X-ray structures, photophysical characterization, and computational analysis of geometrically constrained copper(I) -Phenanthroline complexes

Article abstract of DOI:10.1021/ic034529j

A series of three geometrically constrained C₂-symmetric Cu(I) mono-phenanthroline complexes were characterized by X-ray structural analysis, and their photophysical properties were investigated by absorption and emission spectroscopy. Visible light excitation yielded metal-to-ligand charge-transfer (MLCT) excited states with luminescence lifetimes up to 155 ns. Ultrafast ransient absorption spectroscopy provided further insights into the excited-state dynamics and suggests for all three complexes the formation of a phenanthroline radical anion. In agreement with electrochemical measurements, the data further indicate that coordinative rearrangements are involved in nonradiative deactivation of the excited states. According to time-dependent density functional theory calculations (B3LYP/6-31G**), the major MLCT transitions are polarized along the C₂ axis of the complex and originate predominantly from the copper d_xz orbital. The computational analysis identifies an excited-state manifold with a number of close-lying, potentially emissive triplet states and is in agreement with the multiexponential decay kinetics of the MLCT luminescence. The relationship between structural and photophysical data of the studied Cu(I) mono-phenanthroline complexes agrees well with current models describing the photophysics of the related Cu(I) bis-diimine complexes.

Full text of DOI:10.1021/ic034529j

Inorg. Chem. 2003, 42, 4918−4929
X-ray Structures, Photophysical Characterization, and Computational
Analysis of Geometrically Constrained Copper(I)−Phenanthroline
Complexes
John Cody,† Jeanette Dennisson,† Joshua Gilmore,† Donald G. VanDerveer,† Maged M. Henary,†
,†
,‡
Alan Gabrielli,† C. David Sherrill,* Yiyun Zhang,‡ Chia-Pin Pan,‡ Clemens Burda,* and
Christoph J. Fahrni*
,†
School of Chemistry and Biochemistry, Georgia Institute of Technology, 770 State Street,
Atlanta, Georgia 30332, and Center for Chemical Dynamics, Department of Chemistry,
Case Western ReserVe UniVersity, 10900 Euclid AVenue, CleVeland, Ohio 44106
Received May 16, 2003
A series of three geometrically constrained C -symmetric Cu(I) mono-phenanthroline complexes were characterized
2
by X-ray structural analysis, and their photophysical properties were investigated by absorption and emission
spectroscopy. Visible light excitation yielded metal-to-ligand charge-transfer (MLCT) excited states with luminescence
lifetimes up to 155 ns. Ultrafast transient absorption spectroscopy provided further insights into the excited-state
dynamics and suggests for all three complexes the formation of a phenanthroline radical anion. In agreement with
electrochemical measurements, the data further indicate that coordinative rearrangements are involved in nonradiative
deactivation of the excited states. According to time-dependent density functional theory calculations (B3LYP/6-
31G**), the major MLCT transitions are polarized along the C axis of the complex and originate predominantly
2
from the copper d orbital. The computational analysis identifies an excited-state manifold with a number of close-
xz
lying, potentially emissive triplet states and is in agreement with the multiexponential decay kinetics of the MLCT
luminescence. The relationship between structural and photophysical data of the studied Cu(I) mono-phenanthroline
complexes agrees well with current models describing the photophysics of the related Cu(I) bis-diimine complexes.
Introduction
third-row transition elements with d6 electron configuration,
such as ruthenium(II), rhenium (I), or iridium(III).6 In search
for more cost-effective alternatives, copper-based systems
have received increased attention over the past two decades.7,8
In particular, copper(I) complexes of phenanthroline deriva-
tives exhibit a strong metal-to-ligand charge-transfer absorp-
tion in the visible range, although the emission properties
are typically compromised by short excited-state lifetimes
and low quantum yields. Several detailed studies have shown
that the two major nonradiative deactivation pathways are
due to flattening distortion and exciplex quenching of the
excited state in polar donor solvents.9-12 Both pathways can
Photoluminescent metal complexes with low-lying metal-
