A definitive example of a geometric entatic state effect: Electron-transfer kinetics for a copper(ll/l) complex involving a quinquedentate macrocyclic trithiaether - Bipyridine ligand

Article abstract of DOI:10.1021/ja068960u

The quinquedentate macrocyclic ligand cyclo-6,6′-[1,9-(2,5,8- trithianonane)]-2,2′-bipyridine ([15]aneS₃bpy = L), containing two pyridyl nitrogens and three thiaether sulfurs as donor atoms, has been synthesized and complexed with copper. The Cu^II/I redox potential, the stabilities of the oxidized and reduced complex, and the oxidation and reduction electron-transfer kinetics of the complex reacting with a series of six counter reagents have been studied in acetonitrile at 25°C, μ = 0.10 M (NaCIO₄). The Marcus cross relationship has been applied to the rate constants obtained for the reactions with each of the six counter reagents to permit the evaluation of the electron self-exchange rate constant, k ₁₁. The latter value has also been determined independently from NMR line-broadening experiments. The cumulative data are consistent with a value of k₁₁ = 1 × 10⁵ M^-1 s¹, ranking this among the fastest-reacting Cu^II/I systems, on a par with the blue copper proteins known as cupredoxins. The resolved crystal structures show that the geometry of the Cu^IIL and Cu^IL complexes are nearly identical, both exhibiting a five-coordinate square pyramidal geometry with the central sulfur donor atom occupying the apical site. The most notable geometric difference is a puckering of an ethylene bridge between two sulfur donor atoms in the Cu^IL complex. Theoretical calculations suggest that the reorganizational energy is relatively small, with the transition-state geometry more closely approximating the geometry of the Cu^IIL ground state. The combination of a nearly constant geometry and a large self-exchange rate constant implies that this Cu^II/I redox system represents a true geometric entatic state.

Full text of DOI:10.1021/ja068960u

Published on Web 03/29/2007
A Definitive Example of a Geometric “Entatic State” Effect:
Electron-Transfer Kinetics for a Copper(II/I) Complex Involving
A Quinquedentate Macrocyclic Trithiaether-Bipyridine Ligand
Gezahegn Chaka,1a Jason L. Sonnenberg,1a H. Bernhard Schlegel,1a
Mary Jane Heeg,1a Gregory Jaeger,1b Timothy J. Nelson,1b
L. A. Ochrymowycz,1b and D. B. Rorabacher*,1a
Contribution from the Departments of Chemistry, Wayne State UniVersity, Detroit, Michigan
48202, and the UniVersity of Wisconsin at Eau Claire, Eau Claire, Wisconsin 54701
Received December 14, 2006; E-mail: dbr@chem.wayne.edu
Abstract: The quinquedentate macrocyclic ligand cyclo-6,6′-[1,9-(2,5,8-trithianonane)]-2,2′-bipyridine
([15]aneS3bpy ) L), containing two pyridyl nitrogens and three thiaether sulfurs as donor atoms, has been
synthesized and complexed with copper. The CuII/IL redox potential, the stabilities of the oxidized and
reduced complex, and the oxidation and reduction electron-transfer kinetics of the complex reacting with
a series of six counter reagents have been studied in acetonitrile at 25 °C, µ ) 0.10 M (NaClO4). The
Marcus cross relationship has been applied to the rate constants obtained for the reactions with each of
the six counter reagents to permit the evaluation of the electron self-exchange rate constant, k11. The
latter value has also been determined independently from NMR line-broadening experiments. The cumulative
data are consistent with a value of k11 ) 1 × 105 M-1 s-1, ranking this among the fastest-reacting CuII/I
systems, on a par with the blue copper proteins known as cupredoxins. The resolved crystal structures
show that the geometry of the CuIIL and CuIL complexes are nearly identical, both exhibiting a five-coordinate
square pyramidal geometry with the central sulfur donor atom occupying the apical site. The most notable
geometric difference is a puckering of an ethylene bridge between two sulfur donor atoms in the CuIL
complex. Theoretical calculations suggest that the reorganizational energy is relatively small, with the
transition-state geometry more closely approximating the geometry of the CuIIL ground state. The
combination of a nearly constant geometry and a large self-exchange rate constant implies that this CuII/I
redox system represents a true geometric “entatic state.”
inorganic CuII/I complexes exhibit significantly smaller k11 values
in the range of 100-104 M-1 s-1;6 and Jordan and co-workers
Introduction
Much attention has been given to the unusual properties
exhibited by the type-1 copper sites in blue copper proteins
including their intense visible absorption peaks, unusual EPR
hyperfine splitting, and high redox potentials.2 However, from
a functional standpoint, their most notable property is their rapid
electron-transfer kinetics.3,4 NMR line-broadening measurements
and related techniques have shown that the electron self-
exchange rate constants (k11) for several blue copper proteins
(where L represents the coordinated protein)
have found that the solvated CuII/I couple in water and in
acetonitrile has values of k11 < 10-6 M-1 s-1 7
.
Vallee and Williams were the first to emphasize that the
electron-transfer lability of the blue copper proteins and other
(5) (a) Dahlin, S.; Reinhammar, B.; Wilson, M. T. Biochem. J. 1984, 218,
609-614. (b) Groeneveld, C. M.; Canters, G. W. Eur. J. Biochem. 1985,
153, 559-564. (c) Groeneveld, C. M.; Dahlin, S.; Reinhammar, B.; Canters,
G. W. J. Am. Chem. Soc. 1987, 109, 3247-3250. (d) Groeneveld, C. M.;
Ouwerling, M. C.; Erkelens, C.; Canters, G. W. J. Mol. Biol. 1988, 200,
189-199. (e) Groeneveld, C. M. Canters, C. W. J. Biol. Chem. 1988, 263,
167-173. (f) Lommen, A.; Canters, G. W. J. Biol. Chem. 1990, 265, 2768-
2774. (g) Dennison, C.; Kyritsis, P.; McFarlane, W.; Sykes, A. G. J. Chem.
Soc., Dalton Trans. 1993, 1959-1963. (h) Dennison, C.; Kohzuma, T.;
McFarlane, W.; Suzuki, S.; Sykes, A. G. J. Chem. Soc., Dalton Trans.
1994, 437-443. (i) Kyritsis, P.; Dennison, C.; Ingledew, W. J.; McFarlane,
W.; Sykes, A. G. Inorg. Chem. 1995, 34, 5370-5374. (j) Dennison, C.;
Van, Driessche, G.; Van Beeumen, J.; McFarlane, W.; Sykes, A. G. Chem.s
Eur. J. 1996, 2, 104-109. (k) Kyritsis, P.; Khozhuma [sic], T.; Sykes, A.
G. Biochim. Biophys. Acta 1996, 1295, 245-252. (l) Dennison, C.;
Kohzuma, T. Inorg. Chem. 1999, 38, 1491-1497. (m) Hunter, D. M.;
McFarlane, W.; Sykes, A. G.; Dennison, C. Inorg. Chem. 2001, 40, 354-
360. (n) van Amsterdam, I. M. C.; Ubbink, M.; Einsle, O.; Messerschmidt,
A.; Merli, A.; Cavazzini, D.; Rossi, G. L.; Canters, G. W. Nat. Struct.
Biol. 2002, 9, 48-52. (o) Dennison, C.; Lawler, A. T.; Kohzuma, T.
Biochemistry 2002, 41, 552-560. (p) Sato, K.; Dennison, C. Biochemistry
2002, 41, 120-130. (q) Sato, K.; Kohzuma, T.; Dennison, C. J. Am. Chem.
Soc. 2003, 125, 2101-2112. (r) Harrison, M. D.; Dennison, C. ChemBio-
Chem. 2004, 5, 1579-1581.
k11
\
*CuIIL + CuIL y
z *CuIL + CuIIL
(1)
containing a single type-1 copper site (cupredoxins) lie within
the range of 105-106 M-1 s-1 5,6
By contrast, the majority of
.
(1) (a) Wayne State University. (b) University of Wisconsin - Eau Claire.
(2) Dennison, C. Coord. Chem. ReV. 2005, 249, 3025-3054.
(3) Vallee, B. L.; Williams, R. J. P. Proc. Natl. Acad. Sci. U.S.A. 1968, 59,
498-505.
(4) (a) Holwerda, R. A.; Wherland, S.; Gray, H. B. Ann. ReV. Biophys. Bioeng.
1976, 5, 363-396. (b) Sykes, A. G. Chem. Soc. ReV. 1985, 14, 283-315.
(c) Sykes, A. G. Struct. Bonding 1991, 75, 175-224. (d) Katz, A. K.;
Shimoni-Livny, L.; Navon, O.; Navon, N.; Bock, C. W.; Glusker, J. P.
HelV. Chim. Acta 2003, 86, 1320-1338.
(6) Rorabacher, D. B. Chem. ReV. 2004, 104, 651-697.
9
10.1021/ja068960u CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 5217-5227
5217

A R T I C L E S
Chaka et al.
metalloenzymes might be the result of a structural entatic (i.e.,
strained) state in which the protein matrix constrains the metal
coordination sphere to adopt a geometry closely approximating
the transition state normally encountered in similar uncon-
strained systems, thereby minimizing the reorganization en-
ergy.3,8 Subsequent structural determinations for several
cupredoxins9-13 have revealed that, in both the oxidized and
reduced forms, the copper exists in a trigonal pyramidal
coordination environment in which two unsaturated nitrogens
(from histidines) and a mercaptide sulfur (cysteine) form the
basal plane with an elongated bond to a thiaether sulfur
(methionine) at the apical position.14,15 In some cupredoxins,
such as azurins, a carbonyl oxygen (from glycine) is coordinated
at the opposite apex to generate a distorted trigonal bipyramidal
coordination sphere which is also maintained in both oxidation
states.11 A related red copper protein has a five-coordinate
copper site exhibiting a square pyramidal coordination geom-
etry.16
doxins are primarily attributable to the covalent nature of the
Cu-S(mercaptide) bond in the basal plane of the active site.
