Influences of ligand environment on the spectroscopic properties and disproportionation reactivity of copper-nitrosyl complexes

Article abstract of DOI:10.1021/ja982172q

In studies of the chemistry of new copper-nitrosyl complexes supported by tris(3-trifluoromethyl)-5-methylpyrazol-1-yl)hydroborate (Tp(CF₃,CH₃) and tris(3-mesitylpyrazol-1-yl)hydroborate (Tp(Ms,H)), significant effects of the scorpio

Full text of DOI:10.1021/ja982172q

11408
J. Am. Chem. Soc. 1998, 120, 11408-11418
Influences of Ligand Environment on the Spectroscopic Properties
and Disproportionation Reactivity of Copper-Nitrosyl Complexes
Jamie L. Schneider, Susan M. Carrier, Christy E. Ruggiero, Victor G. Young, Jr., and
William B. Tolman*
Contribution from the Department of Chemistry and Center for Metals in Biocatalysis,
UniVersity of Minnesota, 207 Pleasant Street S.E., Minneapolis, Minnesota 55455
ReceiVed June 22, 1998
Abstract: In studies of the chemistry of new copper-nitrosyl complexes supported by tris(3-(trifluoromethyl)-
5-methylpyrazol-1-yl)hydroborate (TpCF ,CH ) and tris(3-mesitylpyrazol-1-yl)hydroborate (TpMs,H), significant
effects of the scorpionate ligand substituents on the properties of the {CuNO}11 unit were found that have
implications for environmental influences on similar species in biological and catalytic systems. The
3
3
copper(I) complexes TpMs,HCu(THF) and TpCF ,CH Cu(CH3CN) were structurally characterized by X-ray
crystallography, and their respective CO and NO adducts were studied by FTIR, EPR, NMR, and/or UV-vis
spectroscopies in solution. Both nitrosyl complexes disproportionate in the presence of excess NO to N2O
3
3
and TpR,R′Cu(NO2); an X-ray structure of the latter product supported by TpCF ,CH was determined. Unlike
previously studied paramagnetic [CuNO]11 compounds that exhibit EPR signals with g < 2.0 and large ANO
values at temperatures below ∼40 K (Ruggiero, C. E.; Carrier, S. M.; Antholine, W. E.; Whittaker, J. W.;
Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 1993, 115, 11285-11298), TpMs,HCu(NO) is EPR silent at
4.2 K and exhibits an NMR spectrum (238 K, toluene-d8) with sharp signals. Peak assignments for the NMR
spectrum were deduced from integrated intensities, temperature-dependent isotropic shifts, and the nuclear
relaxation rates. The unique NMR spectral behavior for the TpMs,H complex, which only differs from those
of analogues with simple phenyl substituents by virtue of the shape of the substrate binding pocket enforced
by the mesityl methyl groups, suggests that caution should be exercised in characterizing such adducts in
proteins and heterogeneous systems; subtle environmental effects may determine the applicability of EPR
3
3
versus NMR methods. The electron-withdrawing effects of the trifluoromethyl substituents in TpCF ,CH Cu-
(NO) perturb ν(NO) and the Cu(I) f NO MLCT energy in the respective FTIR and UV-vis spectra and
induce a significant slowing of its disproportionation rate. These results, in conjunction with those obtained
from kinetic and spectroscopic studies on the TpMs,H system, support a mechanism for the disproportionation
involving generation of the CuNO adduct from NO and the Cu(I) precursor in a preequilibrium step, followed
by electrophilic attack of a second NO molecule on the adduct that is rate-controlling.
3
3
Introduction
zeolites have been found to be highly active for the catalytic
decomposition of NO.4 The many potential applications of these
materials in heterogeneous NOx removal from gas streams has
motivated several studies of (zeolite)Cu-NO intermediates by
both experimental and theoretical methods.4,5
In efforts to understand the fundamental chemistry of CuNO
species in detail and, ultimately, to control the reactivity of this
interesting unit, we6 and others7 have targeted discrete copper-
nitrosyl complexes for synthesis and physicochemical charac-
terization.8 Previously, we reported the only X-ray crystal
structure of a mononuclear complex, TptBu,HCu(NO) (Figure 1),9
and defined its electronic structure through spectroscopic and
Copper-nitrosyl (CuNO) adducts are implicated in the
reactions of metalloproteins with bioactive nitric oxide and are
pivotal intermediates in biological and catalytic systems that
participate in environmentally significant nitrogen oxide pro-
cessing. Thus, in bacterial denitrification, a key component of
the global nitrogen cycle whereby NO3- and NO2- are reduced
to gaseous N2O and/or N2, a [CuNO]2+ species is a purported
intermediate during copper nitrite reductase turnover.1 Numer-
ous other copper proteins have been probed by purposeful
addition of NO (e.g., as an O2 surrogate), and some of the varied
effects of biosynthesized2 NO may arise from these or related
Cu-NO adducts.3 For example, the cytotoxicity of NO may
be traced in part to its inhibition of mitochondrial cytochrome
c oxidase, for which both copper and heme-iron adducts have
been identified.3c,d In addition to these biological cases, Cu
(3) Selected lead references: (a) Gorren, A. C. F.; de Boer, E.; Wever,
R. Biochim. Biophys. Acta 1987, 916, 38-47. (b) Musci, G.; Marco, S. D.;
di Patti, M. C. B.; Calabrese, L. Biochemistry 1991, 30, 9866-9872. (c)
Zhao, X.-J.; Sampath, V.; Caughey, W. S. Biochem. Biophys. Res. Commun.
1994, 204, 537-543. (d) Cooper, C. E.; Torres, J.; Sharpe, M. A.; Wilson,
M. T. FEBS Lett. 1997, 414, 281-284. (e) Ehrenstein, D.; Filiaci, M.;
Scharf, B.; Engelhard, M.; Steinbach, P. J.; Nienhaus, G. U. Biochemistry
1995, 34, 12170-12177.
* To whom correspondence should be addressed. Fax: 612-624-7029.
E-mail: tolman@chem.umn.edu.
(1) Averill, B. A. Chem. ReV. 1996, 96, 2951-2964.
(2) (a) Crane, B. R.; Arvai, A. S.; Ghosh, D. K.; Wu, C.; Getzoff, E. D.;
Stuehr, D. J.; Tainer, J. A. Science 1998, 279, 2121-2126. (b) Crane, B.
R.; Arvai, A. S.; Gachhui, R.; Wu, C.; Ghosh, D. K.; Getzoff, E. D.; Stuehr,
D. J.; Tainer, J. A. Science 1997, 278, 425-431. (c) Marletta, M. A. J.
Biol. Chem. 1993, 268, 12231-12234.
(4) Shelef, M. Chem. ReV. 1995, 95, 209-225.
(5) Selected recent lead references: (a) Sojka, Z.; Che, M. J. Phys. Chem.
B 1997, 101, 4831-4838. (b) Schneider, W. F.; Hass, K. C.; Ramprasad,
R.; Adams, J. B. J. Phys. Chem. 1996, 100, 6032-6046. (c) Wichterlova´,
B.; Sobal´ık, Z.; Vondrova´, A. Catal. Today 1996, 29, 149-153.
10.1021/ja982172q CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/21/1998

Copper-Nitrosyl Complexes
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11409
and the possible repercussions of such differences on spectral
properties and reactivity, we sought to examine more closely
the consequences of changing the nature of the pocket sur-
rounding the CuNO unit in the TpR,R′ complexes. Herein we
report the results of studies of NO binding to new Cu(I)
complexes supported by the known scorpionates TpCF ,CH 11 and
TpMs,H 12
The former ligand is only marginally more sterically
3
3
.
hindered than TpCH ,CH but, due to the fluorine substitution, is
significantly more electron withdrawing. The ortho methyl
groups on the phenyl rings of TpMs,H restrict rotation of the
mesityl substituents and induce adoption of an active-site pocket
comprising a “wall” of aromatic groups approximately ortho-
gonal to the pyrazolyl rings. We have found that both
3
3
Figure 1. Chem3D representation of the X-ray crystal structure of
TptBu,HCuNO, with hydrogen atoms omitted for clarity.6c
perturbations, the electron withdrawal by TpCF ,CH and the
particular pocket geometry induced by TpMs,H, impact the
chemistry of the CuNO unit significantly. The observed
influences of the ligand substituents on the spectroscopic features
and/or reactivity of the copper nitrosyls imply that environmental
effects on such species in proteins and heterogeneous systems
may be severe.
ab initio theoretical studies.6a,c A molecular orbital description
of the bonding of the {CuNO}11 species was proposed in which
the copper d orbital manifold is full and the SOMO extends
over the entire unit but has a primary contribution from a NO
π* orbital [i.e., Cu(I)-NO• with significant covalency].6c,10
Consistent with this picture, the complex exhibits an EPR signal
3
3
Cu
at temperatures below ∼40 K with ge ∼ g > g| ) 1.83, A
) 62 × 10-4 cm-1, and ANO ) 27 × 10-4 cm-1, no absorption
or MCD intensity in the metal d f d transition region, and a
Cu f NO MLCT band with λmax ) 494 nm (ꢀ ∼1400 M-1
cm-1). Analogous compounds with similar spectroscopic
properties were prepared using ligands with phenyl (TpPh,Ph) or
Results
Copper(I) Complexes. The starting materials TpMs,H
-
Cu(THF) and TpCF ,CH Cu(CH3CN) were isolated as colorless
crystalline solids in moderate yields by respective treatment in
3
3
methyl (TpCH ,CH ) substituents, but the decreased steric hin-
drance in these instances relative to TptBu,H rendered the
complexes unstable with respect to disproportionation (eq 1).6d
3
3
THF of TlTpMs,H with CuCl and of NaTpCF ,CH with [Cu(CH3-
CN)4]SbF6. Isolation of the TpMs,H complex as a monomer
under these reaction conditions was surprising, both because
structurally defined Cu(I)-THF adducts are rare13 and because
analogous preparations with other TpR,R′ ligands (R ) R′ ) H,
CH3, or Ph; R ) tBu, R′ ) H or CH3) yield dimeric complexes
of formula [TpR,R′Cu]2.14 The THF molecule may be removed
either by recrystallizing the complex in the absence of free THF
or by exposing the solid to a vacuum. The resulting species
analyzes correctly for [TpMs,HCu]x, but its exact nature is not
known.
