Imidazole substituent effects on oxidative reactivity of tripodal(imid)2(thioether)CuI complexes

Article abstract of DOI:10.1021/ic800007t

In the search for new bis(imidazole)thioether (BIT) copper complexes that accurately mimic the electronic and reactivity features of the Cu_M site of copper hydroxylase enzymes, a set of tripodal BIT ligands 4a,b-6a,b has been synthesized that vary according to the imidazole C-(Ph or H) and N-(H or Me) substituents, as well as the position (2- or 4-) of the tripodal attachment. Corresponding [(BIT)Cu(L)](PF₆) complexes 7a,b′, 8a,b′, and 9a′,b′ [L = CO (a), CH₃CN (b)] have been prepared and characterized spectroscopically. The IR spectra of 7a-9a (L = CO), specifically ν(CO), show little variation (2090-2100 cm^-1), suggesting a similar electronic character of the Cu centers. In contrast, cyclic voltammetric analysis of these compounds (L = CH₃CN) reveals quasi-reversible oxidation waves with significant variation of E_pa in the range of + 0.45-0.57 V vs Fc/Fc⁺, depending on the imidazole substituents. Each of the [(BIT)Cu(CH₃CN)]PF₆ complexes reacts with dioxygen to form [(BIT)CU^II₂(μ-OH)₂](PF ₆)₂ derivatives, 10-12, but they vary considerably in their relative reactivity, following the same trend as the ease of their electrochemical oxidation, that is, [(2-BIT^NMe)Cu(CH ₃CN)]⁺ (9b′) > [(4-BIT^Ph,NMe) Cu(CH₃CN)]⁺ (8b′) > [(2-BIT^Ph2,NMe) Cu(CH₃CN)]⁺ (1a′) > [(4-BIT^Ph,NH) Cu(CH₃CN)]⁺ (7b′). Thus, N-Me substitution and 4-tethering on the imidazole unit increase oxidation and oxygenation reactivity, while Ph-substitution and 2-tethering decrease reactivity. PM3 and DFT calculations are employed to analyze the relative stability, the electronic features, the Cu-CO vibrtional frequency, and the electrochemical and oxidative reactivity of the complexes.

Full text of DOI:10.1021/ic800007t

Inorg. Chem. 2008, 47, 4356-4367
Imidazole Substituent Effects on Oxidative Reactivity of
Tripodal(imid)2(thioether)CuI Complexes
Lei Zhou and Kenneth M. Nicholas*
Department of Chemistry and Biochemistry, UniVersity of Oklahoma, Norman, Oklahoma 73019
Received January 2, 2008
In the search for new bis(imidazole)thioether (BIT) copper complexes that accurately mimic the electronic and
reactivity features of the CuM site of copper hydroxylase enzymes, a set of tripodal BIT ligands 4a,b-6a,b has
been synthesized that vary according to the imidazole C-(Ph or H) and N-(H or Me) substituents, as well as the
position (2- or 4-) of the tripodal attachment. Corresponding [(BIT)Cu(L)](PF6) complexes 7a,b′, 8a,b′, and 9a′,b′
[L ) CO (a), CH3CN (b)] have been prepared and characterized spectroscopically. The IR spectra of 7a-9a (L
) CO), specifically ν(CO), show little variation (2090–2100 cm-1), suggesting a similar electronic character of the
Cu centers. In contrast, cyclic voltammetric analysis of these compounds (L ) CH3CN) reveals quasi-reversible
oxidation waves with significant variation of Epa in the range of + 0.45–0.57 V vs Fc/Fc+, depending on the
imidazole substituents. Each of the [(BIT)Cu(CH3CN)]PF6 complexes reacts with dioxygen to form [(BIT)CuII2(µ-
OH)2](PF6)2 derivatives, 10-12, but they vary considerably in their relative reactivity, following the same trend as
the ease of their electrochemical oxidation, that is, [(2-BITNMe)Cu(CH3CN)]+ (9b′) > [(4-BITPh,NMe)Cu(CH3CN)]+ (8b′)
> [(2-BITPh2,NMe)Cu(CH3CN)]+ (1a′) > [(4-BITPh,NH)Cu(CH3CN)]+ (7b′). Thus, N-Me substitution and 4-tethering on
the imidazole unit increase oxidation and oxygenation reactivity, while Ph-substitution and 2-tethering decrease
reactivity. PM3 and DFT calculations are employed to analyze the relative stability, the electronic features, the
Cu-CO vibrtional frequency, and the electrochemical and oxidative reactivity of the complexes.
Introduction
The copper hydroxylase enzymes, dopamine ꢀ-hydroxy-
lase (DꢀH,) and peptidylglycine R-hydroxylating monooxy-
genase (PHM), catalyze the regio- and enantioselective
hydroxylation of somewhat activated C-H bonds by dioxy-
gen.1 The enzymes apparently have very similar active site
Figure 1. Cu-hydroxylase active site.
structures, featuring two Cu centers separated by 12 Å, the
CuH (CuA) site being ligated to three histidine-derived
coordination mode to the CuM center,4–6 contrasting with the
more common side-on mononuclear and bridging bimetallic
bonding modes found in most synthetic complexes and in
imidazoles and the CuM (CuB) site by two histidines and a
methionine residue (Figure 1).2 Spectroscopic and amino acid
mutagenesis studies suggest that upon reduction both dioxy-
gen and the substrate are activated at the CuM site and that
the CuH site serves as an electron transfer shuttle from the
ascorbate reductant to the CuM site.3 A recent X-ray structure
of the O2-bound PHM enzyme has revealed a rare end-on
(3) (a) Prigge, S. T.; Mains, R. E.; Eipper, B. A.; Amzel, L. M. Cell.
Mol. Life Sci. 2000, 57, 1236–1259. (b) Prigge, S. T.; Kolhekar, A. S.;
Eipper, B. A.; Mains, R. E.; Amzel, L. M. Nat. Struct. Biol. 1999, 6,
976–983. (c) Evans, J. P.; Ahn, K.; Klinman, J. P. J. Biol. Chem.
2003, 278, 49691–49698. (d) Chen, P.; Solomon, E. I. J. Am. Chem.
Soc. 2004, 126, 4991–5000.
(4) Prigge, S. T.; Eipper, B. A.; Mains, R. E.; Amzel, L. M. Science 2004,
304, 864–867.
*
To whom correspondence should be addressed. E-mail:
knicholas@ou.edu.
(5) Chaudhuri, P.; Hess, M.; Weyhermuller, T.; Wieghardt, K. Angew.
Chem., Int. Ed. 1999, 38, 1095–1098.
(1) (a) Klinman, J. P. Chem. ReV. 1996, 96, 2541–2562. (b) Lewis, E. A.;
Tolman, W. B. Chem. ReV. 2004, 104, 1047–1076.
(2) Prigge, S. T.; Mains, R. E.; Amzel, L. M. Science 1997, 278, 1300–
1305.
(6) Wurtele, C.; Gaoutchenova, E.; Harms, K.; Holthausen, M. C.;
Sundermeyer, J.; Schindler, S. Angew. Chem., Int. Ed. 2006, 45, 3867–
3869.
4356 Inorganic Chemistry, Vol. 47, No. 10, 2008
10.1021/ic800007t CCC: $40.75  2008 American Chemical Society
Published on Web 04/10/2008

