Facile oxidation-based synthesis of sterically encumbered four-coordinate bis(2,9-di-tert-butyl-1,10-phenanthroline)copper(I) and related three-coordinate copper(I) complexes

Article abstract of DOI:10.1021/ic0615224

A new oxidation-based synthetic route was developed for synthesis of Cu(I) complexes with weakly coordinating ligands, leading to the synthesis of the elusive [Cu(dtbp)₂]⁺ (dtbp, 2,9-di-tert-butyl-1,10- phenanthroline) complex that may be useful as a sensor or as a dye for dye-sensitized solar cells. An acetone solution of either 1 or 2 equiv of dtbp was added to excess Cu(0) and 1 equiv of AgY (Y is O₃SCF ₃^-, BF₄^-, SbFe₆ ^-, or B(C₆F₅)₄^-) in a nitrogen-filled glove box. Following filtration and evaporation under vacuum, crystallization from CH₂Cl₂ and hexanes results in X-ray quality crystals of Cu(dtbp)(O₃SCF₃) (1), Cu(dtbp)(BF ₄) (2), [Cu(dtbp)(acetone)][SbF₆] (3), [Cu(dtbp) ₂]-[B(C₆F₅)₄]·CH ₂Cl₂ (4·CH₂Cl₂), [Cu(dtbp)₂][BF₄]·CH₂Cl₂ (5·CH₂Cl₂), and [Cu(dtbp)₂][SbF ₆]·CH₂Cl₂ (6·CH ₂Cl₂). Complexes 1-6 were characterized by X-ray crystallography and NMR. The Cu atom in complexes 1-3 exhibited distorted trigonal coordination geometries, reflecting the steric effect of the bulky tert-butyl substituents. The structures of the pseudotetrahedral complexes 4, 4·CH₂Cl₂, 5·CH₂Cl₂, and 6·CH₂Cl₂ revealed the longest average Cu-N distances (2.11 A, 2.11 A, 2.10 A, and 2.11 A, respectively) in this class of compounds-longer by more than three standard deviations from the average.

Full text of DOI:10.1021/ic0615224

Inorg. Chem. 2007, 46, 3816−3825
Facile Oxidation-Based Synthesis of Sterically Encumbered
Four-Coordinate Bis(2,9-di-tert-butyl-1,10-phenanthroline)copper(I) and
Related Three-Coordinate Copper(I) Complexes
Bhavesh A. Gandhi, Omar Green, and Judith N. Burstyn*
Department of Chemistry, UniVersity of Wisconsin, Madison, Wisconsin 53706
Received August 11, 2006
A new oxidation-based synthetic route was developed for synthesis of Cu(I) complexes with weakly coordinating
ligands, leading to the synthesis of the elusive [Cu(dtbp)2]+ (dtbp, 2,9-di-tert-butyl-1,10-phenanthroline) complex
that may be useful as a sensor or as a dye for dye-sensitized solar cells. An acetone solution of either 1 or 2 equiv
of dtbp was added to excess Cu(0) and 1 equiv of AgY (Y is O3SCF3-, BF4-, SbF6-, or B(C6F5)4-) in a nitrogen-
filled glove box. Following filtration and evaporation under vacuum, crystallization from CH2Cl2 and hexanes results
in X-ray quality crystals of Cu(dtbp)(O3SCF3) (1), Cu(dtbp)(BF4) (2), [Cu(dtbp)(acetone)][SbF6] (3), [Cu(dtbp)2]-
[B(C6F5)4]
Complexes 1
distorted trigonal coordination geometries, reflecting the steric effect of the bulky tert-butyl substituents. The structures
of the pseudotetrahedral complexes 4, 4 CH2Cl2, 5 CH2Cl2, and 6 CH2Cl2 revealed the longest average Cu
•
CH2Cl2 (4•CH2Cl2), [Cu(dtbp)2][BF4]
•
CH2Cl2 (5•CH2Cl2), and [Cu(dtbp)2][SbF6]
•
CH2Cl2 (6•CH2Cl2).
−
6 were characterized by X-ray crystallography and NMR. The Cu atom in complexes 1
−3 exhibited
•
•
•
−N
distances (2.11 Å, 2.11 Å, 2.10 Å, and 2.11 Å, respectively) in this class of compoundsslonger by more than three
standard deviations from the average.
Introduction
may be tailored to favor the formation of Cu(I) complexes,
forming the basis of most Cu(I) complex synthesis.
Copper(I) has been used extensively in the synthesis of
molecular machines,1-3 bioinspired model complexes,4-7 and
industrial catalysts.8 Cu(I) complexes are challenging to
synthesize due to the intrinsic instability of the cuprous
oxidation state; under many conditions disproportionation
of Cu(I) to Cu(0) and Cu(II) is thermodynamically favored.
Nevertheless, the ligand structure and solution conditions
Copper(I) complexes of the formula [CuL2]+, where L is
a 2,9-disubstituted-phenanthroline, display interesting pho-
tophysical properties that result from a metal-to-ligand
charge-transfer (MLCT) transition.9 In the excited state, the
Cu atom is formally in the +2 oxidation state, and relaxation
occurs either through nonemissive geometric reorganization
toward square planar geometry or through radiative emission.
Maximum radiative emission is achieved by preventing the
nonemissive geometric reorganization path, typically by
increasing the steric bulk of the substituents at the 2 and 9
positions of the phenanthroline ligand. A synthetic goal has
been to maximize the size of the substituents while retaining
the two bidentate phenanthroline ligands on the metal center,
which give rise to the MLCT.9-13
* To whom correspondence should be addressed. Phone: 608-262-0328.
Fax: 608-262-6143. E-mail: burstyn@chem.wisc.edu.
(1) Poleschak, I.; Kern, J.-M.; Sauvage, J.-P. Chem. Commun. 2004, 4,
474-476.
(2) Weber, N.; Hamann, C.; Kern, J.-M.; Sauvage, J.-P. Inorg. Chem.
2003, 42, 21, 6780-6792.
(3) Collin, J.-P.; Dietrich-Buchecker, C.; Gavina, P.; Jimenez-Molero, M.
C.; Sauvage, J.-P. Acc. Chem. Res. 2001, 34, 6, 477-487.
(4) Aboelella, N. W.; Kryatov, S. V.; Gherman, B. F.; Brennessel, W.
W.; Young, V. G., Jr.; Sarangi, R.; Rybak-Akimova, E. V.; Hodgson,
K. O.; Hedman, B.; Solomon, E. I.; Cramer, C. J.; Tolman, W. B. J.
Am. Chem. Soc. 2004, 126, 51, 16896-16911.
(9) Scaltrito, D. V.; Thompson, D. W.; O’Callaghan, J. A.; Meyer, G. J.
Coord. Chem. ReV. 2000, 208, 243-266.
(5) Aboelella, N. W.; Reynolds, A. M.; Tolman, W. B. Science 2004,
304, 5672, 836-837.
(10) Pallenberg, A. J.; Koenig, K. S.; Barnhart, D. M. Inorg. Chem. 1995,
34, 11, 2833-2840.
(6) Lewis, E. A.; Tolman, W. B. Chem. ReV. 2004, 104, 2, 1047-1076.
(7) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004,
104, 2, 1013-1045.
(11) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. J. Am. Chem. Soc.
1999, 121, 17, 4292-4293.
(8) Kim, Y. I.; Salim, S.; Huq, M. J.; Mallouk, T. E. J. Am. Chem. Soc.
1991, 113, 25, 9561-9563.
(12) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. Angew. Chem., Int.
Ed. 1998, 37, 11, 1556-1558.
3816 Inorganic Chemistry, Vol. 46, No. 10, 2007
10.1021/ic0615224 CCC: $37.00
© 2007 American Chemical Society
Published on Web 02/14/2007

2,9-Di-t-Bu-1,10-phenanthroline Copper(I) Complexes
These [CuL2]+ complexes have potential applications as
inexpensive and environmentally benign solar energy con-
version devices or sensors.14-16 In this class of compounds,
the homoleptic complexes [Cu(dnpp)2]+ and [Cu(dsbp)2]+
and the heteroleptic complex [Cu(dtbp)(dmp)]+ employ the
sterically bulkiest substituents at the 2 and 9 positions of
the phenanthroline ligand.17 These complexes demonstrate
the most useful photophysical properties: long excited-state
lifetimes and high quantum efficiencies.10,11 Two phenan-
throline ligands are not necessary; Cu(I) complexes with one
phenanthroline ligand and bulky auxiliary ligands such as
triphenylphosphine and bis[2-(diphenylphosphino)phenyl]
ether also exhibit excellent photophysical properties.18-21 To
optimize excited-state lifetimes and quantum efficiencies, it
is necessary to confront the considerable synthetic challenge
of incorporating highly bulky ligands into Cu(I) complexes.