They also found that the electronic effects arising from this
strong Cu-S bond give rise to the unusual geometry associated
with the type-1 Cu site and also facilitate rapid electron transfer.
The Cu atom can thus be viewed as existing in an electronic
entatic state rather than a simple geometric constraint physically
imposed upon the active site by the protein matrix.
The question remains as to whether geometric constraints
alone can lead to rapid electron transfer in CuII/I systems. Several
attempts have been reported in the literature, by Stanbury19-23
and others,24,25 to generate CuII/I complexes maintaining nearly
constant geometries in both oxidation states. To date, none of
these systems have yielded CuII/I self-exchange rate constants
comparable to those exhibited by the rapidly reacting cupre-
doxins, and some have, in fact, yielded extremely small k11
values.19,20 By contrast, of the inorganic copper complexes that
26-30
have been reported to exhibit the largest k11 values,
all
Both Solomon17 and Ryde18 and their co-workers have carried
out theoretical calculations from which they concluded the
spectral properties and electron-transfer kinetics of the cupre-
but one30 involve the rupture of a single Cu-S(thiaether) bond
upon reduction, a geometric change that is apparently quite
facile.
(7) (a) Sisley, M. J.; Jordan, R. B. Inorg. Chem. 1992, 31, 2880-2884. (b)
Irangu, J.; Ferguson, M. J.; Jordan, R. B. Inorg. Chem. 2005, 44, 1619-
1625.
In the current investigation, we have generated a quinque-
dentate ligand incorporating three thiaether sulfurs and a
bipyridine moiety within a macrocyclic structure, namely, cyclo-
6,6′-[1,9-(2,5,8-trithianonane)]-2,2′-bipyridine ([15]aneS3bpy )
L).31 The coordination geometry of the oxidized complex, [CuII-
(8) Williams, R. J. P. Inorg. Chim. Acta ReV. 1971, 5, 137-155. Compare:
Williams, R. J. P. Eur. J. Biochem. 1995, 234, 363-381.
(9) Colman, P. M.; Freeman, H. C.; Guss, J. M.; Murata, M.; Norris, V. A.;
Ramshaw, J. A. M.; Venkatappa, M. P. Nature 1978, 272, 319-324.
(10) Plastocyanins: (a) Guss, J. M.; Freeman, H. C. J. Mol. Biol. 1983, 169,
521-563. (b) Guss, J. M.; Harrowell, P. R.; Murata, M.; Norris, V. A.;
Freeman, H. C. J. Mol. Biol. 1986, 192, 361-387. (c) Collyer, C. A.; Guss,
J. M.; Sugimura, Y.; Yoshizaki, F.; Freeman, H. C. J. Mol. Biol. 1990,
211, 617-632. (d) Guss, J. M.; Bartunik, H. D.; Freeman, H. C. Acta
Crystallogr., Sect. B 1992, 48, 790-811. (e) Fields, B. A.; Bartsch, H. H.;
Bartunik, H. D.; Cordes, F.; Guss, J. M.; Freeman, H. C. Acta Crystallogr.,
Sect. D 1994, 50, 709-730. (f) Bond, C. S.; Bendall, D. S.; Freeman, H.
C.; Guss, J. M.; Howe, C. J.; Wagner, M. J.; Wilce, M. C. J. Acta
Crystallogr., Sect. D 1999, 55, 414-421.
(11) Azurins: (a) Baker, E. N. J. Mol. Biol. 1988, 203, 1071-1095. (b) Shepard,
W. E. B.; Anderson, B. F.; Lewandowski, D. A.; Norris, G. E.; Baker, E.
N. J. Am. Chem. Soc. 1990, 112, 7817-7819. (c) Nar, H.; Messerschmidt,
A.; Huber, R.; van de Kamp, M.; Canters, G. W. J. Mol. Biol. 1991, 221,
765-772. (d) Li, C. M.; Inoue, T.; Gotowda, M.; Suzuki, S.; Yamaguchi,
K.; Kai, K.; Kai, Y. Acta Crystallogr., Sect. D 1998, 54, 347-354. (e)
Dodd, F. E.; Abraham, Z. H. L.; Eady, R. R.; Hasnain, S. S. Acta
Crystallogr., Sect. D 2000, 56, 690-696.
([15]aneS3bpy)]2+, is square pyramidal, paralleling the geometry
of two rapidly reacting macrocyclic complexes, [CuII([15]-
(18) (a) Olsson, M. H. M.; Ryde, U.; Roos, B. O. Protein Sci. 1998, 7, 2659-
2668. (b) De Kerpel, J. O. A.; Ryde, U. Proteins: Struct., Func., Genet.
1999, 36, 157-174. (c) Ryde, U.; Olsson, M. H. M.; Roos, B. O.; De
Kerpel, J. O. A.; Pierloot, K. J. Biol. Inorg. Chem. 2000, 5, 565-574. (d)
Ryde, U.; Olsson, M. H. M. Int. J. Quantum Chem. 2001, 81, 335-347.
(e) Ryde, U.; Olsson, M. H. M.; Roos, B. O.; Borin, A. C. Theor. Chem.
Acc. 2001, 105, 452-462. (f) Sigfridsson, E.; Olsson, M. H. M.; Ryde, U.
J. Phys. Chem. B 2001, 105, 5546-5552.
(12) Stellacyanin: Hart, P. J.; Nersissian, A. M.; Herrmann, R. G.; Nalbandyan,
R. M.; Valentine, J. S.; Eisenberg, D. Protein Sci. 1996, 5, 2175-2183.
(13) Other cupredoxins: (a) Pseudozaurin: Petratos, K.; Dauter, Z.; Wilson,
K. S. Acta Crystallogr., Sect. B 1988, 44, 628-636. (b) Phytocyanin
(cucumber blue protein): Guss, J. M.; Merritt, E. A.; Phizackerley, R. P.;
Freeman, H. C. J. Mol. Biol. 1996, 262, 686-705. Cf., Guss, J. M.; Merritt,
E. A.; Phizackerley, R. P.; Hedman, B.; Murata, M.; Hodgson, K. O.;
Freeman, H. C. Science 1988, 241, 806-811. (c) Rusticyanin: Harvey, I.;
Hao, Q.; Duke, E. M. H.; Ingledew, W. J.; Hasnain, S. S. Acta Crystallogr.,
Sect. D 1998, 54, 629-635. (d) Plantacyanin: Einsle, O.; Mehrabian, Z.;
Nalbandyan, R.; Messerschmidt, A. J. Biol. Inorg. Chem. 2000, 5, 666-
672. (e) Amicyanin: Mukherjee, M.; Maiti, S.; Ghosh, S.; Woolfson, M.
M. Acta Crystallogr., Sect. D 2001, 57, 1276-1280. (f) Auracyanin: Bond,
C. S.; Blankenship, R. E.; Freeman, H. C.; Guss, J. M.; Maher, M. J.;
Selvaraj, F. M.; Wilce, M. C. J.; Willingham, K. M. J. Mol. Biol. 2001,
306, 47-67. Cf., Lee, M.; Maher, M. J.; Freeman, H. C.; Guss, J. M. Acta
Crystallogr., Sect. D 2003, 59, 1545-1550.
(19) Xie, B.; Elder, T.; Wilson, L. J.; Stanbury, D. M. Inorg. Chem. 1999, 38,
12-19.
(20) Xie, B.; Wilson, L. J.; Stanbury, D. M. Inorg. Chem. 2001, 40, 3606-
3614.
(21) Goodwin, J. A.; Stanbury, D. M.; Wilson, L. J.; Eigenbrot, C. W.; Scheidt,
W. R. J. Am. Chem. Soc. 1987, 109, 2979-2991.
(22) Goodwin, J. A.; Wilson, L. J.; Stanbury, D. M.; Scott, R. A. Inorg. Chem.
1989, 28, 42-50.
(23) Coggin, D. K.; Gonza´lez, J. A.; Kook, A. M.; Bergman, C.; Brennan, T.
D.; Scheidt, W. R.; Stanbury, D. M.; Wilson, L. J. Inorg. Chem. 1991, 30,
1125-1134.
(24) Metelski, P. D.; Hinman, A. S.; Takagi, H. D.; Swaddle, T. W. Can. J.
Chem. 1995, 73, 61-69.
(14) Adman, E. T. AdV. Protein Chem. 1991, 42, 145-197.
(15) In stellacyanin, a glutamine oxygen occupies the apical site; see ref 12.
(16) Lieberman, R. L.; Arciero, D. M.; Hooper, A. B.; Rosenzweig, A. C.
Biochemistry 2001, 40, 5674-5681.
(25) McMaster, J.; Beddoes, R. L.; Collison, D.; Eardley, D. R.; Helliwell, M.;
Garner, C. D. Chem. Eur. J. 1996, 2, 685-693.
(26) Vande Linde, A. M. Q.; Juntunen, K. L.; Mols, O.; Ksebati, M. B.;
Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem. 1991, 30, 5037-
5042.
(17) (a) Solomon, E. I.; Lowery, M. D. Science 1993, 259, 1575-1581. (b)
Guckert, J. A.; Lowery, M. D.; Solomon, E. I. J. Am. Chem. Soc. 1995,
117, 2817-2844. (c) Solomon, E. I.; Penfield, K. W.; Gewirth, A. A.;
Lowery, M. D.; Shadle, S. E.; Guckert, J. A.; LaCroix, L. B. Inorg. Chim.
