Synthetic modeling of nitrite binding and activation by reduced copper proteins. Characterization of copper(I)-nitrite complexes that evolve nitric oxide

Full text of DOI:10.1021/ja952691i

J. Am. Chem. Soc. 1996, 118, 763-776
763
Synthetic Modeling of Nitrite Binding and Activation by
Reduced Copper Proteins. Characterization of
Copper(I)-Nitrite Complexes That Evolve Nitric Oxide
Jason A. Halfen, Samiran Mahapatra, Elizabeth C. Wilkinson,
Alan J. Gengenbach, Victor G. Young, Jr., Lawrence Que, Jr., and
William B. Tolman*
Contribution from the Department of Chemistry, UniVersity of Minnesota,
207 Pleasant Street S. E., Minneapolis, Minnesota 55455
ReceiVed August 8, 1995X
Abstract: In an effort to provide precedence for postulated intermediates in copper-protein-mediated nitrite reduction,
a series of novel complexes containing the CuI-NO2- unit, including monocopper(I), dicopper(I,I), and mixed valence
dicopper(I,II) and copper(I)-zinc(II) species, were prepared, fully characterized, and subjected to reactivity studies
designed to probe their ability to produce nitric oxide. Treatment of solutions of [LCu(CH3CN)]PF6 (L ) Li-Pr
,
3
1,4,7-triisopropyl-1,4,7-triazacyclononane, or LBn , 1,4,7-tribenzyl-1,4,7-triazacyclononane) in MeOH with excess
3
NaNO2 yielded the novel dicopper(I,I) complexes [(LCu)2(µ-NO2)]PF6. The complex with L ) Li-Pr was cleaved
3
i-Pr3
by PPh3 to afford [Li-Pr Cu(PPh3)]PF6 and L Cu(NO2), a structural model for the substrate adduct of copper nitrite
3
reductase. Oxidation of the dicopper(I,I) compound (L ) Li-Pr ) with (Cp2Fe)(PF6) in CH2Cl2 yielded the deep red,
3
mixed valent, dicopper(I,II) species [(Li-Pr Cu)2(µ-NO2)](PF6)2, which was structurally characterized as its [B(3,5-
3
(CF3)2C6H3)4]- salt (crystal data: triclinic space group P1h, a ) 13.439(8) Å, b ) 13.777(5) Å, c ) 14.471(8) Å, R
) 108.22(4)°, â ) 92.08(5)°, γ ) 90.08(4)°, Z ) 1, T ) 177 K, R ) 0.074, and Rw ) 0.070). A diamagnetic
i-Pr3
i-Pr3
heterodinuclear CuIZnII analog, [Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2, was assembled by mixing L Cu(NO2), Zn(O3-
3
I
II
SCF3)2, and Li-Pr and was shown to adopt a structure similar to that of its Cu Cu relative (crystal data: monoclinic
space group P21/c, a ) 10.8752(1) Å, b ) 15.6121(3) Å, c ) 25.8020(5) Å, â ) 90.094(1)°, Z ) 4, R1 ) 0.0472,
and wR2 ) 0.1082). Both compounds exhibit an intense electronic absorption feature that was assigned as a CuI f
3
-
NO2 MLCT transition on the basis of resonance Raman spectroscopic results. Functional modeling of copper
nitrite reductase was accomplished by treating solutions of Li-Pr Cu(NO2) with protonic acids or Me3SiO3SCF3. Nitric
3
i-Pr3
oxide evolution was accompanied by the formation of Li-Pr Cu(O2CCH3)2 and L Cu(O3SCF3)2 when acetic acid or
3
Me3SiO3SCF3 was used. The latter crystallized as a water adduct [Li-Pr Cu(H2O)(O3SCF3)](O3SCF3) (crystal data:
3
monoclinic space group P21/c, a ) 8.59(1) Å, b ) 26.04(1) Å, c ) 12.838(4) Å, â ) 108.26(6)°, Z ) 4, T ) 173
K, R ) 0.067, and Rw ) 0.064). The involvement of the CuICuII species as an intermediate in the reaction of
i-Pr3
Li-Pr Cu(NO2) with Me3SiO3SCF3 at low temperature and a mechanism for NO generation involving both L Cu-
3
(NO2) and [(Li-Pr Cu)2(µ-NO2)] are discussed.
2+
3
Denitrification, the dissimilatory, enzymatic reduction of
nitrate (NO3-) and nitrite (NO2-) to gaseous NO, N2O, and/or
N2, is a key process in the global nitrogen cycle whereby simple
organisms utilize the oxidizing equivalents of nitrogen oxides
to drive metabolic processes.1 The environmental consequences
of denitrification are profound, as it is responsible for the
degradation of the agriculturally useful yet polluting nitrogen
oxides NO3 and NO2 and for the generation of the reactive
gaseous molecules NO and N2O that are implicated in ozone
depletion, atmospheric warming, and air pollution.2 These
environmental issues, coupled with the recently realized im-
portance of nitrogen oxide chemistry in mammalian systems,3
have provided considerable stimulus for research into the
properties of the enzymes responsible for NOx reduction.
The copper-containing nitrite reductases (NiRs) isolated from
bacteria1,4 and, more recently, from fungi5 comprise an important
subclass of the set of denitrification enzymes. Copper NiRs
catalyze the reduction of NO2- to NO, although N2O generation
has been induced under some conditions.6 The active site of
the copper NiR from Achromobacter cycloclastes has been
shown by X-ray crystallography4,7,8 to contain a pair of copper
ions (Figure 1), one of which (Cu-I) has been assigned as a
“green” type 1, electron transfer site on the basis of its typical
(His)2(Cys)(Met) donor set and its combined electronic absorp-
tion [λmax 458 (ꢀ 2530), 585 (ꢀ 1890) nm], EPR (axial, A| ) 69
× 10-4 cm-1), and resonance Raman [multiple Cu-S(Cys)
modes between 250 and 500 cm-1] spectroscopic properties.9,10
-
-
(4) Adman, E. T.; Turley, S. K. In Bioinorganic Chemistry of Copper;
Karlin, K. D., Tyekla´r, Z., Eds.; Chapman & Hall, Inc.: New York, 1993;
pp 397-405.
(5) Kobayashi, M.; Shoun, H. J. Biol. Chem. 1995, 270, 4146-4151.
(6) Jackson, M. A.; Tiedje, J. M.; Averill, B. A. FEBS Lett. 1991, 291,
41-44.
(7) Godden, J. W.; Turley, S.; Teller, D. C.; Adman, E. T.; Liu, M. Y.;
Payne, W. J.; LeGall, J. Science 1991, 253, 438-442.
(8) A similar structure has been deduced for the enzyme from Alcaligenes
faecalis S-6: Kukimoto, M.; Nishiyama, M.; Murphy, M. E. P.; Turley,
S.; Adman, E. T.; Horinouchi, S.; Beppu, T. Biochemistry 1994, 33, 5346-
5252.
* Author to whom correspondence should be addressed.
X Abstract published in AdVance ACS Abstracts, January 1, 1996.
(1) (a) Kroneck, P. M. H.; Beuerle, J.; Schumacher, W. In Degradation
of EnVironmental Pollutants by Microorganisms and their Metalloenzymes;
Sigel, H., Sigel, A., Eds.; Marcel Dekker: New York, 1992; Vol. 28, pp
455-505.
(2) Denitrification, Nitrification, and Atmospheric Nitrous Oxide; Del-
wiche, C. C., Ed.; John Wiley & Sons: New York, 1981.
(3) The Biology of Nitric Oxide; Moncada, S., Marletta, M. A., Hibbs,
J. B., Higgs, E. A., Eds.; Portland Press: London, 1992; Vols. 1 and 2.
0002-7863/96/1518-0763$12.00/0 © 1996 American Chemical Society

764 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
Although reasonable, the sequence depicted in Figure 2 had
little firm precedent in copper coordination chemistry at the
outset of the present work.16,17 In particular, while several
mononuclear CuII-NO2- complexes with biomimetic ancillary
ligands had been described,17b,18 there existed no example of a
CuI-NO2- complex in the literature. In view of its potentially
central role in the enzymatic nitrite reduction pathway, we
targeted such a species for synthetic, structural, spectroscopic,
and reactivity studies. Controversy surrounding the existence
-
of CuI-NO2 interactions in copper proteins other than NiR
provided further rationale for synthetic modeling,19 putative
nitrite adducts of hemocyanin being noteworthy in this regard.20
Herein we report details of the synthesis and full characterization
of a series of novel biomimetic complexes containing the CuI-
Figure 1. Chem3D view of the active site of nitrite reductase (NiR)
from Achromobacter cycloclastes (generated from pdb 1afn.full).
-
NO2 moiety supported by sterically hindered triazacy-
clononanes, including monocopper(I), dicopper(I,I), and mixed
valence dicopper(I,II) and copper(I)-zinc(II) species. We also
report reactivity and mechanistic studies that demonstrate the
feasibility of key aspects of the pathway shown in Figure 2.
Portions of this work were communicated previously.21
Figure 2. Proposed mechanism for nitrite reduction by NiR.
The other site (Cu-II), which is linked to Cu-I via a dipeptide
fragment (Cys136-His135; Cu-I‚‚‚Cu-II ) 12.5 Å), is an
unusual, distorted type 2 site, in which the copper ion is bound
at the interface between subunits of the trimeric protein by a
facial array of three histidine residues. A fourth ligand, refined
as a water molecule in the crystallographic analysis,4,7 completes
the pseudotetrahedral environment about this copper site.
On the basis of the crystallographic analysis of nitrite-soaked
NiR,4,7 activity studies on type 2 depleted enzyme,11 EPR and
ENDOR data for the related copper NiR from Alcaligenes
xylosoxidans,12 effects of site-directed mutagenesis of copper
ligands,8 and pulse radiolysis experiments interpreted to support
intramolecular electron transfer,13 Cu-II has been suggested to
be the locus for nitrite binding and subsequent activation, with
Cu-I acting to shuttle electrons, perhaps via the Cys-His bridge,
during catalysis. Drawing an analogy to mechanisms deduced
for the more thoroughly studied iron heme-containing NiRs for
which numerous synthetic models exist,14 Averill and co-
workers proposed that a nitrite adduct to Cu-II in its reduced
state [i.e., a CuI-NO2- complex] is a key reaction intermediate
(Figure 2).6,11,15 This intermediate may be envisioned to form
via reduction of a CuII-NO2- precursor or directly, via substrate
binding to the prereduced CuI site. Subsequent dehydration to
yield an electrophilic copper-nitrosyl (originally written as
CuI-NO+)14a followed by release of NO, the principal enzyme
product, is then postulated to generate the oxidized Cu-II center
ready for reentry into the catalytic process.
Results and Discussion
Syntheses. Reference to the crystallographic data for copper
NiR suggested that use of a ligand(s) that would provide three
nitrogen donor atoms disposed in a facially-coordinating array
and that would enforce a pseudotetrahedral copper ion geometry
would be a prerequisite for accurate modeling of the Cu-II
catalytic site. Our previous successful preparations of copper-
(II)-nitrite17b and copper(I)-nitrosyl17a adducts supported by
sterically hindered tris(pyrazolyl) hydroborates (TpRR′, where
R and R′ are substituents at the 3- and 5-pyrazolyl positions,
respectively)22 suggested that such ligands might also allow the
synthesis of the desired CuI-NO2- species. This strategy was
not successful, however, despite numerous attempts either to
(16) (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.
(17) (a) 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. (b) Ruggiero, C. E.; Carrier, S. M.; Tolman, W. B. Angew. Chem.,
Int. Ed. Engl. 1993, 33, 895-897.
(18) (a) Jiang, F.; Conry, R. R.; Bubacco, L.; Tyekla´r, Z.; Jacobson, R.
R.; Karlin, K. D.; Peisach, J. J. Am. Chem. Soc. 1993, 115, 2093-2102.
(b) Tolman, W. B. Inorg. Chem. 1991, 30, 4878-4880 and references cited
therein. (c) Komeda, N.; Nagao, H.; Adachi, G.; Suzuki, M.; Uehara, A.;
Tanaka, K. Chem. Lett. 1993, 1521-1524. (d) Komeda, N.; Nagao, H.;
Kushi, Y.; Adachi, G.; Suzuki, M.; Uehara, A.; Tanaka, K. Bull. Chem.
Soc. Jpn. 1995, 68, 581-589.
(19) (a) Spira, D. J.; Solomon, E. I. Biochem. Biophys. Res. Commun.
1983, 112, 729-736. (b) Schoot Uiterkamp, A. J. M. S.; van der Deen,
H.; Berendsen, H. C. J.; Boas, J. F. Biochim. Biophys. Acta 1974, 372,
407-425.
(20) (a) van der Deen, H.; Hoving, H. Biochemistry 1977, 16, 3519-
3525. (b) Himmelwright, R. S.; Eickman, N. C.; Solomon, E. I. Biochem.
Biophys. Res. Commun. 1978, 81, 237-242. (c) Himmelwright, R. S.;
Eickman, N. C.; Solomon, E. I. Biochem. Biophys. Res. Commun. 1979,
86, 628-634. (d) Himmelwright, R. S.; Eickman, N. C.; Solomon, E. I. J.
Am. Chem. Soc. 1979, 101, 1576-1586. (e) Verplaetse, J.; Tornout, P.
V.; Degreyn, G.; Witters, R.; Lontie, R. Eur. J. Biochem. 1979, 95, 327-
331. (f) Tahon, J.-P.; Hoof, D. V.; Vinckier, C.; Witters, R.; de Ley, M.;
Lontie, R. Biochem. J. 1988, 249, 891-896. (g) Salvato, B.; Giacometti,
G. M.; Beltramini, M.; Zilio, F.; Giacometti, G.; Magliozzo, R. S.; Peisach,
J. Biochemistry 1989, 28, 680-84. (h) Magliozzo, R. S.; Bubacco, L.;
McCracken, J.; Jiang, F.; Beltramini, M.; Salvato, B.; Peisach, J. Biochem-
istry 1995, 34, 1513-1523. (i) Bubacco, L.; Magliozzo, R. S.; Wirt, M.
D.; Beltramini, M.; Salvato, B.; Peisach, J. Biochemistry 1995, 34, 1524-
1533.
(21) (a) Halfen, J. A.; Mahapatra, S.; Olmstead, M. M.; Tolman, W. B.
J. Am. Chem. Soc. 1994, 116, 2173-2174. (b) Halfen, J. A.; Tolman, W.
B. J. Am. Chem. Soc. 1994, 116, 5475-5476.
(22) (a) Trofimenko, S. Chem. ReV. 1993, 93, 943-980. (b) Kitajima,
N.; Tolman, W. B. Prog. Inorg. Chem. 1995, 43, 419-531. (c) Parkin, G.
AdV. Inorg. Chem. 1995, 42, 291-393.
(9) (a) Lu, Y.; LaCroix, L. B.; Lowery, M. D.; Solomon, E. I.; Bender,
C. J.; Peisach, J.; Roe, J. A.; Gralla, E. B.; Valentine, J. S. J. Am. Chem.
Soc. 1993, 115, 5907-5918 and references cited therein. (b) Han, J.; Loehr,
T. M.; Lu, Y.; Valentine, J. S.; Averill, B. A.; Sanders-Loehr, J. J. Am.
Chem. Soc. 1993, 115, 4256-4263. (c) Andrew, C. R.; Yeom, H.;
Valentine, J. S.; Karlsson, B. G.; Bonander, N.; van Pouderoyen, G.; Canters,
G. W.; Loehr, T. M.; Sanders-Loehr, J. J. Am. Chem. Soc. 1994, 116,
11489-11498.
(10) Abraham, Z. H. L.; Lowe, D. J.; Smith, B. E. Biochem. J. 1993,
295, 587-593.
(11) Libby, E.; Averill, B. A. Biochem. Biophys. Res. Commun. 1992,
187, 1529-1535.
(12) Howes, B. D.; Abraham, Z. H. L.; Lowe, D. J.; Bruser, T.; Eady,
R. R.; Smith, B. E. Biochemistry 1994, 33, 3171-3177.
(13) Suzuki, S.; Kohzuma, T.; Deligeer; Yamaguchi, K.; Nakamura, N.;
Shidara, S.; Kobayashi, K.; Tagawa, S. J. Am. Chem. Soc. 1994, 116,
11145-11146.
(14) (a) Weeg-Aerssens, E.; Tiedje, J. M.; Averill, B. A. J. Am. Chem.
Soc. 1988, 110, 6851-6856. (b) Garber, E. A. E.; Hollocher, T. C. J. Biol.
Chem. 1982, 257, 8091-8097. (c) Fanning, J. C. Coord. Chem. ReV. 1991,
110, 235-273 and references cited therein.
(15) (a) Hulse, C. L.; Averill, B. A.; Tiedje, J. M. J. Am. Chem. Soc.
1989, 111, 2322-2323. (b) Ye, R. W.; Toro-Suarez, I.; Tiedje, J. M.;
Averill, B. A. J. Biol. Chem. 1991, 266, 12848-12851.