The most common method for the synthesis of Cu(I)
complexes has been addition of the desired ligand to [Cu-
a less bulky phenanthroline ligand, i.e., dmp, to a solution
containing dtbp and [Cu(NCCH3)4]+ allows for the synthesis
of the heteroleptic complex [Cu(dtbp)(dmp)]+.11 This ap-
proach, dubbed HETPHEN, has been used to great advantage
in the preparation of supramolecular assemblies.28-30
Other commonly used synthetic methods for the prepara-
tion of Cu(I) complexes include reduction of Cu(II), com-
proportionation, and metathesis. Often used in the syntheses
of Cu(I) phenanthroline complexes is the reduction of Cu-
(II) starting materials by L-ascorbic acid in the presence of
ligands (reaction 2).13,24,25 Since this reaction is carried out
in alcohol and water mixtures, it is not suitable for preparing
water sensitive compounds. Some Cu(I) complexes are
nL + CuSO4 + NaY + L-ascorbic acid H O/CH OH8
2
3
[CuLn]Y + DHA + 1/2Na2SO4 (2)
CuY2 + Cu(s) + nL f 2[CuLn]Y
CuLX + AgY f CuLY + AgX
(3)
(4)
-
-
-
-
(NCCH3)4]Y (where Y is PF6 , ClO4 , SbF6 , or SO3CF3 ,
reaction 1). This synthetic method has been used successfully
to prepare many sterically congested Cu(I) systems,22,23
including both homoleptic and heteroleptic bis(phenanthro-
line)-Cu(I) complexes.10,11,12,24-26 For example, this route was
used to synthesize the homoleptic complexes [Cu(phen)2]+,
[Cu(dmp)2]+, [Cu(dpp)2]+,
-
-
-
-
Y ) SO3CF3 , BF4 , SbF6 , B(C6F5)4 , DHA )
dehydroascorbic acid, X ) halide
prepared by the comproportionation of Cu(II) and Cu(0) in
the presence of an appropriate ligand (reaction 3).25,26
Because the formation of Cu(I) from Cu(II) and Cu(0) is
unfavorable,31 these methods rely heavily on the ability of
the ligand to stabilize the Cu(I) ion. Finally, metathesis is
broadly used to exchange anionic ligands: a halide ion is
abstracted from the metal-halide precursor of the desired
complex by a Ag(I) salt, producing a Cu(I) complex with a
different coordinated anion (reaction 4). This method is less
often employed in Cu(I) complex syntheses; Hamilton and
co-workers provide the only successful implementation of
metathesis reactions for the synthesis of Cu(I)-phenanthroline
complexes.32
In this paper we present an oxidation-based method for
the synthesis of Cu(I) complexes. This new approach allows
weakly binding ligands access to the metal center of Cu(I)-
(dtbp) complexes to form three- and four-coordinate Cu(I)
complexes. Our synthetic route has enabled us to generate a
series of three-coordinate Cu(I)-phenanthroline complexes,
a class of complexes which are challenging to prepare. Only
4.3% of structurally characterized Cu(I) complexes are three-
coordinate.33 The [Cu(dtbp)2]+ complex, which was believed
to be synthetically inaccessible, was also synthesized. This
[Cu(NCCH3)4]Y + nL f [LnCu]Y + 4CH3CN (1)
[Cu(bcp)2]+, and [Cu(bfp)2]+.12,27 Larger substituents at the
2 and 9 positions of the phenanthroline, i.e., tert-butyl, impair
the ability of a second dtbp ligand to compete effectively
with acetonitrile for coordination to Cu(I), rendering [Cu-
(dtbp)2]+ inaccessible.10,11 However, subsequent addition of
(13) McMillin, D. R.; Buckner, M. T.; Ahn, B. T. Inorg. Chem. 1977, 16,
4, 943-945.
(14) Rogers, C. W.; Wolf, M. O. Coord. Chem. ReV. 2002, 233-234, 341-
350.
(15) O’Regan, B.; Graetzel, M. Nature 1991, 353, 6346, 737-740.
(16) Miller, M. T.; Karpishin, T. B. Sens. Actuators, B 1999, B61, 1-3,
222-224.
(17) phen, 1,10-phenanthroline; dnpp, 2,9-dineopentyl-1,10-phenanthroline;
dsbp, 2,9-di-sec-butyl-1,10-phenanthroline; dtp, 2,9-di-tert-butyl-1,-
10-phenanthroline; dmp, 2,9-dimethyl-1,10-phenanthroline; dpp, 2,9-
diphenyl-1,10-phenanthroline; bcp, 2,9-dimethyl-4,7-diphenyl-1,10-
phenanthroline; bfp, 2,9-bis-trifluoromethyl-1,10-phenanthroline.
(18) Rader, R. A.; McMillin, D. R.; Buckner, M. T.; Matthews, T. G.;
Casadonte, D. J.; Lengel, R. K.; Whittaker, S. B.; Darmon, L. M.;
Lytle, F. E. J. Am. Chem. Soc. 1981, 103, 19, 5906-5912.
(19) Palmer, C. E. A.; McMillin, D. R. Inorg. Chem. 1987, 26, 23, 3837-
3840.
(20) Cuttell, D. G.; Kuang, S.-M.; Fanwick, P. E.; McMillin, D. R.; Walton,
R. A. J. Am. Chem. Soc. 2002, 124, 1, 6-7.
(21) Kuang, S.-M.; Cuttell, D. G.; McMillin, D. R.; Fanwick, P. E.; Walton,
R. A. Inorg. Chem. 2002, 41, 12, 3313-3322.
(22) Spencer, D. J. E.; Reynolds, A. M.; Holland, P. L.; Jazdzewski, B.
A.; Duboc-Toia, C.; Le Pape, L.; Yokota, S.; Tachi, Y.; Itoh, S.;
Tolman, W. B. Inorg. Chem. 2002, 41, 24, 6307-6321.
(23) Spencer, D. J. E.; Aboelella, N. W.; Reynolds, A. M.; Holland, P. L.;
Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 10, 2108-2109.
(24) Kovalevsky, A. Y.; Gembicky, M.; Coppens, P. Inorg. Chem. 2004,
43, 26, 8282-8289.
(25) Kovalevsky, A. Y.; Gembicky, M.; Novozhilova, I. V.; Coppens, P.
Inorg. Chem. 2003, 42, 26, 8794-8802.
(26) Hirsch, J.; George, S. D.; Solomon, E. I.; Hedman, B.; Hodgson, K.
O.; Burstyn, J. N. Inorg. Chem. 2001, 40, 10, 2439-2441.
(27) Ruthkosky, M.; Castellano, F. N.; Meyer, G. J. Inorg. Chem. 1996,
35, 22, 6406-6412.
(28) Kalsani, V.; Schmittel, M.; Listorti, A.; Accorsi, G.; Armaroli, N. Inorg.
Chem. 2006, 45, 5, 2061-2067.
(29) Schmittel, M.; Ganz, A.; Fenske, D. Org. Lett. 2002, 4, 14, 2289-
2292.
(30) Schmittel, M.; Kalsani, V.; Fenske, D.; Wiegrefe, A. Chem. Commun.
2004, 5, 490-491.
(31) Cotton, F. A.; Wilkinson, G.; Bochmann, M.; Murillo, C. AdVanced
Inorganic Chemistry, 6th ed.; 1998.
(32) Hamilton, C. W.; Laitar, D. S.; Sadighi, J. P. Chem. Commun. 2004,
14, 1628-1629.
(33) A Cambridge Structural Database (updated November 2004) search
of three-coordinate Cu(I) complexes resulted in 195 complexes, and
a search for all Cu(I) complexes resulted in 4530 complexes.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3817

Gandhi et al.
of Cu powder (approximately 1 g). The mixture was stirred for 1
h and was then filtered through Celite and glass wool. The black-
brown solid was removed, yielding a yellow solution that was
evaporated to dryness under vacuum. The resulting yellow powder
was dissolved in CH2Cl2 and filtered again to clarify, and the solvent
was removed. The yellow solid was redissolved in acetone;
recrystallization from acetone/hexanes yielded 196 mg (88%) of
latter complex may have potential as a dyestuff for dye-
sensitized solar cells and as a sensor for small molecules.