Tripodal(imid)2(thioether)CuI Complexes
the O2-binding Cu-protein hemocyanin.7 The mechanistic
details of the substrate hydroxylation are unclear, but rate-
limiting C-H bond-breaking with a significant tunneling
component by a Cu-oxygen species is supported by the large
intrinsic primary D-isotope effect (10.6) for both DꢀH and
PHM.8 Recent computational studies suggest that either a
side-on Cu-superoxo9 or a Cu-oxo species10 is the likely
H-abstracting agent.
these also form dinuclear Cu2L2-type complexes, mono-
nuclear adducts [(imid3CR)CuL]Z are produced with L)
acetonitrile, carbon monoxide, and t-BuNC. Reactions of
dioxygen with these dinuclear complexes are sluggish,
producing from the latter [(imid)3CH]CuII(acetone)(H2O)]Z2.
Few complexes have been reported that possess the
(amine)2thioether coordination sphere relevant to the CuM
site of the copper hydroxylases. Karlin and co-workers
recently described CuI,II complexes of (non-imidazole) tri-
dentate (pyridine)(amine)(thioether) ligands; oxygenation of
[(N2S)CuI]+ at room temperature leads to ligand sulfoxida-
tion, demonstrating O2 activation and an oxygenation path-
way not displayed by the enzymes.16 They also reported a
N3S-thioether-Cu(I) complex that undergoes low temperature
oxygenation to form a spectroscopically characterized end-
on peroxodicopper(II) derivative.17 Electrochemical studies
of these systems found that the more easily oxidized
complexes were also more reactive toward dioxygen. A few
CuII complexes with mixed polydentate imidazole/thioether
ligands have been characterized,11 but until our recent
studies, only one such CuI derivative had been reported.11
In recent reports, we disclosed the preparation of tripodal
Although many synthetic model complexes for these (and
other) copper enzymes have been prepared with a variety of
polydentate amine-, pyridine-, and pyrazolyl-based ligands,7
remarkably few have incorporated the biologically most
relevant imidazole (from histidine) donors.11,12 Because of
the significantly different basicity (pKa imidazolium, 6.8;
pyrazolium, 2.6; pyridinium, 5.2, tertiary ammonium, 9–10)13
and donor/acceptor properties14 of imidazole vis a vis the
other common nitrogen ligands, we propose that the most
accurate functional mimics for histidine-rich metalloenzymes
will be provided by incorporation of the actual biological
donor set. Regarding imidazole-based models for the CuH
site, Sorrell and co-workers first reported copper(I) com-
plexes of tripodal tris(N-methyl-2-imidazolyl)carbinol12a and
tris(2-imidazolyl)phosphines.12d Dinuclear complexes of the
former ligand, [(tripod)2CuI2]Z2, are unreactive toward di-
oxygen, whereas a mononuclear CuI-complex of the latter
ligand reacts with dioxygen to produce a peroxodicopper(II)
adduct. We recently investigated the chemistry of CuI
complexes of the related, sterically demanding tripodal tris-
(N-methyl-4,5-diphenyl-imidazolyl)methane ligands.15 While
bis-(4,5-diphenyl-2-imidazolyl)thioether (2-BITPh2,SR,OH(Me)
)
ligands and the corresponding Cu(I) complexes, [(2-
BITPh2,SMe,OH(Me))Cu(L)]PF6 [L) CO (1a), CH3CN (1a′)]. The
IR ν(CO) absorptions of the [(2-BITPh2,SMe)Cu-CO]+ com-
plexes 1a were found to be very similar to the enzymes
Cu(M)-CO, suggesting close electronic similarity of the
synthetic and natural Cu centers.19 In contrast to the enzymes,
oxygenation of 1a′ affords a binuclear Cu(II)-hydroxo
complex, [(2-BITPh2,SMe)2Cu2(µ-OH)2](OTf)2 (2a). Copper(I)
complexes of the sterically hindered 2-BITPh2,SR’ ligands 1b′
(R′) t-Bu) and 1c′ (R’) CPh3) were found to undergo oxy-
enation by contrasting pathways.20 The former (1b′) reacts
slowly relative to 1a′, producing two (2-BITPh2,SBu-t)2Cu2(µ-
OH)2-type complexes, 2b + 2b′, the latter of which is ortho-
oxygenated on a neighboring aryl group (Scheme 1). On the
other hand, oxygenation of the CuI complex derived from
2-BITPh2,SCPh3 (1c′) produced a novel dinuclear disulfide
complex, [(BITS)2Cu2(µ-OH)2](PF6)2 (2c).
(7) (a) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004,
104, 1013–1045. (b) Lewis, E. A.; Tolman, W. B. Chem. ReV. 2004,
104, 1047–1076. (c) Fujisawa, K.; Tanaka, M.; Moro-oka, Y.;
Kitajima, N. J. Am. Chem. Soc. 1994, 116, 12079–12080. (d) Kitajima,
N.; Fujisawa, K.; Fujimoto, C.; Morooka, Y.; Hashimoto, S.; Kitagawa,
T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am. Chem. Soc. 1992,
114, 1277–91. (e) Hatcher, L. Q.; Karlin, K. D. J. Biol. Inorg. Chem.
2004, 9, 669–683. (f) Hatcher, L. Q.; Vance, M. A.; Sarjeant, A. A. N.;
Solomon, E. I.; Karlin, K. D. Inorg. Chem. 2006, 45, 3004–3013. (g)
Aboelella, N. W.; Kryatov, S. V.; Gherman, B. F.; Brennessel, W. W.;
Young, V. G.; 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. (h) Itoh, S.; Fukuzumi, S. Bull.
Chem. Soc. Jpn. 2002, 75, 2081–2095.
(8) (a) Takahashi, K.; Onami, T.; Noguchi, M. Biochem. J. 1998, 336,
131–137. (b) Francisco, W. A.; Merkler, D. J.; Blackburn, N. J.;
Klinman, J. P. Biochemistry 1998, 37, 8244–8252. (c) Miller, S. M.;
Klinman, J. P. Biochemistry 1985, 24, 2114–2127. (d) Klinman, J. P.;
Humphries, H.; Voet, J. G. J. Biol. Chem. 1980, 255, 11648–11651.
(9) Chen, P.; Solomon, E. I. J. Am. Chem. Soc. 2004, 126, 4991–5000.
(10) (a) Kamachi, T.; Kihara, N.; Shiota, Y.; Yoshizawa, K. Inorg. Chem.
2005, 44, 4226–4236. (b) Yoshizawa, K.; Kihara, N.; Kamachi, T.;
Shiota, Y. Inorg. Chem. 2006, 45, 3034–3041.
Although the bis(imidazole)(thioether)Cu complexes in-
vestigated to date more accurately represent the donor ligand
set of the CuM enzyme site than do non-imidazole N2S
ligands, significant differences between the active site models
and enzymatic systems still exist. In addition to the C-Ar
and N-Me groups of the synthetic ligands, which are absent
(11) Cao, Y. D.; Zheng, Q. Y.; Chen, C. F.; Hu, H. M.; Huang, Z. T. Inorg.
Chim. Acta 2004, 357, 316–320.
(16) Lee, Y.; Lee, D.-H.; Narducci-Sarjeant, A. A.; Zakharov, L. N.;
Rheingold, A. L.; Karlin, K. D. Inorg. Chem. 2006, 45, 10098–10107.
(17) Hatcher, L. Q.; Lee, D.-H.; Vance, M. A.; Milligan, A. E.; Sarangi,
R.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Karlin, K. D. Inorg.
Chem. 2006, 45, 10055–10057.
(12) (a) Sorrell, T. N.; Borovik, A. S. J. Am. Chem. Soc. 1987, 109, 4255–
4260. (b) Tang, C. C.; Davalian, D.; Huang, P.; Breslow, R. J. Am.
Chem. Soc. 1978, 100, 3918–3922. (c) Breslow, R.; Hunt, J. T.; Smiley,
R.; Tarnowski, T. J. Am. Chem. Soc. 1983, 105, 5337–5342. (d) Sorrell,
T. N.; Allen, W. E.; White, P. S. Inorg. Chem. 1995, 34, 952–960.
(13) Bruice, P. Y. Organic Chemistry, 4th ed.; Prentice-Hall: Trenton, NJ,
2004; pp A8–9.
(18) (a) Tran, K. C.; Battioni, J. P.; Zimmermann, J. L.; Bois, C.; Koolhaas,
G. J. A. A.; Leduc, P.; Mulliez, E.; Boumchita, H.; Reedijk, J.;
Chottard, J. C. Inorg. Chem. 1994, 33, 2808–2814. (b) Zoeteman, M.;
Bouwman, E.; De Graff, R. A. G. D.; Driessen, W. L.; Reedijk, J.;
Zanello, P. Inorg. Chem. 1990, 29, 3487–3492. (c) Dagdigian, J. V.;
Reed, C. A. Inorg. Chem. 1979, 18, 2623–2626.
(14) (a) Marques, H. M.; Munro, O. Q.; Munro, T.; Wet, M. D.; Vashi,
P. R. Inorg. Chem. 1999, 38, 2312–2319. (b) Johnson, C. R.;
Henderson, W. W.; Shepherd, R. E. Inorg. Chem. 1984, 23, 2754–
2763. (c) Peters, L.; Hubner, E.; Burzlaff, N. J. Organomet. Chem.
2005, 690, 2009–2016.
(19) Zhou, L.; Powell, D.; Nicholas, K. M. Inorg. Chem. 2006, 45, 3840–
3842.
(15) Zhou, L.; Powell, D.; Nicholas, K. M. Inorg. Chem. 2007, 46, 2316–
2321.
(20) Zhou, L.; Powell, D.; Nicholas, K. M. Inorg. Chem. 2007, 46, 7789–
7799.
Inorganic Chemistry, Vol. 47, No. 10, 2008 4357

Zhou and Nicholas
Scheme 1
Scheme 2
in the protein-derived imidazole from histidine, the tripodal
alkyl tether is attached at C-2 for the synthetic N2S complexes
while the histidine-derived imidazole is alkyl-substituted at
C-4,5. What effects will such differences have on the
electronic character and, particularly, the reactivity of the
respective metal centers? While the steric and electronic
effects of the type of C- and N-substitution could perhaps
be qualitatively anticipated, the effects of the C-2 versus
C-4,5 position of substitution are less obvious. A study by
Breslow and co-workers21 showed that [(4-imid)3COH]MII
complexes are significantly more stable (Ka is ∼102 greater)
than the [(2-imid)3COH]MII. What would be the impact of
such substitutional differences on the (BIT)Cu-O2 reactivity?
In the present report, we address these questions with the
preparation of a new set of (imid)2(thioether)CuI complexes
and investigate their electronic characteristics through spec-
troscopic characterization of their CO adducts, their elec-
trochemical redox behavior and reactivity toward dioxygen,
and computational studies.
Scheme 3
solution. Alcohol 4a was converted into N-methylated
tripod 4-BIT2-Ph,NMe,OH (5a) (41%) and methoxy compound
4-BIT2-Ph,NMe,OMe (5b) was produced from 5a (70%) by reaction
with NaH/MeI. A new pair of N2S-ligands 2-BITNH,OH(Me) (6a,
6b), lacking aryl groups and tethered at the C-2 position, was
also efficiently prepared by addition of 2-lithio-N-methyl
imidazole to the thioether ester (Scheme 3), followed by
O-alkylation for the preparation of 6b. All of the ligands thus
produced were characterized spectroscopically and were ob-
tained in an analytically pure form.
Preparation and IR spectra of N2S-Tripod Cu(I)-
Carbonyl Complexes. The IR Cu-CO vibrational fre-
quency, affected by the balance between σ-CO-to-Cu dona-
tion and Cu-to-CO π* backbonding, has been considered to
be a measure of the electronic character of the Cu center of
LCu-CO complexes.23 In addition, metal–carbonyl deriva-
tives have been used as models to gauge M-O2 complex
formation/stability.24 To use this probe to detect electronic
differences among the various imidazole-substituted ligands,
Results
Ligand Synthesis. The abbreviated general formula for
the bis-imidazole-thioether (BIT) ligands prepared is:
#-BIT#-R,NR,SMe,OR, sequentially indicating the position of tripodal
attachment, and the imidazole (to C, N), the thioether, and the
tripodal C-substituents, respectively. In all of the present set of
ligands and complexes the thiomethyl unit is present so the SMe
superscript will be omitted for brevity. To prepare 4-tethered
ligands the conveniently introduced and removed ethoxymethyl
(EOM) group was employed to direct imidazole lithiation/
alkylation at the 4-position22 and to enable easy removal. The
EOM-protected compounds 5-BIT2-Ph,NEOM,OH(Me) (3a, 3b) were
thus prepared by double addition of lithiated N-EOM-imidazole
to the ester MeSCMe2CO2Et (57% from 2-Ph-1-EOM-imida-
zole to alcohol, 3a), followed by O-methylation under basic
conditions (Scheme 2). The EOM-protected tripodal com-
pounds 3a and 3b were converted into the N-H derivatives
5-BIT2-Ph,NH,OH(Me) (4a, 4b) by refluxing in acidic MeOH
(21) Tang, C. C.; Davalian, D.; Huang, P.; Breslow, R. J. Am. Chem. Soc.
1978, 100, 3918–3922.
(23) (a) Imai, S.; Fujisawa, K.; Kobayashi, T.; Shirasawa, N.; Fujii, H.;
Yoshimura, T.; Kitajima, N.; Moro-oka, Y. Inorg. Chem. 1998, 37,
3066–3070. (b) Kujime, M.; Kurahashi, T.; Tomura, M.; Fujii, H.
Inorg. Chem. 2007, 46, 541–551.
(22) Breslow, R.; Hunt, J. T.; Smiley, R.; Tarnowski, T. J. Am. Chem.
Soc. 1983, 105, 5337–5342.
4358 Inorganic Chemistry, Vol. 47, No. 10, 2008