3
3
TpR,R′Cu(NO) + 2NO f TpR,R′Cu(NO2) + N2O (1)
Preliminary kinetics and isotope labeling studies confirmed the
reaction stoichiometry and showed that the decay rate depen-
dence on [TpCu(NO)] was first-order, thus ruling out a rate-
determining dimerization but providing little information on the
order with respect to NO and the nature of the critical N-N
bond formation step.
The structures of TpMs,HCu(THF) and TpCF ,CH Cu(CH3CN)
were confirmed by X-ray crystallography; drawings are shown
in Figures 2 and 3, crystallographic data are presented in Table
1, and selected interatomic distances and angles are listed in
Table 2. Like other monomeric Cu(I) complexes of TpR,R′
3
3
In view of the various different environments (multiple
proteins, zeolites) within which Cu-NO adducts may reside
(6) (a) Carrier, S. M.; Ruggiero, C. E.; Tolman, W. B.; Jameson, G. B.
J. Am. Chem. Soc. 1992, 114, 4407-4408. (b) Tolman, W. B.; Carrier, S.
M.; Ruggiero, C. E.; Antholine, W. E.; Whittaker, J. W. In Bioinorganic
Chemistry of Copper; Karlin, K. D., Tyekla´r, Z., Eds.; Chapman & Hall,
Inc.: New York, 1993; pp 406-418. (c) Ruggiero, C. E.; Carrier, S. M.;
Antholine, W. E.; Whittaker, J. W.; Cramer, C. J.; Tolman, W. B. J. Am.
Chem. Soc. 1993, 115, 11285-11298. (d) Ruggiero, C. E.; Carrier, S. M.;
Tolman, W. B. Angew. Chem., Int. Ed. Engl. 1993, 33, 895-897. (e)
Tolman, W. B. In Mechanistic Bioinorganic Chemistry; Thorp, H. H.,
Pecoraro, V. L., Eds.; ACS Symposium Series 246, American Chemical
Society: Washington, DC, 1995; pp 195-217.
(7) (a) Paul, P. P.; Tyekla´r, Z.; Farooq, A.; Karlin, K. D.; Liu, S.; Zubieta,
J. J. Am. Chem. Soc. 1990, 112, 2430-2432. (b) Paul, P. P.; Karlin, K. D.
J. Am. Chem. Soc. 1991, 113, 6331-6332. (c) Mercer, M.; Fraser, R. T.
M. J. Inorg. Nucl. Chem. 1963, 25, 525-534. (d) Fraser, R. T. M. J. Inorg.
Nucl. Chem. 1961, 17, 265-272. (e) Doyle, M. P.; Siegfried, B.; Hammond,
J. J. J. Am. Chem. Soc. 1976, 98, 1627-1629.
ligands,15 TpMs,H and TpCF ,CH bind in tridentate fashion to the
metal ion which has an additional fourth donor, THF and CH3-
CN, respectively. In TpMs,HCu(THF) (Figure 2), there is a
crystallographic 3-fold axis passing through Cu(1) and B(1),
3
3
(11) (a) Bucher, U. E.; Currao, A.; Nesper, R.; Ru¨egger, H.; Venanzi,
L. M.; Younger, E. Inorg. Chem. 1995, 34, 66-74. (b) Bucher, U. E. Ph.D.
Thesis, Swiss Federal Institute of Technology, 1993.
(12) Rheingold, A. L.; White, C. B.; Trofimenko, S. Inorg. Chem. 1993,
32, 3471-3477.
(13) (a) Munakata, M.; Wu, L. P.; Kuroda-Sowa, T.; Maekawa, M.;
Suenaga, Y.; Furuichi, K. J. Am. Chem. Soc. 1996, 118, 3305-3306. (b)
El-Amouri, H.; Bahsoun, A. A.; Fischer, J.; Osborn, J. A.; Youinou, M.-T.
Organometallics 1991, 10, 3582-3588. (c) Zakharov, L. N.; Saf’yanov,
Y. N.; Struchkov, Y. T.; Abakumov, G. A.; Cherkasov, V. K.; Garnov, V.
A. Koord. Khim. 1990, 16, 802.
(14) (a) Mealli, C.; Arcus, C. S.; Wilkinson, J. L.; Marks, T. J.; Ibers, J.
A. J. Am. Chem. Soc. 1976, 98, 711-718. (b) Carrier, S. M.; Ruggiero, C.
E.; Houser, R. P.; Tolman, W. B. Inorg. Chem. 1993, 32, 4889-4899. (c)
Yoon, K.; Parkin, G. Polyhedron 1995, 14, 811-821. (d) Kiani, S.; Long,
J. R.; Stavropoulos, P. Inorg. Chim. Acta 1997, 263, 357-366.
(15) (a) Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Moro-oka, Y.;
Hashimoto, S.; Kitagawa, T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J.
Am. Chem. Soc. 1992, 114, 1277-1291. (b) Dias, H. V. R.; Lu, H.-L. Inorg.
Chem. 1995, 34, 5380-5382. (c) Dias, H. V. R.; Kim, H.-J.; Lu, H.-L.;
Rajeshwar, K.; de Tacconi, N. R.; Derecskei-Kovacs, A.; Marynick, D. S.
Organometallics 1996, 15, 2994-3003. (d) Keyes, M. C.; Chamberlain,
B. M.; Caltagirone, S. A.; Halfen, J. A.; Tolman, W. B. Organometallics
1998, 17, 1984-1992.
(8) Examples of reactions of NO with copper complexes wherein
formation of an adduct is implicated but is not amenable to direct
characterization: (a) Tran, D.; Ford, P. C. Inorg. Chem. 1996, 35, 2411-
2412. (b) Deters, D.; Weser, U. BioMetals 1995, 8, 25-29. (c) Tran, D.;
Skelton, B. W.; White, A. H.; Laverman, L. E.; Ford, P. C. Inorg. Chem.
1998, 37, 2505-2511.
(9) (a) Abbreviations used (according to ref 9b): TptBu,H ) tris(3-tert-
butylpyrazol-1-yl)hydroborate; TpCH ,CH ) tris(3,5-dimethylpyrazol-1-yl)-
3
3
hydroborate; TpCF ,CH ) tris(3-(trifluoromethyl)-5-methylpyrazol-1-yl)-
3
3
hydroborate; TpMs,H ) tris(3-mesitylpyrazol-1-yl)hydroborate; TpPh,Ph
)
tris(3,5-diphenylpyrazol-1-yl)hydroborate. (b) Trofimenko, S. Chem. ReV.
1993, 93, 943-980.
(10) Other theoretical calculations on bare {CuNO}11 agree with this
view: Thomas, J. L. C.; Bauschlicher, C. W., Jr.; Hall, M. B. J. Phys.
Chem. A 1997, 101, 8530-8539 and references therein.

11410 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Table 1. X-ray Crystallographic Data
Schneider et al.
TpMs,HCu(THF)
TpCF ,CH Cu(CH3CN)
TpCF ,CH Cu(NO2)
3
3
3
3
empirical formula
formula weight
crystal system
space group
a (Å)
b (Å)
c (Å)
C40H48BCuN6O
703.19
trigonal
C17H16BCuF9N7
563.72
monoclinic
P21/c
11.936(2)
8.954(2)
21.890(4)
90
102.80 (3)
90
2281.4(8)
4
1.641
293(2)
C15H13BCuF9N7O2
568.67
orthorhombic
Pnma
17.7811(5)
13.4460(4)
8.6908(2)
90
90
90
2077.8(1)
4
1.818
R3c
11.822(2)
11.822(2)
45.376(9)
90
90
120
5492(2)
6
1.276
173(2)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
d(calcd), g cm-3
temp (K)
173(2)
crystal size (mm)
diffractometer
0.60 × 0.50 × 0.30
Enraf-Nonius CAD4
0.637
2.18-25.99
-14,-14,0 to 14,14,55
7289
1225 (Rint ) 0.065)
0.0369/0.0891
0.0438/0.0929
1.062
0.50 × 0.50 × 0.50
Enraf-Nonius CAD4
1.051
2.29-25.04
0,0,-25 to 14,10,25
4511
4008 (Rint ) 0.1723)
0.0565/0.1474
0.1202/0.1890
1.010
0.22 × 0.22 × 0.03
Siemens SMART
1.161
2.29-25.06
0,0,0 to 21,16,10
10311
1936 (Rint ) 0.0342)
0.0374/0.0771
0.0485/0.0810
1.093
abs coeff (mm-1
θ range (deg)
hkl ranges
)
no. of reflns collected
no. of indep reflns
R1/wR (I > 2σ(I))
R1/wR2 (all data)
goodness-of-fit (F2)
largest diff features (e Å-3
)
0.328 and -0.735
0.932 and -0.613
0.425 and -0.405
Figure 3. Representation of the X-ray crystal structure of TpCF ,CH
-
3
3
Cu(CH3CN) (30% ellipsoids; hydrogen atoms omitted for clarity;
heteroatoms labeled).