Experimental Section
General Methods and Materials. All chemicals were purchased
from Aldrich and used without further purification unless otherwise
stated. K[B(C6F5)4] was purchased from Boulder Scientific Co.,
copper powder from STREM, and MnO2 from Fluka; all were used
as received. Solvent grade acetone and hexanes were purchased
from Columbus Chemical Industries, Inc., and spectrophotometric
grade CH2Cl2 was from Burdick and Jackson. All syntheses and
crystallizations were performed in a glove box under a nitrogen
atmosphere. Acetone, distilled from Drierite, and CH2Cl2 and
hexanes, distilled from calcium hydride, were degassed before use.
The ligand, dtbp, was synthesized as described previously.10 AgBF4
and Ag[B(C6F5)4] were synthesized from AgF and (CH3CH2)2O•BF3
and AgNO3 and K[B(C6F5)4], respectively.34 1H and 13C NMR
spectra were recorded at room temperature (22 °C) on a Varian
Mercury-300 MHz spectrometer. Chemical shifts were referenced
to the residual protons in the deuterated solvent or to the solvent
carbons and are reported in parts per million versus Me4Si. Infrared
spectra were recorded on a Bruker Vertex 70 FT-IR spectrometer.
Elemental analyses were performed by Desert Analytics.
Synthesis of Cu(dtbp)(CF3SO3) (1). An acetone solution (5 mL)
of dtbp (100 mg, 0.342 mmol) was added to a vial charged with
87.9 mg (0.342 mmol) of AgCF3SO3 and an excess of Cu powder
(approximately 1 g). The mixture was allowed to stir for 30 min.
The brown slurry was filtered through Celite and glass wool to
remove the black-brown solid. The resulting orange-yellow filtrate
was evaporated to dryness under vacuum. The orange solid was
dissolved in CH2Cl2 and filtered through Celite and glass wool to
clarify. Slow evaporation of a CH2Cl2/hexanes solution yielded 158
mg (92%) of large (4 mm × 4 mm × 1 mm) light-orange X-ray
quality plates of 1. 1H NMR (300 MHz, CD2Cl2): δ 1.756 (s, 18H,
CH3), δ 7.853 (s, 2H, CH), δ 8.024 (d, J ) 8.4 Hz, 2H, CH), δ
8.410 (d, J ) 8.7 Hz, 2H, CH) ppm. 13C NMR (75 MHz, CD2Cl2)
δ 30.70, 38.64, 122.36, 126.15, 127.55, 138.95, 143.72, 170.11 ppm.
IR (cm-1): CF3SO3-, 1234, 1217, 1180, 1161, 1028, 630. Anal.
Calcd for C21H24N2CuSO3F3: C, 49.94; H, 4.79; N, 5.55. Found:
C, 49.58; H, 4.91; N, 5.31.
Synthesis of Cu(dtbp)(BF4) (2). An acetone solution (5 mL) of
dtbp (100 mg, 0.342 mmol) was added to a vial containing 66.3
mg (0.342 mmol) of AgBF4 and an excess of Cu powder
(approximately 1 g). The mixture was allowed to stir for 30 min.
The brown slurry was filtered through Celite and glass wool to
remove the black-brown solid. The resulting yellow filtrate was
evaporated to dryness under vacuum. The orange solid was
dissolved in CH2Cl2 and filtered again to remove any residual solid.
This complex was challenging to purify. Precipitation from
CH2Cl2/hexanes yielded 143 mg (94.4%) of small thin yellow
needles of impure 2. A poor yield of X-ray quality plates was
obtained upon slow evaporation of a CH2Cl2/hexanes solution. 1H
NMR (300 MHz, CD2Cl2): δ 1.731 (s, 18H, CH3), δ 7.917 (s, 2H,
CH), δ 8.064 (d, J ) 8.7 Hz, 2H, CH), δ 8.479 (d, J ) 8.4 Hz,
2H, CH) ppm. IR (cm-1): BF4-, 1138, 1089, 1054, 1024, 1013.
This compound could not be obtained in high analytical purity,
and, thus, a satisfactory elemental analysis was not obtained.
Synthesis of [Cu(dtbp)((CH3)2CO)][SbF6] (3). An acetone
solution (5 mL) of dtbp (100 mg, 0.342 mmol) was added to a
vial containing 119.6 mg (0.349 mmol) of AgSbF6 and an excess
1
thin yellow needles of 3. H NMR (300 MHz, CD2Cl2): δ 1.677
(s, 18H, CH3), δ 2.460 (s, 6H, CH3), δ 7.966 (s, 2H, CH), δ 8.091
(d, J ) 8.7 Hz, 2H, CH), δ 8.537 (d, J ) 8.4 Hz, 2H, CH) ppm.
13C NMR (75 MHz, CD2Cl2) δ 30.64, 32.44, 38.27, 122.60, 126.56,
128.02, 139.91, 143.58, 169.41, 185.99 ppm. IR (cm-1): SbF6
,
-
654. Anal. Calcd for C22H30N2CuOSbF6: C, 42.51; H, 4.65; N,
4.31. Found: C, 42.79; H, 4.44; N, 4.15.
New Method for Synthesis of [Cu(dtbp)(dmp)][BF4]. An
acetone solution (5 mL) of dtbp (50 mg, 0.171 mmol) was added
to a vial containing 33.1 mg (0.171 mmol) of AgBF4 and an excess
of Cu powder (approximately 1 g). The mixture was allowed to
stir for 1 h and was filtered through Celite and glass wool. A black-
brown solid was removed, yielding a yellow solution of [Cu(dtbp)-
((CH3)2CO)][BF4] (3). Upon addition of 2,9-dimethyl-1,10-
phenanthroline (35.6 mg, 0.171 mmol), an intense orange-red color
resulted. Crystallization from acetone/hexanes yielded 97.3 mg
(87.4%) of orange blocks of [Cu(dmp)(dtbp)][BF4]. The composi-
tion was confirmed by comparison with previously reported NMR
spectra.11
Synthesis of [Cu(dtbp)2][B(C6F5)4]•CH2Cl2 (4•CH2Cl2). An
acetone solution (5 mL) of dtbp (89.6 mg, 0.254 mmol) was added
to a vial containing 100 mg (0.127 mmol) of Ag[B(C6F5)4] and an
excess of Cu powder (approximately 1 g). The mixture was allowed
to stir for 30 min and was filtered through Celite and glass wool.
A black-brown solid was removed, yielding a yellow-orange
solution that was evaporated to dryness under vacuum. The orange
solid was dissolved in CH2Cl2 and filtered to clarify, producing a
clear orange solution. Large bright orange X-ray quality blocks of
4•CH2Cl2 (146.3 mg, 81.5%) were isolated by layering with
hexanes or by addition of hexanes followed by slow evaporation.
1H NMR (300 MHz, CD2Cl2): δ 1.214 (s, 36H, CH3), δ 7.996 (s,
4H, CH), δ 8.071 (d, J ) 9.0 Hz, 4H, CH), δ 8.484 (d, J ) 8.7
Hz, 4H, CH) ppm. 13C NMR (75 MHz, CD2Cl2) δ 30.69, 39.17,
124.60, 127.55, 129.49, 138.76, 143.61, 169.01 ppm. IR (cm-1):
[B(C6F5)4]-, 1511, 1462, 1274, 1087, 979, 774, 768, 756, 683, 660.
Anal. Calcd for C64H48N4CuBF20•1/3CH2Cl2: C, 56.99; H, 3.62;
N, 4.13. Found: C, 57.03; H, 3.83; N, 4.24.
Synthesis of [Cu(dtbp)2][BF4]•CH2Cl2 (5•CH2Cl2). An acetone
solution (5 mL) of dtbp (100 mg, 0.342 mmol) was added to a
vial containing 33.2 mg (0.171 mmol) of AgBF4 and an excess of
Cu powder (approximately 1 g). The mixture was allowed to stir
for 30 min and then filtered through Celite and glass wool. The
resulting clear orange solution was evaporated to dryness under
vacuum. The orange solid was dissolved in CH2Cl2 and filtered to
clarify. Layering with hexanes yielded 130.6 mg (93.1%) of large
bright orange needles of 5•CH2Cl2. 1H NMR (300 MHz, CD2Cl2):
δ 1.225 (s, 36H, CH3), δ 8.043 (s, 4H, CH), δ 8.096 (d, J ) 8.7
Hz, 4H, CH), δ 8.534 (d, J ) 8.7 Hz, 4H, CH) ppm. 13C NMR (75
MHz, CD2Cl2) δ 30.73, 39.13, 124.62, 127.63, 129.60, 138.85,
169.26 ppm. IR (cm-1): BF4-, 1072.73, 1057.27, 1030.78, 633.99.
Anal. Calcd for C40H48N4CuBF4: C, 60.04; H, 6.15; N, 6.83.