Tripodal(imid)2(thioether)CuI Complexes
Scheme 4
Table 1. Summary of IR, CV, and Oxygen Reactivity of [BIT)CuL]+ Complexes 1a′, 7′–9′
a
IR νCuCO (cm-1
)
CV Epa (mV)b
(L ) CH3CN) 1a′, 7b′-9b′
O2 reactivity
(L ) CH3CN) 1a′, 7b′-9b′
[(BIT)CuL]PF6 complex
(L ) CO) 1a, 7a–9a
[(2-BITPh2,NMe,OHMe))Cu(L)]+ (1a,1a′)
[(4-BITPh,NH,OHMe))Cu(L)]+ (7a,7b′)
[(4-BITPh,NMe,OHMe))Cu(L)]+ (8a,8b′)
[(2-BITNMe,OHMe))Cu(L)]+ (9a′,9b′)
2097
2094
2094
2105
+ 0.54 V
+ 0.57 V
2 h at rt
12 h at rt
5 min at -50 °C
2 min at -70 °C
+ 0.51 V
+ 0.45 V
a KBr pellet. b Pt working and auxiliary electrodes and Ag/AgCl reference electrodes: CH2Cl2,Bu4NPF6, Epa - E1/2(Cp2Fe); 1 atm O2, CH2Cl2 solvent.
Scheme 5.
we prepared three new copper-carbonyl complexes, 7a, 8a,
and 9a′, via reactions of the respective ligands 4a, 5a, and
6b with [Cu(CH3CN)4]PF6, followed by exposure of the
intermediate [(BIT)Cu(CH3CN)]PF6 species (not isolated) to
carbon monoxide (Scheme 4). The CO adducts 7a-9a′ were
2-tethered Ph2-derivative 1a′; the N-Me derivative 8b′ easier
isolated by ether precipitation and characterized by NMR,
than the N-H complex 7b′; and the 2-tethered non-Ph
ESI-MS, and IR spectroscopy. The IR Cu-CO vibrational
complex 9b′ is more easily oxidized than the 2-BITPh2
frequencies of these compounds, together with that of the
complex 1a′.
previously prepared 2-tethered-4,5-diphenylated derivative
Dioxygen Reactivity. To evaluate the reactivity of these
1a, are listed in Table 1. It can be seen that these vary only
various complexes with dioxygen the Cu complexes of the
by (10 cm-1, revealing no systematic trend nor significant
4-tethered BITPh,NH,OMe (7b′), BITPh,NMe,OMe (8b′), and 2-teth-
effects of replacing C-Ph for C-H, N-Me/H, or 2/4-tethered
ered BITNMe,OMe (9b′) ligands were generated under argon
tripodal substitution.
at low temperature by combining [Cu(CH3CN)4]PF6 with the
corresponding BIT ligand (1:1) in CH2Cl2, followed by
exposure to dry dioxygen (1 atm). Oxygenation was signaled
by a change in the solution color from nearly colorless/light
Cyclic Voltammetry. To further probe the electronic
characteristics of the various [N2S-Cu(CH3CN)]PF6 com-
plexes, cyclic voltammetry studies were carried out. The
complexes 1a′, 7b′, 8b′, and 9b′ (L ) CH3CN) for CV
analysis were produced in situ by combination of the
respective BITOMe ligands with [Cu(CH3CN)4]PF6 (1:1) in
CH2Cl2 solution (1.0 mM) under nitrogen with 0.1 M
NBu4PF6 as the supporting electrolyte. Electrochemical
oxidation/reduction was effected by Pt (working /auxiliary)
and Ag/AgCl reference electrodes with final referencing to
added ferrocene(o,+). The phenylated complexes 1a′, 7b′, and
8b′ exhibit quasireversible oxidation waves (CuI/II) with the
return reduction peak having a lower current and separated
from the oxidation peak by 150–190 mV. Oxidation of the
non-Ph complex 9b′ was essentially irreversible (Epa - Epc
≈ 600 mV). Therefore, the potentials reported are based on
the oxidation peak, Epa, relative to E1/2 for Cp2Fe oxidation
(Table 1).16,25 From these data, it is seen that the ease of
oxidation is in the order 9b′ > 8b′ > 1a′ > 7b′. Thus the
4-tethered-Ph-complex 8b′ is more easily oxidized than the
yellow (CuI) to green or blue (CuII), characteristic of the
hydroxo dimer complexes, [(BIT)2Cu2(OH)2](PF6)2 (10-12,
vide infra, Scheme 5.). If no change was observed after a
few hours at -60 °C, the reaction mixture was allowed to
warm to room temperature over several hours. In no case
was a distinct transient color observed that would suggest
appreciable buildup of an intermediate complex, for example,
[(BIT)Cu(O2)]+. Importantly, the four complexes 1a′, 7b′,
8b′, and 9b′ react at remarkably different rates. Thus, the
oxygenation reactions of the 4-tethered-2-Ph-NMe complex
8b′ and the 2-tethered non-Ph complex 9b′ changed color
within minutes at -50 °C, while 4-BIT2-Ph,NH complex 7b′,
similar to the original 2-BITPh2,NMe derivative 1a′, reacts with
O2 only slowly upon warming to room temperature oVer
2–12 h. The relative reactivity of 1a′ and 7b′-9b′ toward
dioxygen thus parallels their ease of electrochemical oxida-
tion.
The hydroxo dimers, [L2Cu2(OH)2](PF6)2 10-12, were
isolated in 60–70% yield upon ether diffusion into the
reaction mixtures. These complexes were identified by the
presence of appropriate molecular ions in their ESI-mass
spectra with the Cu2 isotopic signature, IR spectra (νOH at
3390–3430 cm-1), and electronic spectra (327–370 and 540
nm). The structural assignments were confirmed by X-ray
diffraction of each complex, Figures 2-4. The two Ph-
(24) (a) Kretzer, R. M.; Ghiladi, R. A.; Lebeau, E. L.; Liang, H.-C.; Karlin,
K. D. Inorg. Chem. 2003, 42, 3016–3025. (b) Karlin, K. D.; Tyeklar,
Z.; Farooq, A.; Haka, M. S.; Ghosh, P.; Cruse, R. W.; Gultneh, Y.;
Hayes, J. C.; Toscano, P. J.; Zubieta, J. Inorg. Chem. 1992, 31, 1436–
1451. (c) Nelson, S. M.; Lavery, A.; Drew, M. G. B. J. Chem. Soc.,
Dalton Trans. 1986, 5, 911–920.
(25) (a) Hatcher, L. Q.; Vance, M. A.; Narducci Sarjeant, A. A.; Solomon,
E. I.; Karlin, K. D. Inorg. Chem. 2006, 45, 3004–3013. (b) Fujisawa,
K.; Ono, T.; Ishikawa, Y.; Amir, N.; Miyashita, Y.; Okamoto, K.-I.;
Lehnert, N. Inorg. Chem. 2006, 45, 1698–1713.
Inorganic Chemistry, Vol. 47, No. 10, 2008 4359

Zhou and Nicholas
and the apex by the thioether sulfur. Interestingly, the
nonphenylated dinuclear dication of 12 displays C2 symmetry
with both Cu-centers again having square pyramidal geom-
etry (Figure 4). The corresponding bond lengths (Å) and
angles (deg) in the three complexes are very similar and
unexceptional.
Computational Studies. To further probe the effects of
substitutional variations of the tripodal ligands on the
electronic character of the Cu(I)-L+ complexes, we carried
out semiempirical (PM3) and ab initio (DFT B3LYP) MO
calculations. The faster PM3 method was best suited for the
larger arylated complexes 17 (model for 1a) and 18 (model
for 8a),26 while the processing-intensive DFT method was
utilized for the nonarylated derivatives, 13-16, models for
synthetic complexes 7a, 8a, and 9a. In addition to molecular
geometry, other properties calculated that are relevant to
electronic character and reactivity include total energy, IR
ν(CO), atomic charges, frontier MO energies, and distributions.
Figure 2. X-ray structure of 10. Selected bond lengths (Å) and angles.
Cu(1)-O(1) ) 1.9208(16), Cu(1)-N(3) ) 1.961(2), Cu(1)-S(18) )
2.7263(13), Cu(1)-Cu(1′) ) 2.9985(12); O(1′)-Cu(1)-O(1) ) 77.38(12),
O(1)-Cu(1)-N(3) ) 96.30(9), O(1)-Cu(1)-S(18) ) 98.18(3), N(3)-
Cu(1)-S(18) ) 86.52(7), O(1)-Cu(1)-Cu(1′) ) 38.69(6).
Figure 3. X-ray structure of 11. Selected bond lengths (Å) and angles.
Cu(1)-O(33) ) 1.941(6), Cu(1)-N(1) ) 1.984(8), Cu(1)-N(7) ) 2.006(7),
Cu(1)-S(19) ) 2.757(3), Cu(1)-Cu(1′) ) 3.028(2); O(33)-Cu(1)-
N(1) ) 174.9(3), O(33)-Cu(1)-N(7) ) 96.5(3), N(1)-Cu(1)-N(7) )
88.4(3), O(33)-Cu(1)-S(19) ) 92.81(19), N(1)-Cu(1)-S(19) ) 86.4(3),
N(7)-Cu(1)-S(19) ) 84.7(3).
To evaluate the effects of the N-substituent (H vs Me)
and the tethering position (2- vs 4-), DFT calculations were
carried out on complexes 13-16, following initial geometry
optimization with the PM3 method. The resulting DFT-
determined total energies, HOMO E and calculated ν(CO)
are listed in Table 2. The calculated total energies of the
pairs of isomeric 2- and 4-tethered tripod complexes (13 and
15, 14 and 16), which should reflect thermodynamic stability
(Ka), indicate that the 4-tethered complexes are more stable
than the 2-substituted isomers by 5.6 and 11.9 kcal,
respectively. The calculated HOMOs for these complexes
have similar characteristics (Figure 5), being primarily
S-Cu-imidazole π-bonding in character with a small Cu-C
π-(back)bonding contribution but differing significantly in
energy (the PM3-calculated HOMOs are qualitatively the
same). The HOMOs for the 4-tethered complexes 13 and
14 were found at higher energy than the corresponding
2-tethered isomers 15 and 16 (by 3.7 and 3.7 kcal). A
comparison of the N-Me versus N-H substituent effect on
the HOMO energies, that is, 14 versus 13 and 16 versus 15,
shows a higher energy of 3.9 and 4.1 kcal/mol, respectively,
Figure 4. X-ray structure of 12. Selected bond lengths (Å) and angles.
Cu(1)-O(14) ) 1.947(6), Cu(1)-N(1) ) 1.944(5), Cu(1)-S(12) )
2.901(5); N(1)-Cu(1)-O(14) ) 96.5(3), N(1)-Cu(1)-O(14′) ) 95.9(3),
N(1)-Cu(1)-S(12) ) 81.83(18), O(14)-Cu(1)-S(12) ) 94.3(9), O(14′)-
Cu(1)-S12 ) 114.610.
derivatives (N-Me (11), N-H (10)) are, as expected, isos-
tructural and very similar to the previously characterized
2-tethered tripod derivative 2a. The dinuclear dication in each
has central inversion symmetry. Each Cu atom has a square
pyramidal geometry with the base defined by the two
imidazole nitrogens and the two bridging hydroxo oxygens
(26) Cundari, T. R.; Deng, J. J. Chem. Inf. Comput. Sci. 1999, 39, 376–
381.
4360 Inorganic Chemistry, Vol. 47, No. 10, 2008