Acta 1996, 243, 67-78. (d) Solomon, E. I.; LaCroix, L. B.; Randall, D.
W. Pure Appl. Chem. 1998, 70, 799-808. (e) Solomon, E. I.; Lowery, M.
D.; Guckert, J. A.; LaCroix, L. B. In Electron Transfer Reactions; Advances
in Chemistry Series, Vol. 253; American Chemical Society: Washington,
DC, 1997; pp 317-330. (f) Randall, D. W.; Gamelin, D. R.; LaCroix, L.
B.; Solomon, E. I. J. Biol. Inorg. Chem. 2000, 5, 16-29. (g) Randall, D.
W.; George, S. D.; Hedman, B.; Hodgson, K. O.; Fujisawa, K.; Solomon,
E. I. J. Am. Chem. Soc. 2000, 122, 11620-11631. (h) Solomon, E. I.;
Szilagyi, R. K,; George, S. D.; Basumallick, L. Chem. ReV. 2004, 104,
419-458.
(27) Vande Linde, A. M. Q.; Westerby, B. C.; Ochrymowycz, L. A.; Rorabacher,
D. B. Inorg. Chem. 1993, 32, 251-257.
(28) Krylova, K.; Kulatilleke, C. P.; Heeg, M. J.; Salhi, C. A.; Ochrymowycz,
L. A.; Rorabacher, D. B. Inorg. Chem. 1999, 38, 4322-4328.
(29) Leggett, G. H.; Dunn, B. C.; Vande Linde, A. M. Q.; Ochrymowycz, L.
A.; Rorabacher, D. B. Inorg. Chem. 1993, 32, 5911-5918.
(30) Pulliam, E. J.; McMillin, D. R. Inorg. Chem. 1984, 23, 1172-1175: TAAB
) tetrabenzo[b,f,j,n]-[1,5,9,13]tetraazacyclohexadecine.
(31) This ligand can also be designated as 2,5,8[9](6,6′)2,2′-bipyridinophane
based on the nomenclature adopted by Weber and Vo¨gtle for similar
macrocyclic ligands containing pyridines and thiaether sulfurs: Weber, E.;
Vo¨gtle, F. Liebigs Ann. Chem. 1976, 891-915.
9
5218 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Geometric “Entatic State” Effect
A R T I C L E S
basal plane, and a crystal structure of the reduced [CuI([15]aneS3-
bpy)]+ complex shows the geometry to be five-coordinate, little
changed from that of the oxidized species. The nearly invariant
geometry for the two oxidation states meets the criterion defined
by Vallee and Williams in their original concept of a geometric
entatic state.3 Reduction and oxidation cross-reaction kinetic
studies as well as NMR line-broadening measurements have
been conducted on this system in acetonitrile, yielding the largest
self-exchange rate constant reported to date for a coordination-
invariant CuII/I system devoid of a Cu-mercaptide sulfur bond.
In support of the current investigation, theoretical calculations
employing density functional theory (DFT) were carried out to
determine the relative energy barrier associated with the small
ligand reorganization accompanying electron transfer in
[CuII/I([15]aneS3bpy)]2+/+. The results of these calculations
provide insight into the probable reaction pathway for the overall
electron-transfer reaction. Additional calculations have also been
made on previously reported coordination-invariant CuII/I sys-
tems for comparative purposes.
Experimental Section
Ligand Synthesis. The synthesis of [15]aneS3bpy was carried out
using the generalized high-dilution cyclocondensation methodology
employed to prepare macrocyclic polythiaethers, including the general
separation and characterization techniques previously described.33 The
essential synthones6,6′-bis(chloromethyl)-2,2′-bipyridineswas pre-
pared by chlorination of 6,6′-dimethyl-2,2′-bipyridine,34 which in turn
was prepared by the in situ generated nickel(0) coupling of 2-bromo-
6-methylpyridine.35 In 0.5 L of anhydrous DMF were dissolved 16.0 g
(0.063 mol) of 6,6′-bis(chloromethyl)-2,2′-bipyridine along with 9.8 g
(0.063 mol) of 2-mercaptoethyl sulfide (Aldrich Chemical Co., #M400-
7). The solution was added under an argon atmosphere at a rate of 0.5
mL/min to a stirred suspension of 20.7 g (0.15 mol) of powdered
anhydrous potassium carbonate in 1 L of DMF, maintained at 85-100
°C for the reaction duration. The filter cake and vacuum evaporation
residue from the reaction were dispersed in 1 L of methylene chloride
and washed with three 300-mL portions of 2 M sodium chloride brine,
dried with MgSO4, and vacuum filtered and reconcentrated to 21 g of
semisolid amber residue. The residue was flash eluted to remove higher
polymers with 1.3 L of toluene through a 3 × 40 cm silica gel column.
The 13 g of recovered elution residue were fractionally chromato-
graphed with 50:50 toluene/hexane through a 3 cm × 50 cm silica gel
column. By TLC analysis the major product fractions, Rf ) 0.55, 5:95
ethyl acetate/toluene, were combined and recrystallized twice from 30:
70 toluene/hexane to yield 5.13 g (24.4%) of colorless plates, mp )
149-151 °C. 13C NMR (100.6 MHz, CDCl3) δ in ppm (multiplicity):
31.99 (t), 32.79 (t), 36.63 (t), 120.01 (d), 122.54 (d), 137.74 (d), 155.10
(s), 160.51 (s). EI-MS, m/z (relative intensity): 334 M+ (67), 301 (14),
273 (88), 248 (100), 214 (78), 184 (33), 154 (19); electrospray MS,
m/z: 335.100 MH+, 357.06 MNa+, 373.04 MK+, 691.00 M2Na+. Anal.
Calcd for C16H18N2S3: C ) 57.45; H ) 5.42, S ) 28.76. Found: C )
57.22, H ) 5.40; S ) 28.50.
Figure 1. Ligands mentioned in this work.
Figure 2. Schematic representation of geometric changes observed to
accompany electron transfer for (A) [CuII/I([14]aneS4)] (top) and (B)
CuII/I([15]aneS5)]. Atom designations are as follows: black ) Cu, red )
S, blue ) O, white ) C. Hydrogen atoms have been omitted for clarity.
Reagents. The reducing agent [CoII(dmg)3(BC6H5)2] (dmg ) di-
methylglyoxime; BC6H52+ ) phenylborate ion) was prepared using the
method of Borchardt and Wherland36 as modified by Stanbury and co-
workers.19 All other counter reagents were prepared by literature
aneS5)]2+ and [CuII([15]aneNS4)]2+ 27,32 (see Figure 1) which
are among the handful of inorganic systems exhibiting values
of k11 g 105 M-1 s-1. Both of these latter complexes involve
the rupture of a labile Cu-S(thiaether) bond in the front of the
basal plane (Figure 2) to generate a tetrahedral CuIL species.26,27
In the current system, the more rigid bipyridine moiety makes
it difficult to rupture a Cu-N bond in the frontal portion of the
(33) Aronne, L.; Dunn, B. C.; Vyvyan, J. R.; Souvignier, C. W.; Mayer, M. J.;
Howard, T. A.; Salhi, C. A.; Goldie, S. N.; Ochrymowycz, L. A.;
Rorabacher, D. B. Inorg. Chem. 1995, 34, 357-369.
(34) Newkome, G. R.; Puckett, W. E.; Kiefer, G. E.; Gupta, V. K.; Xia, Y.;
Coreil, M.; Hackney, M. A. J. Org. Chem. 1982, 47, 4116-4120.
(35) Tiecco, M.; Testaferri, L.; Tingoli, M.; Chianelli, D.; Montanucci, M.
Synthesis 1984, 736-738.
(32) Corfield, P. W. R.; Ceccarelli, C.; Glick, M. D.; Moy, I. W.-Y.;
Ochrymowycz, L. A.; Rorabacher, D. B. J. Am. Chem. Soc. 1985, 107,
2399-2404.
(36) Borchardt, D.; Wherland, S. Inorg. Chem. 1986, 25, 901-905.
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5219

A R T I C L E S
Chaka et al.
methods as previously described.37-39 Copper(II) perchlorate was
prepared by the slow addition of perchloric acid to solid copper
carbonate40 on an ice bath and was then recrystallized from acetonitrile
to remove the waters of hydration. (Warning! All metal perchlorate
salts are potentially explosiVe and should be handled with extreme care.
These compounds should be prepared in small quantities and should
neVer be dried.) Copper(I) perchlorate was prepared by crystallization
from acetonitrile using the method of Hathaway et al.41 All aqueous
solutions were prepared using conductivity grade distilled-deionized
water. Acetonitrile solutions were prepared with HPLC grade solvent
(Fisher Scientific) which has been shown to contain less than 0.02%
water (by wt).38 The addition of small amounts of added water had no
effect upon the experimental results; therefore, no attempt was made
to reduce the residual water content. The concentrations of all counter
reagent solutions were determined spectrophotometrically. Copper
solutions in both water and acetonitrile were first oxidized to CuII and
then standardized by EDTA titration. Ligand solutions were standard-
ized by potentiometric titrations with a standard mercury(II) solution
using a mercury pool indicating electrode.
X-Ray Structural Determinations. Diffraction data for [CuII-
([15]aneS3bpy)](ClO4)2 were collected on a Bruker P4/CCD diffrac-
tometer equipped with Mo radiation and a graphite monochromator at
room temperature. Data collection proceeded on a red needle ap-
proximately 0.4 × 0.06 × 0.04 mm3. A sphere of data was measured
at 10 s/frame and 0.3° between frames. A total of 1650 frames was
collected, yielding 14834 reflections, of which 4876 were independent.