Found: C, 60.29; H, 6.57; N, 6.70.
Synthesis of [Cu(dtbp)2][SbF6]•CH2Cl2 (6•CH2Cl2). An acetone
solution (5 mL) of dtbp (100 mg, 0.342 mmol) was added to a
vial containing 59.8 mg (0.171 mmol) of AgSbF6 and an excess of
(34) Green, O.; Santiago-Cintron, M.; Burstyn, J. N. Manuscript in
preparation.
3818 Inorganic Chemistry, Vol. 46, No. 10, 2007

2,9-Di-t-Bu-1,10-phenanthroline Copper(I) Complexes
Table 1. X-ray Crystallographic Data for Cu(dtbp)(OTf) (1), Cu(dtbp)(BF4) (2), [Cu(dtbp)((CH3)2CO)][SbF6] (3),
[Cu(dtbp)2][B(C6F5)4]•CH2Cl2(4•CH2Cl2), [Cu(dtbp)2][BF4]•CH2Cl2(5•CH2Cl2), and [Cu(dtbp)2][SbF6]•CH2Cl2(6•CH2Cl2)
1
2
3
4‚CH2Cl2
5‚CH2Cl2
6‚CH2Cl2
formula
formula weight
T (K)
C21H24N2CuSO3F3
505.02
100(2)
0.71073
monoclinic
P21/n
11.1025(7)
10.3858(7)
19.158(1)
90
97.848(1)
90
2188.4(2)
4
1.533
1.144
C20H24N2CuBF4
442.76
100(2)
0.71073
monoclinic
P21/c
6.800(1)
19.943(3)
14.035(2)
90
92.709(2)
90
1901.1(4)
4
1.547
1.194
C22H30N2CuOSbF6 C65H50N4CuBF20Cl2 C41H50N4CuBF4Cl2 C41H50N4CuSbF6Cl2
649.78
100(2)
0.71073
monoclinic
P21/c
1412.34
100(2)
0.71073
monoclinic
P21/c
820.10
100(2)
0.71073
triclinic
P1h
884.11
100(2)
0.71073
triclinic
P1h
λ (Å)
crystal system
space group
a (Å)
14.337(2)
6.8138(9)
26.391(3)
90
104.818(2)
90
14.704(2)
25.882(3)
16.704(2)
90
108.057(2)
90
14.754(1)
15.138(1)
17.956(1)
90.459(1)
91.190(1)
98.885(1)
3961.3(4)
4
12.200(1)
13.628(1)
14.403(1)
92.833(1)
109.737(1)
107.814(1)
2114.3(2)
2
b (Å)
c (Å)
R (°)
â (°)
γ (°)
V (Å3)
Z
2492.4(6)
4
6043.8(10)
4
D
calc (g/cm3)
1.732
1.552
1.375
1.389
µ (mm-1
F(000)
)
2.000
0.558
0.740
1.199
1040
912
1296
2864
1712
900
crystal size (mm3) 0.34 × 0.22 × 0.14 0.32 × 0.12 × 0.11 0.40 × 0.17 × 0.08 0.25 × 0.20 × 0.10
0.45 × 0.45 × 0.09 0.18 × 0.15 × 0.11
R1, wR2 [I>(2σI)] 0.0309, 0.0787
R1, wR2 (all data) 0.0375, 0.0827
0.0489, 0.1170
0.0683, 0.1278
0.0389, 0.0913
0.0514, 0.0976
0.0552, 0.1274
0.0882, 0.1447
0.0539, 0.1425
0.0637, 0.1489
0.0315, 0.0753
0.0398, 0.0785
Cu powder (approximately 1 g). The mixture was allowed to stir
for 30 min and then filtered through Celite and glass wool. The
resulting clear orange solution was evaporated to dryness under
vacuum. The orange solid was dissolved in CH2Cl2 and filtered to
clarify. Layering with hexanes yielded 132.4 mg (87.6%) of large
bright orange crystals of 6•CH2Cl2. 1H NMR (300 MHz, CD2Cl2):
δ 1.221 (s, 36H, CH3), δ 8.028 (s, 4H, CH), δ 8.088 (d, J ) 8.4
Hz, 2H, CH), δ 8.517 (d, J ) 8.7 Hz, 2H, CH) ppm. 13C NMR (75
MHz, CD2Cl2) δ 29.74, 36.22, 123.67, 126.45, 128.52, 137.66 ppm.
IR (cm-1): SbF6-, 656.32. Anal. Calcd for C24H48N4CuSbF6•
1/2CH2Cl2: C, 52.50; H, 5.33; N, 6.05. Found: C, 52.04; H, 5.39;
N, 5.76.
unit cell parameters obtained from CellNow and corrected for
absorption effects using TWINABS. The solution and refinement
were carried out as described for 1-4•CH2Cl2 once the corrected
reflection file was obtained.
There were two partially occupied solvate molecules of CH2Cl2
present in the asymmetric unit of 6•CH2Cl2. A significant amount
of time was invested in identifying and refining the disordered
molecules. Bond-length restraints were applied to model the
molecules, but the resulting unusually large isotropic displacement
coefficients suggested the molecules were mobile. Option SQUEEZE
of program PLATON was used to correct the diffraction data for
diffuse scattering effects and to identify the solvate molecule.37
PLATON calculated the upper limit of volume that can be occupied
by the solvent to be 366.1 Å3 or 17.3% of the unit cell volume.
The program calculated 90 electrons in the unit cell for the diffuse
species. This approximately corresponds to two solvate CH2Cl2
molecules in the unit cell (84 electrons). It is very likely that this
solvate molecule is disordered over several positions. All derived
results in the following tables are based on the known contents.
No data are given for the diffusely scattering species.
X-ray Structure Determination. Suitable crystals were selected
under oil in air at room temperature. The crystals were mounted
on the tip of a Nylon loop and immediately placed in a stream of
nitrogen at 100(2) K. The data collection was performed on a Bruker
CCD-1000 diffractometer with Mo KR (λ ) 0.71073 Å) radiation.
The detector was placed at a distance of 4.9 cm from the crystal.
The data frames were integrated with the Bruker SAINT-Plus
software package and corrected for absorption effects using
SADABS.35 Crystal structures for compounds 1-4 and 4•CH2Cl2
were solved by direct methods, and all non-hydrogen atoms were
identified on the initial electron density map. The non-hydrogen
atoms were subsequently refined by full-matrix least-squares
methods with anisotropic displacement coefficients. All hydrogen
atoms were calculated at idealized positions and were refined as
riding atoms with individual isotropic coefficients. Further details
of the data collection and refinement are listed in Table 1.
Results
Facile Synthesis of Cu(I) Complexes with Hindered
Phenanthroline Ligands. Controlled oxidation of metallic
copper by stoichiometric amounts of silver salt in the
presence of ligand results in the facile synthesis of Cu(I)
complexes of hindered phenanthrolines. When excess copper
metal is stirred with 1 equiv of a Ag(I) salt and 1 or 2 equiv
of 2,9-di-tert-butyl-1,10-phenathroline (dtbp), Cu(I) com-
plexes bearing either one or two bulky phenanthroline ligands
may be isolated in high yield (reaction 5). Reaction progress
may be easily monitored by eye; reduction of the Ag(I) is
observable via formation of a black Ag(0) precipitate, while
oxidation of the Cu(0) results in formation of the yellow or
orange Cu(I)LnY (n ) 1 or 2) species.
Under the microscope, several seemingly suitable single crystals
of 5•CH2Cl2, all of which showed strong reflections with no
evidence of twinning, were selected under oil in air at room
temperature. Bruker’s SMART software was unable to determine
a unit cell after matrix collections for any of the crystals. After a
full data collection of one crystal, CellNow was executed, finding
a two part twinned crystal.36 The two domains of 52.6% and 47.4%
proportions were related by 180° rotation. The data frames were
integrated with the Bruker SAINT-Plus software package and the
Cu(s) + AgY + nL acetone8 [CuLn]Y + Ag(s)
(5)
This method provides a facile route to Cu(I)(dtbp)Y
(35) Bruker-AXS. SADABS V.2.05, T., SAINT V.6.22, SHELXTL V.6.10,
& SMART V.5.622 Software Reference Manuals; Bruker-AXS: Madi-
son, WI, U.S.A., 2000-2004.
(36) Sheldrick, G. M. CellNow; University of Go¨ttingen: Go¨ttingen,
Germany, 2005.
complexes, where Y- is one of a variety of weakly
(37) Van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A: Found.