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 765
Scheme 1
add nitrite salts to TpRR′CuI compounds23 or to reduce preformed
analysis of the reaction head space) and gradual development
of a green, presumably copper(II)-containing solution was
observed. In contrast, when excess NaNO2 was added to
solutions of [LCu(CH3CN)]PF6 in MeOH, the dicopper(I,I)
-
TpRR′CuII-NO2 complexes.17b,18b Reasoning that nitrite co-
ordination to a TpRR′CuI fragment might be disfavored in the
nonpolar solvents most amenable for the handling of hindered
TpRR′ complexes because of charge considerations (TpRR′ is a
monoanion, making the copper(I) fragment neutral and a derived
nitrite adduct anionic), we turned to the sterically hindered,
neutral macrocycles 1,4,7-triisopropyl-1,4,7-triazacyclononane
Bn3
complexes [(LCu)2(µ-NO2)]PF6 (L ) Li-Pr or L ) were isolated
as a yellow crystalline solids in high yield (Scheme 1). These
products were characterized by analytical, spectroscopic, and,
3
for the complex containing Li-Pr , X-ray crystallographic meth-
3
24
25
3
(Li-Pr
)
and 1,4,7-tribenzyl-1,4,7-triazacyclononane (LBn
)
ods. No evidence for the formation of mononuclear complexes
3
recently reported by Wieghardt and co-workers. The trialkyl
substituents borne by these molecules render them and their
complexes soluble in many organic solvents and prevent
deleterious bis(ligand) complex (L2Cu) formation, as seen with
the unsubstituted 1,4,7-triazacyclononane.26 Although a variety
of copper(II) complexes have been prepared with triazacy-
clononanes,27 few examples of copper(I) complexes of these
ligands existed at the outset of our work.28
was found (1H NMR), even though excess NaNO2 was used in
the synthesis. Nonetheless, regioselective cleavage of [(Li-Pr
3
-
Cu)2(µ-NO2)]PF6 was effected by adding 1 equiv of PPh3,
i-Pr3
affording analytically pure [Li-Pr Cu(PPh3)]PF6 and L Cu(NO2)
3
in 97% and 63% yields, respectively, after fractional crystal-
lization. Identification of Li-Pr Cu(NO2) as the first example of
3
a mononuclear copper(I)-nitrite complex was confirmed by
X-ray crystallography. Coproduct [Li-Pr Cu(PPh3)]PF6 was
3
Bn3
Stirring Li-Pr or L with [Cu(CH3CN)4]PF6 in THF gener-
3
identified by comparison of its spectroscopic properties to
material synthesized independently via the reaction of [Li-Pr
Cu(CH3CN)]PF6 with PPh3.
3
-
ated the highly air and water sensitive copper(I) complexes
Bn3
[LCu(CH3CN)]PF6 (L ) Li-Pr or L ), which were isolated as
solids and characterized by elemental analysis and FTIR and
NMR spectroscopy (Scheme 1). Because of concerns that protic
solvents might react with nitrite adducts of these starting
materials, we initially attempted to synthesize such species by
treating [LCu(CH3CN)]PF6 with soluble (PPN)(NO2)29 in CH3-
CN. However, only slow evolution of NO (1 equiv by GC
3
The observation of a chemically reversible one-electron
oxidation in the cyclic voltammogram of [(Li-Pr Cu)2(µ-NO2)]PF6
3
(E1/2 ) 0.07 V vs SCE in CH2Cl2; see below) indicated that
isolation of a mixed valence nitrite-containing complex would
be possible. Indeed, addition of (Cp2Fe)(PF6) to [(Li-Pr Cu)2-
3
(µ-NO2)]PF6 in CH2Cl2 afforded the deep red crystalline
dicopper(I,II) complex [(Li-Pr Cu)2(µ-NO2)](PF6)2 (Scheme 1),
3
(23) Carrier, S. M.; Ruggiero, C. E.; Houser, R. P.; Tolman, W. B. Inorg.
Chem. 1993, 32, 4889-4899 and references cited therein.
(24) Haselhorst, G.; Stoetzel, S.; Strassburger, A.; Walz, W.; Wieghardt,
K.; Nuber, B. J. Chem. Soc., Dalton Trans. 1993, 83-90.
(25) Beissel, T.; Vedova, B. S. P. C. D.; Wieghardt, K.; Boese, R. Inorg.
Chem. 1990, 29, 1736-1741.
which was characterized by analytical and spectroscopic meth-
ods. We were unable to elucidate the structure of this compound
or its CF3SO3- salt via X-ray crystallography due to extensive
disorder problems, and attempts to obtain the material with
-
BPh4 as counterion failed because of decomposition of the
(26) Nonoyama, M. Trans. Met. Chem. 1976, 1, 70-74.
(27) (a) Wieghardt, K.; Chaudhuri, P. Prog. Inorg. Chem. 1988, 35, 329-
436 and references cited therein. (b) Chaudhuri, P.; Oder, K. J. Chem.
Soc., Dalton Trans. 1990, 1597-1605. (c) Chaudhuri, P.; Winter, M.;
Vedova, B. P. C. D.; Bill, E.; Trautwein, A.; Gehring, S.; Fleischhauer, P.;
Nuber, B.; Weiss, J. Inorg. Chem. 1991, 30, 2148-2157. (d) Chaudhuri,
P.; Karpenstein, I.; Winter, M.; Butzlaff, C.; Bill, E.; Trautwein, A. X.;
Florke, U.; Haupt, H.-J. J. Chem. Soc., Chem. Commun. 1992, 321-22.
(e) Chaudhuri, P.; Karpenstein, I.; Winter, M.; Lengen, M.; Butzlaff, C.;
Bill, E.; Trautwein, A. X.; Florke, U.; Haupt, H. Inorg. Chem. 1993, 32,
88-894.
complex via an apparent redox reaction(s) [bleaching of color
indicative of the formation of copper(I) species]. However,
metathesis with the more redox inert Na[B(3,5-(CF3)2C6H3)4]30
in CH2Cl2 was successful and the resulting crystalline product
proved to be amenable to crystallographic analysis. The
Bn3
dicopper(I,I) complex of LBn , [(L Cu)2(µ-NO2)]PF6, was less
3
well-behaved electrochemically (irreversible oxidation at Epa
)
0.41 V vs SCE in CH2Cl2), and attempts to prepare a mixed
valuence complex via oxidation with (Cp2Fe)(PF6) or (NO)-
(28) Chaudhuri, P.; Oder, K. J. Organomet. Chem. 1989, 367, 249-
258.
(29) Stevens, R. E.; Gladfelter, W. L. Inorg. Chem. 1983, 22, 2034-
2042.
(30) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992,
11, 3920-3922.

766 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
(PF6) failed. Parenthetically, we note that reactivity differences
arising from the divergent steric and/or electronic properties of
Li-Pr and LBn have also arisen in ongoing studies of the
interactions of [LCu(CH3CN)]X (X ) PF6 or other anions) with
O2.31
3
3
Issues raised during mechanistic studies of NO evolution from
Li-Pr Cu(NO2) (see below) raised the question of whether the
3
2+
mixed-valent dimer [(Li-Pr Cu)2(µ-NO2)] could be accessed
3
by treatment of mononuclear Li-Pr Cu(NO2) with a L Cu
i-Pr3
II
3
fragment that contained labile counterion or solvent coligands.
Such a precursor, Li-Pr Cu(O3SCF3)2, was prepared by admixture
3
of Li-Pr and Cu(O3SCF3)2 in dry THF. Recrystallization from
3
wet solvents or performing the reaction in wet acetone yielded
the aquo complex, [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3, which was
3
characterized by X-ray crystallography. Analogous facile
interconversions among Li-Pr Cu(O3SCF3)2 and mono- and
3
disolvated forms were supported by conductivity measurements,
which indicated that the complex is neutral in CH2Cl2 (i.e., both
CF3SO3- anions coordinated), a 1:1 electrolyte in acetone, and
a 2:1 electrolyte in CH3CN. Ready displacement of the
-
coordinated CF3SO3 anions in anhydrous Li-Pr Cu(O3SCF3)2
3
in CH2Cl2 by the exposed nitrite oxygen atom lone pairs in
Li-Pr Cu(NO2) also was possible, affording mixed valent
3
[(Li-Pr Cu)2(µ-NO2)](O3SCF3)2 in quantitative yield (UV-vis and
3
EPR spectroscopy; see below). Analogous synthetic approaches
to multimetal cyanide-bridged assemblies have been reported.32
A diamagnetic, heterodinuclear analog of the dicopper(I,II)
i-Pr3
species, [Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2, was prepared simi-
3
larly, via admixture of Li-Pr , Zn(O3SCF3)2, and L Cu(NO2)
in CH2Cl2 (Scheme 1). As described in detail below, compari-
son of the properties of the copper(I)-zinc(II) complex with
those of the structurally analogous dicopper(I,II) compound
proved to be useful for assignment of key spectroscopic features
arising from their respective nitrite-bridged cores.
i-Pr3
3
Figure 3. Representations of the X-ray crystal structures of (a) the
i-Pr
cation in [(Li-Pr Cu)2(µ-NO2)]PF6 and (b) L Cu(NO2), showing all
3
3
nonhydrogen atoms as 50% ellipsoids.
or absence of an additional nitrite O-bound metal ion and the
corresponding nitrite binding mode. As in other triazacy-
clononane complexes of copper(I),28 the coordination geometries
of the CuI ions in all four molecules are best described as C3V-
distorted tetrahedral, the principal distortions from idealized
Finally, a monomeric copper(II)-acetate complex, Li-Pr Cu-
3
(O2CCH3)2, was prepared by mixing Li-Pr and cupric acetate in
3
MeOH. This hygroscopic complex was characterized by
spectroscopy, conductivity measurements (neutral in CH3CN),
and X-ray crystallography, allowing its identification as a
tetrahedral local symmetry being (i) intraligand (Li-Pr ) N-Cu-N
3
bond angles significantly smaller than those between the Li-Pr
3
byproduct of the NO-evolving reaction of Li-Pr Cu(NO2) with
3
N atoms and the fourth nitrite donor [average NL-Cu-NL )
86.1° vs average NL-Cu-N(O)nitrite ) 127.7°] and (ii) similar
Cu-NL bond distances (average ) 2.13 Å) which are typical
for CuI-N bond lengths in 4-coordinate complexes (∼2.1 Å)23
but which are significantly longer than the Cu-Onitrite or Cu-
Nnitrite bonds.
acetic acid (see below).
X-ray Crystallography. The X-ray crystal structures of
21a
21b
[(Li-Pr Cu)2(µ-NO2)]PF6, Li-Pr Cu(NO2),
and Li-Pr Cu(O2-
3
3
3
CCH3)221b were reported previously; for reference, drawings are
provided in Figures 3 and 4 (see the supporting information
associated with the original references for the complete data).
Consideration of the metrical parameters of the mono- and
dicopper-nitrite units within each complex reveals distinct
features associated with the various nitrite binding modes. In
New crystallographic data for [(Li-Pr Cu)2(µ-NO2)][B(3,5-
3
(CF3)2C6H3)4]2, [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3, and [L Cu-
i-Pr3
3
(µ-NO2)ZnLi-Pr ](O3SCF3)2 are listed in Tables 1 and 2, with
3
3
the simplest species, Li-Pr Cu(NO2), nitrite is bound to a single
copper(I) ion solely via its N atom (“nitro” coordination),33 with
drawings shown in Figures 4-6. Complete listings of positional
parameters, temperature factors, and bond distances and angles
for these compounds are provided as supporting information.
a short Cu-Nnitrite bond [1.903(4) Å]. Transition metal M-Nni
-
trite bond lengths are usually somewhat longer,33,34 although
examples of short distances have been reported [cf. (TpivPP)-
Fe(NO2), with Fe-Nnitrite ) 1.849(6) Å].35 The short distance
i-Pr3
A. Nitrite Complexes [(Li-Pr Cu)2(µ-NO2)]PF6, L Cu-
3
(NO2), [(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2, and [L
-
i-Pr3
3
Cu(µ-NO2)ZnLi-Pr ](O3SCF3)2. The cores of these molecules
are grouped for comparison in Figure 6, with selected bond
distances and angles noted. In each complex a pseudotetrahedral
CuI ion is bonded to the N atom of a nitrite ion. The key
differentiating features among the compounds are the presence
in the latter species and in Li-Pr Cu(NO2) presumably reflects
3
3
an increased M-NO2 bond order arising from π-back-donation
from filled metal orbitals to the lowest energy nitrite π* orbital
(Vide infra). Significantly longer Cu-Nnitrite bonds (g2.0 Å)
have been reported for various salts of the Cu(II) species
x-
Cu(NO2)n (n ) 5, x ) 3; n ) 6, x ) 4), although disorders
(31) (a) Mahapatra, S.; Halfen, J. A.; Wilkinson, E. C.; Que, L., Jr.;
Tolman, W. B. J. Am. Chem. Soc. 1994, 116, 9785-9786. (b) Mahapatra,
S.; Halfen, J. A.; Wilkinson, E. C.; Pan, G.; Cramer, C. J.; Que, L., Jr.;
Tolman, W. B. J. Am. Chem. Soc. 1995, 117, 8865-8866.
(32) (a) Scott, M. J.; Holm, R. H. J. Am. Chem. Soc. 1994, 116, 11357-
11367. (b) Scott, M. J.; Lee, S. C.; Holm, R. H. Inorg. Chem. 1994, 33,
4651-4662.
(33) Hitchman, M. A.; Rowbottom, G. L. Coord. Chem. ReV. 1982, 42,
55-132.
(34) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A.
G.; Taylor, R. J. Chem. Soc., Perkin Trans. 1987, S1-S19.
(35) Nasri, H.; Wang, Y.; Huynh, B. H.; Scheidt, W. R. J. Am. Chem.
Soc. 1991, 113, 719-721.