The frame data were indexed and integrated with the manufacturer’s
SMART, SAINT, and SADABS software.42 All structures were refined
using Sheldrick’s SHELX-97 software.43 The hydrogen atoms were
placed in calculated positions. All samples examined exhibited some
degree of twinning, and this best data set was the result of the extraction
of the major component with the program GEMINI. The complex
crystallizes with two perchlorate counterions.
Diffraction data for the [CuI([15]aneS3bpy)]ClO4 complex were
measured on a Bruker APEX-II κ geometry diffractometer42 with Mo
radiation and a graphite monochromator at 100 K. A green-brown rod
approximately 0.45 × 0.08 × 0.08 mm3 was used for data collection.
A total of 1083 frames was collected at 10 s/frame, yielding 12392
reflections of which 4463 were independent. Frames were collected as
a series of sweeps with the detector at 40 mm and 0.3° between each
frame. Hydrogen atoms were placed in observed positions. The complex
crystallizes with one perchlorate counterion, which showed typical
perchlorate disorder. The oxygen atoms were modeled in partially
occupied sites held isotropic.
NMR Measurements. The NMR spectrum for [CuI([15]aneS3bpy)]+
and line-broadening measurements in deuterated acetonitrile were made
with a Mercury 400 NMR spectrometer using an appropriate pulse
sequence. A solution of CuIL was prepared by conproportionation by
stirring a CuIIL solution with excess fine copper shot. The solution
was then decanted, and the resultant CuIL concentration was determined
spectrophotometrically at 390 nm.
Figure 3. Cyclic voltammogram for [CuII([15]aneS3bpy)] in acetonitrile
at ambient temperature, µ ) 0.10 M (ClO4-), ν ) 20 mV s-1. Potentials
shown are vs Ag/AgCl reference electrode (E1/2 )-0.127 V vs ferrocene).
Instrumentation. The potentials of all reagents were determined
by slow-scan cyclic voltammetry (CV) in acetonitrile using a BAS-
100 electrochemical analyzer (Bioanalytical Systems, West Lafayette,
IN) equipped with a 3-mm glassy-carbon-disk working electrode, a
platinum wire auxiliary electrode, and an aqueous Ag/AgCl reference
electrode. Ferrocene (Sigma Chemical Co) was used as an internal or
external standard. All kinetic measurements were made using a Durrum
D-110 stopped-flow spectrophotometer that incorporated a modified
flow system from Tritech Scientific Ltd. (Winnipeg, Manitoba, Canada)
with an A/D interface to a PC. The modified flow system contained
all Kel-F and glass fittings to prevent degradation when using
acetonitrile as the solvent. This instrument was thermostatted at 25.0
( 0.2 °C, using a circulating constant temperature bath. Ionic strength
was maintained at 0.10 M using NaClO4 recrystallized from acetonitrile.
Computational Details. All calculations were performed using the
development version of the Gaussian suite44 employing the TPSS meta-
GGA density functional.45 Scalar-relativistic effects were taken into
account via the relativistic effective-core potential of Dolg et al. for
the copper center,46 while spin-orbit effects have been ignored. Ligand
orbitals were described by the 6-311+G(d) basis set.47-51 Geometries
were optimized for all CuI and CuII complexes using the Berny
algorithm.52 Analytic harmonic vibrational frequencies were computed
to ensure that the optimized structures were minima on the potential
energy surface and to provide zero-point energy corrections.
Results
Redox Potentials. As illustrated in Figure 3, the cyclic
voltammogram for [CuII/I([15]aneS3bpy)]2+/+ is reversible at
slow scan rates (20 mV s-1) in acetonitrile with a value of E1/2
) -0.127 V vs ferrocene (Fc).53,54 Formal potential values for
[RuIII/II(NH3)5py]3+/2+, [CoIII/II(phen)3]3+/2+, and [CoIII/II(bpy)3]3+/2+
(37) (a) [Ru(NH3)5isn](ClO4)2: Kuehn, C. G.; Taube, H. J. Am. Chem. Soc.
1976, 98, 689-702. (b) [Ru(NH3)5py](ClO4)2: Gaunder, R. G.; Taube, H.
Inorg. Chem. 1970, 9, 2627-2639, with modifications as described by
Cummins, D.; Gray, H. B. J. Am. Chem. Soc. 1977, 99, 5158-5167. (c)
[Ru(NH3)4bpy](ClO4)2: Brown, G. M.; Sutin, N. J. Am. Chem. Soc. 1979,
101, 883-892. (d) [Co(bpy)3]((ClO4)2: Burstall, F. H.; Nyholm, R. S. J.
Chem. Soc. 1952, 3570-3579; [Co(phen)3](ClO4)2 was prepared by an
analogous approach.
(44) Frisch, M. J.; et al. Gaussian DV, revision E.05; Gaussian, Inc.: Walling-
ford, CT, 2006. [See Supporting Information for complete list of authors.]
(45) Tao, J. M.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys. ReV.
Lett. 2003, 91, 146401.
(46) Dolg, M.; Wedig, U.; Stoll, H.; Preuss, H. J. Chem. Phys. 1987, 86, 866-
872.
(38) Dunn, B. C.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem. 1995,
34, 1954-1956.
(47) Curtiss, L. A.; McGrath, M. P.; Blaudeau, J. P.; Davis, N. E.; Binning, R.
C., Jr.; Radom, L. J. Chem. Phys. 1995, 103, 6104-6113.
(48) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980,
72, 650-654.
(49) McGrath, M. P.; Radom, L. J. Chem. Phys. 1991, 94, 511-516.
(50) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72, 5639-5648.
(51) Binning, R. C., Jr.; Curtiss, L. A. J. Comp. Chem. 1990, 11, 1206-1216.
(52) Schlegel, H. B. J. Comp. Chem. 1982, 3, 214-218.
(53) The use of ferrocene as a standard for potentials in nonaqueous media
follows the recommendations of the International Union of Pure and Applied
Chemistry (IUPAC): Gritzner, G.; Kuta, J. Pure Appl. Chem. 1984, 56,
461-466.
(39) Meagher, N. E.; Juntunen, K. L.; Salhi, C. A.; Ochrymowycz, L. A.;
Rorabacher, D. B. J. Am. Chem. Soc. 1992, 114, 10411-10420.
(40) Cooper, T. H.; Mayer, M. J.; Leung, K.-H.; Ochrymowycz, L. A.;
Rorabacher, D. B. Inorg. Chem. 1992, 31, 3796-3804.
(41) Hathaway, B. J.; Holah, D. G.; Postlethwaite, J. D. J. Chem. Soc. 1961,
3215-3218.
(42) APEX II, SMART, SAINT, GEMINI and SADABS collection and processing
programs are distributed by the manufacturer: Bruker AXS Inc., Madison,
WI.
(43) Sheldrick, G. SHELX-97; University of Go¨ttingen: Germany, 1997.
9
5220 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Geometric “Entatic State” Effect
A R T I C L E S
I
Table 1. Physical Parameters for All Reagents in Acetonitrile at
25 °C, µ ) 0.10 M
for this term to obtain an initial estimate of KCu L′; this estimated
stability constant value was then used to correct for the amount
of uncomplexed CuI (i.e., [CuI] ) CCu - [Cu L]) in an iterative
manner until a consistent value of KCu L′ was obtained. In the
E f
V vs Fc
108 rb λmax
cm nm
10-3 ꢀ
k22
I
I
1
1
reagenta
M-1 cm-
M-1 s-
I
[CuII/I([15]aneS3bpy)]2+/+ -0.127c 4.4c 390c
2.54c
presence of large CuI concentrations, the absorbance values
tended to decrease abnormally, thereby causing a plot of bCL/A
vs [CuI]-1 to become nonlinear. After extensive investigation,
it was determined that this problem resulted from an apparent
autoxidation of the CuIL complex under these conditions
(perhaps by O2) since the initial absorbance could be regenerated
upon electrolysis. Therefore, data from solutions containing large
[CuI] values were not included in the final data analysis. From
the linear portion of the plot of eq 2, the slope and intercept
yielded the following values in acetonitrile at 25.0 °C, µ ) 0.10
Reductants
[RuII(NH3)5py]2+
- 0.088c 3.8d 407e
- 0.212f 5.9g 354f
7.8e
4.7 × 105 d
[CoII(dmg)2(BC6H5)2]
6.05f 1.18 × 102 f
3.31e 5.5 × 105 h
Oxidants
[RuIII(NH3)4bpy]3+
[RuIII(NH3)5isn]3+
[CoIII(phen)3]3+
[CoIII(bpy)3]3+
0.127h 4.4i 522e
-0.012h 3.8e 478 j 11.9 j
4.7 × 105 h
-0.035c 7.0k 330l
4.68l 1.28c
-0.076c 7.0k 317m 31.2m 0.645h
a For ligand abbreviations, see text. b Effective ionic radius. c This work.
d Assumed to be identical to the corresponding isonicotinamide complex.
e Referemce 37c. f Reference 36. g Murguia, M. A.; Wherland, S. Inorg.