Crystallogr. 1990, A46, 3, 194-201.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3819

Gandhi et al.
may be isolated in high yield and purity, improving on a
prior synthesis. Reaction of 1 equiv AgBF4 and 1 equiv of
dtbp in the presence of excess Cu(0) in acetone produces
[Cu(dtbp)((CH3)2CO)][BF4]. Direct addition of 1 equiv of
dmp to the yellow [Cu(dtbp)((CH3)2CO)][BF4] solution
resulted in the desired deep orange complex [Cu(dmp)(dtbp)]-
Figure 1. Packing diagram of the unit cell of three-coordinate Cu(dtbp)-
(BF4) (2) shown along the c-axis with 50% probability thermal ellipsoids.
Note the occurrence of π-stacking between the phenanthroline planes of
two inverted cations. H atoms have been omitted for clarity.
1
[BF4] in 87% yield. The H NMR of this product showed
no evidence of [Cu(dmp)2]+, an impurity reported in the prior
synthesis.11 The order of ligand addition and the steric bulk
of dtbp are key to the success of this method: when 1 equiv
of AgCF3SO3 and 1 equiv of dmp were mixed together and
allowed to react with excess Cu(0), the only product was
1/2 of an equiv of [Cu(dmp)2][CF3SO3].
Remarkably, this oxidation-based method can be used to
prepare the heretofore elusive complex cation [Cu(dtbp)2]+.
Mixtures of the silver salts Ag[B(C6F5)4], AgBF4, or AgSbF6
with dtbp in a 1:2 ratio react with excess Cu(0) to afford
the [Cu(dtbp)2]+ complexes 4•CH2Cl2, 5•CH2Cl2, and
6•CH2Cl2 in good yields (82%, 93%, and 88%, respectively).
The process by which the [Cu(dtbp)2]+ cation is formed is
complex and highly solvent dependent. The [Cu(dtbp)2]+
product is only formed if the oxidation reaction is carried
out in acetone, ethanol, or THF. In CH2Cl2 or toluene no
color change is observed on the same time scale. The initial
product in successful reactions is the [Cu(dtbp)(solvato)]+
complex; the [Cu(dtbp)2]+ species is formed when the
solvent is removed. In acetone, the initial product is pale
yellow, similar to acetone solutions of 3. The second ligand
appears to bind as acetone is removed: as the solvent volume
decreases the solution darkens, tending toward orange, and
the residual solid varies from dark yellow to orange. This
solid dissolves to produce a bright orange solution in
CH2Cl2 and yields bright orange crystals upon precipitation
with hexanes.
Figure 2. Face-on molecular drawing of three-coordinate Cu(dtbp)(SO3-
CF3) (1) shown with 50% probability thermal ellipsoids. H atoms have been
omitted for clarity.
Structures of Cu(dtbp)(CF3SO3) (1) and Cu(dtbp)(BF4)
(2). The structures of 1 and 2 confirm the 1:1 Cu:dtbp
stoichiometry and reveal that the counterion binds to the
metal. Compounds 1 and 2 crystallize in the P21/n and P21/c
space groups, respectively, with four formula units occupying
the unit cell and no solvent molecules present. The unit cell
compositions are thus consistent with a +1 oxidation state
for the metal ion. Furthermore, the anomalous scattering
properties of the heavy atoms and the elemental analyses
are consistent with the lighter Cu atom and not the heavier
Ag atom in the complexes. Both compounds participate in
π-stacking but in different ways. The crystal packing of 2,
illustrated in Figure 1, reveals that the phenanthroline rings
engage in infinitely long π-stacking interactions. Between
any two stacking molecules lies an inversion center; thus,
each molecule is inverted with respect to both of its stacking
partners. The phenanthroline planes are stacked at average
perpendicular distances of 3.4 Å and are offset with respect
to one another. The packing diagram of complex 1 (Figure
S1) reveals that the molecules engage only in pairwise
π-stacking interactions. In each pair, the two molecules are
offset and inverted with respect to one another.
Figure 3. Side-on molecular drawing of three-coordinate Cu(dtbp)(BF4)
(2) shown with 50% probability thermal ellipsoids. Note the orientation of
the tert-butyl methyl groups and the bending of the Cu atom and BF4- ion
away from the upward-facing methyl groups.
coordinating anions. Mixtures of AgCF3SO3, AgBF4, or
AgSbF6 with dtbp in a 1:1 molar ratio react with Cu(0) to
give the mono-dtbp Cu(I) complexes 1-3 (92%, 94%, and
88% yields, respectively). These are solvent-facilitated
reactions; the redox process will not occur efficiently unless
the solvent is of modest polarity and coordinating ability. In
acetone and ethanol the redox reaction is complete in
minutes, whereas in THF the reaction occurs over the course
of hours, and in dichloromethane the reaction may go to
completion in days to weeks. We believe that the reaction
product is a solvato complex in coordinating solvents, but
the coordinated solvent molecule may be lost upon workup
if an alternate ligand is present, as occurs in complexes 1
and 2.
The method is particularly well-suited to the clean
preparation of mixed ligand complexes of the hindered
phenanthroline ligand, dtbp. The complex, [Cu(dmp)-
(dtbp)]+, where dmp is 2,9-dimethyl-1,10-phenanthroline,
The geometry about the metal ions is distorted trigonal
planar. The metal ion in each complex is three coordinate,
3820 Inorganic Chemistry, Vol. 46, No. 10, 2007

2,9-Di-t-Bu-1,10-phenanthroline Copper(I) Complexes
Table 2. Significant Bond Distances (in Å) of Cu(dtbp)(OTf) (1), Cu(dtbp)(BF4) (2), [Cu(dtbp)((CH3)2CO)][SbF6] (3), [Cu(dtbp)2][B(C6F5)4] (4),
[Cu(dtbp)2][B(C6F5)4]•CH2Cl2(4•CH2Cl2), [Cu(dtbp)2][BF4]•CH2Cl2(5•CH2Cl2), and [Cu(dtbp)2][SbF6]•CH2Cl2(6•CH2Cl2)
1
2
3
4
4•CH2Cl2
5•CH2Cl2
2.081(2)
2.105(2)
2.092(2)
2.114(2)
6•CH2Cl2
Cu-N(1)
Cu-N(2)
Cu-N(3)
Cu-N(4)
Cu-F or O
av Cu-N
2.102(2)
1.990(2)
2.040(3)
2.019(3)
2.056(3)
2.012(3)
2.130(1)
2.096(1)
2.103(1)
2.120(1)
2.096(3)
2.115(3)
2.076(3)
2.139(3)
2.108(2)
2.105(2)
2.104(2)
2.111(2)
2.072(2)
2.079(2)
2.145(2)
2.149(2)
1.927(2)
2.0462(16)
2.012(2)
2.030(3)
1.929(3)
2.034(3)
2.1121(12)
2.107(3)
2.103(2)
2.1108(19)
with the two nitrogens of the phenanthroline (dtbp) and
counterion is 5.072 Å; the shortest distance between the Cu
atom and the Sb atom is 6.764 Å.
-
-
either an O atom (CF3SO3 , 1) or an F atom (BF4 , 2) as
ligands. The coordination environment of the Cu(I) is shown
in Figure 2 for complex 1. The Cu-N(1) and Cu-N(2)
distances are 2.102(2) Å and 1.990(2) Å, respectively, for 1
and 2.040(3) Å and 2.019(3) Å, respectively, for 2. The
distances from the Cu(I) to the coordinating atoms of the
anions are 1.927(2) Å for 1 (O(1)) and 2.012(2) Å for 2
(F(1)). These bond lengths are listed in Table 2. The metal
ion and counterion-derived ligand reside above the plane of
the phenanthroline, as shown in the packing diagram of 2 in
Figure 1, and when the molecule is viewed along the plane
of the phenanthroline ligand in Figure 3 (complex 1, Figure
S2). The plane containing the Cu and the two phenanthroline
nitrogen atoms is coplanar neither with the phenanthroline
aryl rings nor with the Cu-X (anion atom) bond. The angle
between the N-Cu-N plane and the phenanthroline aryl
plane is 22.4° in 1 and 14.2° in 2. The angle between the
N-Cu-N plane and the Cu-X bond is 16.3° in 1 and 10.9°
in 2. Two pairs of methyl groups point downward, away
from the counterion ligand, while the third pair of methyl
groups points upward. The counterion is nestled into the cleft
created by the upward facing methyl groups. Similar to the
The coordination sphere of the metal ion in 3 is composed
of the two nitrogen atoms of the phenanthroline and the
carbonyl oxygen atom of the acetone molecule. The Cu-
N(1) and Cu-N(2) distances are 2.056(3) Å and 2.012(3)
Å, respectively; the Cu(I)-O bond distance is 1.929(3) Å.