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 767
Table 1. Selected Bond Lengths (Å) and Angles (deg) for
Complexes Characterized by X-ray Crystallographya
[Li-Pr Cu(H2O)(O3SCF3)]O3SCF3
3
Cu(1)-O(1)
2.014(5) Cu(1)-O(2)
2.046(5) Cu(1)-N(4)
2.180(5) S(1)-O(2)
1.438(5) S(1)-O(4)
2.052(4)
2.070(5)
1.472(4)
1.431(4)
Cu(1)-N(1)
Cu(1)-N(7)
S(1)-O(3)
O(1)-Cu(1)-O(2)
O(1)-Cu(1)-N(4)
O(2)-Cu(1)-N(1)
86.9(2)
94.9(2)
91.8(2)
O(1)-Cu(1)-N(1) 178.6(2)
O(1)-Cu(1)-N(7)
93.9(2)
O(2)-Cu(1)-N(4) 157.7(2)
O(2)-Cu(1)-N(7) 113.2(2)
N(1)-Cu(1)-N(4)
N(4)-Cu(1)-N(7)
O(2)-S(1)-O(4)
Cu(1)-O(2)-S(1)
86.3(2)
88.9(2)
112.8(3)
140.5(3)
N(1)-Cu(1)-N(7)
O(2)-S(1)-O(3)
O(3)-S(1)-O(4)
86.8(2)
114.0(3)
117.6(3)
[(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2
3
Cu(1′)-O(1)
Cu(1)-N(1)
2.047(9) Cu(1′)-O(2)
2.13(1)
2.069(5)
2.051(6)
1.30(1)
1.780(8) Cu(1)-N(2)
2.104(6) Cu(1)-N(4)
Cu(1)-N(3)
O(1)-N(1)
1.26(1) O(2)-N(1)
O(1)-Cu(1′)-O(2)
60.2(3)
O(1)-Cu(1′)-N(2′) 153.5(3)
O(1)-Cu(1′)-N(4′) 113.1(3)
O(2)-Cu(1′)-N(3′) 122.3(3)
N(1)-Cu(1)-N(2) 126.4(4)
N(1)-Cu(1)-N(4) 132.0(4)
O(1)-Cu(1′)-N(3′) 107.9(3)
O(2)-Cu(1′)-N(2′) 93.5(3)
O(2)-Cu(1′)-N(4′) 148.7(3)
N(1)-Cu(1)-N(3) 120.8(4)
N(2)-Cu(1)-N(3)
N(3)-Cu(1)-N(4)
Cu(1′)-O(2)-N(1)
88.8(2)
89.1(2)
92.3(7)
N(2)-Cu(1)-N(4)
Cu(1′)-O(1)-N(1)
86.9(2)
97.4(7)
Cu(1)-N(1)-O(2) 125.4(9)
Cu(1)-N(1)-O(1) 124.3(9)
O(1)-N(1)-O(2)
110.1(7)
i-Pr
[Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2
3
3
Zn(1)-N(5)
Zn(1)-N(6)
Zn(1)-N(7)
Cu(1)-N(2)
Cu(1)-N(4)
N(1)-O(2)
1.998(3) Zn(1)-O(1)
2.031(3)
2.039(3) Zn(1)-O(2)
2.044(3)
1.848(3)
2.050(3)
1.269(4)
99.43(11)
2.161(3) Cu(1)-N(1)
2.095(3) Cu(1)-N(3)
2.143(3) N(1)-O(1)
1.274(4) N(5)-Zn(1)-O(1)
N(5)-Zn(1)-N(6)
88.34(12) O(1)-Zn(1)-N(6) 156.05(11)
N(5)-Zn(1)-O(2) 161.71(11) O(1)-Zn(1)-O(2)
N(6)-Zn(1)-O(2) 108.30(11) N(5)-Zn(1)-N(7)
O(1)-Zn(1)-N(7) 113.84(11) N(6)-Zn(1)-N(7)
62.28(10)
88.64(12)
88.80(12)
O(2)-Zn(1)-N(7)
98.98(11) N(5)-Zn(1)-N(1) 130.49(11)
N(6)-Zn(1)-N(1) 136.39(11) N(7)-Zn(1)-N(1) 108.27(11)
N(1)-Cu(1)-N(3) 125.03(14) N(1)-Cu(1)-N(2) 135.57(14)
Figure 4. Representations of the X-ray crystal structures of (a) Li-Pr
-
3
Cu(O2CCH3)2 and (b) the cation in [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3,
showing all nonhydrogen atoms as 50% ellipsoids.
3
N(3)-Cu(1)-N(2)
N(3)-Cu(1)-N(4)
O(1)-N(1)-O(2)
87.9(2)
87.78(12) N(2)-Cu(1)-N(4)
111.9(3)
N(1)-Cu(1)-N(4) 120.20(12)
86.01(12)
93.2(2)
O(1)-N(1)-Cu(1) 122.5(2)
O(2)-N(1)-Cu(1) 125.3(2)
N(1)-O(1)-Zn(1)
involving possible static and/or dynamic Jahn-Teller distortions
complicate comparisons to crystallographic data obtained for
many of these compounds.36 Finally, as is typical for transition
N(1)-O(2)-Zn(1) 92.5(2)
a Estimated standard deviations are given in parentheses.
metal nitro complexes,33 in Li-Pr Cu(NO2), the nitrite O-N-O
3
nitrite bridges in which the anti (instead of the syn) lone pair
on the O atom is coordinated occur in the presence of additional
bridging ligands.39 In the only other example of a complex
containing a CuI-Onitrite bond, [(PPh3)2Cu(η2-O,O-NO2)],40 the
Cu-O distance [2.191(4) Å] is significantly longer than in
angle [116.0(2)°] and N-O distances [1.253(5) and 1.238(5)
Å] differ only slightly from those of the free nitrite ion [114.9-
(5)° and 1.240(3) Å, respectively],37 the approximate equiva-
lence of the distances being consistent with the expected lack
of bond alternation.
[(Li-Pr Cu)2(µ-NO2)]PF6 [1.968(2) Å], perhaps reflecting the
3
Increased N-O bond alternation is evident in the structure
contrasting η2 vs η1 binding modes and/or divergent interligand
steric influences. The Cu-Nnitrite bond length in the dinuclear
of [(Li-Pr Cu)2(µ-NO2)]PF6, in which a nitrite ion bridges two
3
copper(I) ions via its N atom and a syn lone pair on one O
atom to afford a planar CuI-(µ-NO2)-CuI array [N-O )
1.235(3) and 1.279(3) Å; ∆(N-O) ) 0.044 Å vs 0.015 for
complex, 1.899(2) Å, is identical to that in Li-Pr Cu(NO2) within
3
experimental error. Thus, the binding of an additional copper-
(I) ion to the O atom of the nitrite ligand appears to have a
negligible effect on the Cu-Nnitrite bonding, despite its influence
Li-Pr Cu(NO2)]. Weak or nonexistent bonding between O1 and
3
the metal ions is evident from the approximately C3V-symmetric
copper(I) geometries and the long O1-Cu distances [2.786(2)
and 2.820(2) Å to Cu1 and Cu2, respectively]. The µ-(η1-N:
η1-O)-nitrite coordination mode observed for the complex is
unusual in transition metal chemistry;38 related µ-(η1-N:η1-O)-
-
on the degree of NO2 N-O bond alternation.
(38) (a) Finney, A. J.; Hitchman, M. A.; Raston, C. L.; Rowbottom, G.
L.; White, A. H. Aust. J. Chem. 1981, 34, 2163. (b) Drew, M. G. B.;
Goodgame, D. M. L.; Hitchman, M. A.; Rogers, D. J. Chem. Soc., Chem.
Commun. 1965, 477. (c) Strouse, C. E.; Swanson, B. I. J. Chem. Soc.,
Chem. Commun. 1971, 55.
(36) (a) Ozarowski, A.; Allmann, R.; Pour-Ibrahim, A.; Reinen, R. Z.
Anorg. Allg. Chem. 1991, 592, 187-201. (b) Klein, S.; Reinen, D. J. Solid
State Chem. 1980, 32, 311-319. (c) Cullen, D. L.; Lingafelter, E. C. Inorg.
Chem. 1971, 10, 1264-1268. (d) Takagi, S.; Joesten, M. D.; Lenhert, P.
G. J. Am. Chem. Soc. 1975, 97, 444-445. (e) Takagi, S.; Joesten, M. D.;
Lenhert, P. G. Acta Crystallogr. 1975, B31, 596-598.
(39) (a) Johnson, B. F.; Sieker, A.; Blake, A. J.; Winpenny, R. E. P. J.
Chem. Soc., Chem. Commun. 1993, 1345-1346. (b) Thewalt, U.; Marsh,
R. E. Inorg. Chem. 1970, 9, 1604-1610. (c) Goodgame, D. M. L.;
Hitchman, M. A.; Marsham, D. F.; Phavanantha, P.; Rogers, D. J. Chem.
Soc., Chem. Commun. 1969, 1383-1384.
(40) Halfen, J. A.; Tolman, W. B. Acta Crystallogr. 1995, C51, 215-
217.
(37) Kay, M. I.; Frazer, B. C. Acta Crystallogr. 1961, B24, 56.