Table 2. Selected Bond Lengths (Å) and Angles (deg)a
TpMs,HCu(THF)
Cu(1)-O(51)
Cu(1)-N(1)
2.004(7)
2.061(3)
O(51)-Cu(1)-N(1) 129.6(4)
TpCF ,CH Cu(CH3CN)
3
3
Cu-N(1)
1.875(5)
2.075(4)
134.8(2)
90.0(2)
Cu-N(22)
2.057(4)
2.200(4)
128.5(2)
108.6(2)
88.8(2)
Cu-N(32)
Cu-N(12)
N(1)-Cu-N(22)
N(22)-Cu(N(32)
N(22)-Cu-N(12)
N(1)-Cu-N(32)
N(1)-Cu-N(12)
N(32)-Cu-N(12)
92.8(2)
TpCF ,CH Cu(NO2)
3
3
Figure 2. Representations of the X-ray crystal structure of TpMs,H
-
Cu(1)-N(2)
Cu(1)-N(4)
N(5)-O(1)
1.970(2)
2.195(3)
1.265(3)
Cu(1)-O(1)
2.003(2)
2.439(4)
Cu(THF) (50% ellipsoids; hydrogen atoms omitted for clarity); (a) full
complex, heteroatoms labeled, with only one of three positions for the
disordered THF molecule shown; (b) complex viewed along the
Cu(1)-B(1) axis without the THF ligand.
Cu(1)-N(5)
N(2)-Cu(1)-N(2A) 89.77(13)
N(2)-Cu(1)-O(1A) 160.65(9) N(2)-Cu(1)-O(1)
O(1A)-Cu(1)-O(1) 62.33(13) N(2)-Cu(1)-N(4)
102.30(9)
90.79(9)
132.58(7)
106.13(13)
O(1)-Cu(1)-N(4)
O(1)-N(5)-O(1A) 110.0(3)
a Estimated standard deviations indicated in parentheses.
103.93(10) N(2)-Cu(1)-N(5)
N(4)-Cu(1)-N(5)
but the THF molecule is displaced from this axis and thus is
disordered over three positions. Because the THF oxygen atom
O(51) resides only slightly off-axis, it displayed unreasonable
anisotropic thermal parameters, so it was refined isotropically.
substituents lie orthogonal to their respective pyrazolyl rings
to provide a substrate binding pocket in the fourth copper
coordination position that is enclosed by three aromatic “walls”
Importantly, as in other structures with TpMs,H 12
, and similar to
the case of complexes of a related ligand with 9-anthryl
substituents on the pyrazolyl groups (TpAnt),16 the mesityl
(Figure 2b). The complex TpCF ,CH Cu(CH3CN) (Figure 3) lacks
crystallographic 3-fold symmetry, has disordered CF3 groups,
3
3
(16) Han, R.; Parkin, G. Polyhedron 1995, 14, 387-391.

Copper-Nitrosyl Complexes
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11411
Table 3. FTIR and UV-Vis Data for TpR,R′Cu(L) (L ) NO, CO)
freshly prepared solutions were amenable to spectroscopic
characterization.
ν(CO)
ν(14NO) ν(15NO)
λmax (nm)
L ) NO
(cm-1
)
(cm-1
)
(cm-1
)
ref
UV-vis spectra of the solutions generated from exposure of
the Cu(I) precursors with NO contain a feature at 472 nm
ligand
TpCH ,CH
2060
2069
2079
2086
2100
2109
2137
a
1712
1712
1720
a
1679
1682
1687
494
494
472
474
6d, 14a
6a,c
this work
15a, 6c
15b
this work
15c
3
3
(TpMs,H) or 436 nm (TpCF ,CH ) that we assign by analogy to the
TptBu,H
TpMs,H
TpPh,Ph
3
3
previously more thoroughly studied TptBu,H case as a Cu f NO
MLCT transition (Table 3). For the case with TpCF ,CH , we
have not been able to determine an extinction coefficient because
of our inability to assay the extent of NO binding; manometry
results were unreliable due to decomposition of the complex in
the time required for equilibration of the apparatus, and EPR
and NMR methods were inapplicable (vide infra). For the
3
3
TpCF ,H
3
TpCF ,CH
1753
1722
436
3
3
TpCF ,CF
3
3
a Low stability of this complex prevented acquisition of FTIR data.
1
TpMs,H case, by correlating H NMR integration data obtained
and exhibits Cu-N bonds of lengths that vary significantly to
yield a structure distorted from ideal C3V symmetry. Thus, the
Cu(1)-N(1) distance is short [1.875(5) Å], as in other Cu(I)-
CH3CN complexes,15a,c,17 and the Cu(1)-Npyrazolyl distances
range from 2.057(4) to 2.200(4) Å. Also contributing to the
distortion from C3V symmetry is the skewing of the CH3CN axis
away from the B-Cu vector.
at 238 K that indicated ∼90% NO binding (vide infra) with
UV-vis absorption intensity information under identical condi-
tions, we estimated an extinction coefficient of 1200 ( 120
M-1 cm-1. This absorptivity agrees well with that reported
previously for TptBu,HCu(NO) and TpPh,PhCu(NO) (∼1400 M-1
cm-1).6c Using this extinction coefficient, we calculated that
approximately 59% of TpMs,HCu(THF) in THF binds to NO
(1 atm) at 297 K. The corresponding equilibrium constant, Keq,
defined by eq 2, was then estimated using the extinction
Adoption of monomeric C3-symmetric or fluxional lower
symmetry structures for TpMs,HCu(THF) and TpCF ,CH Cu(CH3-
3
3
1
CN) in solution is indicated by H NMR spectra in toluene-d8
that contain sharp, single sets of the appropriate pyrazolyl and
substituent resonances. The CH3CN peak of the latter complex
is broadened and shifted from that of free CH3CN, consistent
with rapid dissociation and reassociation that are common for
this class of compound. Corroborating the elemental and X-ray
crystallagraphic analysis, 1 equivalent of THF per copper in
TpMs,HCu(THF) was detected by 1H NMR spectroscopy.18 The
1H NMR spectrum of [TpMs,HCu]x in toluene-d8 closely re-
sembles that of TpMs,HCu(THF), implying similar fluxionality
and/or structures in solution for these species. Since solutions
of [TpMs,HCu]x and TpMs,HCu(THF) react identically with CO
and NO (vide infra), we have elected to suspend further studies
aimed at comparing their solution structures.
[TpMs,HCu(NO)]
[TpMs,HCu][NO]
Keq
)
(2)
coefficient and intial concentrations to calculate [TpMs,HCu] and
[TpMs,HCu(NO)] and determining [NO] in THF through ex-
trapolation of solubility data in the literature for cyclohexane.19
At 297 K, Keq ) 85 ( 25 M-1, with the large estimated error
arising from the inexact extinction coefficient and NO concen-
tration values. A similar value of Keq ) 140 ( 10 M-1 was
reported previously for NO binding to [TptBu,HCu]2 in toluene
at 296 K.6c
Consistent with the MLCT assignment for the optical
absorption feature of the nitrosyl adducts, as the electron-
withdrawing capability of the TpR,R′ ligand and the ν(CO) value
for its Cu(I)-CO complex increase down the series listed in
Table 3, so does the energy of the absorption band of the nitrosyl
complex. The relative electron-withdrawing influences of the
TpR,R′ ligands are also reflected by the ν(NO) values in the FTIR
spectra, which were identified as nitrosyl stretches by their
15NO isotope shifts that matched those calculated theoretically
for a NO harmonic oscillator [Table 3; observed ν(14NO)/
ν(15NO) ) 1.018-1.020; calculated ) 1.019].
Using an established protocol,15b,c the relative electronic
influences of TpMs,H and TpCF ,CH on a coordinated Cu(I) ion
were assessed by preparing CO adducts and examining their
ν(CO) values by FTIR spectroscopy (Table 3). The high value
3
3
of 2109 cm-1 for TpCF ,CH Cu(CO) reflects the strong electron-
withdrawing property of its scorpionate ligand, which is
3
3
surpassed only by TpCF ,CF [ν(CO) ) 2137 cm-1 for its copper
3
3
carbonyl complex].15b,c As expected, the value for TpMs,H
-
Cu(CO) (2079 cm-1) falls just below that of the TpPh,Ph analogue
(2086 cm-1),15a indicative of only minor electronic differences
between these two aryl-substituted ligands. As shown in Table
Surprisingly, the solutions obtained upon reactions of NO
with TpMs,HCu(THF) and TpCF ,CH Cu(CH3CN) were EPR silent
between room temperature and 4.2 K; only signals correspond-
ing to free NO plus Cu(II) impurities that integrated for <10%
of the available copper in the samples were observed at 4.2 K.
This lack of an EPR signal stands in distinct contrast to the
cases of the nitrosyl adducts formed in Cu-zeolites and of those
supported by TptBu,H and TpPh,Ph, which exhibit characteristic
signals with g < 2.0 and large Cu and NO hyperfine splitting.5,6
We did note that the color of the solutions deepened consider-
ably upon freezing, suggestive of a chemical change (dimer-
ization or binding of a second NO molecule?)20 that might result
in loss of EPR activity. Unfortunately, the 1H NMR spectrum
3
3
3, the alkyl-substituted compounds with TptBu,H 14b and
14a
TpCH ,CH
exhibit the lowest ν(CO) values and thus are the
3
3
most electron rich.
Nitric Oxide Adducts. Upon addition of NO (1 atm), the
colorless solution of TpMs,HCu(THF) or TpCF ,CH Cu(CH3CN)
in toluene, THF, or CH2Cl2 immediately became orange or
yellow, respectively, indicative of adduct formation (albeit not
quantitative, as seen previously6 and described below). For the
TpMs,H complex, the NO adduct could be obtained as a dark
orange precipitate, but NO loss from this solid was too facile
to allow drying and elemental analysis. We were unable to
3
3
obtain solid samples of the NO adduct in the TpCF ,CH case.
Both nitrosyl complexes decayed over time (vide infra), but
3
3
of TpCF ,CH Cu(CH3CN) under 1 atm of NO in THF-d8 only
shows Cu(I) starting material between 213 K and room
3
3
(17) But see: Tyeklar, Z.; Jacobson, R. R.; Wei, N.; Murthy, N. N.;
Zubieta, J.; Karlin, K. D. J. Am. Chem. Soc. 1993, 115, 2677-2689.
(18) Due to a long T1 of 26 s for the THF hydrogens (measured by the
inversion recovery method), observation of full intensity for the THF signal
necessitated using a relaxation delay time of ∼90 s in the FT-NMR
experiment.