The coordination sphere of the metal center is distorted
trigonal planar, with the metal ion and the acetone ligand
above the plane of the phenanthroline. In 3, the angle
between the N-Cu-N plane and the phenanthroline aryl
plane is 19.6°, and the angle between the N-Cu-N plane
and the Cu-O bond is 23.1°. Similar to the structures of 1
and 2, the methyl groups of the tert-butyl substituents are
rotated such that there is only one methyl group on the face
of the phenanthroline where the Cu(I)-acetone moiety resides.
As expected, the chemical shift of the methyl group of the
dtbp in complex 3 is also shifted downfield (δ 1.677 ppm)
compared to that of free dtbp. The protons of the ligated
acetone can been observed by NMR at δ 2.460 ppm.
Structures of [Cu(dtbp)2][B(C6F5)4]•CH2Cl2 (4•CH2Cl2),
[Cu(dtbp)2][BF4]•CH2Cl2 (5•CH2Cl2), and [Cu(dtbp)2]-
[SbF6]•CH2Cl2 (6•CH2Cl2). The previously elusive Cu(I)-
(dtbp)2 cation is largely similar in structure to other members
of this class of compounds; the exceptional features are the
elongated Cu-N bonds. As observed in the illustration of a
single formula unit in Figure 6, two bulky dtbp ligands
coordinate the Cu(I) atom. Crystals of 4•CH2Cl2 contain four
[Cu(dtbp)2]+ cations, four [B(C6F5)4]- anions, and four
CH2Cl2 solvent molecules in the unit cell in the P21/c space
group. The contents of the unit cell are consistent with a +1
oxidation state of the Cu center. Elemental analysis and the
anomalous scattering of the heavy atom are consistent with
the presence of a Cu atom. The bulky cation and anion are
well separated from one another; the distance of closest
approach between the Cu atom and an F atom of [B(C6F5)4]-
is 5.500 Å. The shortest distance between the centroids of
the cation and anion is 9.416 Å.
1
previous H NMR observations of dnpp complexes, the
chemical shifts of the methyl groups of the substituents of
dtbp in 1 (δ 1.756 ppm) and 2 (δ 1.731 ppm) are shifted
downfield with respect to that of free dtbp ligand (δ 1.58
ppm); this shift is expected due to the inductive effect upon
complexation of the ligand to a Lewis acidic, electropositive
center such as Cu(I).10,11
Structure of [Cu(dtbp)((CH3)2CO)][SbF6] (3). In con-
trast, the counterion in 3 does not serve as a ligand, rather,
a solvent molecule binds to the metal. Compound 3 crystal-
lized in a P21/c space group, with four formula units, i.e.,
four [Cu(dtbp)((CH3)2CO)]+ cations and four SbF6- anions,
occupying the unit cell. The complex cation is shown in
Figure 4 and the packing diagram with the unit cell is shown
in Figure 5. The anomalous scattering properties of the heavy
atom and elemental analysis are only consistent with the
presence of a Cu atom in the complex, and the contents of
the unit cell are again consistent with the Cu(I) oxidation
state of metal ion. Comparison of the packing arrangements
between 2 (Figure 1) and 3 (Figure 5), which crystallize in
the same space group, reveals similar infinite stacking
interactions of the Cu-phenanthroline units: the distance
between the phenanthroline planes of the two inverted cations
in the center of Figure 5 is 3.4 Å. Interestingly, however,
the stacking axes are different between the two crystal
structures (a for 2 and b for 3). In 3, the distance of closest
The coordination geometry of the metal is pseudotetra-
hedral. The complex exhibits elongated Cu-N bonds result-
ing in a significant D2d distortion along one axis. The Cu-
N(x) (where x ) 1-4) bond distances are 2.096(3) Å,
2.115(3) Å, 2.076(3) Å, and 2.139(3) Å, respectively, for
an average distance of 2.11 Å. The intraligand N-Cu-N
angles are 84.53(10)° and 84.43(10)°, while the interligand
N-Cu-N angles are 123.97(10)°, 125.12(10)°, 121.09(10)°,
and 122.70(10)°. One phenanthroline ligand is slightly
distorted from its position in an idealized D2d geometry. As
illustrated in Figure 7, the plane of one phenanthroline is
tilted slightly downward and displaced to one side. Another
-
approach between the Cu atom and an F atom of the SbF6
Inorganic Chemistry, Vol. 46, No. 10, 2007 3821

Gandhi et al.
Figure 7. Drawing of the [Cu(dtbp)2]+ cation of complex 4•CH2Cl2
shown with 50% probability thermal ellipsoids. Only slight distortions from
pseudotetrahedral D2d geometry are observed, with one dtbp ligand tilted
down and displaced sideways. H atoms have been omitted for clarity.
Figure 4. View of the three-coordinate [Cu(dtbp)((CH3)2CO)][SbF6] (3)
shown with 50% probability thermal ellipsoids. The Cu atom and the acetone
ligand are out of the plane of the phenanthroline ligand, similar to complex
2, as shown in Figure 3. H atoms have been omitted for clarity.
[Cu(dtbp)2]+ cations, four BF4 anions, and four CH2Cl2
-
solvent molecules in a P1h space group; the asymmetric unit
contains two formula units. The average Cu-N bond distance
in the two independent molecules is 2.10 Å; individual bond
distances are listed in Table 2. The structure of the cation in
5•CH2Cl2 is similar to the second, less common isomorph
of 4, which has a more idealized D2d geometry. The distances
between the cations and anions in the structure of 5•CH2Cl2
are shorter than those of 4•CH2Cl2, as expected for the
smaller anion: Cu-F is 6.29 Å and 7.72 Å; Cu-B is 7.35
Å and 8.80 Å, respectively, for the two molecules in the
asymmetric unit. A picture of one of the molecules of the
asymmetric unit is included in the Supporting Information
(Figure S3).
Figure 5. Packing diagram of the unit cell of the three-coordinate [Cu-
(dtbp)((CH3)2CO)][SbF6] shown along the b-axis with 30% probability
thermal ellipsoids. Note the occurrence of π-stacking between the phenan-
throline planes of two inverted cations in the center. H atoms have been
omitted for clarity.
The unit cell of 6•CH2Cl2 contains two formula units in
-
the P1h space group, i.e., two [Cu(dtbp)2]+ cations, two SbF6
anions, and two CH2Cl2 solvent molecules. The average
Cu-N bond distance is 2.11 Å, although the individual bond
distances (Table 2) span the widest range of the above three
complexes. The distortion of the two phenanthroline rings
from idealized D2d geometry is also greater than those of
the above three complexes. The Cu-F distance is 5.56 Å,
and the Cu-Sb distance is 7.08 Å. Figure S4 illustrates the
[Cu(dtbp)2]+ cation of 6•CH2Cl2. The chemical shifts of the
methyl groups of the dtbp ligand in 4•CH2Cl2, 5•CH2Cl2,
and 6•CH2Cl2 (δ 1.214, 1.225, and 1.221 ppm, respectively)
are shifted upfield relative to that of free dtbp ligand. This
upfield shift phenomenon has previously been attributed to
ring current effects on the alkyl groups.10,11
Figure 6. View of the four-coordiante [Cu(dtbp)2][B(C6F5)4]•CH2Cl2
(4•CH2Cl2) formula unit shown with 50% probability thermal ellipsoids.
The two dtbp ligands are arranged in a pseudotetrahedral fashion. H atoms
have been omitted for clarity.
crystal form of 4, without a solvent molecule, was obtained;
the structure of the cation was similar, though the extent of
distortion of the phenanthroline ligand was different. This
second structure was of a more idealized D2d cation; minimal
distortion of the phenanthroline ligands and smaller devia-
tions among the Cu-N bond distances were observed (Table
S1). The average Cu-N bond length in this second structure
was 2.11 Å. This second crystal form was a less frequently
observed morphology for crystals of this compound.
Discussion
We report a novel synthetic method allowing access to
Cu(I) complexes with weakly coordinated ligands, complexes
that may be useful in molecular machines, biomimetic
models, catalysts, or solar energy conversion devices.