768 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
Table 2. Summary of X-ray Crystallographic Data
i-Pr
3
3
3
3
complex
emp form
fw
[Li-Pr Cu(H2O)(O3SCF3)]O3SCF3
[(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2
[Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2
C17H35CuF6N3O7S2
635.14
C94H90B2Cu2F48N7O2
2410.43
C32H66CuF6N7O8S2Zn
983.95
cryst syst
space group
a (Å)
b (Å)
c (Å)
monoclinic
P21/c
8.59(1)
26.04(1)
12.838(4)
triclinic
P1h
13.439(8)
13.777(5)
14.471(8)
108.22(4)
92.08(5)
monoclinic
P21/c
10.8752(1)
15.6121(3)
25.8020(5)
R (deg)
â (deg)
108.26(6)
90.094(1)
γ (deg)
90.08(4)
2543(4)
V (Å3)
2727(6)
4380.78(13)
Z
4
1
4
density (calcd)
temp (K)
1.547 g/cm3
173(2)
1.574 g/cm3
177(2)
1.492 g/cm3
173(2)
crystal size (mm)
diffractometer
radiation
0.57 × 0.47 × 0.31
Enraf-Nonius CAD-4
Mo KR (λ ) 0.71073 Å)
10.27 cm-1
56.0
8540
6704
4111 [I > 2σ(I)]
326
0.067
0.064
0.60 × 0.35 × 0.30
Enraf-Nonius CAD-4
Mo KR (λ ) 0.71069 Å)
5.53 cm-1
52.0
10436
9956
6113 [I > 2σ(I)]
715
0.074
0.070
0.50 × 0.16 × 0.12
Siemens SMART
Mo KR (λ ) 0.71073 Å)
1.207 mm-1
48.26
17862
6851
6849
622
0.0472
absn coeff
2θ max (deg)
no. of reflns colld
no. of ind reflns
no. of obsd reflns
params
R1a [I > 2σ(I)]
Rwa
wR2b [I > 2σ(I)]
goodness-of-fit
largest diff peak and
0.1082
1.122
0.438; -0.332
1.61
1.10; -1.27
1.80
0.65; -0.54
hole (e Å-3
)
2
2
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|; Rw ) [(∑w(|Fo| - |Fc|)2/∑wFo∞)]1/2, where w ) 4Fo /σ2(Fo ); σ2(Fo ) ) [S2(C + R2B) + (pFo )2]/Lp2; S ) scan rate,
C ) total integrated peak count, R ) ratio of scan time to background counting time, B ) total background count, Lp ) Lorentz-polarization
factor, and p ) p factor. b wR2 ) [∑[w(Fo - Fc2)2]/∑[w(Fo )2]]1/2, where w ) q/σ2(Fo ) + (aP)2 + bP.
2
2
2
NO2)][B(3,5-(CF3)2C6H3)4]2 (Figures 5 and 6). The complex
crystallizes in the centrosymmetric space group P1h with a
crystallographic inversion center within the cation, thus indicat-
ing “end-for-end” disorder. Location of peaks in difference
Fourier maps corresponding to the two possible orientations of
the bridging nitrite ligand allowed successful modeling of the
disorder (see the Experimental Section). The novel µ-(η1-N:
η2-O,O)-nitrite bonding geometry in the complex is character-
ized by a remarkably short Cu-Nnitrite bond distance [1.780(8)
Å], only slightly asymmetric O,O-chelation to the other copper
center [Cu1′-O1 ) 2.047(9) Å; Cu1′-O2 ) 2.13(1) Å], and a
planar Cu2(NO2) core with a Cu‚‚‚Cu separation of 4.31 Å. This
nitrite binding mode is, to our knowledge, unprecedented in
transition metal chemistry; the few known examples of such a
topology involve PbII or CdII ions.41
2+
A description of crystalline [(Li-Pr Cu)2(µ-NO2)] as com-
3
pletely valence localized (class I)42 appears appropriate on the
basis of the divergent coordination geometries of the copper
ions that are characteristic for their respective 1+ and 2+
oxidation levels. Thus, we assign a 1+ oxidation state to Cu1
because it is roughly tetrahedral, with L-Cu-L bond angles
similar to those observed for Li-Pr Cu(NO2) and [(L Cu)2(µ-
NO2)]PF6. A distorted 5-coordinate square pyramidal geometry
for Cu1′ is consistent with its CuII designation. This preference
for coordination of the nitrite N atom to CuI and the O atoms
to CuII is in accord with simple hard-soft acid-base consid-
erations.
i-Pr3
3
Our assessment of the valence localization in the dicopper-
(I,II) complex is corroborated by the congruence between its
Figure 5. Representations of the X-ray crystal structures of the cations
of (a) [(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2 and (b) [Li-Pr Cu(µ-
3
3
NO2)ZnLi-Pr ](O3SCF3)2, showing all nonhydrogen atoms as 50%
3
(41) (a) Shvelashvili, A. E.; Porai-koshits, M. A. Koord. Khim. 1975, 1,
467. (b) Nardelli, M.; Pelizzi, G. Inorg. Chim. Acta 1980, 38, 15. (c)
Takagi, S.; Joesten, M.; Lenhert, P. G. Acta Crystallogr. 1976, B32, 326,
1278. (d) Cullen, D. L.; Lingafelter, E. C. Inorg. Chem. 1971, 10, 1264-
1268.
(42) Robin, M. B.; Day, P. AdV. Inorg. Chem. Radiochem. 1967, 10,
247-422.
ellipsoids.
A shift in the binding mode of the bridging nitrite induced
by one-electron oxidation of the dicopper(I,I) complex is evident
from the X-ray crystal structure of mixed valent [(Li-Pr Cu)2(µ-
3

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 769
i-Pr
Figure 6. Drawings with selected bond distances (Å) and angles (deg) noted of the cores of (a) [(Li-Pr Cu)2(µ-NO2)]PF6, (b) L Cu(NO2), (c)
3
3
i-Pr
i-Pr
3
[(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2, and (d) [L Cu(µ-NO2)ZnL ](O3SCF3)2.
3
3
series43 and with the known greater stability of bidentate
coordination of nitrate to N3MII moieties when MII ) CuII vs
ZnII.44 As described below, this qualitative bonding description
(Figure 7) also accounts for observed differences in spectro-
scopic properties among the various copper(I) nitrite complexes.
B. [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3. The structure of this
3
compound (Figure 4b) reveals a typical 5-coordinate distorted
square pyramidal copper(II) ion with one Li-Pr N atom (N7)
3
occupying the pseudoaxial position at a Cu-N distance [2.180-
(5) Å] significantly longer than the other two (average 2.06 Å).
A water molecule and a monodentate triflate ion complete the
coordination sphere of this additional member of a growing class
of crystallographically characterized copper(II) triflates sup-
ported by tridentate N-donor ligands.18b,32 The aquo ligand
appears to be hydrogen-bonded to both the coordinated and
uncoordinated triflate groups, as indicated by the O1‚‚‚O3 and
O1‚‚‚O7 distances of 2.73 and 2.72 Å, respectively (aquo H
atoms not found). Similar patterns of hydrogen bonding to
triflate ions in aquo complexes have been reported.45 Also
similar to other transition metal triflate complexes,47 the S-O
distances reflect O atom coordination (S1-O2 longer than S1-
O3 or S1-O4 by >0.03 Å).
Figure 7. Qualitative view of the back-bonding interaction between
I
II
I
CuI and nitrite in Li-Pr Cu(NO2) and the mixed valent Cu Cu and Cu -
ZnII complexes, with related structural and spectroscopic features listed
(see Discussion).
3
i-Pr3
structure and that of [Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2 (Figures
5 and 6), which is constrained to a CuIZnII formulation by the
inaccessibility of oxidation states other than 2+ for zinc. The
high degree of similarity between the two structures is striking,
the only significant difference between them being the Cu-
3
N
nitrite bond length [1.780(8) vs 1.848(3) Å for the CuICuII and
CuIZnII complexes, respectively]. Consideration of these dif-
NMR and EPR Spectroscopy. 1H and 13C NMR spectros-
copy were found to be particularly useful for elucidation of the
structures of the diamagnetic copper(I) complexes in solution.
We assign C3V-distorted pseudotetrahedral geometries to
ferent Cu-Nnitrite bond distances, in conjunction with the
corresponding 1.903(4) Å bond length in Li-Pr Cu(NO2), reveals
3
a clear trend that reflects the nature of the O,O-nitrite bonding.
The shortest distance is found for the case where CuII is
coordinated to the nitrite O atoms and the longest occurs for
the monomer, which has no O-bound metal ions, with the
distance for the CuIZnII complex falling between these extremes.
Single O-nitrite coordination as in the dicopper(I,I) complex
apparently does not perturb the system sufficiently to affect the
Cu-Nnitrite bond distance. The correlation between O,O-nitrite
coordination and CuI-Nnitrite bond length can be rationalized
by invoking an effect of O,O-binding on the π-back-bonding
between the copper(I) ion and the N-bound nitrite ligand (Figure
7). As the strength of O,O-nitrite coordination by the divalent
metal ion increases (CuII > ZnII > none), the energy of the
nitrite π* orbital of appropriate symmetry to interact with a filled
copper(I) orbital would be expected to decrease due to the
electron-withdrawing influence of the O,O-bound metal, thus
resulting in stronger CuI f Nnitrite back-bonding and a cor-
respondingly shorter CuI-Nnitrite bond length. Although direct
Bn3
i-Pr3
[LCu(CH3CN)]PF6 (L ) Li-Pr or L ) and [L Cu(PPh3)]PF6
on the basis of the number of resonances in their 13C{1H} NMR
spectra, the pattern of ligand substituent peaks in their 1H NMR
spectra (cf. one doublet/heptet set for the isopropyl substituents
3
for L ) Li-Pr ), and diagnostic features in NMR and IR spectra
3
consistent with coordinated CH3CN or PPh3. In addition,
1
comparison of the H NMR data for the free ligands to those
for the complexes reveals differences in the methylene backbone
region that are diagnostic for metal ion binding. For example,
a single peak at 2.5 ppm for Li-Pr is split into multiplets centered
3
at 2.80 and 2.55 ppm upon binding to copper(I) to yield
[Li-Pr Cu(CH3CN)]PF6. NMR spectra similar to those observed
3
for the copper(I) starting materials were recorded for the nitrite
(43) Sigel, H.; McCormick, D. B. Acc. Chem. Res. 1970, 3, 201-208.
(44) (a) Han, R.; Parkin, G. J. Am. Chem. Soc. 1991, 113, 9707-9708.
(b) Looney, A.; Parkin, G. Inorg. Chem. 1994, 33, 1234-1237.
(45) For a recent example, see: Stang, P. J.; Cao, D. H.; Poulter, G. T.;
Arif, A. M. Organometallics 1995, 14, 1110-1114 and references cited
therein.
experimental evidence for the stronger bonding to Li-Pr Cu(NO2)
3
II
i-Pr3
II
by the Li-Pr Cu fragment compared to L Zn is lacking, such
3
a bond strength order is consistent with the Irving-Williams
(46) Lawrance, G. A. Chem. ReV. 1986, 86, 17-33.