(19) (a) Tsiklis, D. S.; Svetlova, G. M. Zh. Fiz. Khim. 1958, 32, 1476-
1480. (b) Solubility Data Series: Oxides of Nitrogen; Young, C. L., Ed.;
Pergamon Press: Oxford, U.K., 1981; Vol. 8.
(20) UV-vis or FTIR data acquired at cryogenic temperatures (<77 K)
would be informative, but we have been unable to make these measurements.

11412 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Schneider et al.
Figure 5. Temperature dependence of isotropic shifts for TpMs,HCuNO
(peak assignments as shown in Figure 4).
Table 4. Nuclear Relaxation Rates and Estimated N(nitrosyl)‚‚‚H
Distances (r) for TpMs,HCuNO
hydrogensa
T1M (s)b
T2M (s)c
r (Å)d
a
b
c
d
e
0.027(2)
0.21(2)
0.29(4)
0.080(8)
0.38(4)
0.019
0.069
0.065
0.038
0.050
4.9(11)
6.0(8)
6.3(2)
5.0(3)
6.6(2)
a Labels refer to assignments indicated in Figure 4. b Measured using
the inversion recovery method at 223 K (500 MHz). Estimated standard
deviations are given in parentheses. c Measured from the peak width
at half-height. d Estimated by superimposing the coordinates for the
CuNO fragment from the X-ray structure of TptBu,HCuNO6c on those
from the structure of TpMs,HCu(THF).
Figure 4. 1H NMR spectra (THF-d8, 238 K, 300 MHz) with peak
assignments of (top) TpMs,HCu(THF) and (bottom) TpMs,HCu(THF) +
NO (1 atm).
temperature, similar to the spectra of the EPR-active TptBu,H
and TpPh,Ph systems.6 Thus, it seems that the chemical changes
at the low temperatures (<40 K) required to observe an EPR
signal6 in conjunction with paramagnetic properties that preclude
observation of NMR data (e.g., a long electron relaxation
correlation time, τs) at high temperatures (>200 K) conspire to
intensity of the most downfield resonance at 12.6 ppm of the
major species under NO. This result also allowed assignment
of the 12.6 ppm peak to the 5′-pyrazolyl hydrogen in the major
1
species. Thus, we ascribe the H NMR spectrum of TpMs,H
-
Cu(THF) under NO to paramagnetic TpMs,HCu(NO) (major) in
equilibrium with the starting Cu(I) complex (minor). Evidently,
the electron relaxation correlation time (τs) for the nitrosyl
supported by TpMs,H (NMR active) is shortened compared to
those of the TptBu,H and TpPh,Ph complexes (EPR active, albeit
only at temperatures <40 K).6
prevent further EPR and NMR spectroscopic characterization
21
of TpCF ,CH Cu(NO).
3
3
However, the 1H NMR spectrum of TpMs,HCu(THF) in THF-
d8 under 1 atm of NO at 238 K (temperature chosen to increase
NO binding and inhibit decomposition; vide infra) contained a
new set of peaks along with those of the Cu(I) starting material
in an ∼9:1 ratio (Figure 4). By a comparison of the peak
intensities to those of the added ferrocene standard, g95% of
the TpMs,H present in the sample was accounted for by these
two sets of signals. The number of peaks for the major species
under NO indicates that it is effectively C3-symmetric, with
equivalent environments for the substituted pyrazolyl groups.
The peak positions and widths indicate the influence of a
The indicated peak assignments for TpMs,HCu(NO) were
deduced from integrated intensities, isotropic shifts and their
temperature dependencies (Figure 5), and the nuclear relaxation
rates (Table 4). Also listed in Table 4 are approximate distances
(r) between the TpMs,H hydrogens and the unpaired spin, which
was assumed to be localized on the nitrosyl nitrogen. Because
the SOMO has a large component on N, this is a reasonable
qualitative assumption based on theory,6c but a more sophisti-
cated treatment should take into account delocalization of the
electron over the entire CuNO unit. The distances r (averaged
over chemically equivalent hydrogen nuclei) were estimated by
superimposing the CuNO core coordinates determined by X-ray
crystallography6c for TptBu,HCu(NO) on those for the TpMs,HCu
fragment from the X-ray structure of TpMs,HCu(THF). The
nuclear relaxation rates T1M and T2M were measured at 223 K
by using the inversion recovery method and the line width at
half-height, respectively (Table 4).22 Consistent with a dominant
paramagnetic center. The presence of peaks due to TpMs,H
-
Cu(THF) in addition to those due to the major species rules
out nonspecific effects of free NO in solution on the NMR
properties of TpMs,HCu(THF) as being responsible for the new
signals. Moreover, dynamic equilibration between TpMs,H
-
Cu(THF) and the major species was indicated by saturation
transfer experiments, wherein saturation of the 5′-pyrazolyl
hydrogen signal of TpMs,HCu(THF) at 7.7 ppm caused loss of
(21) We also have been frustrated in our attempts to obtain an EPR signal
for TpCH ,CH Cu(NO), but for different reasons. It disproportionates to
(22) The absolute T2M values are not necessarily accurate, as the relatively
narrow line widths observed render the method used to obtain them suspect
(see pp 253-254 in ref 23). Still, it is worth considering the relative
magnitudes of the estimated T2M values listed in Table 4, which follow the
trends for the T1M values.
3
3
TpCH ,CH Cu(NO2) and N2O very rapidly, about as fast as it is formed. As
3
3
a result, we have been unable to avoid a large Cu(II) signal due to TpCH ,CH
3
3
-
Cu(NO2) that overwhelms the EPR spectrum and prevents observation of
the nitrosyl complex signal in this case.

Copper-Nitrosyl Complexes
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11413
Figure 6. Representation of the X-ray crystal structure of TpCF ,CH
-
3
3
Cu(NO2) (50% ellipsoids; hydrogen atoms omitted for clarity; hetero-
atoms labeled).
Figure 7. Representative pseudo-first-order plot of kinetic data for
the loss of TpMs,HCuNO (O) and the formation of TpMs,HCu(NO2)
(4) under 1.5 atm of NO(g) at 297 K. Linear fits to the data are shown,
with slopes corresponding to kobs1 ) 2.76(2) × 10-4 s-1 (R ) 0.9998)
and kobs2 ) 2.75(2) × 10-4 s-1 (R ) 0.9999).
dipolar contribution to the relaxation of the hydrogen nuclei,
T1M tracks approximately with r-6 23
,
but there is significant
scatter about a linear fit to a plot of T1M versus r-6. We
attribute this poor fit to the spread of hydrogen positions for
each chemically equivalent site and error inherent in the
assumption of spin localization on N in the CuNO unit. As is
evident from Figure 5, the 5′-pyrazolyl hydrogen most distant
from the paramagnetic locus exhibits the largest isotropic shift
with the greatest temperature dependence, with much smaller
shifts evident for all of the other hydrogen atoms. These results
implicate the dominance of a contact interaction between the
5′-pyrazolyl hydrogen and the unpaired spin, presumably via a
σ and/or π delocalization mechanism.23 Much weaker contact
and/or dipolar interactions apparently are responsible for the
smaller isotropic shifts for the other hydrogen atoms in the
ligand. The orthogonality of the pyrazolyl and mesityl rings
coupled with the large number of bonds separating the mesityl
hydrogens from the paramagnetic center would attenuate σ or
π delocalization onto them, consistent with their small isotropic
shifts.
-1
Table 5. NO Disproportionation Rate Constants for the Decay of
a
TpR,R′CuNO (kobs1) and the Formation of TpR,R′Cu(NO2) (kobs2
)
4
ligand
104kobs1 (s-1
)
kobs2 (s-1
>30b
)
ref
TpCH ,CH
>30b
3.9(2)
2.0(1)
6d
3
3
TpMs,H d
TpPh,Ph
3.9(1)
1.8(1)
0.40(3)
this work
c
TpCF ,CH
0.65(3)
this work
6c
3
3
TptBu,H
<0.16e
<0.16e
a Conditions: 2 atm of NO(g), 297 K, CH2Cl2. Estimated standard
deviations are indicated in parentheses. b Estimated from data previously
reported using different NO pressures. c Ruggiero, C. E. Ph.D. Thesis,
University of Minnesota, 1995. d In THF. e Upper bound estimated from
observed stability of TptBu,HCu(NO); N2O generation not confirmed.
compounds contain square pyramidal Cu(II) ions with a
symmetrically coordinated nitrite ion O,O-bound in the basal
plane. In contrast, the Cu(II) geometry in the TptBu,H complex
is distorted trigonal bipyramidal and the Cu(II)-Onitrite bond
distances differ considerably [1.976(5) and 2.169(6) Å]. These
Disproportionation Reactivity. Like the NO adducts sup-
6d
ported by TpPh,Ph and TpCH ,CH
,
the nitrosyls TpMs,HCu(NO)
structural differences have spectral consequences; most notably,
3
3
CH3,CH3
the TpCF ,CH and Tp
compounds exhibit quite similar axial
3
3
and TpCF ,CH Cu(NO) decompose under excess NO to yield N2O
and the respective Cu(II)-nitrite complex TpR,R′Cu(NO2).
Production of 0.6-1.0 equiv (based on Cu) of N2O was
confirmed by GC analysis of the headspace gases. In situ UV-
vis or EPR monitoring of the reactions revealed quantitative
3
3
EPR signals6d but the signal for the TptBu,H complex is
rhombic.24 Overall, the analogous properties of the Cu(II)-
-
NO2 complexes supported by TpCF ,CH and TpCH ,CH are
consistent with similar steric effects for these two scorpionates,
with the large tert-butyl substituents in TptBu,HCu(NO2) being
responsible for its divergent structural and spectral features.