Tripodal(imid)2(thioether)CuI Complexes
Table 2. Computational Data for [BIT)CuCO)]+ Complexes
ET(Hc) or
L/complex
∆Hf (kcal/mol)
νCu-CO (cm-1
)
FMO energy (eV)
ESCe Cu, N1, N3
[(4-BITNH)CuCO]+ (13)a
[(4-BITNMe)CuCO]+ (14)a
[(2-BITNH)CuCO]+(15)a
-2796.073 H
-2874.703 H
-2796.064 H
-2874.684 H
52.5 kcal
2063 2071d)
2059 (2066)d
2071 (2070)d
2065 (2071)d
2166
-8.99 (HOMO)
+0.31,-0.22, -0.09
+0.30,-0.20, +0.51
+0.32,-0.23, -0.26
+0.31,-0.20, +0.27
-0.72, +0.25, -0.28
-0.77, +0.37, +0.82
-8.81 (HOMO)
-9.15 (HOMO)
>8.97 (HOMO)
-11.68(HOMO) -12.49 (HOMO-5)
-11.66 (HOMO)-12.37(HOMO-5)
[(2-BITNMe)CuCO]+(16)a
[(2-BIT4-Ph,NMe)CuCO]+ (17)b
[(4-BIT2-Ph,NMe)CuCO]+ (18)b
40.6 kcal
2156
a Calculated with DFT B3LYP. method. b Calculated with PM3 method. c ET ) total electron energy, 1 H (Hartree) ) 627.5 kcal. d Calculated with BP
6–31G** method. e Electrostatic charge.
Figure 6. HOMOs of 18 and 17 and the HOMO-5 of 17.
tions on the X-rayed [(2-BITPh2,NMe,OH)Cu(CO)]PF6 (1a). The
overall trigonal pyramidal geometry of 1a was reasonably
reproduced as is generally found with the PM3 method.23b,26
Figure 5. HOMOs for 2- and 4-tethered complexes 15 and 13, respectively,
(H’s omitted).
The calculated bond lengths and angles were generally within
5% of the experimental, but significant differences (10–12%)
for the Me derivatives. The computed Cu-CO IR frequencies
were found in a few metrics, notably the Cu-S distances
and the C-Cu-N angles. A selection of the bond lengths
and angles around the Cu center from the X-ray structure
(and calculated) of 1a follows: Cu-C ) 1.811(2) (1.905),
Cu-N(1) ) 2.0005(19) (1.916), Cu-N(2) ) 2.0259(17)
(1.916), Cu-S ) 2.4392(7) (2.178), C-O ) 1.123(2)
(1.156); C-Cu-N(2) ) 122.90(9) (119.1), C-Cu-N(1) )
132.84(10) (119.5), C-Cu-S ) 118.81(9) (120.7), N-Cu-N
) 90.67(7) (94.1), N(2)-Cu-S ) 93.37(6) (101.6),
N(1)-Cu-S ) 87.33(5) (96.4), Cu-C-O ) 177.4(2)
(179.2).
To assess the effects of the tripodal 2/4-tethering and aryl
group positions the isomeric complexes 17 and 18 were
compared (Table 2). As was the case with the DFT results
on the nonarylated derivatives, the 4-tethered complex 18 is
calculated to be significantly more stable (∼12 kcal) than
the 2-tethered 17. Once again, the 4-tethered 18 is calculated
to have a modestly lower (∼10 cm-1) Cu-CO vibrational
frequency, but in contrast to the results on the nonarylated
complexes, little difference in the HOMO energies (0.02 eV)
is calculated. Inspection of the HOMO distribution in the
arylated complexes (Figure 6) found it to be dependent on
the tripodal tethering position and, for the 2-tethered deriva-
tive 17, to be quite different from that of the nonaryl
derivatives. The HOMO of 4-tethered 18, like the nonphenyl
derivatives 13-16, is primarily S-Cu π-bonding in character
with a small Cu-CO backbonding component. In contrast,
the HOMO of the 2-tethered 17 (Figure 6), is largely
localized on the aryl groups and the imidazoles with modest
Cu, S contributions. At lower energy, HOMO-5 of 17 is
similar in character to the HOMO of the nonarylated
derivatives, with S-Cu-imid π-bonding and a small Cu-CO
backbonding interaction. Comparing the relative energies of
HOMO-5 and the computed Cu-CO vibrational frequencies
for the four complexes varied modestly between 2059 and
2071 cm-1 (Table 2) and were inversely correlated with the
respective HOMO energy, in the order 15 > 16 > 13 > 14.
Because the effectiveness of different DFT methods for
accurately calculating transition metal–carbonyl vibrational
frequencies (i.e., agreement with experiment) varies consid-
erably,27,28 we also employed the BP (6–31G**) method to
calculate the vibrational spectra of 13-16 (Table 2). In this
case, virtually no difference in the Cu-CO vibrational
frequency among the complexes is calculated, varying only
between 2066 and 2071 cm-1, in agreement with the
experimentally determined IR spectra of the synthetic
complexes 1a, 7a, 8a, and 9a. Interestingly, although there
is little difference in the calculated electrostatic charges at
Cu (0.32, 0.31) or at the coordinated N (-0.23, -0.22)
among the 4- vs the 2-tethered complexes 13 and 15, the
charges at the remote N (-0.25,-0.09) are quite different.
Because the CV experiments were conducted on the aceto-
nitrile adducts, parallel PM3 calculations on [(2-
BITNH)Cu(CH3CN)]+ and [(4-BITNH)Cu(CH3CN)]+ revealed
the same energy/HOMO effects operating as with the
(BIT)Cu-COs, that is, the 4-tethered complex is of lower
total energy (-2.5 kcal) and has a higher energy HOMO
(+3.9 kcal).
The electronic character of the phenylated complexes was
probed by semiempirical PM3 level calculations. Because
the structures of the newly prepared [(BIT)CuL]+ complexes
have not been determined crystallographically, the effective-
ness of the PM3 method to determine molecular geometry
for this class of complexes was first evaluated from calcula-
(27) Bolsjakov, V. I.; Rossikhin, V. V.; Voronkov, E. O.; Okovytyy, S. I.;
Leszczynski, J. J. Comput. Chem. 2007, 28, 778–782.
(28) Feng, X.-J.; Gu, J.-D.; Xie, Y.-M.; King, B. R.; Schaefer III, H. F.
J. Chem. Theory Comput. 2007, 3, 1580–1587.
Inorganic Chemistry, Vol. 47, No. 10, 2008 4361