770 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Table 3. Summary of Selected Spectroscopic Data
Halfen et al.
complex
EPRa
UV-visb
FTIRc
resonance Ramanc
[(Li-Pr Cu)2(µ-NO2)]PF6
338 (3000), 380 (2200)
336 (3700)
1352 (1333), 1185 (1168)
1347 (1331), 1202 (1174)
3
[(LBn Cu)2(µ-NO2)]PF6
3
[(Li-Pr Cu)2(µ-NO2)](PF6)2
g ) 2.07, g| ) 2.24, 314 (3900), 356 (sh, 3000), 1235 (1214)
1238 (1218), 876 (874)
1238 (1214), 872 (868)
3
A| ) 136 × 10-4
444 (2500), 668 (320)
406 (2200), 358 (1900)
308 (3800)
i-Pr
[Li-Pr Cu(µ-NO2)ZnL ](CF3SO3)2
d
3
3
Li-Pr Cu(NO2)
1306,e 1268 (1247)
3
Li-Pr Cu(O2CCH3)2
g ) 2.06, g| ) 2.30, 298 (4900), 684 (110)
3
A| ) 154 × 10-4
Li-Pr Cu(O3SCF3)2
g ) 2.09, g| ) 2.28, 300 (3600), 718 (100)
A| ) 160 × 10-4
3
a A| values quoted in cm-1; measured at 77 K as 1:1 CH2Cl2/toluene glasses. b λmax, nm (ꢀ/complex, M-1 cm-1); measured at room temperature
-
as solutions in CH2Cl2. c 15N-sensitive vibrations quoted in cm-1; measured at room temperature as KBr pellets. d 15N-sensitive NO2 vibrations
-
obscured by counterion stretches. e The corresponding 15NO2 stretch was obscured by other vibrations.
i-Pr3
complexes Li-Pr Cu(NO2), [(LCu)2(µ-NO2)]PF6 (L ) L
or
3
LBn ), and [L Cu(µ-NO2)ZnL ](O3SCF3)2. Despite the
i-Pr3
i-Pr3
3
asymmetry inherent to the solid state structure of [(Li-Pr Cu)2-
3
(µ-NO2)]PF6, only one set of ligand resonances appeared in its
NMR spectrum, suggesting either coincidental overlap of peaks
associated with the Nnitrite- and Onitrite-bound copper units or,
more likely, rapid fluxionality involving interchange of Li-Pr
3
-
Cu environments. Although such an interchange may involve
mononuclear intermediates, none were observed. In contrast,
1
i-Pr3
the H and 13C NMR spectra of [Li-Pr Cu(µ-NO2)ZnL ](O3-
3
SCF3)2 in CD2Cl2 contained distinct and sharp Li-Pr resonances
3
for the zinc(II) and copper(I) sites, consistent with a lack of
exchange of Li-Pr between the CuI and ZnII ions.
3
The mixed valent complex [(Li-Pr Cu)2(µ-NO2)](PF6)2 exhib-
3
ited an axial signal in its 77 K X-band EPR spectrum with
parameters typical for a square pyramidal CuII site with mixed
N,O coordination (Table 3).47 Double integration confirmed
that the signal arose from one unpaired electron per complex.
The nature of this signal, which showed no evidence for
additional hyperfine coupling to a second copper nucleus, is
consistent with localization of unpaired spin density on the
5-coordinate copper ion as suggested by the X-ray crystal-
lographic evidence cited above. EPR spectra typical for
5-coordinate CuII ions in square pyramidal geometries were also
Figure 8. UV-vis spectra of CH2Cl2 solutions Li-Pr Cu(NO2) (‚‚‚),
3
[(Li-Pr Cu)2(µ-NO2)](O3SCF3)2 (s), and [L Cu(µ-NO2)ZnL ](O3-
i-Pr
i-Pr
3
3
3
SCF3)2 (- - -).
the nitrite ligand. To investigate whether this transition should
-
-
be designated as a CuI f NO2 MLCT or a NO2 f CuII
ligand-to-metal charge transfer (LMCT), we prepared and
characterized the copper(I)-zinc(II) analog, for which the latter
type of transition would be impossible because both metal ions
have filled d shells. Like the CuICuII compound, the CuIZnII
complex is deep red, exhibits an intense electronic absorption
at 406 nm (ꢀ/complex 2200 M-1 cm-1) in CH2Cl2 (Figure 8),
and gives rise to 15NO2- isotope sensitive bands in its resonance
Raman spectrum upon 458 nm excitation (see below). As
expected, the low-energy ligand field band observed for the
dicopper(I,II) compound (λmax ) 668 nm) is absent in the
spectrum of the heterodinuclear analog. Taken together, these
observations allow definitive attribution of the 406 nm band in
recorded for solutions of the mononuclear compounds Li-Pr Cu-
3
(O3SCF3)2 and Li-Pr Cu(O2CCH3)2 in 1:1 CH2Cl2/toluene (Table
3
3).
Electronic Absorption Spectroscopy. In contrast to the
colorless [LCu(CH3CN)]X starting materials and [Li-Pr Cu-
3
(PPh3)]PF6, the copper(I) nitrite adducts are yellow to yellow-
orange due to intense absorptions between 308 and 338 nm (ꢀ/
complex 3000-3800 M-1 cm-1) (Table 3). We suggest that
these features are CuI f NO2- metal-to-ligand charge transfer
(MLCT) transitions, although such assignments remain tentative
in the absence of more conclusive evidence such as that which
would be provided by resonance Raman spectroscopic experi-
ments (planned). One-electron oxidation of yellow-orange
-
the copper-zinc salt as a CuI f NO2 MLCT transition and
provide convincing, albeit indirect, support for a similar
assignment to the 444 nm absorption for the dicopper(I,II)
complex.
[(Li-Pr Cu)2(µ-NO2)]PF6 caused a color change to deep red due
3
to a much lower energy absorption in the mixed valent dicopper-
(I,II) product [Figure 8 and Table 3; λmax ) 444 nm (ꢀ/complex
2500 M-1 cm-1)]. An additional much weaker absorption
feature at 668 nm (ꢀ/complex 320 M-1 cm-1) attributable to a
ligand field (dd) transition corroborates the presence of a copper-
(II) ion in this compound. Resonance Raman spectra of solid
Agreement between electronic absorption spectroscopic and
X-ray structural features lends support for the bonding model
cited above for the copper(I) nitrite unit. As depicted in Figure
7, the CuI-Nnitrite bond length is inversely correlated with MLCT
energies in the series (ordered according to the bond distance)
[(Li-Pr Cu)2(µ-NO2)](PF6)2 (KBr pellet) obtained using 458 nm
3
i-Pr3
i-Pr3
Li-Pr Cu(NO2) ≈ [(L Cu)2(µ-NO2)]PF6 > [L Cu(µ-NO2)-
3
-
excitation contained intense features that shifted upon 15NO2
i-Pr3
ZnLi-Pr ](O3SCF3)2 > [(L Cu)2(µ-NO2)](PF6)2. This correla-
tion may be explained by arguing that, as the divalent metal
O,O-nitrite coordination strengthens in the series, more efficient
electron withdrawal by the divalent metal ion causes the nitrite
π* orbital to decrease in energy, resulting in a stronger CuI f
3
substitution (see below), clearly indicating that the 444 nm
absorption band arises from a charge transfer transition involving
(47) Peisach, J.; Blumberg, W. E. Arch. Biochem. Biophys. 1974, 165,
691-708.

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 771
Figure 9. Resonance Raman spectra obtained using 457.9 nm excitation for the 14NO2- and 15NO2- isotopomers of the mixed valent dicopper(I,II)
and copper(I)-zinc(II) complexes, with difference spectra displayed at the bottom of each set (* denotes the sulfate standard peak).
NO2- back-bonding interaction, a shorter Cu-Nnitrite bond, and
transitions. The largest isotope shifts occurred for the 1238
cm-1 peaks (∆15N/14N ) 20 and 24 cm-1, respectively, for the
CuICuII and CuIZnII compounds), the magnitudes of which argue
-
a lower energy (longer wavelength) CuI f NO2 MLCT
transition.
-
for their assignment as NO2 νs (typical ∆15N/14N values for
Infrared and Resonance Raman Spectroscopy. Pairs of
infrared vibrations sensitive to 15NO2- substitution between 1150
nitro νs are ∼20 cm-1).48 This band also was identified in the
IR spectrum of the dicopper(I,II) complex (1235 cm-1; ∆15N/
14N ) 21 cm-1). The frequencies and much less pronounced
isotope shifts for the sharp 876 cm-1 (CuICuII; ∆15N/14N ) 2
cm-1) and 872 cm-1 (CuIZnII; ∆15N/14N ) 4 cm-1) bands are
consistent with their assignment as deformation modes (δs); in
other nitro complexes, these typically occur at ∼830 cm-1 with
and 1400 cm-1 were identified for Li-Pr Cu(NO2) and [(LCu)2-
3
(µ-NO2)]PF6 (L ) Li-Pr and LBn ) (Table 3). We tentatively
3
3
assign the 1306 and 1268 cm-1 15N-sensitive bands in Li-Pr Cu-
3
(NO2) to nitrite asymmetric (νas) and symmetric (νs) stretches,
respectively, on the basis of the similarity of their frequencies
and/or isotope shifts to features identified in other transition
metal η1-NO2- (nitro) complexes such as [Cu(NO2)6]4- (∼1330
and ∼1280 cm-1).33,48 The difference in energy between νas
and νs for the dicopper(I,I) complexes is notably greater than
∆15N/14N ) 4-5 cm-1 33,48
Finally, a third distinctive feature
.
is apparent in the resonance Raman spectra, a broad peak at
low energy [422 cm-1 (CuICuII) and 430 cm-1 (CuIZnII)] which
did not appear to shift upon 15NO2- substitution. This feature
may be attributed either to a CuI-Nnitrite vibration (typical M-N
for nitro complexes occur at 300-400 cm-1 with ∆15N/14N )
for Li-Pr Cu(NO2), which is consistent with greater N-O bond
3
alternation in the former compounds (see ref 33 for a discussion
of the correlation between N-O bond distance differences and
the νas/νs vibrational frequency separation in metal nitrite
compounds). More definitive assignments for all of the copper-
(I) compounds await a full analysis incorporating results from
resonance Raman spectroscopy using suitable excitation wave-
lengths in the UV region. In any case, the similarity of the IR
4 cm-1 33,48 or to a symmetric vibration of the three-atom MII-
)
O,O-nitrito unit (MII ) CuII or ZnII), which would be expected
to have the observed minimal 15N isotope sensitivity but distinct
MII dependence. We cannot distinguish between these pos-
sibilities at this point in our studies, however.
data for the dicopper(I,I) complexes capped by Li-Pr and LBn
3
3
represents important evidence for the postulate of congruent
structures for the two molecules.
Electrochemistry. Cyclic voltammetry experiments revealed
that [LCu(CH3CN)]PF6 (L ) Li-Pr and LBn ) in CH3CN with
3
3
For the mixed valent dicopper(I,II) and copper(I)-zinc(II)
complexes, 458 nm irradiation into their absorption bands (λmax
) 444 and 406 nm, respectively) afforded resonance-enhanced
Raman features (Figure 9). The obvious overall similarity of
the data for both compounds corroborates the X-ray crystal-
lographic results, indicating near identity of their structures. Two
of the features in the resonance Raman spectra at 1238 and ∼875
0.5 M (Bu4N)(PF6) undergoes electrochemically quasi-reversible
Bn3
[∆Ep (0.1 V/s) ) 110 mV (Li-Pr ) and 310 mV (L )] and
3
chemically reversible (ipc ≈ ipa) one-electron (verified by
coulometry) redox transformations with E1/2 ) 0.36 V (Li-Pr
)
3
and 0.42 V (LBn ) vs SCE. The more positive redox potential
for the latter couple (indicative of relatively greater stabilization
of the reduced compared to the oxidized state in the former) is
consistent with the respective electronic influences of the
supporting ligand substituents (Bn is more electron withdrawing
than i-Pr).49 Both potentials are less than those measured for
analogous TpRR′Cu(CH3CN) complexes,23 presumably because
the sp3 amine donors in the triazacyclononane complexes favor
the CuII state more than the sp2 pyrazolyl donors in the TpRR′
3
-
cm-1 shifted upon 15NO2 substitution. As described above,
this finding, in conjunction with observation of decreased
resonance enhancement of these vibrations relative to a sulfate
standard as the excitation wavelength was increased from 457
to 613 nm, lends important support for assignment of the
-
absorption bands for the complexes as CuI f NO2 MLCT
(48) Cleare, M. J.; Griffith, W. P. J. Chem. Soc. 1967, 1144-1147.

772 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
Scheme 2
Scheme 3
red color bleached within seconds even at low temperature, with
concomitant evolution of both NO (65%) and a small amount
of N2O (<5%). When Me3SiO3SCF3 (a “proton acid equiva-
lent”) was used instead, the red solution persisted for ca. 5-10
min at -80 °C before turning green and evolving NO and, in
this instance, (Me3Si)2O, allowing acquisition of UV-vis and
EPR spectroscopic data. Unfortunately, rather than indicating
the presence of a nitrosyl intermediate, these data matched those
compounds, and in spite of the overall 1+ charge for the former.
Divergent propensities of the triazacyclononane and TpRR′
ligands to support nontetrahedral metal ion geometries may also
be responsible for the redox potential differences, i.e., the TpRR′
ligands are arguably greater “tetrahedral enforcers”,22 especially
when the 3-pyrazolyl substituents are large, and thus drive the
redox potential to a higher value because of their tendency to
favor geometries suitable for copper(I).
for the mixed valent cation [(Li-Pr Cu)2(µ-NO2)]2+. We suggest
3
Differences in electrochemical behavior arising from i-Pr
versus Bn substitution also were observed in cyclic voltammo-
grams of the dicopper(I,I) complexes [(LCu)2(µ-NO2)]PF6. As
that this compound is formed in a side reaction of Li-Pr Cu-
3
II
(NO2) starting material and the Li-Pr Cu fragment resulting from
NO loss and that the dicopper(I,II) complex also reacts with
H+ or Me3Si+ to afford the observed final products (Scheme
3
already mentioned above, for L ) Li-Pr , the complex undergoes
3
an electrochemically quasi-reversible [∆Ep (0.1 V/s) ) 110 mV]
and chemically reversible (ipa/ipc ≈ 1) one-electron oxidation
with E1/2 ) 0.07 V vs SCE (0.1 M TBAP in CH2Cl2), which is
associated with a shift in the coordination mode of the bound
nitrite ion and the generation of a stable mixed valent, dicopper-
3). In support of these postulates, [(Li-Pr Cu)2(µ-NO2)](O3-
3
SCF3)2, which exhibits spectroscopic properties in CH2Cl2
identical to those of the red reaction solution, was successfully
i-Pr3
prepared independently from Li-Pr Cu(NO2) and L Cu(O3-
3
SCF3)2 and was found to yield NO (68%) and Li-Pr Cu(O3SCF3)2
3
(I,II) complex. In contrast, [(LBn Cu)2(µ-NO2)PF6 does not
3
upon treatment with Me3SiO3SCF3. Thus, as outlined in
exhibit such clean electrochemical behavior; it is irreversibly
oxidized at Epa ) +0.41 V vs SCE. We currently do not
understand why our efforts to generate a stable mixed valence
II
Scheme 3, the Li-Pr Cu fragment resulting from nitrite dehydra-
3
tion and NO loss from Li-Pr Cu(NO2) appears to undergo
3
competitive trapping by either starting material or the conjugate
base of the acid reagent; in the case of HOAc, the latter
predominates, preventing observation of the mixed valent
species, whereas for Me3SiO3SCF3, this species forms at low
temperature and persists for sufficient time prior to its decom-
position to NO to be spectroscopically identified.
species supported by LBn either electrochemically or chemically
3
have not met with success; clearly, these and other observa-
tions31 demonstrate that differences in the steric and/or electronic
Bn3
properties of Li-Pr and L significantly affect the stability and
3
reactivity of their respective copper complexes.
Reactivity: NO Generation. An investigation of the reactiv-
Biological Relevance. The complexes described herein
represent the first examples of discrete copper(I)-nitrite adducts
to be identified; thus, they provide important precedents for
nitrite binding to reduced copper sites in proteins. Such binding
is most significant in the catalytic cycle of nitrite reductase,
which has been postulated to coordinate NO2- at Cu-II (Figure
1). The pseudotetrahedral geometries of the copper(I) ions in
the mono- and dinuclear nitrite complexes supported by the
hindered triazacyclononane ligands mimic possible topologies
for the reduced, tris(histidyl)-ligated Cu-II site. Our observation
ity of Li-Pr Cu(NO2) with proton sources was undertaken in order
3
to probe the viability of the complex as a functional NiR model,
our primary objectives being to observe NO generation and
intermediates along the pathway thereto. Decomposition of
(Bu4N)(NO2) by a mixture of a dicopper(I) complex and HPF6‚-
Et2O has been reported, but instead of evolving NO, a dicopper-
(II)-nitrosyl complex formulated as having a CuII-(µ-NO-)-
CuII core was generated.16b Clean evolution of 1 equiv of NO
and, presumably, H2O was observed (GC) when a solution of
Li-Pr Cu(NO2) in CH2Cl2 was treated with 2 equiv of glacial
3
-
of both O- and N-NO2 coordination to CuI in the synthetic
acetic acid (Scheme 2). Nitric oxide gas production was
analogs renders either mode a viable option in the biological
system. Not surprisingly considering the well-known soft,
strong-field nature of the nitrite ion when utilizing its N atom
to bind to transition metals,33 the latter “nitro” mode appears to
be thermodynamically favored in the copper(I) complexes. The
preferential oxidation of the O-bound site to yield a localized
mixed valent species that retains the CuI-N-NO2- unit can be
construed as support for its stability. This argument, in
conjunction with the demonstration of quantitative NO evolution
accompanied by the quantitative formation of blue Li-Pr Cu(O2-
3
CCH3)2, which was identified by comparison of its UV-vis
and EPR spectra with those of independently synthesized
material (Vide infra). Presumably, the reaction proceeds via
nitrite protonation and dehydration to afford a [CuNO]10
intermediate (CuI-NO+ T CuII-NO, in valence bond terms);
subsequent ejection of NO from this intermediate and coordina-
tion of acetate to the resulting CuII fragment (not necessarily in
that order) would account for the observed final products. We
were not able to observe any intermediates in this instance, even
upon addition of 2 equiv of protons to Li-Pr Cu(NO2), provides
3
-
a basis for suggesting that the substrate adduct having N-NO2
when HOAc was added to Li-Pr Cu(NO2) at -80 °C, suggesting
3
coordination is the most important enzyme intermediate. Clean
dehydration of O- or O,O-bound nitrite in a CuII complex has
not been reported,18 making such a species a less likely candidate
for the direct precursor of NO. Linkage isomerism (O-to-N-
NO2-) was observed in a copper(II) system that promoted nitrite
reduction when reduced electrochemically.18c,d Thus, reduction
that the latter steps (NO evolution/CuII trapping) were efficient
with the acetate anion. Thus, several other protonic acids or
analogs with poorly ligating conjugate bases were screened for
their ability to stabilize the postulated nitrosyl intermediate.
Addition of excess HBF4‚Et2O to a solution of Li-Pr Cu(NO2)
3
at -80 °C initially yielded a promising red solution, but the
-
of an initially formed CuII-NO2 moiety with concomitant
(49) The carbonyl stretching frequencies in IR spectra of [LCu(CO)]+
isomerization to a CuI-N-NO2- intermediate that subsequently
(2088 cm-1 for L ) LBn vs 2067 cm-1 for L ) Li-Pr ) corroborate the
3
3
3
i-Pr3
generates NO is a reasonable possibility. Presumably, Li-Pr
-
suggested order of electron-withdrawing capability LBn > L
A.; Mahapatra, S.; Tolman, W. B. Unpublished results).
(Halfen, J.
3
Cu(NO2) and its enzyme congener evolve NO via a [CuNO]10