3
3
3
3
generation of TpCF ,CH Cu(NO2) or TpMs,HCu(NO2); these prod-
ucts also were isolated from the reaction mixtures as crystalline
solids in 62% and quantitative yields, respectively. In addition,
TpMs,HCu(NO2) was prepared independently via admixture of
TlTpMs,H, CuCl2, and NaNO2. Identification of this compound
rested on its proper elemental analysis, spectroscopic features
consistent with a mononuclear 5-coordinate Cu(II) complex
[UV-vis λ (ꢀ) 300 (1500 M-1 cm-1), 390 (sh, 260), 778 (70)
nm; EPR (9.47 GHz, 77 K) g| ) 2.31, g ) 2.06, A|Cu ) 135
3
3
The rates of disproportionation of TpMs,HCu(NO) and TpCF ,CH
-
3
3
Cu(NO) were measured by monitoring the decay of the Cu f
NO MLCT band and/or the growth of the d f d absorption
feature of the product nitrite complex under conditions of excess
NO. First-order plots of the data over 1-6 half-lives were linear
(R > 0.997; for a representative plot, see Figure 7), providing
observed rate constants for nitrosyl complex decay (kobs1) and
nitrite complex production (kobs2). The rate constants obtained
under 2.0 atm of NO are listed with the values obtained
G, A|N ) 10 G, A ) 12 G], and NO2 ligand vibrations in
the FTIR spectrum that shifted appropriately when 15NO2- was
incorporated. Analogous analytical and spectral data were
N
-
previously6d for the TpCH ,CH , TpPh,Ph, and TptBu,H systems in
Table 5. The kobs1 and kobs2 values for each run generally were
the same within experimental error. However, in some instances
3
3
obtained for TpCF ,CH Cu(NO2), which also was characterized
by an X-ray crystal structure (Figure 6 and Tables 1 and 2).
3
3
kobs1 * kobs2, particularly for the TpCF ,CH case at low NO
pressure, indicative of a metastable intermediate (yet to be
observed by spectroscopy) or possible mass transfer effects
(diffusion of gas into solution) under these particular conditions.
Both steric and electronic effects on the rate of NO dispropor-
tionation are evident from the data in Table 5. Particularly
illustrative comparisons are (a) the different rates for the
3
3
The topology of TpCF ,CH Cu(NO2) is analogous to that of its
3
3
TpCH ,CH analogue reported previously6d but differs from that
3
3
of TptBu,HCu(NO2).24 Both of the TpCF ,CH and TpCH ,CH
3
3
3
3
(23) Bertini, I.; Luchinat, C. In NMR of Paramagnetic Substances; Lever,
A. B. P., Ed.; Coordination Chemistry Reviews, Vol. 150; Elsevier: New
York, 1996.
(24) Tolman, W. B. Inorg. Chem. 1991, 30, 4878-4880.

11414 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Schneider et al.
Figure 8. Proposed mechanisms for the disproportionation of nitric oxide mediated by TpR,R′CuI complexes.
electronically similar systems with the sterically hindered TptBu,H
(slow) and much less hindered TpCH ,CH (fast) and (b) the
different rates for the compounds supported by the similarly
3
3
sized but relatively electron-donating TpCH ,CH (fast) and
3
3
electron-withdrawing TpCF ,CH (slow).
3
3
Shown in Figure 8 are the two simplest mechanisms that
would account for (a) the existence of a rapid equilibrium
between the Cu(I) precursor and its NO adduct (Keq, eq 2) and
(b) the observed first-order dependence of the disproportionation
rate on the copper complex concentration. In pathway A,
equilibration between the starting Cu(I) complex and its NO
adduct is followed by some type of rate-determining unimo-
lecular isomerization (k2A). Loss of a pyrazolyl arm of the
scorpionate ligand is one attractive possibility, as such arm
dissociation/reassociation paths have been identified in other
reactions of Tp complexes.14b,25 Subsequent fast and kinetically
invisible steps involving binding of a second NO molecule,
N-N bond formation, and oxygen atom transfer would ensue.
The rate law for mechanism A is described by eqs 3 and 4,
rate ) kobs[Cutot]
k2Keq[NO]
(3)
(4)
kobs
)
1 + Keq[NO]
where [Cutot] ) [TpCu(THF)] + [TpCu(NO)] (for derivations,
see Supporting Information). The alternative pathway B differs
only by virtue of the rate-controlling step, which here involves
attack at the nitrosyl adduct by a second NO molecule, either
at bound NO or at Cu, to yield a cis-dinitrosyl or a hyponitrite
species, respectively, in a bimolecular step (k2B). Again, the
final products, N2O and TpR,R′Cu(NO2), would then be generated
in subsequent fast and ill-defined steps. The expression for kobs
for mechanism B is given by eq 5.
Figure 9. Plots of kobs1(1 + Keq[NO])/Keq versus [NO] (a) and [NO]2
(b) for the NO disproportionation reaction mediated by TpMs,HCuNO,
where Keq is that determined from NMR and UV-vis measurements
(85 ( 25 M-1). In (a), the parabola is a fit of the data to a second-
order polynomial (y ) a + bx + cx2; a ) 1.38 × 10-6, b ) -5.4 ×
10-5, c ) k2 ) 0.0158(5) s-1; R ) 0.996). In (b), a linear fit is shown
(R ) 0.996; slope ) k2 ) 0.0146(5) s-1).
k2Keq[NO]2
kobs
)
(5)
1 + Keq[NO]
Comparison of eqs 4 and 5 indicates differences in the
dependence on the concentration of NO for mechanisms A and
B, suggesting that they may be distinguished experimentally
by measuring kobs as a function of [NO]. We performed the
experiments for the TpMs,H case, as these were facilitated by
the close correspondence between kobs1 and kobs2 combined with
the convenient rates at room temperature. The disproportion-
ation rate constant was found to be dependent on the NO
pressure (509-1500 Torr) as shown in Figure 9. Here, kobs1(1
+ Keq[NO])/Keq (see eqs 4 and 5) is plotted versus [NO] (Figure
9a) and [NO]2 (Figure 9b), where kobs1 is multiply determined
for three to nine runs at each NO pressure, and the Keq value is
(25) (a) Bucher, U. E.; Currao, A.; Nesper, R.; Ru¨egger, H.; Venanzi,
L. M.; Younger, E. Inorg. Chem. 1995, 34, 66-74. (b) Looney, A.; Parkin,
G. Polyhedron 1990, 9, 265-276. (c) Keyes, M. C.; Young, V. G., Jr.;
Tolman, W. B. Organometallics 1996, 15, 4133-4140.

Copper-Nitrosyl Complexes
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11415
that determined from NMR and UV-vis measurements (85 (
25 M-1). The good fits of the data in Figure 9a (vs [NO]) to
a parabola [y ) a + bx + cx2, c ) k2 ) 0.0158(5) s-1; R )
0.996] and in Figure 9b (vs [NO]2) to a straight line [slope )
k2 ) 0.0146(5) s-1; R ) 0.996] are strong evidence in support
of eq 5 (mechanism B) and argue against eq 4 (mechanism A).26
Similar results also were obtained for TpPh,Ph (data not shown).
Thus, the data best support the pathway wherein reaction of
NO with TpR,R′Cu(NO) occurs in the rate-determining step.
relatively slightly) by the paramagnetism arising from the
unpaired spin density of the {CuNO}11 core. We hypothesize
that the distinctly contrasting spectroscopic behavior derives
from a shortening of the electron spin correlation time, τs, of
the TpMs,H complex compared to that supported by TpPh,Ph 23
.
This shorter τs renders TpMs,HCu(NO) amenable to NMR
characterization. Although it is difficult to ascertain the cause
of the short τs value, we speculate that relaxation in the TpMs,H
case is enhanced by an Orbach process involving low-lying
excited states.27 Perhaps this process is promoted by interactions
of the bound NO with the π electron systems of the specifically
oriented aromatic rings enclosing the TpMs,H binding pocket.
Whatever the specific rationale, the finding that a simple
perturbation of aromatic ring geometry surrounding the bound
NO in the TpPh,Ph and TpMs,H complexes induces a drastic change
in the EPR and NMR properties is significant, with particular
implications for spectroscopic studies of {CuNO}11 adducts in
zeolites and proteins. Thus, care should be exercised in
characterizing such adducts, as subtle environmental effects may
determine the applicability of EPR versus NMR methods.
Whereas unique EPR and NMR properties are observed for
TpMs,HCu(NO) due to its specifically shaped substrate binding
pocket, the electron-withdrawing influences of the trifluoro-
Discussion
Nitric oxide reacts with the copper(I) complexes TpMs,H
-
Cu(THF) and TpCF ,CH Cu(CH3CN) to yield mononuclear mono-
nitrosyl adducts, identified as such principally on the basis of
optical and infrared absorption features and disproportionation
reactivity analogous to that previously identified for structurally
3
3
defined TptBu,HCu(NO) (Figure 1) and congeners supported by
6
TpPh,Ph and TpCH ,CH
Thus, each member of the set of TpR,R′
-
3
3
.
Cu(NO) species exhibits a Cu f NO MLCT band with λmax
)
430-500 nm (ꢀ ∼1200 M-1 cm-1) and a single ν(NO) between
1712 and 1753 cm-1 that shifts appropriately upon substitution
with 15NO (Table 3). In addition, all of the adducts, except the
most stable one supported by TptBu,H, react further with NO to
yield N2O and the respective nitrite complex TpR,R′Cu(NO2).
These important similarities in spectral properties and reactivity
allow the nitrosyl complexes of the various TpR,R′ ligands to be
identified as constituents of a topologically analogous class
comprising the novel {CuNO}11 unit. Despite these similarities
that provide convincing evidence of structural congruence,
however, the different substituents on the various TpR,R′ scaffolds
induce significant disparities in key properties of the nitrosyl
complexes, as illustrated in particular by the studies of the TpMs,H
methyl substituents of TpCF ,CH are dominant in affecting other
properties of its complexes.11 The electron-withdrawing effect
3
3
of TpCF ,CH is manifested clearly by high ν(CO) and ν(NO)
values for its respective copper carbonyl and nitrosyl complexes
and a short λmax for the Cu f NO MLCT feature for the latter
(Table 3). Consideration of the similar structures of the copper-
3
3
(II)-nitrite complexes supported by TpCF ,CH (Figure 6) and
3
3
TpCH ,CH 6d supports comparable effective sizes for the binding
pockets of these ligands, the substituents of the former being
only slightly larger than the latter. While steric factors influence
the rate of NO disproportionation by the copper complexes of
3
3
and TpCF ,CH cases reported herein.