Zhou and Nicholas
for the 4- versus 2-tethered complexes (Table 2), we find,
as with the nonarylated complexes, that the 4-tethered
derivative 18 has a higher energy partially backbonding FMO
and corresponding lower-energy calculated Cu-CO vibra-
tion. Because the oxygenation reactions of 1a′ and 7b′-9b′
are likely to involve initial association of O2 with a
coordinatively unsaturated [(BIT)Cu]+ species, the relative
stability and FMO energies of the isomeric 2- and 4-tethered
complexes are potentially relevant to understanding their
relative reactivity. A PM3-level comparison of the nonphe-
result of significant structural reorganization of the [(BIT)-
CuIIL]2+ species occurring upon oxidation, from chemical
instability of the [(BIT)CuIIL]2+or from slow electron-transfer
kinetics on the electrode surface. Since true E1/2 values could
not be determined, the potentials corresponding to CuI/CuII
oxidation peaks (Epa) relative to E1/2 for ferrocene were used
to enable semiquantitative comparisons of the ease of
[(BIT)CuIL]+ oxidation among this set of complexes.16,25
The order of the Epa values, 7b′ > 1a′ > 8b′ > 9b′, parallels
the anticipated electronic effects of the various substituents.
Thus, complex 8b′, having the electron-releasing N-Me
group, is easier to oxidize than the N-H relative 7b′. A
comparison of the 2-tethered, N-Me complexes 1a′ and 9b′,
with and without Ph-groups, shows the latter to be more
easily oxidized than the former, consistent with a net
electron-withdrawing effect of the Ph-group. The phenylated
4-tethered tripod complex 8b′ is more easily oxidized than
the 2-tethered relative (1a′). This could be ascribed in part
to the electronic differences derived from the tethering
position but also probably reflects the lesser net electron
withdrawal from two phenyl groups of 8b′ versus the four
phenyls of 1a′. It is not possible to quantitatively compare
the oxidation potentials of the present [(BIT)CuL]+ com-
plexes with the Karlin tridentate [N2S-Cu-L]+ compounds
that have a pyridine-alkyl amine-thioether donor set since
nylated [(2-BITNH,OH)Cu]+ (Etot ) 32.28 kcal, EHOMO
)
-11.87 eV) and [(4-BITNH,OH)Cu]+ (29.77 kcal, -11.70 eV)
found, as with the [(BIT)Cu(L)]+ derivatives, that the
4-tethered complex was lower in energy than the 2-isomer
and had a higher-energy HOMO.
Discussion
We have prepared a new set of tripodal N2S-(BIT) ligands
and derived CuI complexes that possess the biologically
relevant imidazole/thioether donor set of the CuM site of the
Cu-hydroxylase enzymes. Our objective was to ascertain the
influence of ligand-derived steric and electronic effects on
the electronic character and oxidative reactivity of the
resulting Cu(I) complexes. The ligands differ in the steric/
electronic character and position of the imidazole substitu-
ents, specifically the C-2/4-position of the tripodal tether,
the presence (absence) of C-phenyl and of N-Me versus N-H
substituents. The more histidine-like 4-tethered ligands were
efficiently prepared using a regioselective metalation/protec-
tion sequence that provides access to the imidazole N-Me
and N-H and the tripodal C-OH and C-OMe derivatives
selectively. Electronically speaking, one would expect N- or
C-alkyl substituents on the imidazole units to increase their
basicity (and σ-donor ability). This is supported by the
determined pKa values of alkyl-imidazoles,29 reported Kassoc
of some imidazole-metal ion complexes,30 and the Breslow
study showing that 4-tethered (imid)3COH form more stable
complexes with Zn(II) and Cu(II).24 While aryl- for H-
substitution will clearly increase steric bulk in the vicinity
of the Cu-L center, the electronic impact is less certain,
with a balance between electron-donating inductive/field
effects and electron-releasing resonance effect.31
Each of the complexes readily forms an isolable CO
adduct. Several studies of N3-Cu-CO derivatives have shown
a correlation between total complex charge and L-donor/
acceptor properties with ν(CO), with neutral complexes and
those with more electron-donating ligands, showing lower
ν(CO), consistent with increased backbonding.23 Somewhat
unexpectedly, therefore, we found that the Cu-CO ν(CO)
hardly varied among the complexes 1a and 7a-9a. Although
this could be taken to indicate there was little electronic
difference among these complexes, the CV, dioxygen
reactivity, and MO calculations indicate otherwise.
the latter are reported to undergo irreversible oxidation at
16
about +400–500 mV vs/Fc(0,+)
.
The differences in reactivity of 1a′ and 7b′-9b′ toward
oxygen, signaled by the CuI/CuII color change, are quite
remarkable, ranging from minutes at -50 to -70 °C for
8b′ and 9b′ to hours at room temperature for 7b′ and 1a′.
These qualitative reactivities reflect the composite rate of
reaction over several steps, presumably including [(BIT)Cu-
CH3CN)]+ dissociation, O2 association with [(BIT)Cu]+ to
form [(BIT)Cu-O2]+, and subsequent steps (in unknown
order), involving binuclear complex formation and H-atom
abstraction, to form the copper hydroxo-dimer products
10-12. Although we have no information yet on the kinetics
or the detailed mechanism of these oxygenations, the absence
of obvious transient colored species that would signal long-
lived intermediates, suggests that an early step, for example,
[(BIT)Cu(CH3CN)]+ dissociation or [(BIT)Cu]+ + O2 as-
sociation, is likely rate-limiting. If the latter is the case, one
would expect that the more electron-rich [(BIT)Cu]+ species
would be more reactive (and form a more stable complex)
with the electrophilic dioxygen.24b Likewise, the same factor
is expected to facilitate electrochemical oxidation of the
[(BIT)Cu-L]+ complexes since both processes fundamen-
tally involve partial/complete electron removal from the
Cu(I). Indeed, a correlation between redox potentials and
dioxygen reactivity has been noted in other studies of Cu(I)
(29) (a) El Ghomari, M. J.; Mokhlisse, R.; Laurence, C.; Le Questel, J.-
Y.; Berthelot, M. J. Phys. Org. Chem. 1997, 10, 669–674. (b) Ramsey,
B. G. J. Org. Chem. 1979, 44, 2093–2097.
The electrochemical E1/2 oxidation potential reflects the
energy needed for electron removal/addition at the electrode
surface/solution interface in converting [(BIT)CuIL]+ to
[(BIT)CuIIL]2+ and back. The quasi-reversible nature of the
CV waves for 1a′, 7b′, and 8b′ (L ) CH3CN) may be the
(30) Paiva, A. C. M.; Juliano, L.; Boschcov, P. J. Am. Chem. Soc. 1976,
98, 7645–7648.
(31) March, J. AdVanced Organic Chemistry, 4th ed; Wiley: New York,
1992; pp. 280–284.
4362 Inorganic Chemistry, Vol. 47, No. 10, 2008

Tripodal(imid)2(thioether)CuI Complexes
complexes.32 However, unless Cu-O2 association involves
initial electron transfer,33 followed by recombination to form
the Cu-O2 complex, the electronic details of the two
processes would not be identical and a perfect correlation
would not be expected. Aside from the electronic effects
underlying the redox and oxygenation reactivity differences
among 1a′ and 7b′-9b′, the generally lower oxygenation
reactivity of the phenylated complexes vis a vis the non-
phenylated ones, may in part result from steric inhibition,
kinetic or thermodynamic, to oxygen association (and
subsequent dinuclear complex formation).
The oxygenation results of the [(BIT)CuL]+ complexes
can be compared and contrasted with Karlin’s non-imidazole
N2S-Cu complexes16 and other N3-Cu complexes. Two of
the former were found to be unreactive with O2 between
-40 and -80 °C, and their oxygenation at room temperature
over minutes/hour afforded complexes in which thioether
S-oxidation had occurred, a pathway not observed with the
BIT-Cu complexes. The [(BIT)CuL]+ compounds 8b′ and
9b′ thus appear to be the most reactive toward dioxygen of
the N2S-Cu complexes prepared to date and are comparable
to several of the reported N2- and N3-Cu model sys-
tems.12,15,19,20 The fundamental question raised here is what
electronic parameters control these two related reactions,
ionization and oxygenation, and the IR Cu-CO vibrational
frequency?
The computational results on the various [(BIT)Cu-CO]+
complexes aid to some extent in clarifying the importance
of various electronic factors that determine the experimental
measurables of ν(Cu-CO), redox potential and oxygenation
reactivity. Before addressing these points, however, we note
the consistent calculational finding that CuI complexes of
the 4-tethered (histidine-like) N2S-ligands are of lower energy
(i.e., are more stable) than the 2-tethered derivatives; the
calculated energy difference ranges from 5.7 kcal from DFT
on 13/15 to 11.9 kcal by PM3 on the phenylated 17/18.
Although experimental data supporting this result for the
[(BIT)CuIL]+ complexes are not yet available, the greater
formation constants of the 4- versus 2-tethered (imid)3COH
ligands with some MII ions was established by Breslow et
al.21 A simple electronic explanation for this effect is not
obvious because the pKa values of alkyl-substituted imida-
zoles show the 2-substituted derivatives to be more basic
(∼0.5 pKa units) than the 4-substituted ones.29 In contrast,
the Breslow study showed the greater basicity of the
4-tethered over the 2-tethered tris(imidazolyl) ligands, ascrib-
ing it to greater (stabilizing) charge separation in the
protonated (and coordinated) 4-tethered derivatives. Although
there is little difference in the calculated electrostatic charges
of the 2- versus 4-tethered complexes at copper or the
coordinated nitrogen, the calculated charges at the remote
nitrogen are quite different, consistent with the Breslow
explanation. However, the total energy/stability of each
complex is the result of the sum of all the electron energies,
which include multiple electronic interactions, and hence,
the origin of the greater stability of the 4-tethered ligand
complexes is not readily explained by a single dominant
interaction.
The electrochemical oxidation potential for the
[(BIT)Cu-L]+ complexes is a measure of the free energy
change accompanying the removal of a valence electron from
[(BIT)CuIL]+ to produce [(BIT)CuIIL]2+ in solution. Accurate
calculations of E1/2 would thus require the energies of both
initial and final copper-containing species in solution.
Because of uncertainties associated with obtaining accurate
energies of the paramagnetic N2S-CuII species, including
solvent effects and electrode surface overpotentials, and the
nonideal electrochemical behavior for the present complexes,
rigorous calculations of the electrochemical oxidation po-
tentials were judged impractical.34 To the first approximation,
however, trends in Epa could be estimated from the E(HO-
MO)35 of the [(BIT)Cu-L]+ complexes because this ap-
proximates the energy of the electron to be removed.
Indeed, the calculated HOMO energies among the various
[(BIT)Cu(CO)]+ complexes parallel their ease of electro-
chemical oxidation. The imidazole-substitutional effects on
the [(BIT)CuL]+ complexes suggested by the computational
studies apparently hold for both the carbonyl and the
acetonitrile complexes since PM3 calculations on [(2-
BIT)Cu(CH3CN)]+ and [(4-BIT)Cu(CH3CN)]+ found the
same total energy/HOMO effects as the [(BIT)CuCO]+, that
is, the 4-tethered complex is of lower energy and has a
higher-energy HOMO. Comparison of the HOMO energies
of the phenylated and nonphenylated complexes is not
possible because of the use of different computational
methods.
A thorough computational analysis of [(BIT)Cu(O2)]+
reactivity should consider the energy profile of the complete
stepwise conversion of [(BIT)Cu(L)]+ + O2 to products,
especially the rate-limiting step. Given the present lack of
mechanistic information on these reactions, such an analysis
is beyond the scope of the current study but is the subject of
an ongoing investigation. However, if [(BIT)Cu(L)]+ + O2
association or electron transfer is rate-limiting (consistent
with the apparent absence of long-lived intermediates), then
the energy of the HOMO of sixteen electron [(BIT)Cu]+ (for
O2 association) or eighteen electron [(BIT)Cu(L)]+ (for ET)
should be an important factor in determining reactivity, that
is, the complexes with a higher E(HOMO) should be more
effective (stabilizing) backbonders to O2 in [(BIT)Cu(O2)]+
and also be more easily oxidized. For the [(BIT)Cu(L)]+
complexes, the HOMO energies of the various derivatives
(Table 2) do parallel their reactivity toward dioxygen (Table
1), with the 4-tethered, N-methylated or nonphenylated
derivatives having the higher HOMO energies and being
more readily oxygenated. This appears to be the case for
(32) (a) Baldridge, K. K.; Donovan-Merkert, B. T.; O’Connor, J. M.; Lee,
L. I.; Closson, A.; Fandrick, D.; Tran, T.; Bunker, K. D.; Fouzi, M.;
Gantzel, P. Org. Biomol. Chem. 2003, 1, 763–766. (b) Klinman, J. P.
J. Biol. Chem. 2006, 281, 3013–3016.
(34) (a) Riahi, S.; Moghaddam, A. B.; Ganjali, M. R.; Norouzi, P.; Latifi,
M. THEOCHEM 2007, 807, 137–145. (b) Reynolds, C. A. J. Am.
Chem. Soc. 1990, 112, 7545–7551.
(33) Kamachi, T.; Kihara, N.; Shiota, Y.; Yoshizawa, K. Inorg. Chem. 2005,
44, 4226–4236.
(35) Ue, M.; Murakami, A.; Nakamura, S. J. Electrochem. Soc. 2002, 149,
A1572–A1577.
Inorganic Chemistry, Vol. 47, No. 10, 2008 4363