Chem. 1991, 30, 139-144. h Dunn, B. C.; Ochrymowycz, L. A.; Rorabacher,
(ClO4-): KCu L′ ) (3.15 ( 0.14) × 104 M-1, ꢀCu L ) (2.72 (
I
I
j
0.04) × 103 M-1 cm-1 (390 nm). (This value is similar to that
D. B. Inorg. Chem. 1997, 36, 3253-3257. i Reference 60. Shepherd, R.
found for [CuI([14]aneS4)]2+ in acetonitrile: KCu L′) 1.7 × 104
E.; Taube, H. Inorg. Chem. 1973, 12, 1392-1401. k Tsukahara, K.; Wilkins,
R. G. Inorg. Chem. 1985, 24, 3399-3402. l Przystas, T. J.; Sutin, N. J.
Am. Chem. Soc. 1973, 95, 5545-5546. m Martin, B.; Waind, G. M. J. Chem.
Soc. 1958, 4284-4288.
I
M-1.)33 An independent measurement of the molar absorptivity
was obtained by adding large excess amounts of ligand to
solutions of CuI. A series of such measurements yielded an
average value of ꢀCu L ) 2.54 × 103 M-1 cm-1. The latter value
I
in acetonitrile (py ) pyridine; phen ) 1,10-phenanthroline; bpy
) 2,2′-bipyridine) were determined in a similar fashion as part
was used for all subsequent calculations.
of the current work. Potential values for [RuIII/II(NH3)4bpy]3+/2+ 38
,
The stability constant for the oxidized complex in acetonitrile
was calculated from the measured CuII/IL redox potential using
the Nernst equation in the form,
[RuIII/II(NH3)5isn]3+/2+ 38 and [CoIII/II(dmg)3(BC6H5)2]+/0 19 were
,
taken from the literature (isn ) isonicotinamide, dmg )
dimethylglyoxime). The physical parameters for all reagents
used in this work are listed in Table 1.
KCuIIL
′
2.303RT
nF
Ef ) E0solv′ -
log
(3)
Stability Constants. As noted in reports on previously studied
CuII/IL systems,33,55 the stability of CuIIL complexes is enhanced
in acetonitrile relative to water, whereas the affinity of CuI for
acetonitrile causes the reduced complexes to decrease in
stability. The diminished stability of the [CuI([15]aneS3bpy)]+
complex, combined with its sizable absorbance in the vicinity
of 390 nm, made it possible to determine the stability constant
of the reduced complex in acetonitrile by the spectrophotometric
approach of McConnell and Davidson56 as represented by the
relationship:57
KCuIL
′
where Ef represents the formal potential value obtained in
acetonitrile (vs ferrocene) for the CuII/IL redox couple from
slow-scan cyclic voltammetry, E0solv′ ) 0.660 V (vs ferrocene)
represents the potential for the solvated CuII/I couple in
acetonitrile obtained in a similar manner in 0.10 M NaClO4,
and the other terms have their conventional meanings. Equation
3 yielded a value of KCu L′ ) 6.3 × 1017 M-1
.
II
Reaction Kinetics. The kinetics for CuIL oxidation were
studied with a series of four oxidants (AOx); and the kinetics of
the reduction reactions for CuIIL were similarly measured using
two counter reductants (ARed) according to the general reaction:
bCL
A
1
ꢀCuIL
1
)
+
(2)
ꢀCuILKCuIL′[CuI]
k12
y\
CuIIL + ARed
z CuIL + AOx
(4)
In this expression, b represents the cell path length, CL is the
k21
I
total ligand concentration in solution, ꢀCu L is the molar
I
absorptivity of the reduced complex, KCu L′ is the conditional
The reactions involving the oxidation of CuIL by [CoIII(phen)3]3+
and [CoIII(bpy)3]3+ were run under pseudo-first-order conditions
with the counter reagent present in excess. For reactions in which
the cobalt reagent was present in less than 10-fold excess, only
the initial portion of the reaction was used in analyzing the data.
Pseudo-first-order conditions were also used in studying the
CuIIL reduction kinetics with [CoII(dmg)3(BC6H5)2] except that,
in that case, the CuIIL species was present in large excess. The
reduction study involving CuIIL reacting with [RuII(NH3)5py]2+
and the oxidation studies in which [RuIII(NH3)4bpy]3+ and
[RuIII(NH3)5isn]3+ served as the counter reagents were carried
out under second-order conditions, the latter two reactions being
very rapid. Each reaction was run approximately 10 times for
each set of reaction concentrations. The resolved second-order
rate constants obtained for all six cross-reactions are listed in
Table 2. (Individual rate constant data may be accessed as
Supporting Information.)
CuIL stability constant,58 and [CuI] is the concentration of
uncomplexed copper(I). Because the value of [CuI] was not
known initially, the total CuI concentration (CCu ) was substituted
I
(54) For comparison, the potential values for [CuII/I([15]aneS5)]2+/+ and
[CuII/I([15]aneNS4)]2+/+ are about 0.31 and 0.09 V vs ferrocene, respectively,
based on (a) their potential values relative to [CuII/I([14]aneS4)]2+/+ in
aqueous solution [0.68, 0.457, and 0.58 V (vs SHE), respectively: Bernardo,
M. M.; Heeg, M. J.; Schroeder, R. R.; Ochrymowycz, L. A.; Rorabacher,
D. B. Inorg. Chem. 1992, 31, 191-198] and (b) the potential of the latter
system in acetonitrile [0.205 V (vs Fc): ref 38].
(55) Kulatilleke, C. P.; Goldie, S. N.; Ochrymowycz, L. A.; Rorabacher, D. B.
Inorg. Chem. 1999, 38, 5906-5909.
(56) McConnell, H.; Davidson, N. J. Am. Chem. Soc. 1950, 72, 3164-3167.
Compare: Benesi, H. A.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71,
2703-2707.
(57) Sokol, L. S. W. L.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem.
1981, 20, 3189-3195.
(58) The conditional stability constant, KCu L′, includes any adducts of CuIL
I
formed with perchlorate ion as previously documented for CuIIL complexes
in water: ref 40. Compare: Nazarenko, A. Y.; Izatt, R. M.; Lamb, J. D.;
Desper, J. M.; Matysik, B. E.; Gellman, S. H. Inorg. Chem. 1992, 31, 3990-
3993.
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5221

A R T I C L E S
Chaka et al.
Table 2. Cross-Reaction Rate Constants and Experimental and
Calculated Self-Exchange Rate Constants in Acetonitrile at 25 °C,
µ ) 0.10 M
6
10- k12 or k21
log k11
1
a
1
reagent
(M-1 s-
)
(M-1 s-
)
NMR line-broadening
5.3, 5.6
Reductions
0.21(3)
0.0167(6)
[RuII(NH3)5py]2+
5.07
4.54
[CoII(dmg)2(BC6H5)2]
Oxidations
[RuIII(NH3)4bpy]3+
[RuIII(NH3)5isn]3+
[CoIII(phen)3]3+
[CoIII(bpy)3]3+
29(16)
3.2(4)
0.00159(6)
0.00055(1)
5.19
5.01
4.80
4.76
a Values in parentheses represent standard deviations in terms of the last
digits shown: e.g., 0.21(3) ) 0.21 ( 0.03.
Self-Exchange Rate Constant Values. The second-order
cross-reaction rate constants, k12 and k21, were inserted into the
Marcus cross-relationship59 as rearranged to the form,6
Figure 4. NMR spectrum for [CuI([15]aneS3bpy)]+ in acetonitrile at
ambient temperature, µ ) 0.10 M (ClO4-).
(k12)2
k22K12f12(W12)2
(k21)2
k22K21f21(W21)2
check on the value of the self-exchange rate constant for
[CuII/I([15]aneS3bpy)]2+/+ was conducted using NMR line-
broadening measurements in deuterated acetonitrile at 25 °C.
Small amounts of CuII were added to a stock CuIL solution,
and the line widths of the selected peaks were measured. The
ionic strengths of the solutions were calculated from the known
concentrations of the constituent species and corrected to 0.10
M as previously described.26 The data were then plotted in the
form:26
k11
)
OR k11
)
(5)
to obtain estimates of the electron self-exchange rate constant
for the [CuII/I([15]aneS3bpy)]2+/+ redox couple, k11, from each
of the six cross-reactions. The individual terms in eq 5 include
the self-exchange rate constant for each counter reagent, k22
(Table 2), and the reaction equilibrium constant, K12 or K21, as
calculated from the difference in the redox potentials of the two
reactants involved (Table 2). Of the remaining terms, f12 and
f21 represent nonlinear correction terms, and W12 and W21 are
electrostatic work terms that can be calculated as previously
described.39,60 The values for the effective ionic radius (r) for
each reactant, as used to calculate the work term values,60 are
included in Table 1. (It should be noted that the k11 value
obtained from the oxidation of CuIL by [RuIII(NH3)4bpy]3+ is
the least reliable since the reaction rate constant was exception-
ally large (k21 ≈ 107 M-1 s-1) relative to the detection limits of
the stopped-flow instrument used.)61
WDPQπ ) WDQπ + k11[CuIIL]
(6)
In this expression, WDP is the peak width at half-height of the
selected singlet proton resonance peak for a solution containing
both the diamagnetic (CuIL) and paramagnetic (CuIIL) species,
WD is the corresponding value for a solution containing only
the diamagnetic species, and Q ()1.97) is a factor used to
correct the extent of outer-sphere complex formation at the
specific ionic strength used to a value of µ ) 0.10 M.26
Preliminary NMR line-broadening measurements, based on
the line width of the 7.6 ppm peak, yielded k11 ) (2.2 ( 0.1)
× 105 M-1 s-1 for 25 °C, corrected to µ ) 0.10 M.26 A more
carefully prepared series of solutions run under similar condi-
tions yielded k11 ) (4.2 ( 0.6) × 105 M-1 s-1 based on the 7.6
ppm peak width and (4.0 ( 0.4) × 105 M-1 s-1 based on the
peak at 4.2 ppm. (Experimental data for the latter series are
available as Supporting Information.)