Conventional synthetic methods are insufficient for the
synthesis of such complexes. Our method combines the
oxidizing power of Ag(I) and the elegance of compropor-
tionation reactions. Via this reaction, we synthesized Cu(I)-
The [Cu(dtbp)2]+ cation in 5•CH2Cl2 is similar to that in
4•CH2Cl2. The unit cell contains four formula units, i.e., four
3822 Inorganic Chemistry, Vol. 46, No. 10, 2007

2,9-Di-t-Bu-1,10-phenanthroline Copper(I) Complexes
(dtbp) complexes with either weakly coordinating counte-
rions (1 and 2) or the weakly coordinating solvent acetone
(3). Most surprisingly, the [Cu(dtbp)2]+ complexes (4-6),
which were previously reported to be impossible to synthe-
size, were easily and cleanly prepared via this method. The
difficulty in forming [Cu(dtbp)2]+, because of the great steric
bulk of the substituents at the 2 and 9 positions, is evidenced
by the impressive Cu(I)-N distancessthe longest for this
class of complexessin the molecular structures of 4-6.
Previous studies of complexes with the general formula
[Cu(dap)(B)]+ (dap, 2,9-dialkyl-1,10-phenanthroline; B, a
less bulky phenanthroline or chloride) showed that the three-
coordinate chloride adduct Cu(dap)Cl could only be formed
with the bulky dtbp and dnpp ligands; reactions with all
other substituted dialkylphenanthrolines resulted in only the
[Cu(dap)2]+ complexes.10,11 With dnpp, both Cu(dnpp)Cl and
[Cu(dnpp)2]Cl could be isolated. These observations revealed
that only the tert-butyl groups are sufficiently bulky to inhibit
the binding of a second dtbp ligand, suggesting that the
second bidentate dtbp is a weaker ligand than monodentate
chloride in these systems.
In an effort to synthesize Cu(I)(dtbp) complexes with
weakly coordinated ligands, several synthetic routes were
evaluated. Direct reaction of dtbp with [Cu(CH3CN)4][PF6]
only led to the formation of the three-coordinate complex
[Cu(dtbp)(CH3CN)][PF6], regardless of the ratio of dtbp
used. (Data collection, refinement details, and a crystal
structure of [Cu(dtbp)(CH3CN)][PF6] are provided in the
Supporting Information in Table S1 and Figure S5.) The tert-
butyl substituents of the dtbp ligand impede effective
competition with CH3CN for ligation to the Cu(I) center.
These observations demonstrate that the second bidentate
dtbp is a weaker ligand than monodentate CH3CN, which
led previous researchers to believe that the homoleptic [Cu-
(dtbp)2]+ complex was impossible to synthesize.11 Since our
goal was to prepare three-coordinate complexes with very
weakly coordinating ligands, it was obvious that ligand
displacement on [Cu(dtbp)(CH3CN)]+ was not a viable
synthetic strategy.
by a Cu(II) salt in the presence of dtbp, proved fruitless for
this system. Reactions of this type in a variety of solvents
were unpredictable and never resulted in the desired products;
the reaction solutions were consistently blue to green in color,
indicative of Cu(II) products. This observation suggests that
Cu(II) is not a strong enough oxidant to oxidize Cu(0) to
Cu(I) in the presence of dtbp. It should be noted that the
only Cu(II) source tested was the commercially available
Cu(BF4)2•6H2O, and the waters of hydration may inhibit
comproportionation.
The logical substitution of Cu(II) with Ag(I) as an
oxidizing agent for excess Cu(0) provides a novel and
straightforward path to Cu(I) complexes.38 In this high-
yielding one-pot reaction, Cu(0) is oxidized to Cu(I), the
ligand is bound to Cu(I), the Cu(I) complex is charge-
balanced with a desirable counterion, and, conveniently, Ag-
(0) and the remaining Cu(0) are removed by filtration. The
reaction proceeds more quickly in a polar coordinating
solvent such as acetone than in a less polar solvent such as
CH2Cl2. The fact that the solvato complex 3 is the product
in reaction of AgSbF6, dtbp and Cu(0) strongly suggests
that acetone temporarily binds to and stabilizes the newly
formed Cu(I) ion. The absence of solvent coordination likely
explains the slow rate of reaction in CH2Cl2. The beauty of
this reaction is that it allows the synthesis of a variety of
coordinatively unsaturated Cu(I) complexes via a simple
reaction in which unwanted species are easily separated. Our
method appears to be general and should be applicable to
the preparation of otherwise inaccessible Cu(I) complexes
with a variety of ligands. A further advantage is the wide
variety of commercially available silver salts that may be
used as oxidants and provide potential anionic ligands or
counterions.
The use of excess Cu(0) was not a necessity for the success
of this reaction, rather the small scale of these reactions
would have required measuring inconveniently small quanti-
ties of Cu(0). One can imagine that on an industrial scale,
this reaction could be performed in an environmentally
friendly manner, producing no waste: Cu(0) could be used
stoichiometrically and Ag(0) and solvent could be recovered,
with the former used for the regeneration of AgY starting
material.
Attempted alternative routes to the desired three-coordinate
complexes were as follows: (1) preparation of a halide
complex and metathesis with a silver salt of the desired
anionic ligand or (2) comproportionation of Cu(0) with a
Cu(II) salt. Metathesis proved ineffective due to incomplete
chloride abstraction; reaction of Cu(dtbp)Cl with various Ag-
(I) salts did not produce the desired Cu(dtbp)Y (Y is
The size and composition of the anion plays an important
role in the structure of the resulting complex: specifically,
whether the anion will bind to the Cu(I) center. The
complexes Cu(dtbp)(O3SCF3) (1) and Cu(dtbp)(BF4) (2)
contain bound anions. In (2), the relatively small BF4- anion
sits comfortably in the cleft created by the tert-butyl groups.
-
-
-
-
SO3CF3 , BF4 , SbF6 , or B(C6F5)4 ) complexes. The
reaction of Cu(dtbp)Cl with stoichiometric AgSbF6 produced
a chloride-bridged π-stacked [(Cu(dtbp))2Cl]+ species. (Data
collection, refinement details, and a crystal structure of [(Cu-
(dtbp))2Cl][SbF6] are in the Supporting Information in Table
S1 and Figure S6.) In these systems it was thermodynami-
cally unfavorable for stoichiometric chloride abstraction to
take place. When the Ag(I) salt was present in excess, Ag-
(I) ions in solution inevitably underwent reduction, resulting
in oxidation of Cu(I) to Cu(II), as observed by the color
change from yellow to green. Similarly, the commonly
utilized comproportionation strategy, where Cu(0) is oxidized
-
The larger CF3SO3 anion sits in a similar position but is
more substantially displaced above the phenanthroline plane.
In these complexes, electronic and steric effects are in
competition. As oxygen is a better σ-donor than fluorine,
complex 1 may be considered to contain the strongest ion-
(38) The reduction potentials of Ag(I), Cu(II), and Cu(I) in acetone were
not found in the literature, but the standard reduction potentials do
offer some explanation. The standard one-electron reduction potentials
of Ag(I), Cu(II), and Cu(I) are 0.799 V, 0.159 V, and 0.520 V,
respectively. See ref 31.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3823

Gandhi et al.
-
Table 3. Average Cu-N Distances (in Å) of Other
pair, despite the larger size of the O3SCF3 . This conclusion
Bis-(2,9-R2-1,10-phenanthroline)Cu(I) Complexes
is borne out in the structures of 1 and 2: the Cu-O distance
is 1.927(2) A, while the Cu-F distance in 2 is 2.012(2) A.
For a larger anion, such as SbF6 , the open coordination site
av Cu-N
complex
[Cu(dmp)2]Xa
distance, Å
refs
-
2.04
2.06
2.08
2.06
2.06
31, 40, 41, 13 , 21
on the Cu(I)(dtbp) center is not large enough, and the
complex contains a noncoordinating ion pair. Presumably,
as the distance between the metal center and the counterion
increases, the stability of the ion pair decreases, allowing
solvent molecules to remain coordinated to the metal. The
solvato complex 3, which was obtained from a CH2Cl2/
hexanes solution in poor yield, shows an acetone molecule
filling the third coordination site. Since acetone is a ligand
to the Cu center, isolation from acetone/hexanes resulted in
significant improvement in the yield of 3.
[Cu(dnpp)2][PF6]
42
8
43
34
[Cu(dmp)(dtbp)][PF6]
[Cu(dpp)2]Xa
[Cu(2,9-C6F5-1,10-phen)2][SbF6]•
CH2Cl2
[Cu(xop)2][PF6]•CH3OHb
2.02
44
a This complex has been crystallized with various counterions. The
reported value is an average of all Cu-N distances found. b Most abbrevia-
tions are defined in footnote 17. xop, 2-(2-methylphenyl)-9-(2,6-dimeth-
ylphenyl)-1,10-phenanthroline.
Structural Database by more than three standard deviations.39
The average Cu(I)-N distances of select bis(2,9-R2-1,10-
phenanthroline)Cu(I) complexes are listed in Table 3 for
comparison.