Zhou and Nicholas
the electronically unsaturated [(BIT)Cu]+ complexes as well
because a PM3 calculational comparison of the sixteen
electron [(2-BITNH)Cu]+ and [(4-BITNH)Cu]+ shows that the
4-tethered form is more stable but has the higher-energy
HOMO. A more rigorous analysis will include computations
(and structures, stabilities) on the [(BIT)Cu(O2)]+ complexes
that are in progress. The above computational analysis
ignores steric effects, which are likely to be minimal in
comparing the N-Me to the N-H derivatives and the 2- versus
4-tethered tripodal ligands. However, complexes of the
phenylated ligands (e.g., 1, 7, 8) could be of attenuated
oxygenation reactivity relative to the nonarylated derivative
9 because of a combination of stabilizing electronic effects
and kinetically retarding steric effects to the approach of O2.
Finally, we have found that the calculational methods
employed vary in their effectiveness of reproducing the
experimentally observed invariance of the Cu-CO vibra-
tional frequencies among 1a and 7a–9a. The PM3 method
calculates ν(Cu-CO) values substantially higher (60–70 cm-1)
than experimentally observed and predicts a significant effect
(10 cm-1) of the position of the tripodal imidazole alkyl
tether. The DFT methods, B3LYP and BP applied to 13–16,
fare better in reproducing the experimental ν(Cu-CO) values
for [(BIT)Cu(CO)]+ (1a, 7a–9a), both with respect to
magnitude (underestimating by 25–30 cm-1) and to the
substituent effects. The B3LYP method calculates a small
effect (∼5 cm-1) of N-Me versus N-H and 4-BIT versus
2-BIT substitution, following a trend expected from simple
inductive effects, but only over a 12 cm-1 range for 13–16.
The BP DFT method, found to more accurately calculate
Conclusions
New bis(imidazole)thioether (N2S)-copper complexes that
incorporate the ligand set of the CuM site of the copper
hydroxylase enzymes have been prepared and characterized.
This investigation has shown that imidazole-based substituent
variations on the BIT ligands can have important effects on
the oxygenation and electrochemical oxidation of corre-
sponding [(BIT)Cu-L]+ complexes. In general, 4-tethering,
analogous to that found in histidine-coordinated metalloen-
zymes, and N-methylation increase electrochemical oxidation
and oxygenation reactivity, while C-arylation decreases
reactivity. The effects of imidazole alkyl substitution on the
stability and reactivity of the (BIT)Cu(I) complexes examined
herein will enable us to more accurately mimic the active
site of copper redox enzymes. On the other hand, within this
set of complexes the Cu-CO IR vibrational frequency is a
poor predictor of oxidative reactivity. Finally, in spite of
strong evidence that dioxygen coordination/activation occurs
at the N2S-CuM site of the hydroxylase enzymes, there has
been a perception from some model studies that N2S-Cu
complexes are inherently less reactive toward dioxygen than
N3-Cu species. Our findings, which include rapid oxygenation
reactions of two [(BIT)CuL]+ complexes at low temperature,
indicate that this is not necessarily so. Further reactivity and
mechanistic studies of these systems and the development
of still more accurate structural and functional mimics of
the copper hydroxylase enzymes are continuing.
Experimental Section
Materials and Methods. All operations were carried out under
nitrogen or argon by means of standard Schlenk and vacuum-line
techniques. CH2Cl2 was dried over CaH2 and distilled under N2;
THF was dried over Na using benzophenone as indicator and
distilled under N2. Glassware was oven-dried at 150 °C overnight.
A Vacuum Atmospheres drybox was used in the handling of air-
sensitive solids. IR spectra were recorded in either CH2Cl2 solution
or in KBr pellets with a Perkin-Elmer 283-B infrared spectropho-
ν
(M-CO) values of binary metal–carbonyls,27,28 predicts virtu-
ally no difference in ν(Cu-CO) among 13–16 (total range ) 5
cm-1) and no systematic substituent effects, in accord with
the experimental IR data for 1a, 7a–9a. Although chemists
often consider the M-CO frequency as being largely
determined by the balance between two interactions, the CO-
to-M σ-donation and the M-to-CO π-backbonding, extensive
MO mixing and multiple orbital contributions to the M-CO
interaction may complicate quantitative predictions. The
pitfalls of the two (or one) interaction explanation for IR
Cu-CO frequencies in the present systems is also compli-
cated by the changing character (atomic distribution) of the
FMOs between the phenylated versus nonphenylated com-
plexes and the differences in the FMO energies among the
isomeric 4- versus 2-tethered derivatives. Although the EHOMO
for the 4-tethered complexes was consistently found at higher
energy than for the 2-tethered compounds with all the
computational methods, the partially “backbonding” Cu-CO
orbital is the HOMO for the nonphenylated complexes
(13–16) and for the 4-tethered phenylated 18, but it is
“buried” as HOMO-5 in the 2-tethered Ph-complex 17.
Regardless of the electronic reason for the nonvariability of
the experimental Cu-CO absorption in the series, our
experimental and calculational findings caution against using
the M-CO IR vibrational frequency as a reliable indicator
of electronic character (e.g., metal electron density, single
orbital backbonding).
tometer (resolution 4 cm-1). The H and 13C NMR spectra were
1
recorded on a Varian Mercury-300 spectrometer. Mass spectra were
obtained on a Finnigan TSQ 700 ESI instrument in methanol or
acetonitrile solution. Elemental analyses were performed by
Midwest Microlab, LLC. Electrochemical measurements were
performed in freshly distilled CH2Cl2 with a BAS CV-50W
instrument. A three-electrode cell was used (Pt working and
auxiliary electrodes, Ag/AgCl reference electrode). Potentials were
recorded versus ferrocene. Scans were run at 50–400 mV/s under
a nitrogen atmosphere. The solutions were ca. 1.0 mM in analyte
in 10 mL of ca. 0.1 M NBu4PF6 as the supporting electrolyte.
Nitrogen gas was bubbled through the solution for about 10 min
before each set of measurements and was passed continuously over
the surface of the solution during the measurements. N-Ethoxym-
ethyl-2-phenylimidazole22 and compounds 1a, 1a′19 were prepared
according to reported procedures. [Cu(CH3CN)4]PF6 was obtained
commercially.
1,1-Bis(N-ethoxymethyl-2-phenyl-5-imidazolyl)-2-methyl-2-
methylthio-propanol (3a, BIT2-Ph,NEOM,OH). To a solution of
N-ethoxymethyl-2-phenylimidazole (2.42 g, 12.0 mmol) in 30 mL
of dry THF under N2 cooled to -78 °C was added n-BuLi dropwise
(6.30 mL in hexane, 10.8 mmol). The mixture was stirred at this
temperature for 2 h to give a dark green solution. After dropwise
4364 Inorganic Chemistry, Vol. 47, No. 10, 2008