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 773
Experimental Section
intermediate, which to date remains an elusive synthetic target
and an unsubstantiated metalloprotein moiety. Such a unit has
been proposed to form upon reaction of various copper(II)-
containing proteins and complexes with NO (see references cited
in 17a), but supporting evidence beyond reversible bleaching
of CuII-associated UV-vis and EPR features is lacking in most
instances and the only well-characterized mononuclear copper
nitrosyl has a [CuNO]11 electron inventory (it is generated upon
treatment of a CuI precursor with NO).17
There are many analogies between iron- and copper-contain-
ing dissimilatory nitrite reductases with regard to proposed
mechanisms and relevant synthetic model complexes. For both,
it is believed that nitrite binds to the reduced metal ion (FeII or
CuI) and undergoes dehydration to an unstable, electrophilic
nitrosyl ([FeNO]6 or [CuNO]10) that subsequently evolves NO
and provides the oxidized metal (FeIII or CuII).14 Like the
compounds we have characterized, an iron(II) porphyrin syn-
thetic model of the protein substrate adduct exhibits nitro
coordination characterized by a short Fe-Nnitrite bond that gives
rise to distinct spectroscopic behavior.35 Both iron and copper
nitrosyls proposed to form upon protonation of the nitrite
complexes are less stable than their one-electron reduced
counterparts ([FeNO]7 or [CuNO]11), although in contrast to the
case for copper, synthetic porphyrin- and corrole-iron nitrosyls
at both oxidation levels have been well-characterized.50
General Procedures. All reagents and solvents were obtained from
commercial sources and used as received unless otherwise noted.
Solvents were dried according to published procedures51 and distilled
under N2 immediately prior to use. 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
Bn
techniques. The ligands Li-Pr and L were synthesized according to
the published procedures.24,25 Although a variety of syntheses of the
parent macrocycle 1,4,7-triazacyclononane have appeared in the
literature, we have found that the following methods for the preparation
of synthetic intermediates are most efficient. Ethylene glycol ditosylate
and diethylenetriamine tris(tosylate) were prepared according to the
methods of Ouchi et al.52 and Verkade et al.,53 respectively, and both
were recrystallized from CH3CN. The closure of the macrocyclic ring
and the subsequent acidic hydrolysis of the product tris(tosylate) were
performed according to the method of Searle and Geue.54 The tris-
(hydrochloride) salt of 1,4,7-triazacyclononane was isolated by cautious
addition of 6 M HCl to the sulfuric acid detosylation mixture, followed
by precipitation with Et2O. Isolation of the 1,4,7-triazacyclononane
free base was accomplished following the method of Verkade et al.53
3
3
55
The salts [Cu(CH3CN)4]PF6 and Na[B(3,5-(CF3)2C6H3)4]30 were
prepared as described in the literature. Gas chromatography experi-
ments were performed by using a Hewlett-Packard 5890 Series II gas
chromatograph with a HP 3396 Series II integrator, Porpak Q column
(6 ft, 20 mL/min flow rate, 30 °C, helium carrier gas), and TCD
detector. NO quantitation was performed by calibrating detector
response with known concentrations of NO mixed with N2; molar
quantities were calculated using the ideal gas equation. Electrochemical
experiments were performed using an EG & G Princeton Applied
Research (PAR) VeraStat potentiostat driven by EG & G PAR Model
250 software. The cell used was described previously.23 Measurements
were conducted under an Ar atmosphere, using tetrabutylammonium
hexafluorophosphate (TBAP) as the supporting electrolyte. Conduc-
tance measurements were made using a YSI Model 35 cnductance meter
and analyzed according to a published protocol.56 IR spectra were
recorded on a Perkin Elmer 1600 Series FTIR spectrophotometer as
KBr pellets. Resonance Raman spectra were collected on a Spex 1403
spectrometer interfaced with a DM3000 data collection system using
Spectra-Physics Models 2030-15 argon ion and 375B CW dye
(Rhodamine 6G) lasers. Spectra were obtained at room temperature
as KBr pellets, and Raman frequencies are referenced to internal Na2-
SO4 [ν1(SO42-) ) 998 cm-1]. Electronic absorption spectra were
recorded on a Hewlett-Packard HP8452A spectrophotometer (190-
820 nm scan range) interfaced with a microcomputer for data acquistion
and analysis, with low-temperature data collected by using a custom
manufactured vacuum Dewar with quartz windows. NMR spectra were
recorded on Varian VXR-300 or VXR-500 spectrometers; 1H and 13C-
{1H} chemical shifts are reported versus tetramethylsilane and refer-
enced to the residual solvent peak(s). 31P{1H} NMR spectra are
referenced to external 85% H3PO4. X-band EPR spectra were recorded
on a Bruker ESP300 spectrometer fitted with a liquid nitrogen finger
dewar (77 K, ∼9.44 GHz). Fast atom bombardment (FAB) mass
spectra were determined using a VG 7070E-HF mass spectrometer with
xenon gas for ionization. Electrospray ionization mass spectral data
were collected using a Sciex API III mass spectrometer, with scanning
performed in 0.2 Da steps and a 5-10 ms dwell time per step. Signal
averaging was used to enhance the signal-to-noise ratio. Elemental
A second type of copper protein nitrite adduct is a mixed
valent nitritohemocyanin (halfmetNO2Hc) that has been exten-
sively studied because of the potential for insight into the factors
that control dioxygen binding and/or activation in this and
related proteins such as tyrosinase.20 The nitrite-bridged CuI-
CuII complexes we have described represent possible models
of this adduct. Despite some recent controversy,20g it has
become evident that the initial report by Himmelwright et al.20d
that nitrite is tightly bound to the active site of halfmetNO2Hc
is correct.20h Recent spectroscopic studies on halfmetNO2Hc20h
and a mononuclear CuII form of Hc both with and without bound
nitrite20i have been interpreted to indicate that nitrite coordinates
via an O atom to a single CuII ion ligated to three histidine
residues in a distorted tetragonal geometry. Because of similari-
ties in EPR and ESEEM spectroscopic properties between the
mononuclear CuII Hc and halfmetNO2Hc and the fact that the
latter is completely valence localized, the presence of an
additional interaction of nitrite with the second CuI ion has been
questioned.20h,i Our discovery of a nitrite-bridged CuICuII
compound that also is valence localized and that exhibits an
EPR signal typical for a mononuclear CuII site implies that
interaction of nitrite with a CuI ion may be feasible in the
protein. However, this idea is mitigated by the distinct differ-
ences between the UV-vis and EPR properties of [(Li-Pr Cu)2-
3
(µ-NO2)]2+ and halfmetNO2Hc, particularly the lack of the
MLCT band in the absorption spectrum of the latter, that argue
against congruent structures for their dimetal cores.
In sum, the work described herein provides direct experi-
-
mental support for the viability of CuI-NO2 adducts in
analyses were performed by Atlantic Microlabs of Norcross, GA.
i-Pr
[Li-Pr Cu(CH3CN)]PF6. To a solution of L (0.421 g, 1.65 mmol)
in THF (5 mL) was added [Cu(CH3CN)4]PF6 (0.611 g, 1.64 mmol) as
a solid. The resultant mixture was stirred for 1 h and then filtered to
remove any remaining solid. Pentane (10 mL) was then added to the
biological denitrification processes. Demonstration of their
competence to evolve nitric oxide upon protonation further
corroborates their suggested significance as intermediates in
dissimilatory nitrite reduction by copper proteins. Several issues
remain unresolved, however, such as the mechanism of N2O
production6 and the intermediacy of [CuNO]10 species;15a these
are the focus of ongoing efforts in our laboratory.
3
3
(51) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon Press: New York, 1988.
(52) Ouchi, M.; Inoue, Y.; Liu, Y.; Nagamune, S.; Nakamura, S.; Wada,
K.; Hakushi, T. Bull. Chem. Soc. Jpn. 1990, 63, 1260-1262.
(53) White, D. W.; Karcher, B. A.; Jacobsen, R. A.; Verkade, J. G. J.
Am. Chem. Soc. 1979, 101, 4921-4925.
(50) (a) Autret, M.; Will, S.; Caemelbecke, E. V.; Lex, J.; Gisselbrecht,
J.-P.; Gross, M.; Vogel, E.; Kadish, K. M. J. Am. Chem. Soc. 1994, 116,
9141-9149 and references cited therein. (b) Ozawa, S.; Sakamoto, E.;
Ichikawa, T.; Watanabe, Y.; Morishima, I. Inorg. Chem. 1995, 34, 6362-
6370.
(54) Searle, G. H.; Geue, R. J. Aust. J. Chem. 1984, 37, 959-970.
(55) Kubas, G. J. Inorg. Synth. 1979, 19, 90-92; 1990, 28, 68-70.
(56) Geary, W. J. Coord. Chem. ReV. 1971, 7, 81.