NMR Measurements. As illustrated in Figure 4, the NMR
spectrum of [CuI([15]aneS3bpy)]+ exhibits a doublet centered
at 3.1 ppm (a) and singlets at 4.2 (b), 7.6 (c), and 8.1 ppm (d)
with peak ratios of 4:2:1:2, respectively. This spectrum is
consistent with the peak assignments shown below assuming
full complexation by all five donor atoms: A limited independent
Structural Determinations. In conjunction with the current
investigation, we have determined the crystal structures for the
perchlorate salts of both the oxidized and reduced complexes:
[CuII([15]aneS3bpy)](ClO4)2 and [CuI([15]aneS3bpy)](ClO4). In
both complexes the copper atom is coordinated to all five donor
atoms, in agreement with the NMR spectrum. Crystallographic
data are listed in Table 3. The bond lengths and bond angles of
primary interest for both structures are presented in Table 4.
Figure 5 shows the superimposed ORTEP views of the
[CuII([15]aneS3bpy)]2+ and [CuI([15]aneS3bpy)]+ cationic units.
In both cases, the perchlorate oxygens are more than 4 Å away
from the copper atom.
(59) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta. 1985, 811, 265-322.
(60) Martin, M. J.; Endicott, J. F.; Ochrymowycz, L. A.; Rorabacher, D. B.
Inorg. Chem. 1987, 26, 3012-3022.
(61) We have previously demonstrated our ability to measure second-order rate
constants as large as 108 M-1 s-1: Dunn, B. C.; Meagher, N. E.; Rorabacher,
D. B. J. Phys. Chem. 1996, 100, 16925-16933. However, the current
measurements in acetonitrile were carried out using an instrument that was
modified for use with acetonitrile. This instrument has a significantly slower
mixing time (∼10 ms), thereby reducing our ability to study exceptionally
fast reactions.
Theoretical Calculations. Previous attempts within our
laboratory to make meaningful theoretical calculations regarding
9
5222 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Geometric “Entatic State” Effect
A R T I C L E S
Table 3. Crystal Parameters and Experimental Data for X-ray
electron localized on *CuL, must equal the system’s energy with
the electron localized on CuL. As a result of the Cs symmetry
of the self-exchange system’s potential energy surface, in the
absence of a perturbing field, this energetic coincidence will
occur when both oxidation states obtain the coincidence
geometry, qc. It is important to note that identical coincidence
geometries for both oxidation states is not a requirement for
electron transfer but, rather, a manifestation of the underlying
particle indistinguishability in self-exchange reactions. Assigning
a diabatic potential energy surface to the reactants, q1, and
products, q2, we arrive at the standard Marcus two-state energy
diagram for a symmetric (∆G° ) 0) electron-transfer reaction
in Figure 6a.
To probe the energetics associated with merging the two
oxidation-state geometries toward qc, we followed the four-point
method of Nelsen et al.63 employing a complete set of redundant
internal coordinates. Energies were calculated at seven equi-
distant points linearly interpolated between the CuIIL and CuIL
ground-state geometries to determine the electronic component
of ∆G* for both the C1 and Cs surfaces. Since two surfaces are
available and energetically accessible for the [15]aneS3bpy
complexes, transition states (TSs) connecting the C1 and Cs
ground states were located for both oxidation states (see Table
6). Although both surfaces are accessible at room temperature,
the CuIL barrier for C1 to Cs interconversion is 3 kJ mol-1 lower
than the CuIIL barrier.
Diffraction Measurements on the Copper(II) and Copper(I)
a
Complexes Formed with [15]aneS3bpy
parameter
[CuII([15]aneS3bpy)](ClO4)2
[CuI([15]aneS3bpy)]ClO4
emp. form.
fw
space group
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
Z
CuC16H18S3Cl2O8
596.94
C2/c
38.152(13)
5.4157(15)
20.548(7)
90
98.457(9)
90
4199(2)
8
CuC16H18S3ClO4
479.49
C2/c
24.0998(11)
14.1272(6)
12.6599(5)
90
120.609(2)
90
3709(3)
8
T, K
295(2)
1.888
1.643
0.0694
0.1672
100(2)
1.782
1.686
0.0461
0.1231
F
calcd, g cm-1
µ, mm-1
R(F), I > 2σ b
Rw(F2), I > 2σ c
a λ ) 0.71703 Å. b R(F) ) Σ||Fo| - |Fc||/∑|Fo| for 2σ(I) reflections.
2
2
2
c Rw(F2) ) [∑w(Fo - Fc )2/∑w(Fo )2]1/2 for 2σ(I) reflections.
the energy barriers associated with electron transfer in CuII/I
systems had been relatively qualitative, being limited both by
the low level of theory used and the difficulty of treating the
impact of bond formation/rupture during the electron-transfer
process. The fact that the [CuII/I([15]aneS3bpy)]2+/+ self-
exchange reaction occurs without significant changes in coor-
dination geometry suggested that more sophisticated theoretical
calculations might be profitable.
For comparative purposes, the same optimization and redox-
reaction path approaches were applied to the five-coordinate
invariant CuII/I systems involving the acyclic ligands previously
reported by Stanbury and co-workers, designated as (py)2DAP,21
(imidH)2DAP,21,22 (5-Meimid)2DAP,23 and (imidR)2DAP,21 for
which the results are listed in Table 5.
Electronic structure calculations were computed by using the
TPSS meta-GGA density functional45 and a basis set of the
quality recommended by Solomon and co-workers.17 It was
recently shown that meta-GGA functionals are well suited for
organometallic complexes and generally outperform standard
GGA and hybrid functionals.62 The C1 and Cs geometries were
optimized for both [CuI([15]aneS3bpy)]+ and [CuII([15]aneS3-
bpy)]2+ complexes to produce four different optimized struc-
tures. The minimized gas-phase C1 and Cs structures for
[CuI([15]aneS3bpy)]+ and [CuII([15]aneS3bpy)]2+, respectively,
are in excellent agreement with the resolved crystal structures.
It is interesting to note that, when the ligand conformation goes
from asymmetric (C1) to symmetric (Cs), the Cu-S(2) gas-phase
bond shortens by 0.06 and 0.02 Å for oxidation states I and II,
respectively. As the oxidation state goes from I to II for a given
ligand conformation, the Cu-S(2) bond always lengthens by
∼0.2 Å, while the other four coordinate bonds contract by ∼0.1
Å. This trend is seen in both the computed gas-phase structures
and the crystal structures. The electronic energies of the
optimized C1 and Cs geometries for both oxidation states are
presented in Table 6. With an energy difference of less than 6
kJ mol-1, both the C1 and Cs complexes are readily accessible
for each oxidation state at room temperature.
Discussion
For all six cross-reactions studied in this work, the calculated
electron self-exchange rate constants are included in Table 2
as logarithmic k11 values. It will be noted that all such values
are within an order of magnitude of the corresponding values
obtained from NMR line-broadening measurements. This rep-
resents an acceptable level of agreement, considering the large
number of variables involved in the calculations and the fact
that any errors in the experimental k12 or k21 values become
squared when calculating k11 (eq 5).6 From the combined self-
exchange and cross-reaction kinetic data, we conclude that k11
) 1 × 105 M-1 s-1 for the [CuII/I([15]aneS3bpy)]2+/+ system.
In several CuII/IL systems previously studied, the k11 values
calculated from oxidation reactions differed from those calcu-
lated from reductions.29,39,60,65-72 This phenomenon has been
(63) Nelsen, S. F.; Blackstock, S. C.; Kim, Y. J. Am. Chem. Soc. 1987, 109,
677-682.
(64) The calculated geometries at the sequential points are based on taking the
differences in the bond lengths and bond angles between the ground-state
C1 geometry of CuIL and the ground-state Cs geometry of CuIIL and altering
each length and angle by one-eighth of these differences for each sequential
point.
(65) Ambundo, E. A.; Yu, Q.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg.
Chem. 2003, 42, 5267-5273.
(66) Meagher, N. E.; Juntunen, K. L.; Heeg, M. J.; Salhi, C. A.; Dunn, B. C.;
Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem. 1994, 33, 670-
679.
(67) Salhi, C. A.; Yu, Q.; Heeg. M. J.; Villeneuve, N. M.; Juntunen, K. L.;
Schroeder, R. R.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem.
1995, 34, 6053-6064.
(68) Wijetunge, P.; Kulatilleke, C. P.; Dressel, L. T.; Heeg, M. J.; Ochrymowycz,
L. A.; Rorabacher, D. B. Inorg. Chem. 2000, 39, 2897-2905.
While it is not possible to model directly the CuIIL/CuIL redox
reaction energetics due to the complications of properly
representing a solvated electron, it is possible to investigate those
parts of the reaction path not involving the electron-transfer step
itself. In order for the outer-sphere self-exchange electron
transfer (eq 1) to occur, the energy of the system containing
two Cu centers, denoted *CuL + CuL, with the redox active
(62) Schultz, N. E.; Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2005, 109,
11127-11143.
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5223

A R T I C L E S
Chaka et al.