Tripodal(imid)2(thioether)CuI Complexes
addition of dry ethyl 2-methyl-2-methylthiopropinate (0.79 mL, 5.4
mmol), the mixture was allowed to warm to room temperature
overnight to give an orange solution. The reaction was quenched
with 80 mL of water, extracted with diethyl ether, and the organic
phase was dried over MgSO4. Flash column chromatography (silica)
using petroleum ether/ethyl acetate (1:1) as eluant gave 1.60 g (57%)
1260, 1228, 1137, 1083, 694. Anal. Calcd for C24H26N4OS: C,
68.87; H, 6.26; N, 13.39. Found: C, 67.98; H, 6.20; N, 13.10.
1,1-Bis-(N-methyl-2-phenyl-4-imidazolyl)-2-methyl-2-methyl-
thio-propanol (5a, 4-BIT2-Ph,NMe,OH). Sodium hydride (55 mg, 60%
mineral oil suspension, 1.3 mmol) was suspended in dry THF (15
mL) under N2. A solution of 4a (160 mg, 0.40 mmol) dissolved in
dry THF (15 mL) was added dropwise to the NaH suspension over
2 min, and the resulting mixture was stirred for 2 h at room
temperature until no bubbles appeared. Methyl iodide (82 µL, 1.3
mmol) was added dropwise, and then the reaction mixture was
stirred for 24 h to give a yellow solution. The reaction was quenched
with 4 mL of methanol. Rotary evaporation of the solvent and then
flash column chromatography (silica) of the residue using ethyl
acetate as eluant gave 70 mg (41%) of 5a as a light yellow solid.
1H NMR (300 MHz, CD2Cl2): δ 1.48(s, 6H), 1.84(s, 3H), 3.71(s,
6H), 5.17(s, 1H), 7.27(s, 2H), 7.40–7.48(m, 6H), 7.63–7.66(m, 4H).
This isomer was confirmed by 1D-NOESY, which showed an
interaction between N-Me and imidazole vinyl H and an interaction
between N-Me and C-Ph when the N-Me was irradiated. 13C NMR
(75 MHz, CD2Cl2): δ 145.6, 144.4, 131.5, 129.0, 128.9, 128.8,
121.3, 77.8, 60.8, 35.0, 25.0, 13.1. HRMS (ESI) m/z 433.2104 (M
+ 1); calcd for C25H29N4OS 433.2062. IR (KBr; cm-1): 3149, 3063,
2970, 2926, 2861, 1473, 1388, 1296, 1137, 1075, 775, 698. Anal.
Calcd for C25H28N4OS: C, 69.41; H, 6.52; N, 12.95. Found: C,
69.31; H, 6.36; N, 12.64.
1
of the alcohol 3a as a white solid. H NMR (300 MHz, CD2Cl2):
δ 1.06(t, J ) 7 Hz, 6H), 1.61(s, 6H), 1.99(s, 3H), 3.27(m, 4H),
5.11(d, J ) 8 Hz, 2H), 5.30(d, J ) 8 Hz, 3H, CH2 + COH), 7.31(s,
2H), 7.43–7.47(m, 6H), 7.63–7.66(m, 4H). 13C NMR (75 MHz,
CD2Cl2): δ 150.9, 133.5, 131.3, 131.2, 129.8, 129.6, 128.9, 80.1,
74.6, 63.9, 56.7, 26.0, 15.0, 13.4. HRMS (ESI) m/z 521.2574 (M
+ 1); calcd for C29H37N4O3S 521.2586. IR (KBr; cm-1): 2973, 2927,
2857, 1652, 1559, 1508, 1458, 1090. Anal. Calcd for C29H36N4O3S:
C, 66.89; H, 6.97; N, 10.76. Found: C, 66.71; H, 7.03; N, 10.58.
1-Methoxy-2-methyl-1,1-bis-(N-ethoxymethyl-2-phenyl-5-
imidazolyl)-2-methylthiopropane (3b, BIT2-Ph,NEOM,OMe). Sodium
hydride (170 mg, 60% mineral oil suspension, 4.30 mmol) was
suspended in dry THF (20 mL) under N2. Compound 3a (1.80 g,
3.46 mmol) dissolved in dry THF (20 mL) was added dropwise to
the NaH suspension over 2 min, and the resulting mixture was
stirred for 2 h at room temperature until no bubbles appeared.
Methyl iodide (260 µL, 4.30 mmol, 1.20 equiv) was added
dropwise, and then the reaction mixture was stirred for 24 h. Then
the reaction was quenched with 40 mL of deionized water and
extracted twice with 25 mL portions of ether, and the organic phase
was dried over MgSO4. Rotary evaporation of the ether gave 0.78 g
of 3b as a light yellow solid (42%) which was used in the next
step without further purification.
1,1-Bis(N-H-2-phenyl-4(5)-imidazolyl)-2-methyl-2-methyl-
thio-propanol (4a, 4-BIT2-Ph,NH,OH). Compound 3a (100 mg, 0.19
mmol) was dissolved in the mixture of 5 mL of methanol and 10
mL of a 3 M HCl aqueous solution. The resultant mixture was
refluxed under N2 for 18 h to give a colorless solution. After it
was cooled, the reaction mixture was quenched with 40 mL of brine,
neutralized with concentrated ammonium hydroxide, and extracted
twice with 30 mL portions of CH2Cl2, and the organic phase was
dried over MgSO4. Rotary evaporation of CH2Cl2 gave 70 mg of
4a as a light yellow solid (90%). 1H NMR (300 MHz, CD2Cl2): δ
1.44(s, 6H), 1.80(s, 3H), 3.42(s, 1H), 7.23(s, 2H), 7.35–7.49(m,
6H), 7.84–7.86(m, 4H). 13C NMR (75 MHz, CD2Cl2): δ 145.4,
130.8, 129.4, 129.1, 125.4, 76.9, 60.2, 24.9, 13.3. HRMS (ESI)
m/z 405.1740 (M+1); calcd for C23H25N4OS 405.1749. IR (KBr;
cm-1): 3399, 2985, 2828, 1700, 1652, 1560, 1459, 1260, 1083,
694. Anal. Calcd for C23H24N4OS: C, 68.29; H, 5.98; N, 13.85.
Found: C, 68.83; H, 6.29; N, 13.75.
1-Methoxy-2-methyl-1,1-bis-(N-methyl-2-phenyl-4-imida-
zolyl)-2-methylthiopropane (5b, 4-BIT2-Ph,NMe,OMe). Sodium hy-
dride (20 mg, 60% mineral oil suspension, 0.50 mmol) was
suspended in dry THF (10 mL) under N2. Compound 5a (140 mg,
0.32 mmol) dissolved in dry THF (10 mL) was added dropwise to
the NaH suspension over 2 min, and the resulting mixture was
stirred for 2 h at room temperature until no bubbles appeared.
Methyl iodide (30 µL, 0.38 mmol) was added dropwise, and then
the mixture was stirred for 24 h to give a yellow solution. The
reaction was quenched with 40 mL of brine and extracted twice
with 20 mL portions of CH2Cl2, and the organic phase was dried
over MgSO4. Rotary evaporation of the solvent and then flash
column chromatography (silica) of the residue using ethyl acetate
1
as eluant gave 100 mg (70%) of 5b as a white solid. H NMR
(300 MHz, CD2Cl2): δ 1.48(s, 6H), 1.81(s, 3H), 3.22(s, 3H), 3.76(s,
6H), 7.40–7.48(m, 8H), 7.60–7.70(m, 4H). 13C NMR (75 MHz,
CD2Cl2): δ 146.1, 141.1, 131.9, 129.1, 129.0, 128.7, 125.1, 85.5,
64.1, 35.1, 25.5, 13.1. HRMS (ESI) m/z 447.2242 (M + 1); calcd
for C26H31N4OS 447.2219. IR (KBr; cm-1): 2972, 2930, 2861,
14753, 1076, 942, 775, 703. Anal. Calcd for C26H30N4OS: C, 69.92;
H, 6.77; N, 12.54. Found: C, 69.45; H, 6.82; N, 12.66.
1-Methoxy-2-methyl-1,1-bis-(N-H-2-phenyl-4(5)-imidazolyl)-
2-methylthiopropane (4b, 4-BIT2-Ph,NH,OMe). Compound 3b (210 mg,
0.39 mmol) was dissolved in a mixture of 20 mL of freshly dry
methanol and 4 mL of trifluoroacetic acid (TFA). The resultant
mixture was refluxed under N2 for 12 h to give a colorless solution.
After it was cooled, the reaction mixture was quenched with 40
mL of brine, neutralized with concentrated ammonium hydroxide,
and extracted twice with 30 mL portions of CH2Cl2, and the organic
phase was dried over MgSO4. Flash column chromatography (silica)
using petroleum ether:ethyl acetate (1:1) as eluant gave 80 mg
1,1-Bis-(N-methyl-2-imidazolyl)-2-methyl-2-methylthio-
propanol (6a, 2-BITNMe,OH). To a solution of N-methylimidazole
(1.03 g, 12.5 mmol) in 20 mL dry THF under N2 cooled to –78 °C
was added n-BuLi dropwise (8.90 mL in hexane, 12.5 mmol). The
mixture was stirred at this temperature for 2 h to give a light yellow
solution. After addition of dry ethyl 2-methyl-2-methylthiopropinate
dropwise (0.93 mL, 6.25 mmol), the mixture was allowed to warm
to room temperature overnight to give an orange solution. The
mixture was quenched with 80 mL of water and extracted with
CH2Cl2, and the organic phase was dried over MgSO4. Rotary
evaporation of the solvent gave 1.0 g of 6a as colorless crystals
1
(49%) of 4b as a white solid. H NMR (300 MHz, CD3OD): δ
1
1.31(s, 6H), 1.93(s, 6H), 3.22(s, 3H), 7.19(s, 2H), 7.38–7.50(m,
6H), 7.84–7.92(m, 4H). 13C NMR (75 MHz, CD3OD): δ 129.1,
128.7, 125.2, 85.0, 52.8, 52.4, 24.4, 12.4. HRMS (ESI) m/z
419.1823 (M + 1); calcd for C24H27N4OS 419.1906. IR (KBr;
cm-1): 3071, 2985, 2927, 2828, 1700, 1652, 1560, 1509, 1459,
(57%). H NMR (300 MHz, CDCl3): δ 1.71(s, 6H), 1.73(s, 3H),
3.12(s, 6H), 5.30(s, 1H), 6.76(d, J ) 1.2 Hz, 2H), 6.98(d, J ) 1.2
Hz, 2H). 13C NMR (75 MHz, CDCl3): δ 146.7, 126.0, 123.5, 80.9,
54.7, 34.3, 25.4, 13.5. HRMS (ESI) m/z 281.1421 (M + 1); calcd
for C13H21N4OS 281.1436. IR (KBr; cm-1): 3233, 2998, 2967, 2931,
Inorganic Chemistry, Vol. 47, No. 10, 2008 4365