3
3
The X-ray structures of TpMs,HCu(THF) (Figure 2) and of
other reported complexes of the TpMs,H ligand12 show that it
provides a relatively rigid cavity of defined shape, lined by arene
groups, about the remaining substrate binding site. The phenyl
substituents in the related ligand TpPh,Ph afford a less well-
defined binding pocket of different effective shape due to the
ability of the phenyl groups to rotate about the C-C bond
connecting them to their respective pyrazolyl groups.14b Not
surprisingly, electronic differences between TpMs,H and TpPh,Ph
are insignificant, as shown by the similar CO and NO stretching
frequencies and MLCT energies in the respective copper adducts
(Table 3). A large effect on the NO disproportionation rate
due to the different substituent environments of these two ligands
is also absent (Table 5), further supporting the notion that steric
and electronic disparities between them are subtle. However,
while previous work showed that TpPh,PhCu(NO) exhibits a
distinct EPR signal below 40 K with g < 2.0 and large NO
hyperfine splitting features like that of TptBu,HCu(NO),6 here
we report that the TpMs,H analogue is EPR silent at 4.2 K and
exhibits a sharp NMR spectrum (at 238 K) perturbed (albeit
the TpR,R′ ligands (Table 5), the large difference in kobs values
CH3,CH3
for the similarly sized TpCF ,CH and Tp
cases (>45-fold)
3
3
implicates a significant electronic effect as well. In other words,
the experimental data show that increased steric hindrance and
electron withdrawal by the supporting ligand slow the NO
disproportionation rate considerably. Consistent with this result,
mechanistic studies show that attack of a second, electrophilic
NO molecule on the initially and reversibly formed mono-
copper-mononitrosyl adduct is rate-controlling (mechanism B,
Figure 8). These findings also have implications in protein-
and zeolite-based Cu/NO reactions wherein multiple NO
decomposition pathways may compete.3-5 Extrapolation of our
results to these systems suggests that disproportionation of NO
-
to N2O and NO2 would be disfavored in an electron-poor
environment (e.g., such as in the oxygen-rich zeolites).
Conclusions
Copper(I) complexes of TpMs,H and TpCF ,CH have been
characterized and their reactions with NO explored. Mono-
copper-mononitrosyl adducts form in both cases, as shown by
spectroscopic comparisons to previously prepared examples.6
However, these studies also reveal significant effects on the
spectroscopy and reactivity of the {CuNO}11 unit that arise from
3
3
(26) If rate law 4 for mechanism A were operative, the plot in Figure 9a
would be linear and that in Figure 9b would be parabolic. Although the
data may be fit accordingly (R ) 0.98 for linear fit of Figure 9a and R )
0.996 for parabolic fit of Figure 9b), the associated k2 values for these fits
(which should be identical) differ by more than 3 orders of magnitude [6.5
× 10-4 s-1 vs 0.77 s-1, respectively]. Thus, the combination of better fits
of Figure 9a and Figure 9b to a parabola and a straight line, respectively
(R ) 0.996), and the close agreement between the associated k2 values
[0.0146(5) s-1 vs 0.0158(5) s-1] better supports rate law 5 and mechanism
B. While it would be preferable to corroborate this conclusion using data
obtained at high NO pressures near the saturation limit, this has not been
experimentally feasible to date (NO pressures ∼ 7000 Torr would be needed
to increase [NO] sufficiently to make Keq[NO] . 1).
the substituents on TpMs,H and TpCF ,CH . Thus, in contrast to
its relative coordinated to TpPh,Ph that is EPR active and not
amenable to NMR characterization, the nitrosyl complex sup-
3
3
1
ported by TpMs,H exhibits a sharp H NMR spectrum. Not-
withstanding our lack of a definitive rationale for this unique
(27) (a) Orbach, R. Proc. R. Soc. London, Ser. A 1961, 264, 458. (b)
See pp 83-88 in ref 23.

11416 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Schneider et al.
(0.550 g, 1.14 mmol) dissolved in THF (8 mL) was added. After 2-3
h of stirring at room temperature, the solvent was removed under
reduced pressure. The off-white residue was slurried in CH2Cl2 (∼20
mL), the slurry was filtered through Celite, and the filtrate was
concentrated to about 5-10 mL. Slow evaporation under N2 yielded
colorless crystals suitable for X-ray diffraction analysis (0.530 g,
83%): 1H NMR (300 MHz, CD2Cl2) δ 2.23 (s, 3 H), 2.45 (s, 9 H),
6.25 (s, 3 H) ppm; 13C{1H} NMR (75 MHz, CD2Cl2) δ 2.2, 12.3, 103,
121.5 (q, JC,F ) 269 Hz), 141.1 (q, JC,F ) 37 Hz), 145 ppm; FTIR
(KBr) 2558 [ν(BH)], 2277, 1609, 1496, 1461, 1356, 1244, 1188, 1131,
spectral behavior, the discovery that a subtle difference in the
binding pocket provided by the two arene-substituted ligands
induces a drastic change in the applicable spectroscopic method
has important implications for studies of copper nitrosyls in
catalytic and biological systems. The electron-poor nature of
TpCF ,CH is reflected in the spectral features of its copper-
carbonyl and -nitrosyl adducts and results in a significant
slowing of the rate of disproportionation of the latter to N2O
3
3
and TpCF ,CH Cu(NO2). Consistent with this electronic effect
and a rate decrease with increased steric bulk in electronically
similar systems, mechanistic studies support a mechanism for
the disproportionation reaction involving preequilibration of NO
and the Cu(I) precursor with the CuNO adduct, followed by
rate-determining electrophilic attack of a second NO molecule.
Overall, the results reported herein of studies of the TpMs,H and
3
3
1068, 998, 808, 786, 744, 646 cm-1
. Anal. Calcd for C17H16N7-
CuBF9: C, 36.22; H, 2.86; N, 17.39. Found: C, 36.73; H, 2.92; N,
17.65.
TpMs,HCu(THF). TlTpMs,H (1.029 g, 1.33 mmol) and cuprous
chloride (0.145 g, 1.46 mmol) were stirred in THF (15 mL) for 10
min. The solution then was filtered, the volume was reduced under
vacuum to 5 mL, and pentane was allowed to diffuse into the solution
at 253 K to give the product as colorless crystals suitable for X-ray
diffraction analysis (0.490 g, 52%): 1H NMR (500 MHz, toluene-d8,
216 K) δ 1.38 (m, 4 H), 2.12 (s, 18 H), 2.15 (s, 9 H), 3.58 (m, 4 H),
5.87 (d, J ) 2.0 Hz, 3 H), 6.77 (s, 6 H), 7.63 (d, J ) 2.0 Hz, 3 H)
ppm; 13C{1H} NMR (75 MHz, CD2Cl2) δ 20.59, 21.45, 26.13, 68.31,
106.69, 128.39, 130.73, 136.44, 137.72, 137.86, 153.07 ppm; FTIR
(KBr) 2960, 2953, 2916, 2420 [ν(BH)], 1478, 1461, 1345, 1179, 1165,
1033, 771, 743, 709 cm-1. Anal. Calcd for C36H40N6BCu: C, 68.32;
H, 6.88; N, 11.95. Found: C, 68.62; H, 6.84; N, 12.13.
TpCF ,CH systems illustrate the importance of the specific ligand
3
3
environment in affecting copper-nitrosyl complex properties.
Experimental Section
General Procedures. Unless otherwise noted, all reagents, solvents,
and gases used were obtained commercially and were of analytical
grade. When necessary, solvents were dried according to published
procedures28 and distilled under N2 immediately prior to use. Nitric
oxide gas was purified according to published procedures.6c 15NO was
purchased from Cambridge Isotope Laboratories, Inc., and used without
further purification. All air-sensitive reactions were performed either
in a Vacuum Atmospheres inert-atmosphere glovebox under a N2
atmosphere or by using standard Schlenk and vacuum-line techniques.
[TpMs,HCu]x. This complex was obtained either by recrystallizing
TpMs,HCu(THF) via diffusion of pentante into a toluene solution (23%
yield) or by drying TpMs,HCu(THF) under vacuum (95% yield): 1H
NMR (500 MHz, toluene-d8, 213 K) δ 2.04 (s, 18 H), 2.16 (s, 9 H),
5.85 (d, J ) 2.0 Hz, 3 H), 6.79 (s, 6 H), 7.64 (d, J ) 2.0 Hz, 3 H)
ppm; 13C{1H} NMR (125 MHz, benzene-d6) δ 20.3, 20.8, 104.5, 127.9,
132.2, 135.1, 137.2, 137.8, 151.1 ppm; FTIR (KBr) 2972, 2918, 2386
[ν(BH)], 1613, 1486, 1461, 1351, 1200, 1184, 1165, 1071, 1037, 847,
788, 734, 711 cm-1. Anal. Calcd for C36H40N6BCu: C, 68.51; H,
11
3
29
The compounds TlTpMs,H 12
,
NaTpCF ,CH
,
and [Cu(CH3CN)4]SbF6
3
were prepared according to published procedures. Cuprous chloride
was purchased from Mallinckrodt, purified by published procedures,28
and stored under N2. Elemental analyses were performed by Atlantic
Microlabs, Norcross, GA. Spectroscopic and GC/MS data were
collected as described previously.6c,30 Analyses of N2O evolution by
gas chromatography were performed on a Hewlett-Packard 5890 Series
II gas chromatograph with a HP 3396 Series II integrator and a Poropak
Q column (6 ft, 15 mL/min flow rate, 293 K, He carrier gas).