Large anions facilitate the synthesis of [Cu(dtbp)2]+, a
complex previously reported to be impossible to prepare.11
As the size of the counterion is increased from SbF6 in 3
-
to [B(C6F5)4]- in 4 and 4•CH2Cl2, the distance between the
[Cu(dtbp)]+ and the [B(C6F5)4]- counterion becomes so great
that a second dtbp ligand binds to the metal center (Figure
6). The weak ion pairing, and the fact that the anion is unable
to closely approach the metal to satisfy its coordination
requirements, enables formation of the sterically encumbered
bis-dtbp complex [Cu(dtbp)2]+. Notably, complex 4 was the
sole product even when only 1 equiv of dtbp was added
relative to Ag[B(C6F5)4]. In this reaction the product was
1/2 of an equiv of the bis-dtbp complex, [Cu(dtbp)2]-
[B(C6F5)4] (4).
When compared to other complexes of its class, [Cu-
(dtbp)2]+ exhibits the least distortion from an idealized D2d
geometry. Bis complexes of 2,9-disubstituted phenanthroline
ligands exhibit three different types of distortion: rocking,
wagging, and flattening. Kovalevsky et al.24 defined úCD
,
which effectively combines these distinct structural deviations
into a single parameter. A perfect D2d structure would have
a úCD value of 1.00. For the [Cu(dtbp)2]+ cation in complexes
4, 4•CH2Cl2, 5•CH2Cl2 (2 asymmetric units), and 6•CH2Cl2
the úCD values are 0.994, 0.980, 0.987 and 0.982, and 0.962,
respectively (Table S2). Although there are examples of
previously characterized complexes with similarly minimal
distortions, the [Cu(dtbp)2]+ cation exhibits the highest value
of úCD across multiple crystal forms. Another structural
feature of the [Cu(dtbp)2]+ cation is the minimal extent of
π-stacking between the aromatic ring systems (Figure S7).
This minimal π overlap is unlike that in the other complexes
described herein; however, similarly minimal overlap was
observed in multiple structures of the Cu(bfp)2 cation.24
When the orange complex 4 is dissolved in acetone, the
resulting solution is yellow, suggesting that the solution-state
structure is different from the solid-state structure. The yellow
solution contains an acetone solvato complex, analogous to
3, as confirmed by NMR. When an acetone solution of 4 is
concentrated, the concentration of dtbp increases relative
to the concentration of acetone, and the second dtbp is able
to replace the coordinated acetone ligand. The resulting
orange solid [(dtbp)2Cu][B(C6F5)4] forms an orange solution
in CH2Cl2, indicating that the connectivity of the complex
At first glance, it may seem that an extraordinarily large
counterion like [B(C6F5)4]- is necessary to prepare the [Cu-
(dtbp)2]+ complex. However, once 4 was synthesized, simple
adjustments to the stoichiometry of the reaction, i.e., addition
of 2 equiv of dtbp relative to AgBF4 or AgSbF6 in reaction
5, allowed for the preparation of complexes 5•CH2Cl2 and
6•CH2Cl2, using smaller counterions and inexpensive, com-
mercially available starting materials. Though considerably
-
smaller than the [B(C6F5)4]- anion. The SbF6 anion is
sufficiently large to behave in a similar manner to the
-
[B(C6F5)4]- anion. SbF6 is too large to fit into the cleft
created by the two tert-butyl groups of the [Cu(dtbp)]+
moiety, as evidenced by the molecular structure of complex
-
3. The smaller size of BF4 allows it to compete with free
dtbp ligands during reactions for the synthesis of 2.
However, when a second equiv of dtbp ligand is present,
there is no competition between the two; dtbp is the better
ligand due to the chelate effect and the [Cu(dtbp)2][BF4]
complex is easily prepared.
(39) A Cambridge Structural Database (updated November 2004) search
for four-coordinate bis-phenanthroline-Cu(I) complexes found 79
matches. The Cu(I)-N distances for each complex were averaged,
and the average of these distances was calculated to be 2.045 Å with
a standard deviation of 0.017 Å.
The crystal structures of 4, 4•CH2Cl2, 5•CH2Cl2, and
6•CH2Cl2 demonstrate particularly long Cu(I)-N distances
(Table 2) that are due to steric crowding imposed by the
tert-butyl groups of the dtbp ligand. The long Cu(I)-N bond
distances indicate the difficulty of positioning two dtbp
ligands about the Cu center. The average Cu(I)-N distances
of 4, 4•CH2Cl2, 5•CH2Cl2, and 6•CH2Cl2 (Table 2) are the
longest in this class of compounds, longer than the average
Cu(I)-N distance of all other bis(2,9-R2-1,10-phenanthro-
line)Cu(I) complexes (2.045 Å) found in the Cambridge
(40) Pawlowski, V.; Kunkely, H.; Zabel, M.; Vogler, A. Inorg. Chim. Acta
2004, 357, 3, 824-826.
(41) Van Allen, D.; Venkataraman, D. J. Org. Chem. 2003, 68, 11, 4590-
4593.
(42) Eggleston, M. K.; Fanwick, P. E.; Pallenberg, A. J.; McMillin, D. R.
Inorg. Chem. 1997, 36, 18, 4007-4010.
(43) Miller, M. T.; Karpishin, T. B. Inorg. Chem. 1999, 38, 23, 5246-
5249.
(44) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. Inorg. Chem. 1999,
38, 14, 3414-3422.
3824 Inorganic Chemistry, Vol. 46, No. 10, 2007

2,9-Di-t-Bu-1,10-phenanthroline Copper(I) Complexes
is preserved in a noncoordinating solvent. NMR spectral data
confirm that both ligands are bound to the Cu(I) center in
CD2Cl2 solution. Thus, a coordinated dtbp ligand of 4 is
lost in the coordinating solvent acetone but is regained when
the coordinating solvent is removed. This type of ligand-
replacement reactivity may be of value in sensing applica-
tions.
Acknowledgment. We gratefully acknowledge ACS-PRF
grant 42041-AC3 to J.N.B. for the support of this work. The
authors would also like to acknowledge Dr. Ilia Guzei for
the use of his X-ray facility, Prof. Lawrence Dahl and Dr.
Rachel Linck for the helpful discussions, and Betsy True
for help with the cover art.
Supporting Information Available: Crystallographic data for
complex 4, [Cu(dtbp)(CH3CN)][PF6], and [(Cu(dtbp))2Cl][SbF6],
Conclusion
ú
CD calculations of bis(2,9-R2-1,10-phenanthroline)copper(I) com-
We have designed an oxidation-based general synthetic
route for the preparation of Cu(I) complexes, one which is
particularly well-suited for the synthesis of complexes
containing sterically bulky ligands. By a straightforward
oxidation method we have prepared a series of three-
coordinate Cu(I)-phenanthroline complexes containing weak
ligands at the third coordination site, Cu(dtbp)Y (Y is
SO3CF3 , BF4 , or acetone), and a complex that was
previously thought to be impossible to synthesize due to the
steric bulk on the substituents of the phenanthroline ligand,
[Cu(dtbp)2]+. That a complex that is structurally unfavorable,
with extraordinarily long Cu(I)-N distances, is the thermo-
dynamically favored product exemplifies the simple beauty
of this synthetic method. The photophysical properties and
reactivity of these newly synthesized complexes will be
reported elsewhere.
plexes, molecular drawings of complexes 1, 5•CH2Cl2, 6•CH2Cl2,
[Cu(dtbp)(CH3CN)][PF6], and [(Cu(dtbp))2Cl][SbF6], packing
diagrams of complex 1, diagrams of π-stacking between the ligands
of the various [Cu(dtbp)2]+ complexes, and crystallographic data
in CIF format of complexes 1, 2, 3, 4, 4•CH2Cl2, 5•CH2Cl2,
6•CH2Cl2, [Cu(dtbp)(CH3CN)][PF6], and [(Cu(dtbp))2Cl][SbF6].
This material is available free of charge via the Internet at
http://pubs.acs.org. CCDC 619999, CCDC 620000, CCDC
620001, CCDC 620002, CCDC 620003, CCDC 620004, CCDC
620005, CCDC 620006, and CCDC 620007 contain the supple-
mentary crystallographic data for complexes 4, 4•CH2Cl2, 5•CH2-
Cl2, 6•CH2Cl2, [(Cu(dtbp))2Cl](SbF6), 2, 1, 3, and [Cu(dtbp)(CH3-
CN)](PF6), respectively. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
-
-
IC0615224
Inorganic Chemistry, Vol. 46, No. 10, 2007 3825