Zhou and Nicholas
Table 3. X-ray Diffraction Collection Data for 10-12
10
11
12
empirical formula
fw
(C48H52Cu2N8O4S2)(PF6)2
1288.13
(C52H60Cu2N8O4S2)(PF6)2(H2O)0.5
1351.23
(C28H44Cu2N8O4S2)(PF6)2
1037.85
cryst syst
space group
unit cell dimensions
monoclinic
C2/m
monoclinic
C2/c
orthorhombic
Pmmn
a ) 20.615(6) Å
b ) 13.416(4) Å
c ) 11.190(3) Å
R ) 90°
a ) 24.906(12) Å
b ) 14.336(6) Å
c ) 17.529(8) Å
R ) 90°
a ) 17.010(8) Å
b ) 17.712 Å
c ) 6.801(4) Å
R ) 90°
ꢀ ) 118.276(8)°
γ ) 90°
ꢀ ) 100.240(8)°
γ ) 90°
ꢀ ) 90°
γ ) 90°
vol
Z, Z′
density
2725.5(14) Å3
2, 0.25
6159(5). Å3
4, 0.5
2049.0(19) Å3
2, 0.25
1.570 Mg/m3
0.71073 Å
1.457 Mg/m3
0.71073 Å
1.682 Mg/m3
0.71073 Å
wavelength
temp
100(2) K
100(2) K
100(2) K
F(000)
abs coeff
1316
2772
1056
1.008 mm-1
semiempirical equiv
0.974, 0.631
1.89-25.99°
10 781
0.896 mm-1
semiempirical equiv
0.966, 0.797
1.65-25.00°
17 227
1.318 mm-1
semiempirical equiv
0.939, 0.707
2.30-23.26°
10 385
abs correction
max., min. transm
θ range
reflns collected
independent reflns
data/restraints/param
wR (F2 all data)
R (F obsd data)
GOF on F2
2786 [Rint ) 0.0528]
2786/68/261
wR2) 0.1221
R1 ) 0.0435
1.010
8203 [Rint ) 0.0525]
8203/523/439
wR2) 0.2991
R1 ) 0.1228
1.104
1590 [Rint ) 0.0561]
1590/1/157
wR2) 0.2147
R1 ) 0.0718
1.027
obsd data [I > 2σI)]
larg., mean shift (s.u.)
larg. diff. peak, hole
2554
6996
1270
0.001, 0.000
0.000, 0.000
0.000, 0.000
0.926, -0.286 e/Å3
1.621 and -1.031 e/Å3
1.863 and -1.055 e/Å3
1476, 1334, 1278, 1198, 1144, 1066, 935, 738, 688. Anal. Calcd
for C13H20N4OS: C, 55.69; H, 7.19; N, 19.98. Found: C, 56.02; H,
7.22; N, 19.91.
HRMS (ESI) m/z 495.0837; calcd for C24H2463CuN4O2S+ 495.0916.
[(4-BIT2-Ph,NMe,OH)CuCO]PF6 (8a). To 5a (21 mg, 0.050 mmol)
and [Cu(CH3CN)4]PF6 (19 mg, 0.050 mmol) was added dry CH2Cl2
(5 mL), and the resulting light yellow solution was stirred at room
temperature for 1 h under argon. Carbon monoxide was then
bubbled into the solution at rt for 1 h to give a light yellow solution.
Compound 8a was obtained as a white solid by slow diffusion of
ether into the solution at rt under N2 (20 mg, 50%). IR (KBr; cm-1):
3167, 2977, 2939, 2094, 1627, 1477, 1082, 838, 701, 559. 1H NMR
(300 MHz, CD2Cl2): δ 1.31(s, 6H), 1.80(s, 3H), 3.72(s, 6H), 7.31(s,
2H), 7.45–7.65(m, 10H). 13C NMR (75 MHz, CD2Cl2): δ 144.0,
131.4, 131.1, 130.5, 130.3, 128.5, 117.7, 76.6, 52.4, 35.0, 24.8,
16.4. HRMS (ESI) m/z 523.1179; calcd for C26H2863CuN4O2S+
523.1329.
1-Methoxy-2-methyl-1,1-bis-(N-methyl-2-imidazolyl)-2-
methylthiopropane (6b, 2-BITNMe,OMe). Sodium hydride (135 mg,
60% mineral oil suspension, 3.38 mmol) was suspended in dry THF
(20 mL) under N2. Compound 6a (630 mg, 2.25 mmol) was
dissolved in dry THF (10 mL) to give a colorless solution that was
added dropwise to the NaH suspension over 2 min, and the resulting
mixture was stirred for 1 h at rt until no bubbles appeared. Methyl
iodide (210 µL, 3.38 mmol) was added dropwise and then the
reaction mixture was stirred for 24 h. Then the reaction was
quenched with 40 mL of deionized water and extracted twice with
25 mL portions of CH2Cl2, and the organic phase dried over MgSO4.
Rotary evaporation of the CH2Cl2 gave 400 mg of 6b as a white
[(2-BITNMe,OMe)CuCO]PF6 (9a′). To 6b (30 mg, 0.10 mmol)
and [Cu(CH3CN)4]PF6 (38 mg, 0.10 mmol) was added dry CH2Cl2
(3 mL), and the resulting light purple solution was stirred at rt for
10 min under argon. Carbon monoxide was bubbled into the
solution at for 0.5 h to give a light purple solution. Compound 9a′
was obtained as an off-white solid by slow diffusion of ether into
1
solid (61%). H NMR (300 MHz, CDCl3): δ 1.70(s, 3H), 1.87(s,
6H), 3.13(s, 3H), 3.37(s, broad, 6H), 6.76(d, J ) 1.2 Hz, 2H),
6.93(d, J ) 1.2 Hz, 2H). 13C NMR (75 MHz, CDCl3): δ 146.3,
126.6, 123.2, 78.9, 56.7, 40.3, 34.3, 26.4, 12.9. HRMS (ESI) m/z
295.1580 (M + 1); calcd for C14H23N4OS 295.1593. IR (KBr;
cm-1): 2998, 2930, 1405, 1383, 1277, 1199, 1143, 1044, 935, 737,
669. Anal. Calcd for C14H22N4OS: C, 57.11; H, 7.53; N, 19.03.
Found: C, 57.31; H, 7.58; N, 18.94.
[(4-BIT2-Ph,NH,OH)CuCO]PF6 (7a). To 4a (100 mg, 0.25 mmol)
and [Cu(CH3CN)4]PF6 (92 mg, 0.25 mmol) was added dry CH2Cl2
(6 mL), and the resulting light yellow solution was stirred at rt for
2 h under argon. Carbon monoxide was bubbled into the solution
at rt for 1 h to give a cloudy solution. Compound 7a was obtained
as a white solid by slow diffusion of ether into the light yellow
solution under N2 (135 mg, 70%). IR (KBr; cm-1): 3399, 2931,
2857, 2094, 1198, 1144, 1081, 843. 1H NMR (300 MHz, CD3OD):
δ 1.28(s, 6H), 1.74(s, 3H), 5.48(s, 1H), 7.34(s, 2H), 7.50–7.60(m,
6H), 7.75–7.85(m, 4H). 13C NMR (75 MHz, CD3OD): δ 144.0,
131.4, 131.1, 130.5, 130.3, 128.5, 117.7, 76.6, 52.4, 24.8, 16.4.
1
the solution at rt under N2 (40 mg, 59%). H NMR (300 MHz,
CDCl3): δ 1.72(s, 3H), 1.94(s, 6H), 3.14(s, 6H), 4.08(s, 3H), 6.78(d,
2H), 6.96(d, 2H). 13C NMR (75 MHz, CDCl3): δ 145.3, 127.6,
122.2, 76.9, 54.7, 39.3, 30.3, 27.4, 14.9. HRMS (ESI) m/z 357.0822
calcd for C14H2263CuN4OS+ 357.0810; 385.0810 calcd for
C15H2263CuN4O2S+; 385.0859 found. IR (KBr; cm-1): 3163, 3143,
2985, 2105, 1545, 1489, 1283, 1163, 1078, 878, 848, 774.
[(4-BIT2-Ph,NH,OMe)2Cu2(OH)2](PF6)2 (10). To 4b (40 mg, 0.10
mmol) and [Cu(CH3CN)4]PF6 (37 mg, 0.10 mmol) was added dry
methanol (5 mL), and the resulting solution was stirred at rt for
1 h under argon. Then the solution was stirred under an O2
atmosphere (balloon) at rt for 18 h to give a dark green solution.
Green crystals of 10 were obtained by slow diffusion of ether into
the solution over 6 days (60%). HRMS (ESI) m/z 497.0970
(8.6%), calcd for C48H5263Cu2N8O4S22+, 497.1072; 497.5946(4.0%);
4366 Inorganic Chemistry, Vol. 47, No. 10, 2008

Tripodal(imid)2(thioether)CuI Complexes
498.1057(21.7%), calcd for C48H5463Cu2N8O4S22+, 498.1150;
498.6167(2.0%); 499.1148(5.0%). IR (KBr; cm-1): 3387, 2932,
2855, 1700, 1652, 1559, 1459, 1077, 845. UV–vis(CH3OH): 327
nm (ꢀ 6.03 × 103 M-1 cm-1). See X-ray experimental (Table 3)
and Supporting Information for crystallographic data on 10.
[(4-BIT2-Ph,NMe,OMe)2Cu2(OH)2](PF6)2 (11). To 5b (10 mg, 0.020
mmol) and [Cu(CH3CN)4]PF6 (8 mg, 0.020 mmol) was added dry
CH2Cl2 (3 mL), and the resulting solution was stirred at rt for 1 h
under argon. The solution was cooled to -50 °C, and then O2 was
bubbled into it. The colorless solution turned green within 5 min
and was stirred under O2 at -50 °C for 20 h. Green crystals of 11
were obtained by slow diffusion of ether into the green solution at
room temperature (70%). HRMS (ESI) m/z 525.1569 (30%), calcd
for C52H6063Cu2N8O4S22+, 525.1486; 525.6665 (9.3%); 526.1510
(64%), calcd for C52H6263Cu2N8O4S22+ 526.1564; 526.6640 (14%);
527.1560 (26%); 527.6537 (8.1%). IR (KBr; cm-1): 3425, 2938,
2855, 1688, 1676, 1656, 1545, 1207, 1083, 842, 670, 558.
UV–vis(CH2Cl2): 327 nm (ꢀ ) 6.03 × 103 M-1 cm-1), 540 nm (ꢀ
) 281 M-1 cm-1). See X-ray experimental (Table 3) and SI for
crystallographic data on 11.
[(2-BITNMe,OMe)2Cu2(OH)2](PF6)2 (12). To 6b (30 mg, 0.10
mmol) and [Cu(CH3CN)4]PF6 (38 mg, 0.10 mmol) was added dry
CH2Cl2 (5 mL), and the resulting solution was stirred at rt for 10
min under argon. The solution was cooled to -70 °C, and then O2
was bubbled into it. The colorless solution turned blue within 2–3
min and was stirred under O2 at -50 °C for 1 h to give a blue
cloudy solution. After addition of 3 mL of methanol, the solution
turned clear and dark blue. Needlelike blue crystals of 12 were
obtained by slow diffusion of ether into the solution at rt (70%).
LRMS(ESI, MeOH) m/z 388.11 (100.0%, for C15H2563CuN4O2S+),
390.10 (44.6%, for C15H2565CuN4O2S+). IR (KBr; cm-1): 3429,
2931, 1547, 1494, 1283, 1164, 1080, 842, 758, 558. UV–vis
(CH3OH): λ 370 nm (ꢀ 86.3 M-1 cm-1). See X-ray experimental
(Table 3) and SI for crystallographic data on 12.
Computational Studies. PM3 and DFT computations were
carried out using the Spartan 04 software suite (Wave function,
Inc.). The energetically minimized structures (shown) were deter-
mined in the semiempirical PM3 mode, followed by single energy
point calculations with the DFT method (B3LYP with 6–31G basis
set and BP with 6–31G** basis set) to provide final calculated
energies for comparison. A summary of the calculated Cartesian
coordinates, energies and electrostatic charges for 13-18 is
provided in the Supporting Information.
Acknowledgment. We are grateful for partial support of
this project by the American Chemical Society Petroleum
Research Fund. We also appreciate helpful discussions with
Profs. R. Wheeler, R. Houser, and G. Richter-Addo. X-ray
structure determinations by Dr. Douglas Powell and the
assistance of the Richter-Addo group for the electrochemical
experiments are also greatly appreciated.
Supporting Information Available: Cyclic volatammetric data
for 1a′, 7b′-9b′, X-ray crystallographic data and analysis for 10,
11, and 12, and PM3 and DFT computational output (Cartesian
coordinates, calculated energies, and electrostatic charges) for
13-18. This material is available free of charge via the Internet at
http://pubs.acs.org.
IC800007T
Inorganic Chemistry, Vol. 47, No. 10, 2008 4367