774 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
nearly colorless filtrate with rapid stirring, causing the precipitation of
the product as a white powder (0.655 g, 80%): mp 124 °C dec; H
with a solution of Na[B(3,5-(CF3)2C6H3)4] (0.082 g, 0.092 mmol) in
Et2O (2 mL), causing the deposition of a white powder. After standing
for 1 h at -20 °C, the mixture was filtered; storage of the filtrate at
1
NMR (CD2Cl2, 500 MHz) δ 3.05 (septet, J ) 6.5 Hz, 3H), 2.79-2.72
(m, 6H), 2.48-2.42 (m, 6H), 1.94 (s, 3H), 1.20 (d, J ) 6.5 Hz, 18H)
ppm; 13C{1H} NMR (CD2Cl2, 125 MHz): δ 164.6, 58.55, 51.2, 20.0,
3.2 ppm; FTIR (KBr, cm-1) 2974, 2935, 2854, 2267 (CtN), 1496,
1471, 1452, 1386, 1367, 1350, 1324, 1296, 1278, 1164, 1128, 1072,
1050, 967, 955, 877, 840 (PF6-), 779, 759, 719, 556 (PF6-); FAB-MS
(MNBA) m/e (relative intensity) 359 ([M - PF6]+, 15), 318 ([M -
CH3CN - PF6]+, 100). Anal. Calcd for C17H36CuF6N4P: C, 40.43;
H, 7.19; N, 11.09. Found: C, 40.72; H, 6.95; N, 11.23.
-20 °C overnight produced small dark crystals of [(Li-Pr )2Cu2(NO2)]-
3
[B(3,5-(CF3)2C6H3)4]2 (0.105 g, 96%). Crystallization from CH2Cl2/
toluene produced crystals suitable for X-ray diffraction: FTIR (KBr,
cm-1) 2986, 2946, 2877, 1607, 1471, 1355, 1282, 1227, 1186, 1156,
1122, 960, 887, 837, 714, 683, 667. Anal. Calcd for C94H90-
Cu2B2F48N7O2: C, 46.84; H, 3.76; N, 4.07. Found: C, 46.73; H, 3.75;
N, 4.10.
i-Pr
3
[Li-Pr Cu(PPh3)]PF6. To a solution of [L Cu(CH3CN)]PF6 (0.123
g, 0.244 mmol) in THF (5 mL) was added PPh3 (0.064 g, 0.246 mmol).
A fine white solid precipitated almost immediately, which was
recrystallized from CH2Cl2/Et2O to yield the product as colorless needles
(0.123 g, 70%): mp > 250 °C; 1H NMR (CD2Cl2, 300 MHz) δ 7.51-
7.40 (m, 9H), 7.28-7.22 (m, 6H), 2.98 (septet, J ) 6.6 Hz, 3H), 2.88-
2.78 (m, 6H), 2.75-2.65 (m, 6H), 0.98 (d, J ) 6.6 Hz, 18H) ppm;
13C{1H} NMR (CD2Cl2, 75 MHz) δ 133.3, 133.1, 131.0, 129.3, 129.4,
59.5, 36.5, 18.9 ppm; 31P{1H} NMR (CD2Cl2, 121 MHz) δ 5.27 ppm;
FTIR (KBr, cm-1) 3053, 2987, 2937, 2869, 1491, 1479, 1466, 1434,
1388, 1366, 1351, 1328, 1299, 1281, 1265, 1163, 1127, 1092, 1068,
1023, 99i, 965, 877, 838 (PF6-), 787, 756, 744, 719, 696, 585, 556
(PF6-), 531; FAB-MS (MNBA) m/e (relative intensity) 580 ([M -
PF6]+, 100). Anal. Calcd for C33H48CuF6N3P2: C, 54.58; H, 6.66; N,
5.79. Found: C, 55.27; H, 6.80; N, 5.88.
3
Bn
[LBn Cu(CH3CN)]PF6. To a solution of L (0.330 g, 0.83 mmol)
in THF (10 mL) was added [Cu(CH3CN)4]PF6 (0.308 g, 0.83 mmol).
The resultant mixture was stirred for 1 h and then filtered to remove
any remaining solid. Heptane (20 mL) was added to the yellow filtrate
with rapid stirring, causing the formation of an oil. Upon treatment of
this oil with MeOH, the product was obtained as a light tan solid (0.33
g, 62%): 1H NMR (CD2Cl2, 300 MHz) δ 7.34-7.30 (m, 15H), 3.82
(s, 6H), 2.72-2.62 (m, 6H), 2.55-2.45 (m, 6H), 2.35 (s, 3H) ppm;
13C{1H} NMR (CD2Cl2, 75 MHz): δ 141.21, 135.10, 131.2, 128.8,
117.4, 64.3, 52.4, 0.4 ppm; FTIR (KBr, cm-1) 3087, 3059, 3028, 2973,
2935, 2854, 2267 (CtN), 1495, 1471, 1386, 1367, 1277, 1164, 1137,
1128, 1071, 1049, 967, 955, 876, 837 (PF6-), 759, 719 cm-1; FAB-
MS (MNBA) m/e (relative intensity) 462 ([M - CH3CN - PF6]+, 100).
Anal. Calcd for C29H36CuF6N4P: C, 53.69; H, 5.60; N, 8.64. Found:
C, 53.51; H, 5.54; N, 8.69.
3
3
i-Pr
Li-Pr Cu(NO2). To a solution of [(L Cu)2(µ-NO2)]PF6 (0.222 g,
0.27 mmol) in THF (5 mL) was slowly added PPh3 (0.071 g, 0.27
mmol) dissolved in THF (5 mL). The original orange solution gradually
3
3
[(Li-Pr Cu)2(µ-NO2)]PF6. To a solution of [Li-Pr Cu(CH3CN)]PF6
(0.145 g, 0.287 mmol) in MeOH (10 mL) was added NaNO2 (0.140 g,
2.03 mmol) dissolved in MeOH (5 mL). The solution immediately
turned yellow, and a small amount of a dark solid precipitated. After
stirring for 15 min, the solvent was removed under vacuum, leaving
an orange residue. This residue was extracted with CH2Cl2 (15 mL),
the extracts were filtered, and the solvent was removed from the filtrate
to afford an orange solid. Crystallization from THF/pentane yielded
pure product as yellow-orange plates suitable for X-ray crystallographic
analysis (0.106 g, 89%): mp 178 °C dec; 1H NMR (CD2Cl2, 300 MHz)
δ 3.13 (m, 6H), 2.86-2.79 (m, 12H), 2.56-2.49 (m, 12H), 1.17 (d, J
) 6.3 Hz, 36H) ppm; 13C{1H} NMR (CD2Cl2, 75 MHz) δ 58.0, 51.0,
19.7 ppm; FTIR (KBr, cm-1) 2967, 1492, 1466, 1388, 1364, 1352 (1333
15NO2-), 1279, 1262, 1185 (1168 15NO2-), 1151, 1124, 1078, 1050,
1018, 966, 841 (PF6-), 749, 715, 663, 598, 557 (PF6-); FAB-MS
3
3
changed to bright yellow, and a white precipitate of [Li-Pr Cu(PPh3)]-
3
PF6 appeared. After filtering, the filtrate was evaporated to dryness.
The resulting yellow powder was dissolved in CH2Cl2 (5 mL) and
treated with Et2O (20 mL), causing the precipitation of additional
[Li-Pr Cu(PPh3)]PF6. The mixture was then filtered again, and the filtrate
3
volume was reduced to 3 mL causing the appearance of a small
additional amount of the phosphine complex, to give a total yield of
0.190 g (97%). Following a final filtration, the filtrate was evaporated
to dryness and the yellow residue was redissolved in a small amount
of THF. Diffusion of pentane into this THF solution resulted in the
crystallization of Li-Pr Cu(NO2) as yellow plates (0.062 g, 63%): mp
3
147 °C dec; 1H NMR (CD2Cl2, 300 MHz) δ 3.24 (septet, J ) 6.9 Hz,
3H), 2.88-2.81 (m, 6H), 2.52-2.45 (m, 6H), 1.25 (d, J ) 6.9 Hz,
18H) ppm; 13C{1H} NMR (CD2Cl2, 75 MHz) δ 57.8, 50.6, 19.2 ppm;
FTIR (KBr, cm-1) 2968, 2887, 2841, 1493, 1485, 1470, 1453, 1386,
1370, 1344, 1324, 1306 (1258 15NO2-), 1268 (1247 15NO2-), 1169,
1124, 1068, 1047, 1015, 963, 801, 755, 717, 583; FAB-MS (MNBA)
m/e (relative intensity) 364 ([M]+, 5), 318 ([M - NO2]+, 100). Anal.
Calcd for C15H33CuN4O2: C, 49.36; H, 9.11; N, 15.35. Found: C,
48.95; H, 8.96; N, 14.69.
(MNBA) m/e (relative intensity) 682 ([M - PF6]+, 20), 318 ([M -
+
Li-Pr Cu - NO2 - PF6] , 100). Anal. Calcd for C30H66Cu2F6N7O2P:
3
C, 43.47; H, 8.03; N, 11.83. Found: C, 43.57; H, 7.98; N, 11.81.
[(LBn Cu)2(µ-NO2)]PF6. A procedure identical to that for the
3
synthesis of [(Li-Pr Cu)2(µ-NO2)]PF6 was followed, except [L Cu(CH3-
CN)]PF6 (0.234 g, 0.36 mmol) was used as the starting material. This
precursor is insoluble in MeOH, but upon addition of the solution of
NaNO2, it dissolved and the reaction and workup proceeded as for the
Bn
3
3
i-Pr
[Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2. Solid Zn(O3SCF3)2 (0.091 g,
3
3
Li-Pr analog. The product was isolated as yellow-orange needles from
0.25 mmol) as added to a solution of Li-Pr (0.064 g, 0.25 mmol) in
3
3
CH2Cl2/Et2O (0.178 g, 89%): 1H NMR (acetone-d6, 300 MHz) δ 7.57-
7.55 (m, 6H), 7.26-7.24 (m, 9H), 4.02 (s, 6H), 2.83-2.67 (m, 6H),
2.73-2.66 (m, 6H) ppm; 13C{1H} NMR (acetone-d6, 75 MHz) δ 136.5,
131.7, 129.0, 128.7, 64.2, 52.6 ppm; FTIR (KBr, cm-1) 3087, 3059,
3028, 2998, 2969, 2928, 2775, 1496, 1453, 1347 (1331 15NO2-), 1332,
1287, 1202 (1174 15NO2-), 1127, 1104, 1093, 1075, 1060, 1002, 987,
924, 896, 872, 835 (PF6-), 786, 764, 731, 710, 702, 695. Anal. Calcd
for C54H66Cu2F6N7O2P: C, 58.05; H, 5.95; N, 8.78. Found: C, 57.40;
H, 5.87; N, 8.72.
CH2Cl2 (8 mL). After allowing the slurry to stir for 1 h, a solution of
Li-Pr Cu(NO2) (0.046 g, 0.126 mmol) in CH2Cl2 (2 mL) was added a
3
deep red color gradually developed. The solution was stirred overnight
and filtered, and Et2O was diffused into the filtrate to yield the product
as red crystals (0.095 g, 76%): 1H NMR (CD2Cl2, 500 MHz) δ 3.27-
3.20 (m, 6H), 3.04 (s, 12H), 2.91-2.86 (m, 6H), 2.67-2.62 (m, 6H),
1.22 (d, J ) 6.5 Hz, 18H), 1.18 (d, J ) 7.0 Hz, 18H); 13C NMR (CD2-
Cl2, MHz) δ 58.6, 57.5, 51.3, 49.3, 19.8, 18.8 ppm; FTIR (KBr, cm-1
)
2980, 2939, 1492, 1470, 1393, 1377, 1263 (O3SCF3-), 1222, 1149
(O3SCF3-), 1029 (O3SCF3-), 639 (O3SCF3-), 516; ISP-MS m/e (relative
intensity) 832 ([M - O3SCF3]+, 50), 359 (100). Anal. Calcd for
C32H66CuF6N7O8S2Zn: C, 39.13; H, 6.78; N, 9.99. Found: C, 39.18;
H, 6.82; N, 10.00.
[(Li-Pr Cu)2(µ-NO2)](PF6)2. To a solution of [(L Cu)2(µ-NO2)]PF6
(0.158 g, 0.190 mmol) in CH2Cl2 (5 mL) was added ferricenium
hexafluorophosphate (Cp2Fe)PF6, 0.063 g, 0.190 mmol). As the
oxidizing agent slowly dissolved, the color of the solution changed
from orange to deep red. After 30 min of stirring, Et2O (10 mL) was
added, causing the precipitation of a brown powder, which was
recrystallized from CH2Cl2/Et2O to yield the product as nearly black
cubes (0.140 g, 76%): mp 192 °C dec; FTIR (KBr, cm-1) 2978, 1491,
1474, 1389, 1351, 1235 (1214 15NO2-), 1170, 1070, 965, 838 (PF6-),
757, 720, 669, 557 (PF6-). Anal. Calcd for C30H66Cu2F12N7O2P2: C,
36.00; H, 6.83; N, 10.07. Found: C, 36.23; H, 6.72; N, 9.67.
i-Pr
3
3
Li-Pr Cu(O2CCH3)2‚H2O. To a solution of Li-Pr (0.036 g, 0.14
mmol) in MeOH (5 mL) was added solid Cu(O2CCH3)2‚H2O (0.0293
g, 0.15 mmol), causing the solution to become bright blue. Removal
of solvent followed by crystallization from THF/pentane at -20 °C
yielded the product as blue blocks (0.048 g, 77%): mp 112 °C dec;
FTIR (KBr, cm-1) 3445 (H2O), 2973, 1628 [ν(CdO)], 1571, 1501,
1471, 1380, 1320 [ν(CsO)], 1262, 1168, 1148, 1126, 1068, 1054, 1013,
3
3
i-Pr
3
[(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2. A solution of [(L
-
967, 844, 763, 721, 664, 647, 607; ΛM (CH3CN) 8(1) cm2 Ω-1 mol-1
FAB-MS (MNBA) m/e (relative intensity) 377.2 ([M - OAc]+, 100).
;
3
Cu)2(NO2)](PF6)2 (0.045 g, 0.046 mmol) in CH2Cl2 (5 mL) was treated