6.39; N, 13.32. Found: C, 68.85; H, 6.50; N, 13.39.
CF ,CH
3
TpCF ,CH CuCO. Tp
Cu(CH3CN) (0.054 g, 0.096 mmol) was
3
3
3
dissolved under N2 in dry, degassed CH2Cl2 (2 mL), and the solution
was cooled to 263 K. CO gas was bubbled through the solution for
∼5 min. After the solution was warmed to room temperature under a
CO purge, the Schlenk flask was sealed and the reaction mixture was
stirred for an additional 30 min. The solvent was removed under
reduced pressure, the off-white residue was redissolved in CH2Cl2, and
pentane was allowed to diffuse into the solution at room temperature
to yield the product as a white crystalline solid (53 mg, 100%): 1H
NMR (300 MHz, CD2Cl2) δ 2.46 (s, 9 H), 6.32 (s, 3 H); 13C{1H} NMR
(75 MHz, THF-d8) δ 12.5, 105.2, 122.4 (q, JC,F ) 268 Hz), 143.2,
147.2, 168.7 ppm; FTIR (KBr) 2558 [ν(BH)], 2109 [ν(CO)], 1476,
1356, 1265, 1173, 1138, 1068, 998, 786, 646 cm-1. Anal. Calcd for
C16H13N6CuOBF9: C, 34.90; H, 2.38; N, 15.26. Found: C, 35.36; H,
2.51; N, 15.16.
3-(Trifluoromethyl)-5-methylpyrazole. This method combines
steps similar to those used to prepare 3,5-bis(trifluoromethyl)pyrazole.31
1,1,1-Trifluoropentane-2,4-dione (10.0 g, 0.0649 mol) was dissolved
under N2 in degassed absolute EtOH (90 mL), and the solution was
cooled to 273 K. Hydrazine monohydrate (3.9 g, 0.0779 mol) was
added via syringe over 10 min, and the mixture was stirred for ∼2 h
at 273 K. Most of the solvent was removed under reduced pressure to
leave a thick white slurry. The slurry was dissolved in dry, degassed
benzene (200 mL), and the solution was then refluxed for 2 d through
a thimble filled with calcium hydride. Most of the benzene was
removed by distillation under nitrogen, with the last ∼30 mL removed
under reduced pressure without heating to prevent charring of the
product. The remaining white residue was then sublimed under vacuum
(0.05 Torr) through bent glass tubing using a heat gun to separate
evolved H2O from the sublimate. The product was then collected as a
white solid (8.4 g, 86%). GC/MS: tR 5.91 min, m/z (relative intensity)
150 (100, M+). 1H NMR data matched those reported previously.32
TpMs,HCuCO. TpMs,HCu(THF) (0.090 g, 0.127 mmol) was dissolved
in a mixture of toluene (1 mL) and pentane (1 mL) in a Schlenk flask,
and the solution was then exposed to a CO atmosphere, causing the
precipitation of a white powder. Excess CO was removed, the mixture
was filtered, and the product was washed with a small amount of
pentane (0.059 g, 70%): 1H NMR (300 MHz, benzene-d6) δ 1.97 (s,
9 H), 2.04 (s, 18 H), 5.92 (d, J ) 2.1 Hz, 3 H), 6.65 (s, 6 H), 7.56 (d,
J ) 2.1 Hz, 3 H) ppm;13C{1H} NMR (75 MHz, benzene-d6) δ 20.7,
21.1, 105.1, 128.3, 131.4, 134.9, 137.5, 137.6, 151.4 ppm (CO carbon
not observed); FTIR (KBr) 2953, 2920, 2462 [ν(BH)], 2079 [ν(CO)],
1482, 1368, 1352, 1183, 1167, 1046, 777, 745 cm-1. Anal. Calcd for
C37H40N6BCuO: C, 67.42; H, 6.12; N, 12.75. Found: C, 67.31; H,
6.21; N, 12.77.
TpCF ,CH CuCH3CN. [Cu(CH3CN)4]SbF6 (0.530 g, 1.14 mmol) was
3
3
dissolved under nitrogen in dry, degassed THF (8 mL). NaTpCF ,CH
3
3
(28) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: New York, 1988.
(29) Kubas, G. J. Inorg. Synth. 1979, 19, 90-92; 1990, 28, 68-70.
(30) Mahapatra, S.; Halfen, J. A.; Wilkinson, E. C.; Pan, G.; Wang, X.;
Young, V. G., Jr.; Cramer, C. J.; Que, L., Jr.; Tolman, W. B. J. Am. Chem.
Soc. 1996, 118, 11555-11574.
(31) (a) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 3165-3170. (b)
Renn, O.; Venanzi, L. M.; Marteletti, A.; Gramlich, V. HelV. Chim. Acta
1995, 78, 993-1000.
(32) (a) Nishiwaki, T. J. Chem. Soc. B 1967, 885-888. (b) Atwood, J.
L.; Dixon, K. R.; Eadie, D. T.; Stobart, S. R.; Zaworotko, M. J. Inorg.
Chem. 1983, 22, 774-779. (c) Elguero, J.; Yranzo, G. I.; Laynez, J.;
Jime´nez, P.; Mene´ndez, M.; Catala´n, J.; de Paz, J. L. G.; Anvia, F.; Taft,
R. W. J. Org. Chem. 1991, 56, 3942-3947.
TpCF ,CH Cu(NO). Yellow solutions containing TpCF ,CH Cu(NO)
3
3
3
3
were prepared either by vacuum-transferring NO(g) into a Schlenk flask
with a known concentration of TpCF ,CH Cu(CH3CN) dissolved in the
3
3
appropriate dry, degassed solvent or by injecting a solution of TpCF ,CH
3
3
-
Cu(CH3CN) via a gastight syringe into a flask that was thoroughly
purged with NO(g). These yellow solutions were used immediately
or stored at T < 253 K for less than 1 d. Solutions containing TpCF ,CH
3
3
-

Copper-Nitrosyl Complexes
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11417
Cu(NO) rapidly decomposed to TpCF ,CH Cu(NO2) and N2O when
exposed to trace amounts of NO2(g) (formed from O2 and NO), so
purging solutions with NO(g) is not a suitable method for preparing
data: FTIR (KBr) 2969, 2939, 2855, 2472 [ν(BH)], 1614, 1483, 1288
[νa(NO2); νa(15NO2) ) 1266], 1350, 1184 [νs(NO2); νs(15NO2) ) 1156],
1167, 1049, 880 [δa(ONO); δa(O15NO) ) 876], 796, 779, 744, 734
cm-1; UV-vis [THF, nm (ꢀ, M-1 cm-1)] 300 (1500), 390 (sh, 260),
3
3
solutions of TpCF ,CH Cu(NO) because trace amounts of NO2(g) may
be admitted into the flask as the needle is inserted into the septum. 1H
NMR (300 MHz, CD2Cl2 or THF-d8, 213 K < T < 297 K): only signals
3
3
778 (70); EPR (9.47 GHz, CH2Cl2/toluene, 77 K) g| ) 2.31, g
)
N
2.06, A|Cu ) 135 G, A|N ) 10 G, A ) 12 G. Anal. Calcd for
C36H40N7BO2Cu: C, 63.86; H, 5.95; N, 14.48. Found: C, 63.96; H,
6.03; N, 14.46.
corresponding to TpCF ,CH Cu(CH3CN) were observed. FTIR (CH2Cl2,
3
3
cm-1): 1753 [ν(15NO) ) 1722], using a CH2Cl2 solution of TpCF ,CH
-
3
3
Cu(CH3CN) as the spectroscopic blank. UV-vis (CH2Cl2, nm) 300,
436. EPR (1:1 CH2Cl2/toluene or 1:1 THF/toluene, 9.44 GHz,
2 K < T < 297 K): silent.
Determination of N2O Yields. In a typical experiment for the
TpCF ,CH system, NO(g) (1 atm, 49 mL) was vacuum-transferred into
3
3
a solution of TpCF ,CH Cu(CH3CN) (24 mg, 0.043 mmol) in dry, degassed
CH2Cl2 (3.0 mL) in a 25 mL Schlenk flask with a glass joint and glass
stopper (after ∼1-2 days, septa were degraded by NO). After the
solution was stirred for 3 d, the headspace gas and solvent were vacuum-
transferred into a Schlenk flask with a rubber septum. The amount of
N2O produced was determined by injecting 10 µL aliquots of headspace
gas above the solvent onto a GC column and then comparing the area
of the N2O peak to that of a standard curve. The standard curve was
created by adding CH2Cl2 (3.0 mL) to the same Schlenk flask and then
vacuum-transferring in NO(g) (49 mL) plus 0, 1, or 2 mL of pure N2O.
