiminate ligands
that feature alkyl or aryl backbone substituents (e.g., R′ )
Me in Figure 1), those of ligands 4 and 5 feature electron-
deficient metal centers, as best indicated by the high ν(CO)
frequencies for the species LCu(CO). Nonetheless, the
complexes LCu(CH3CN) (L ) 4 or 5) react with O2 at room
(42) Attempts to identify products of oxidation of the solvent via gas
chromatography/mass spectrometry analysis after allowing the inter-
mediate to decompose were unsuccessful, however.
(43) Gilroy, J. B.; McKinnon, S. D. J.; Kennepohl, P.; Zsombor, M. S.;
Ferguson, M. J.; Thompson, L. K.; Hicks, R. G. J. Org. Chem. 2007,
72, 8062–8069.
4522 Inorganic Chemistry, Vol. 48, No. 10, 2009

Effects of Electron-Deficient Ligands on Cu/O2 Chemistry
temperature to yield bis(µ-hydroxo)dicopper complexes
structurally defined by X-ray crystallography and at -80 °C
to yield bis(µ-oxo)dicopper compounds, identified as such
on the basis of UV-vis and rR spectroscopy, EPR silence,
spectrophotometric titration data, and theory. The observation
of bis(µ-oxo)dicopper complex formation upon low-temper-
ature oxygenation of (4)Cu(CH3CN) stands in contrast to
the lack of O2 reactivity reported previously for
(2)Cu(CH3CN), which is similarly electron-poor but is
sterically more encumbered by virture of the diisopropy-
lphenyl versus dimethylphenyl substituents. We surmise that
in both cases O2 binding is thermodynamically unfavorable,
but in the case of the system with the less hindered 4, rapid
irreversible trapping of what is likely a small equilibrium
concentration of a 1:1 adduct drives the reaction forward.
The nitroformazan ligand 5 confers unusual properties to
its bis(µ-oxo)dicopper complex, such as an intense electronic
absorption feature at lower energy than that seen previ-
ously2,36 which is ascribed on the basis of theory to
significant ligand orbital contributions to the [Cu2O2]2+ core
transition. In addition, the decay of (5)2Cu2(µ-O)2 proceeds
via a novel observable intermediate that is assigned as a
metalloverdazyl radical. We postulate that this radical species
is formed by a mechanism involving H-atom abstraction from
the solvent followed by an internal redox isomerization. Such
an intramolecular electron transfer involving the formazan
ligand may be viewed in the context of an extensive literature
on redox noninnocent ligands and the novel reactivity of their
complexes,44 with reports of complexes containing aminyl
(R2N•) radical ligands generated by one-electron oxidation
of the corresponding amido species being particularly rel-
evant.45 Thus, while structurally analogous to diketiminates,
the formazan ligand system exhibits noninnocence and, as a
result, alternate dimensions to its coordination chemistry.
Acknowledgment. We thank the NIH (Grant GM47365
to W.B.T.), NSF (Grant CHE-0610183 to C.J.C. and REU
funds to B.D.N.), NSERC of Canada (R.G.H.), and UM
Lando Undergraduate Research Program (B.D.N.) for finan-
cial support of this research. We also thank William
Antholine for preliminary EPR data and analysis, J. D.
Lipscomb and L. Que, Jr., for providing access to EPR and
rR facilities, and Victor G. Young, Jr., for assistance with
X-ray crystallography.
Supporting Information Available: X-ray crystallagraphic data
in CIF format, tables of coordinates for calculated structures, and
Figures S1-S5. This material is available free of charge via the
Internet at http://pubs.acs.org.
IC9002466
(44) Selected recent examples:(a) Lu, C. C.; Bill, E.; Weyhermuller, T.;
Bothe, E.; Wieghardt, K. J. Am. Chem. Soc. 2008, 130, 3181–3197.
(b) Boyer, J. L.; Cundari, T. R.; DeYonker, N. J.; Rauchfuss, T. B.;
Wilson, S. R. Inorg. Chem. 2009, 48, 638–645. (c) Ketterer, N. A.;
Fan, H.; Blackmore, K. J.; Yang, X.; Ziller, J. W.; Baik, M.-H.;
Heyduk, A. F. J. Am. Chem. Soc. 2008, 130, 4364–4374. (d)
Blackmore, K. J.; Sly, M. B.; Haneline, M. R.; Ziller, J. W.; Heyduk,
A. F. Inorg. Chem. 2008, 47, 10522–10532. (e) Stanciu, C.; Jones,
M. E.; Fanwick, P. E.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007,
129, 12400–12401.
(45) Hicks, R. G. Angew. Chem., Int. Ed. 2008, 47, 7393–7395, and
references cited therein.
Inorganic Chemistry, Vol. 48, No. 10, 2009 4523