Copper(I)-Nitrite Complexes That EVolVe Nitric Oxide
J. Am. Chem. Soc., Vol. 118, No. 4, 1996 775
Anal. Calcd for C19H41CuN3O5: C, 50.15; H, 9.08; N, 9.23. Found:
C, 50.16; H, 9.02; N, 9.33. Although a water of hydration was not
present in the crystal subjected to crystallographic analysis, the complex
is rather hygroscopic and is formulated as a monohydrate on the basis
of FTIR and elemental analysis data.
a glass fiber with heavy-weight oil and transferred to the cold stream
of an Enraf-Nonius CAD-4 diffractometer with graphite monochromated
Mo KR radiation (λ ) 0.710 69 Å). Relevant crystallographic
information is given in Table 2. Cell constants were obtained from a
least squares refinement of the setting angles for 24 carefully centered
reflections in the range 21.00° < 2θ < 43.00°. The intensity data were
collected using the ω-2θ scan technique to a maximum 2θ value of
52.0°. An empirical absorption correction was applied, using the
program DIFABS,57 which resulted in transmission factors ranging from
0.78 to 1.20. The data were corrected for Lorentz and polarization
effects; no decay correction was applied.
Li-Pr Cu(O3SCF3)2 and [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3. To a
3
3
solution of Li-Pr (0.192 g, 0.752 mmol) in THF (5 mL) was added
3
solid Cu(O3SCF3)2 (0.260 g, 0.72 mmol). After being stirred for 2 h,
the mixture was filtered. The filtrate was treated with pentane (10 mL)
and stored at -20 °C to yield the product as green microcrystals (0.242
g, 55%): FTIR (KBr, cm-1) 2980, 1501, 1327, 1279 (O3SCF3-), 1241
(O3SCF3-), 1208 (O3SCF3-), 1172 (O3SCF3-), 1070, 1027 (O3SCF3-),
1024 (O3SCF3-), 961, 771, 724, 635 (O3SCF3-), 576, 518; ΛM (CH2-
Cl2), 2(1) cm2 Ω-1 mol-1; ΛM (acetone) 120(2) cm2 Ω-1 mol-1; ΛM
(CH3CN) 300(2) cm2 Ω-1 mol-1; FAB-MS (MNBA) m/e (relative
intensity) 467 ([M - O3SCF3]+, 75), 318 ([M - 2O3SCF3]+, 100).
Anal. Calcd for C17H33CuF6N3O6S2: C, 33.09; H, 5.39; N, 6.81.
Found: C, 33.23; H, 5.69; N, 6.90. Crystals suitable for X-ray
diffraction were obtained only when using wet solvents. Thus,
crystallization by vapor diffusion of Et2O into a 1:3 (v/v) mixture of
The structure was solved by direct methods58 using the TEXSAN
software package.59 Non-hydrogen atoms were refined anisotropically,
and the hydrogen atoms were placed at calculated positions. The
maximum and minimum peaks on the final difference Fourier map
corresponded to 0.65 and -0.54 e-/Å3, respectively. Neutral-atom
scattering factors60 and anomalous dispersion terms61 were taken from
the literature. Because the apparent space group is P1h, the cation is
end-for-end disordered and the Cu-N,O distances are based on average
positions of the atoms involved, with the exception of the nitrite atoms
(N1, O1, O2), which were located in both disordered positions (but
refined isotropically). There is also disorder in one of the isopropyl
groups (C13) and three of the CF3 groups. The isopropyl group occurs
in all three possible orientations and was modeled with occupancies
for three isopropyl groups of C13-C14-C15 (0.579), C13-C14-C16
(0.149), and C13-C15-C16 (0.272). For the CF3 groups, occupancies
are C117-F111-F112-F113 (0.65)/C119-F117-F118-F119 (0.35),
C128-F124-F125-F126 (0.75)/C129-F127-F128-F129 (0.25), and
C147-F141-F142-F143 (0.67)/C149-F147-F148-F149 (0.33). The
major component in each case was refined anisotropically, and the
minor component was treated as a rigid group with individual isotropic
displacement parameters. Drawings of the cation appear in Figures 5
and 6, and selected bond lengths and angles are presented in Table 1.
Complete structure diagrams and full tables of bond lengths and angles,
atomic positional parameters, and final thermal parameters for non-
hydrogen atoms are given in the supporting information.
acetone (wet)/THF yielded green plates of [Li-Pr Cu(H2O)(O3SCF3)]O3-
3
SCF3, as identified by X-ray crystallography and elemental analysis.
Calcd for C17H35CuF6N3O7S2: C, 32.15; H, 5.55; N, 6.62; S, 10.10.
Found: C, 32.11; H, 5.36; N, 6.62; S, 10.02.
[(Li-Pr Cu)2(µ-NO2)](O3SCF3)2. To a solution of Li-Pr Cu(NO2)
3
3
(0.0062 g, 0.017 mmol) in CH2Cl2 (2 mL) was added a solution of
Li-Pr Cu(O3SCF3)2 (0.0105 g, 0.017 mmol) in CH2Cl2 (2 mL). The
3
solution immediately turned deep red. Diffusion of Et2O directly into
this solution caused the crystallization of the product as small dark
needles in nearly quantitative yield. This compound exhibited UV-
-
vis and EPR spectra identical to those for the PF6 and [B(3,5-
(CF3)2C6H3)4]- salts described above: FTIR (KBr, cm-1) 2977, 2942,
2882, 1493, 1472, 1393, 1373, 1257 (O3SCF3-), 1227 (O3SCF3-), 1159
(O3SCF3-), 1067, 1046, 1028 (O3SCF3-), 964, 869, 842, 761, 720, 668,
637 (O3SCF3-), 572, 517. Anal. Calcd for C32H66Cu2F6N7O8S2: C,
39.14; H, 6.77; N, 9.98. Found: C, 39.13; H, 6.64; N, 9.81.
3
[Li-Pr Cu(H2O)(O3SCF3)]O3SCF3. A green plate crystal of the
3
i-Pr
3
Reaction of Li-Pr Cu(NO2) with CH3CO2H. A solution of L
-
complex was mounted on a glass fiber with heavy-weight oil and
transferred to the cold stream of the Enraf-Nonius CAD4 diffractometer.
Data collection was as described above. Important crystallographic
information is given in Table 2. Cell constants were obtained from a
least squares refinement of the setting angles for 20 carefully centered
reflections in the range 10.00° < 2θ < 48.00°. The intensity data were
collected using the ω-2θ scan technique to a maximum 2θ value of
56.0°. An empirical absorption correction was applied on the basis of
azimuthal scans of several reflections, which resulted in transmission
factors ranging from 0.72 to 1.00. The data were corrected for Lorentz
and polarization effects; no decay correction was applied. A correction
for secondary extinction was applied (coefficient ) 0.159 69 × 10-6).
The structure was solved by direct methods58 using the TEXSAN
software package.59 Non-hydrogen atoms were refined anisotropically,
and the hydrogen atoms were placed at calculated positions. A search
for hydrogen atoms on the coordinated water molecule (O1) was
unsuccessful; no attempt to place such atoms in idealized positions
was made. The maximum and minimum peaks on the final difference
Fourier map corresponded to 1.10 and -1.27 e-/Å3, respectively. An
ORTEP drawing of the structure appears in Figure 4, and selected bond
lengths and angles are presented in Table 1. Full tables of bond lengths
and angles, atomic positional parameters, and final thermal parameters
for non-hydrogen atoms are given in the supporting information.
Cu(NO2) (0.0044 g, 0.012 mmol) in CH2Cl2 (0.15 mL) was prepared
in a small vial capped with a rubber septum. A degassed solution of
CH3CO2H (1.8 µL) in CH2Cl2 (0.05 mL) was then introduced via
syringe at room temperature. The solution changed immediately from
yellow to blue. Analysis of the head space gas by GC indicated that
NO(g) had been generated (0.012 mmol, 100%). The copper-containing
product of the reaction was identified as Li-Pr Cu(O2CCH3)2 by
3
comparison of UV-vis and EPR spectra of the blue solution to those
of independently synthesized material.
Reaction of Li-Pr Cu(NO2) with Me3SiO3SCF3. A procedure
3
identical to that used for the reaction of Li-Pr Cu(NO2) with CH3CO2H
3
(above) was followed. Reaction of Li-Pr Cu(NO2) (0.0066 g, 0.018
3
mmol) in CH2Cl2 (0.15 mL) with Me3SiO3SCF3 (7 µL) in CH2Cl2 (0.05
mL) at room temperature caused a color change from yellow to green.
GC analysis of the head space gas indicated NO(g) generation (0.013
mmol, 70%). The copper-containing product of the reaction was
identified as Li-Pr Cu(O3SCF3)2 by comparison of UV-vis and EPR
3
spectra of the green solution to those of independently synthesized
material. Upon addition of Me3SiO3SCF3 to a solution of Li-Pr Cu-
3
(NO2) in CH2Cl2 at -80 °C, a red color developed which ultimately
changed to green upon warming; UV-vis data acquired at -80 °C
matched those obtained for independently synthesized [(Li-Pr Cu)2(µ-
3
NO2)](O3SCF3)2.
i-Pr
[Li-Pr Cu(µ-NO2)ZnL ](O3SCF3)2. A dark red block-shaped
crystal of the compound was attached to a glass fiber and mounted on
a Siemens SMART system for data collection at 173(2) K. An initial
3
3
Reaction of [(Li-Pr Cu)2(µ-NO2)](O3SCF3)2 with Me3SiO3SCF3. A
3
procedure identical to that described above was followed except using
the dinuclear starting material. Reaction of [(Li-Pr Cu)2(µ-NO2)](O3-
3
SCF3)2 (0.0061 mg, 0.0062 mmol) in CH2Cl2 (0.15 mL) with Me3-
SiO3SCF3 (3 µL) in CH2Cl2 (0.05 mL) at room temperature caused a
color change from red to green. GC analysis of the head space gas
indicated NO(g) generation (0.0042 mmol, 68%). The copper-
(57) Walker, N.; Stuart, D. Acta Crystallogr. 1983, A39, 158-166.
(58) Calabrese, J. C. PHASE: Patterson Heavy Atom Solution Extractor.
Ph.D. Thesis, University of Wisconsin, Madison, WI, 1972.
(59) TEXSAN-Texray Structure Analysis Package; Molecular Structure
Corp.: The Woodlands, TX, 1985.
(60) Cromer, D. T. International Tables for X-ray Crystallography; The
Kynoch Press: Birmingham, U.K., 1974; Vol. IV, Table 2.2A.
(61) (a) Ibers, J. A.; Hamilton, W. C. Acta Crystallogr. 1964, 17, 781.
(b) Cromer, D. T. International Tables for X-ray Crystallography; The
Kynoch Press: Birmingham, U.K., 1974; Vol. IV, Table 2.3.1.
containing product of the reaction was identified as Li-Pr Cu(O3SCF3)2
3
by comparison of UV-vis and EPR spectra of the green solution to
those of independently synthesized material.
X-ray Crystallography. [(Li-Pr Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2.
3
An irregularly shaped brown crystal of the complex was mounted on

776 J. Am. Chem. Soc., Vol. 118, No. 4, 1996
Halfen et al.
set of cell constants was calculated from reflections harvested from
three sets of 30 frames. These initial sets of frames were oriented such
that orthogonal wedges of reciprocal space were surveyed, thus
producing orientation matrices determined from 196 reflections. Final
cell constants were calculated from a set of 8192 strong reflections
from the actual data collection. Final cell constants reported in this
manner usually are about 1 order of magnitude better in precision than
those reported from four-circle diffractometers. The data collection
technique used for this specimen is generally known as a hemisphere
collection. A randomly oriented region of reciprocal space was
surveyed to the extent of 1.3 hemispheres to a resolution of 0.87 Å.
Three swaths of frames were collected with 0.30° steps in ω. This
collection strategy provides a high degree of redundancy in the data
and good ψ input in the event a semiempirical absorption correction is
applied; here, a semiempirical correction afforded minimum and
maximum transmission factors of 0.624 and 0.871, respectively. See
Table 2 for additional crystal and refinement information.
The space group P21/c was determined on the basis of systematic
absences and intensity statistics, and a successful direct-methods
solution was calculated which provided most nonhydrogen atoms from
the E map. The metric parameter â is close to 90.0°, but the Lane
symmetry was determined to be 2/m and not mmm. Several full-matrix
least squares/difference Fourier cycles were performed which located
the remainder of the nonhydrogen atoms, all of which were refined
with anisotropic displacement parameters. All hydrogen atoms were
placed in ideal positions and refined as riding atoms with individual
isotropic displacement parameters. Two isopropyl groups were found
to be disordered; these groups are atoms C7, C8, and C9 and C10,
C11, and C12, respectively. The requisite atoms were refined such
that the occupancy equaled 1.0. C11 and C12 coincidentally appear
in the same location in both orientations; C11′ and C12′ are entered in
the atom list only to provide proper geometry for the calculation of
the associated tertiary riding hydrogen. Drawings of the cation appear
in Figures 5 and 6, and selected bond lengths and angles are presented
in Table 1. Full details of the structure determination, including tables
of bond lengths and angles, atomic positional parameters, and final
thermal parameters for non-hydrogen atoms, are given in the supporting
information. All calculations were performed using SGI INDY R4400-
SC or Pentium computers with the SHELXTL-Plus V5.0 program
suite.62
Acknowledgment. Funding in support of this research was
provided by the National Institutes of Health (GM47365 to
W.B.T., a SURE Supplement to A.G., GM33162 to L.Q., and
a predoctoral traineeship to E.C.W.), the Searle Scholars
Program/Chicago Community Trust, the University of Min-
nesota Undergraduate Research Opportunities Program (A.G.),
and the National Science Foundation (National Young Inves-
tigator Award to W.B.T. and CHE-9413114 for partial funding
for the purchase of the Siemens SMART system). W.B.T. is a
recipient of research fellowships from the Alfred P. Sloan and
Camille and Henry Dreyfus Foundations. We thank Professor
Doyle Britton for his work on the X-ray structures of [(Li-Pr
-
3
Cu)2(µ-NO2)][B(3,5-(CF3)2C6H3)4]2 and [Li-Pr Cu(H2O)(O3-
3
SCF3)]O3SCF3 and Dr. Marilyn M. Olmstead for solving the
X-ray structure of [(Li-Pr Cu)2(µ-NO2)]PF6 (cited in ref 21a).
3
Supporting Information Available: Complete drawings and
full details of the structure determinations, including tables of
bond lengths and angles, atomic positional parameters, and final
thermal parameters for all atoms for [(Li-Pr Cu)2(µ-NO2)][B(3,5-
3
i-Pr3
(CF3)2C6H3)4]2, [Li-Pr Cu(H2O)(O3SCF3)]O3SCF3, and [L Cu-
3
(µ-NO2)ZnLi-Pr ](O3SCF3)2 (51 pages). This material is con-
3
tained in many libraries on microfiche, immediately follows this
article in the microfilm version of the journal, can be ordered
from the ACS, and can be downloaded from the Internet; see
any current masthead page for ordering information and Internet
access instructions.
JA952691I
(62) SHELXTL-Plus V5.0, Siemens Industrial Automation, Inc., Madi-
son, WI.