Samples of the headspace gas (10 µL volume) were injected onto the
GC column, and a plot of the moles of N2O in the flask versus the
area of the N2O peak was used to generate the standard curve. For the
3
3
TpMs,HCu(NO). TpMs,HCu(THF) (0.078 g, 0.111 mmol) was dis-
solved in toluene (5 mL) in a Schlenk flask, and the flask was purged
with NO(g). The solution immediately turned dark orange, and an
orange precipitate formed. The mixture was filtered, and the solid was
briefly dried under a stream of N2, yielding TpMs,HCu(NO) as a powder
(0.060 g, 82% crude yield). Crystalline material was obtained in lower
yield by reacting a concentrated THF/pentane solution of TpMs,H
-
Cu(THF) (0.150 g, 0.211 mmol) with NO, followed by cooling to 193
K (0.054 g, 34% crude yield). Solid TpMs,HCu(NO) was stored and
handled in the glovebox. The complex was unstable as a solid and
decomposed when dried completely either with N2 or by exposure to
a vacuum, as evidenced by bleaching of its color. As a result, attempts
to obtain satisfactory elemental analyses were unsuccessful. The
analogue TpMs,HCu15NO was prepared as a solid in a similar fashion,
except 15NO was vacuum-transferred into the reaction flask. Solutions
containing TpMs,HCu(NO) were prepared by purging NO(g) over known
concentrations of TpMs,HCu(THF) in the appropriate dry, degassed
solvent and were used immediately or stored at T < 253 K for less
than 1 d. Clean solutions of TpMs,HCu(NO) could not be prepared by
dissolving crystals or powders of the complex due to the instability of
the complex in solution in the absence of NO as well as the low
solubility of the solid once it was precipitated from solution. Experi-
mental data: 1H NMR (300 MHz, THF-d8, 238 K) δ 2.56 (s, 18 H),
2.72 (s, 9 H), 6.47 (s, 3 H), 7.12 (s, 6 H), 12.64 (s, 3 H) ppm; FTIR
(KBr) 2975, 2948, 2921, 2472 [ν(BH)], 1711 [ν(15NO) ) 1682], 1484,
1352, 1186, 1167, 1068, 1049, 780, 744, 708 cm-1; UV-vis [THF,
nm (ꢀ, M-1 cm-1)] 344 (sh, 2800), 472 (1200); EPR (1:1 CH2Cl2/
TpCF ,CH system, the yield of N2O measured by this method based on
moles of copper was 60%. In a typical experiment for TpMs,H, a solution
of TpMs,HCu(THF) (30 mg, 0.043 mmol) in THF (3.0 mL) was injected
into a 25 mL Schlenk flask which was purged with NO(g) and the
solution was stirred for 18 h (a rubber septum could be used due to the
3
3
shorter exposure time). The amount of N2O produced was determined
Ms,H
by GC in the same manner as described for TpCF ,CH . For the Tp
system, the yield of N2O measured by this method based on moles of
copper was 97%.
3
3
Kinetics Experiments for the Disproportionation Reactions of
TpR,R′Cu(NO). In a typical experiment, TpCF ,CH Cu(CH3CN) (10 mg,
0.012 mmol) dissolved in CH2Cl2 (4 mL) was transferred to a 10 mL
test tube equipped with a stirbar, a UV cell side arm, and a Teflon
stopcock (total volume of the empty UV flask was 45 mL). The
solution was then frozen in liquid N2, and the tube was evacuated.
Purified NO gas (81 mL, 1.97 atm) was then transferred into the flask.
The solution was quickly warmed to 297(1) K and placed in a constant-
temperature bath for ∼5 min to equilibrate prior to beginning data
acquisition. Solutions of TpMs,HCu(NO) in THF were prepared in a
similar fashion. The electronic spectra were measured every 900 or
3
3
toluene, 9.44 GHz, 4.2 K < T < 297 K) silent.
CF ,CH
TpCF ,CH Cu(NO2). Tp
3
3
3
3
Cu(CH3CN) (0.088 g, 0.156 mmol) was
dissolved under N2 in dry, degassed CH2Cl2 (5 mL). Purified NO (1
atm) was vacuum-transferred into the reaction flask. After 1 week of
stirring, the color changed from yellow to green and a precipitate
formed. The NO atmosphere was replaced with N2, the mixture was
filtered, and the collected solid was washed with Et2O to afford the
product as a green microcrystalline material (0.055 g, 62%). The crystal
used for X-ray crystallography was prepared by allowing a THF solution
300 s for TpCF ,CH or TpMs,H, respectively, with vigorous stirring of
the solution in the bath between each data point collection. For the
3
3
TpCF ,CH system, the data collection was ended after about 5-6 h
(∼1-3 half-lives) because the final product precipitated out of solution.
3
3
(∼3.5 mL) containing TpCF ,CH Cu(NO) [prepared from 0.054 g of
The infinity readings for TpCF ,CH were determined using the Kezdy-
Swinbourne treatment.33 For the TpMs,H system, data were collected
for 4-5 half-lives, and an infinity spectrum was collected after 8-10
half-lives. Slight adjustments (<5%) of the infinity spectrum gave
improved first-order fits. To correct for slight variations in the baseline,
the data sets were adjusted by setting the isosbestic point to zero
absorbance for all scans. The changes in absorbance at 436 and 768
3
3
3
3
TpCF ,CH Cu(CH3CN)] to decompose for 2 d. The solvent and excess
NO(g) were removed under reduced pressure. The green residue was
dissolved in boiling THF (∼4 mL), and Et2O was diffused into this
solution at 278 K to give a few green crystals. Experimental data:
FTIR (KBr) 3142, 2572 [ν(BH)], 1466, 1356, 1265, 1203, 1172, 1146,
1131, 1114, 1067, 1010, 811, 800, 646 cm-1; UV-vis [THF, nm (ꢀ,
M-1 cm-1)] 296 (3000) 768 (90); EPR (1:1 THF/toluene, 9.44 GHz,
77 K) g| ) 2.31, g ) 2.06, A|Cu ) 148 G, A N ) 14 G. Anal. Calcd
for C15H13N7CuBF9O2: C, 31.68; H, 2.3; N, 17.24. Found: C, 31.82;
H, 2.24; N, 17.08.
3
3
nm for TpCF ,CH or 472 and 778 nm for TpMs,H versus time were fit to
first-order plots. Kinetics experiments to determine the [NO] depen-
dence of the reaction were conducted using a 4.78 mM stock solution
of TpMs,HCu(THF) in THF. These experiments were performed as
mentioned above, except the NO concentration in solution was varied
by vacuum-transferring in different amounts of NO to obtain final
pressures of 0.67, 0.97, 1.50, and 1.97 atm.
3
3
TpMs,HCu(NO2). This complex was isolated in quantitative yield
from the reaction of NO with TpMs,HCu(THF) (0.100 g, 0.142 mmol)
in THF (4 mL) for 18 h, followed by removal of excess NO and solvent
under vacuum. In an independent synthesis, a solution of TlTpMs,H
(0.172 g, 0.223 mmol) in THF (10 mL) was stirred with admixed
solutions of CuCl2 (0.090 g, 0.669 mmol) in MeOH (1 mL) and NaNO2
(0.092 g, 1.34 mmol) in MeOH (2 mL) for 30 min. Solvent was
removed under reduced pressure, the solid was extracted with CH2Cl2
(10 mL), and the extract was filtered. Methanol (2-3 mL) was added
to the solution, and the volume was reduced slowly under vacuum until
a light green solid precipitated. The product was collected by filtration
and air-dried (0.125 g, 83%). The 15NO2 compound was prepared by
the same procedure, using Na15NO2 instead of NaNO2. Experimental
Ms,H
X-ray Crystallography. Crystals of TpCF ,CH Cu(CH3CN), Tp
-
3
3
Cu(THF), and TpCF ,CH Cu(NO2) were attached to glass fibers and
mounted on an Enraf-Nonius CAD4 [Cu(I) complexes] or Siemens
SMART system [Cu(II)-nitrite complex] for data collection. For the
structures determined using the CAD4 instrument, cell constants were
determined from a set of 25 reflections determined from a random
search routine. For the structure determined using the SMART system,
3
3
(33) Espensen, J. H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill: New York, 1981, pp. 24-30.

11418 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Schneider et al.
an initial set of cell constants was calculated from reflections harvested
from three sets of 20 frames oriented such that orthogonal wedges of
reciprocal space were surveyed; orientation matrices were determined
from 85 reflections. Final cell constants were calculated from a set of
4008 strong reflections from the actual data collection. Space groups
for all the structures were determined on the basis of systematic
absences and intensity statistics, and successful direct-methods solutions
were calculated which provided most non-hydrogen atoms. The
remaining non-hydrogen atoms were located by several least-squares/
difference Fourier cycles. All non-hydrogen atoms were refined with
anisotropic displacement parameters unless stated otherwise. All
hydrogen atoms were placed in ideal positions and refined as riding
atoms with relative isotropic displacement parameters. All calculations
were performed with an SGI INDY R4400-SC or a Pentium computer
using the SHELXTL V5.0 suite of programs.34 Crystallographic data
are listed in Table 1, selected bond lengths and angles are provided in
Table 2, and full descriptions of each structure are provided as
Supporting Information.
For TpMs,HCu(THF), the THF molecule was disordered over three
positions defined by the 3-fold Cu-B axis. The THF oxygen atom is
offset from the 3-fold axis only slightly, causing unreasonable thermal
parameters. Therefore, this oxygen atom was refined isotropically and
the THF molecule was restrained using SAME.
For TpCF ,CH Cu(NO2), the technique of hemisphere collection was
used in which a randomly oriented region of reciprocal space was
surveyed to the extent of 1.3 hemispheres to a resolution of 0.84 Å.
Three major swaths of frames were collected with 0.30° steps in ω.
3
3
Acknowledgment. We thank the NIH (Grant GM47365),
the NSF (National Young Investigator Award to W.B.T.), and
the Alfred P. Sloan and Camille and Henry Dreyfus Foundations
(fellowships to W.B.T.) for financial support for this research.
We also thank M. Keyes for preliminary work on the X-ray
crystal structure of TpCF ,CH Cu(CH3CN), C. Buss for help with
the X-ray structures, and the reviewers for some excellent
suggestions.
3
3
For TpCF ,CH Cu(CH3CN), each of the disordered CF3 groups was
refined as two parts. The C(17) group was refined with each part half
occupied. The C(27) and C(37) CF3 groups refined as 0.64/0.36 and
0.40/0.60% occupied, respectively; SAME and DELU restraints were
used on the CF3 groups in order to maintain sensible bond lengths and
angles. The EADP and EXYZ constraints were used on carbon atoms
C(17), C(27), and C(37) to keep the positional and displacement
parameters of each part equivalent.
3
3
Supporting Information Available: Text and tables giving
full details and results of the X-ray crystal structure determina-
tions, additional structural diagrams, and the derivation of eqs
3-5 (36 pages, print/PDF). See any current masthead page
for ordering information and Web access instructions.
(34) SHELXTL-Plus V5.0; Siemens Industrial Automation, Inc.: Madison,
WI.
JA982172Q