Table 4. Average Bond Distances and Bond Angles in the Cationic Units of [CuII([15]aneS3bpy)](ClO4)2 and [CuI([15]aneS3bpy)]ClO4
bond distance/angle
[CuII([15]aneS3bpy)](ClO4)2
CuI([15]aneS3bpy)ClO4
Cu-S(1), Å
2.300(3)
2.475(2)
2.314(2)
1.954(7)
1.953(7)
90.91(8)
90.69(8)
99.3(2)
98.9(2)
105.26(9)
85.9(2)
165.2(2)
85.5(2)
165.2(2)
81.9(3)
2.4226(10)
2.3219(10)
2.3881(9)
2.136(3)
2.141(3)
91.92(4)
Cu-S(2), Å
Cu-S(3), Å
Cu-N(1), Å
Cu-N(2), Å
S(1)-Cu-S(2), deg
S(2)-Cu-S(3), deg
S(2)-Cu-N(1), deg
S(2)-Cu-N(2), deg
S(1)-Cu-S(3), deg
S(1)-Cu-N(2), deg
S(1)-Cu-N(1), deg
S(3)-Cu-N(1), deg
S(3)-Cu-N(2), deg
N(1)-Cu-N(2), deg
Cu displacement, Åa
93.36(3)
105.33(7)
113.00(8)
115.95(4)
81.10(8)
154.76(8)
81.87(8)
148.70(8)
75.29(11)
+ 0.369(1)
+ 0.168(3)
a Displacement of the Cu atom from the S2N2 plane. The positive values represent displacement of the Cu toward the axial S donor atom.
Figure 5. Superimposed ORTEP drawings of the cationic unit in the crystal
structure of the oxidized (red) and reduced (blue) complex for [CuII/I
-
([15]aneS3bpy)]2+/+ showing the atom numbering scheme. The most notable
structural change involves a “puckering” of the C1 and C2 carbon atoms
on going from CuIIL to CuIL. Ellipsoids represent 50% probability.
Hydrogen atoms have been omitted for clarity.
attributed to ligand conformational changes occurring in a
sequential, rather than a concerted, manner with the electron-
transfer step. Since the two processes can occur in either order,
two competing mechanistic pathways are accessible, one of
which involves a metastable CuIIL species (Q) and the other a
metastable CuIL analogue (P) as illustrated in Scheme 1. For
most of the studies conducted in our laboratory in which both
pathways have been accessed, the more favored pathway has
been the one designated as Pathway A in Scheme 1.29,39,66-69
This implies that the CuIL metastable intermediate, CuIL(P), is
relatively more stable than the corresponding CuIIL intermediate,
CuIIL(Q), in those systems. However, for complexes with some
tripodal ligands65 and for the bis complexes with some
substituted derivatives of bipyridyl ligands studied by Takagi
and co-workers71,72 Pathway B appears to be more favorable.
In the current system, the same k11 value was obtained for
both oxidation and reduction within experimental error. Under
these circumstances, the data obtained for both the oxidation
Figure 6. Two-state potential energy diagram in (a) two- and (b) three-
dimensions. The energy at the coincidence geometry, qc in (a), is the diabatic
activation energy, ∆G*′, λ is the reorganization energy, and ∆G* is the
activation energy. The grayscale surface in (b) is Ψ1 and the colored surface
is Ψ2. The intersection of Ψ1 and Ψ2 is the seam of all possible coincidence
geometries. The self-exchange ground states, q1 and q2 in part (a), correspond
to points {0,0,0} and {1,1,0} in part (b), respectively. At {0,0,1} both Cu
centers are in the CuIL ground-state geometry, while at {1,1,1} both centers
have adopted the ground-state geometry of CuIIL.
(69) Yu, Q.; Salhi, C. A.; Ambundo, E. A.; Heeg, M. J.; Ochrymowycz, L. A.;
Rorabacher, D. B. J. Am. Chem. Soc. 2001, 123, 5720-5729.
(70) Ambundo, E. A.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem.
2001, 40, 5133-5138.
(71) Koshino, N.; Kuchiyama, Y.; Funahashi, S.; Takagi, H. D. Can. J. Chem.
1999, 77, 1498-1507.
and reduction reactions are presumed to be proceeding by the
same reaction pathway, implying that the ground-state species
(72) Koshino, N.; Kuchiyama, Y.; Ozaki, H.; Funahashi, S.; Takagi, H. D. Inorg.
Chem. 1999, 38, 3352-3360.
9
5224 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Geometric “Entatic State” Effect
A R T I C L E S
have been reported for the CuII/I complexes with the sterically
hindered bidentate ligands designated as dpym- and bimdpk
(Figure 1) which have been shown to remain four-coordinate
Table 5. Comparative Self-Exchange Rate Constants for
Coordination-Invariant Quinquedentate and Quadridentate
Copper(II/I) Systems Plus Rapidly Reacting Systems with Variable
Coordination at 25 °C, µ ) 0.10 M
in both oxidation states.24,25 By contrast, the k11 values reported
1
a
1 b,c
coordinated ligand
log k11 (M-1 s-
)
Calcd ∆E*′ )
(kJ mol-
19
by Stanbury and co-workers for [CuII/I(bib)2]2+/+
and
Coordination Invariant: Five-Coordinate
[CuII/I(bite)]2+/+ 20
, both of which are presumed to remain four-
[15]aneS3bpy
4.5-5.6b
12.6 (15.1)c
16.5
15.3
16.5
18.1
(5-Meimid)2DAP
(imidR)2DAP
(imidH)2DAP
(py)2DAP
4.5d
coordinate upon electron transfer, are remarkably small, an
observation that has not yet been satisfactorily explained.
Recently, Fujisawa and co-workers73 reported the first
electron-transfer kinetic study on a CuII/I complex containing a
Cu-S(mercaptide) bond which maintains four-coordination in
both oxidation states. The complex involves a coordinated
tripodal ligand, hydrotris(3,5-diisopropyl-1-pyrazolyl)borate ligand
(HB(3,5-iPrpy)3 ) L), plus a pentafluorobenzenethiolate ligand
(C6F5S-), the latter of which provides a coordinated mercaptide
sulfur donor atom. On the basis of NMR line-broadening
measurements in deuterated acetone at -20 to -60 °C, they
reported a value of k11 ) 3.14 × 105 M-1 s-1 (corrected to 25
°C).
3.5e
3.2-4.1e,f
2.5-3.2e
Coordination Invariant: Four-Coordinate
HB(3,5-iPr2py)3 + C6F5S-
5.5 (?)g
4.3h
bis-bimdpk
bis-(dpym-)
3.8i
8
W12O40
bis-bib
bite
-
0.4j
- (1.6-0.3)k
- (1.9-1.2)l
Variable Coordination: Five-Coordinate f Four-Coordinate
oxathiane-[12]aneS4
TAAB
[15]aneS5
[15]aneNS4
5.6-5.9m
5.7n
4.9-5.5 (5.0)o,p
4.4-5.1q
a Values in boldface were determined by NMR line-broadening; all other
values are based on application of the Marcus relationship to cross-reactions.
b This work. c ∆E*′ values represent gas-phase calculations except for the
parenthetical value for [15]aneS3bpy which includes the contributions of a
solvent continuum in acetonitrile (see text). d Reference 23. e Reference 21.
f Reference 22. g Reference 73; NMR line-broadening data were nonlinear
and were extrapolated from a very low temperature (see text). h Reference
25. i Reference 24. j Lappin, A. G.; Peacock, R. D. Inorg. Chim. Acta 1980,
46, L71-L72. k Reference 19. l Reference 20. m Reference 28. n Reference
30. o Reference 26. p Reference 60. q Reference 27.
The Supporting Information provided with Fujisawa’s article
reveals that the experimental line width plots were so badly
curved that the asymptotic slopes obtained from the lowest and
highest CuII concentrations differ by an order of magnitude at
each temperature studied. Thus, not only are the individual k11
values for each experimental temperature ill-defined, but the
curvature also leaves some doubt as to whether the data actually
represent line-broadening due to electron exchange. At the very
least, the uncertainty associated with the line width trend at each
temperature, plus the uncertainty inherent in extrapolating the
low-temperature data to +25 °C, suggests that their self-
exchange rate constant is, at best, known only to the nearest
order of magnitude, that is, k11 ≈ 105 M-1 s-1. If the latter
value has validity, it would appear to corroborate Solomon and
Ryde’s conclusions regarding the impact of the electronic effects
induced by the copper-mercaptide sulfur bond upon electron-
transfer kinetics.17,18 However, it leaves open the question as
to whether similar rapid electron-transfer kinetics can be
observed in coordination-invariant CuII/I systems based solely
on the type of geometric entatic state originally envisioned by
Vallee and Williams.
Scheme 1
and the metastable intermediate for the dominant pathway are
in rapid equilibrium relative to the rate of the electron-transfer
step. Under these circumstances it is impossible to determine
which pathway is involved, nor is it possible to distinguish
between a sequential and a concerted pathway. It is interesting
to note, however, that the same pattern of behavior has been
observed with nearly all other CuII/I systems studied to date in
which a quinquedentate ligand is coordinated to the copper
atomsthat is, all such systems have generated a single k11
valuesregardless of whether the system remains five-coordinate
or changes to four-coordinate upon reduction.21-23,26,27
Previous studies on CuII/IL systems maintaining five-
coordinate geometries in both oxidation states, as reported by
Stanbury, Wilson, and co-workers,21-23 are included in Table
5. These closely related acyclic ligandss(py)2DAP, (imidH)2DAP,
(imidR)2DAP, and (5-Meimid)2DAPsinvolve five imine nitro-
gen donor atoms (see Figure 1). Although little cross-reaction
rate data are available for these systems, the reported k11 values
appear to lie within the range of 2 × 103 to 3 × 104 M-1 s-1
as determined from NMR line-broadening studies (Table 5),
that is, they are 1-2 orders of magnitude smaller than the
To understand our theoretical results, we begin by defining
the distortion energy for a particular set of nuclear coordinates
as the electronic energy at that geometry minus the electronic
energy of the ground-state geometry for a particular oxidation
state. In terms of distortion energies, self-exchange electron
transfer occurs when the total distortion energy of the CuIIL +
(73) Fujisawa, K.; Fujita, K.; Takahashi, T.; Kitajima, N.; Moro-oka, Y.;
Matsunaga, Y.; Miyashita, Y.; Okamoto, K. Inorg. Chem. Commun. 2004,
7, 1188-1190 (the experimental linewidth values are provided as Sup-
porting Information). Additional structural and spectroscopic information
on the [CuII/I(HB(3,5-iPrpy)3(SC6F5)] system is reported in: (a) Kitajima,
N.; Fujisawa, K.; Tanaka, M.; Moro-oka, Y. J. Am. Chem. Soc. 1992, 114,
9232-9233. (b) Matsunaga, Y.; Fujisawa, K.; Ibi, N.; Miyashita, Y.;
Okamoto, K. Inorg. Chem. 2005, 44, 325-335.
values obtained from NMR measurements on the [CuII/I
([15]aneS3bpy)]2+/+ complex in this work. Similar k11 values
-
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5225

A R T I C L E S
Chaka et al.
Table 6. Calculated Electronic Energy for [CuI([15]aneS3bpy)] and
[CuII([15]aneS3bpy)] Complexes with Values Relative to the
Ground State for Each Oxidation State
1
relative electronic energy, kJ mol-
molecule
symmetry
gaseous state
with solvation shell
CuIL
C1
Cs
C1
0.00
5.41
18.14
0.00
3.66
18.03
CuIL
CuIL* (TS)a
CuIIL
C1
Cs
C1
0.00
2.63
21.49
0.50
0.00
20.49
CuIIL
CuIIL* (TS)a
a Calculated transition state.
into the relative energetics involved in achieving the common
coincidence geometry that must exist for self-exchange electron
transfer to occur between CuIL and CuIIL.
Figure 7. Total distortion energy for the [CuII([15]aneS3bpy)] +
[CuI([15]aneS3bpy)] self-exchange system when both species adopt the same
geometry. Point 1 represents the electronic reorganization energy required
to distort CuIIL to the C1 ground-state geometry of CuIL for a given
symmetry and phase, while point 9 is the electronic reorganization energy
required to distort CuIL to the Cs ground-state geometry of CuIIL for that
same symmetry and phase. Points 2-8 are equidistant configurations as
the molecular geometry of the CuIL and CuIIL species are stepped from
the CuIL ground state to the CuIIL ground state. The solid curves represent
gas-phase species, while the dashed curves include a dielectric continuum
approximating the acetonitrile solvent shell. The minimum in the total
distortion energy provides an upper boundary to E*′.
For comparative purposes, the computed gas-phase, electronic
components of the diabatic activation energy, ∆E*′, have been
calculated for the other five-coordinate invariant systems
previously studied by Stanbury and co-workers. The results are
summarized in Table 5. Once again, these calculations are based
on the presumption of a concerted mechanism. It is noted that
the acyclic ligands containing imidazole groups bound to Cu
have activation energies clustered around 16 kJ mol-1. Although
the (py)2DAP ligand has the same conformational flexibility as
the imidazole-containing ligands, the activation barrier is ∼2
kJ mol-1 larger. This system has been reported to react more
slowly than its imidazole analogues. Table 5 also reveals that,
among the investigated five-coordinate ligand systems, the
CuIL system is minimized. Since the coincidence geometry is
the same for both oxidation states in self-exchange reactions, it
is very tempting to interpret the intersection of the CuIIL and
CuIL distortion energy curves as the point of electron transfer,
but that ignores the different definitions of zero energy for each
curve. Using the CuIIL and CuIL ground-state geometries plus
the seven interpolated structures, the total distortion energy along
a linear path from reactants to products is plotted for the C1
and Cs surfaces in Figure 7. After fitting polynomials to these
data, the minimum distortion energy and corresponding geom-
etry provide an upper boundary to the electronic component of
the diabatic activation energy, ∆E*′, and a reasonable ap-
proximation to qc, respectively. (See Supporting Information
for individual data and fits.) Location of the true self-exchange
∆E*′ requires optimization along the 3N - 7 seam of intersec-
tion, where N is the number of nuclei in CuL.
In Figure 7, point 1 denotes the energy required to distort
the ground states of CuIIL and CuIL into the ground-state
geometry of CuIL. Since CuIL is already in its ground state,
the energy at point 1 (28.0 and 26.8 kJ mol-1 for Cs and C1,
respectively) is equal to the internal electronic reorganization
energy, λi, for CuIIL. Likewise, point 9 gives λi (23.4 and 22.9
kJ mol-1 for Cs and C1, respectively) for CuIL and indicates
that CuIL is more easily distorted than CuIIL. Since the
corresponding qc values (5.46 and 5.49 for Cs and C1,
respectivelyscorresponding to the abscissa scale in Figure 7)
are greater than five, the coincidence geometry is slightly closer
to the ground-state geometry for CuIIL.
smallest diabatic activation barrier is seen for the [CuII/I
-
([15]aneS3bpy)]2+/+ system, which has the least ligand flexibility
and exhibits the largest k11 value. The rough parallel between
the calculated ∆E*′ values and the experimentally obtained k11
values for these five five-coordinate invariant systems tends to
support the validity of our computational approach.
The foregoing calculations omit any contribution of the outer-
sphere solvent reorganization that may, in fact, be very
significant.74 In an attempt to assess the influence of the
surrounding acetonitrile solvent shell upon the solvent reorga-
nization energy, we repeated the calculations for the [15]aneS3-
bpy system employing the integral equation formalism of the
polarizable continuum model (IEF-PCM) with a dielectric value
of 36.64.75 For both the C1 and Cs surfaces, the relative energy
difference of the two ground-state species decreased slightly
for each oxidation state as shown in Table 6. Interestingly, the
relative energies of the C1 and Cs surfaces reverse so that the
Cs geometry for [CuII([15] aneS3bpy)]2+, which was observed
in the crystal structure, is now the ground state. These new
structures were used to generate solvated total distortion curves
for Figure 7 in the same manner as described above. It should
be noted that the geometries plotted on the abscissa differ for
each phase and geometry. Although the transition states for both
oxidation states have lower energies in solution, the CuI barrier
for C1-to-Cs interconversion is still 3 kJ mol-1 smaller than the
CuII barrier. In the gas-phase, ∆E*′ is 12.60 and 13.01 kJ mol-1
It should be re-emphasized that the calculations outlined
above represent a concerted mechanism on a surface of uniform
symmetry, which does not presume the existence of metastable
intermediates of the type designated in Scheme 1. Since each
oxidation state has different distortion energies for a given
geometry, even concerted mechanisms will rarely have coin-
cidence geometries whose bond lengths and angles are exactly
halfway between those found in the ground state for the two
reacting partners. Nonetheless, our approach does provide insight
(74) Comba, P.; Kerscher, M.; Roodt, A. Eur. J. Inorg. Chem. 2004, 4640-
4645.
(75) (a) Cance`s, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3032-
3041. (b) Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 106, 5151-5158.
(c) Mennucci, B.; Cance`s, E.; Tomasi, J. J. Phys. Chem. B 1997, 101,
10506-10517. (d) Tomasi, J.; Mennucci, B.; Cance`s, E. J. Mol. Struct.
(THEOCHEM) 1999, 464, 211-226.
9
5226 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Geometric “Entatic State” Effect
A R T I C L E S
for the C1 and Cs surfaces, respectively, indicating that the C1
pathway is only slightly more favorable. With the inclusion of
continuum solvation effects, the corresponding activation ener-
gies are 15.08 and 16.12 kJ mol-1 for the C1 and Cs surfaces,
respectively.
neither the theoretical calculations nor the experimental kinetic
data allow us to rule out the possibility of a concerted
mechanism for the [CuII/I([15]aneS3bpy)]2+/+ system in aceto-
nitrile. The relatively small energy barrier associated with the
geometric change (12.6 kJ mol-1 for the system in a vacuum;
15.1 kJ mol-1 when including a solvent continuum) appears to
confirm the hypothesis that the overall reorganizational energy
barrier for electron transfer in geometrically constrained CuII/I
systems can be successfully minimized by appropriate ligand
constraints.
Conclusion
Comba has provided an extensive analysis of the entatic state
concept, including attempts made to generate CuII/IL systems
which provide physical constraints intended to generate geo-
metric entatic states in accordance with the original concept
espoused by Vallee and Williams.76 Of the coordination-
invariant CuII/IL systems for which electron self-exchange rate
constants have been reported to date, the [CuII/I([15]aneS3bpy)]2+/+
system represents the most convincing example of a redox
couple exhibiting rapid electron-transfer kinetics brought about
by a geometrically constrained entatic state. If the redox reaction
proceeds through a metastable intermediate, thereby changing
molecular symmetry before or after the electron transfer, then
Pathway A is the preferred mechanism due to the lower C1 to
Cs barrier for CuIL in both solution and gas phases. However,
Acknowledgment. This work was supported by the National
Science Foundation under Grant CHE-0211696. We thank
Professor John F. Endicott for helpful discussions and Dr. Lew
Hryhorczuk of Wayne State University’s Central Instrument
Facility for assistance in generating electrospray mass spectral
data.
Supporting Information Available: Experimental rate con-
stant data for cross-reactions, NMR line-broadening data, and
tables of DFT calculations. This material is available free of
charge via the Internet at http://pubs.acs.org.
(76) Comba, P. Coord. Chem. ReV. 2000, 200, 217-245.
JA068960U
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5227