to-ligand charge-transfer (MLCT) excited states have been
extensively studied as chemical sensors, catalysts for solar
energy conversion, or components in display devices.1-5 The
majority of work has focused on complexes of polypyridine
or phenanthroline ligands in combination with second- or
* Corresponding authors. E-mail: fahrni@chemistry.gatech.edu (C.J.F.);
sherrill@chemistry.gatech.edu (C.D.S); cxb77@cwru.edu (C.B.).
† Georgia Institute of Technology.
‡ Case Western Reserve University.
(1) Czarnik, A. W. Fluorescent Chemosensors for Ion and Molecular
Recognition; American Chemical Society: Washington, DC, 1993.
(2) Lakowicz, J. R. Topics in Fluorescence Spectroscopy. Volume 4:
Probe Design and Chemical Sensing; Plenum Press: New York, 1994.
(3) Chanon, M. Homogeneous Photocatalysis; Wiley & Sons: Chichester,
1997; Vol. 2.
(6) Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin
Complexes; Academic Press: London, 1997.
(7) Armaroli, N. Chem. Soc. ReV. 2001, 30, 113.
(8) McMillin, D. R.; McNett, K. M. Chem. ReV. 1998, 98, 1201.
(9) Eggleston, M. K.; McMillin, D. R.; Koenig, K. S.; Pallenberg, A. J.
Inorg. Chem. 1997, 36, 172.
(4) Roundhill, D. M. Photochemistry and Photophysics of Metal Com-
plexes; Plenum Press: New York, 1994.
(5) Mauch, R. H.; Gumlich, H.-E. Inorganic and Organic Electrolumi-
nescence; Wissenschaft und Technik Verlag: Berlin, 1996.
(10) Eggleston, M. K.; Fanwick, P. E.; Pallenberg, A. J.; McMillin, D. R.
Inorg. Chem. 1997, 36, 4007.
4918 Inorganic Chemistry, Vol. 42, No. 16, 2003
10.1021/ic034529j CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/18/2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
gel (70-230 mesh). TLC: 0.25 mm, Merck silica gel 60 F254
visualizing at 254 nm or with 2% KMnO4 solution.
,
be rationalized by the increased formal charge on the copper
in the emissive CT state, which is stabilized by a square
planar or five-coordinate coordination environment. Bulky
substituents in the 2- and 9-positions of the phenanthroline
ring greatly improve the quantum yield by both inhibiting
flattening distortion and suppressing ligand-addition reactions
in the excited state.
Luminescent copper(I) phenanthroline complexes might
be suitable as highly specific luminescent probes for copper
itself. All previously characterized luminescent Cu(I) phenan-
throline complexes are based on homo- or heteroleptic
systems (A) with 2:1 stoichiometry and pseudotetrahedral
geometry.7 Since a mixed-ligand environment complicates
Synthesis: (a) 2-(2-Ethylsulfanylmethyl-phenyl)-4,4,5,5-tet-
ramethyl-[1,3,2]dioxaborolane (2). A solution of ethanethiol (270
µL, 0.81 mmol) and sodium hydroxide (80 mg) in 4.0 mL of ethanol
was added dropwise to a solution of (2-bromomethylphenyl)boronic
acid pinacol ester (200 mg, 0.67 mmol) in 4.0 mL of ethanol. After
the reaction mixture was stirred for 30 min at room temperature,
the solvent was evaporated and the residue purified by column
chromatography on silica gel (20:1 hexane/EtOAc) to yield 100
mg (54%) of pinacol ester 2 as a colorless oil. Rf: 0.25 (30:1 hexane/
EtOAc). 1H NMR (CDCl3, 300 MHz): δ 1.20 (t, J ) 7.1 Hz, 3H),
1.35 (s, 12H), 2.40 (t, J ) 7.1 Hz, 2H), 4.03 (s, 2H), 7.21 (td, J )
7.1, 1.1 Hz, 1H), 7.25 (d, J ) 7.1 Hz, 1H), 7.33 (dd, J ) 7.7, 1.6
Hz, 1H), 7.78 (dd, J ) 7.1, 1.6 Hz, 1H). MS (EI): m/z 278 (M+,
37), 217 (35), 191 (32), 149 (73), 59 (42), 40 (100). FAB-HRMS:
m/e calcd for (M + H) C15H23BO2S 278.2202, found 278.1512.
(b) 1-[2-(4,4,5,5-Tetramethyl-[1,3,2]dioxaborolan-2-yl)benz-
yl]-1H-pyrazole (3a). A mixture of pyrazole (950 mg, 14.0 mmol),
sodium hydride (95 mg, 60% dispersion in mineral oil), and (2-
bromomethylphenyl)boronic acid pinacol ester (500 mg, 1.68 mmol)
in 5 mL of anhydrous dimethylformamide (DMF) was heated at
80 °C for 3 h. The solvent was removed under reduced pressure
and the residue diluted with water and extracted twice with EtOAc.
The combined organic layers were dried with MgSO4 and concen-
trated under reduced pressure. The crude product was purified on
silica gel (4:1 hexane/EtOAc) providing 185 mg (0.65 mmol, 39%)
of boron ester 3a as a colorless oil. Rf: 0.51 (1:1 hexane/EtOAc).
1H NMR (CDCl3, 300 MHz): δ 1.35 (s, 12H), 5.66 (s, 2H), 6.25
(t, J ) 2.2 Hz, 1H), 7.01 (d, J ) 7.7 Hz, 1H), 7.28 (td, J ) 7.7, 1.1
Hz, 1H), 7.38 (td, J ) 7.0, 1.6 Hz, 1H), 7.44 (d, J ) 2.2 Hz, 1H),
7.55 (d, J ) 1.1 Hz, 1H); 7.86 (dd, J ) 7.1, 1.6 Hz, 1H). MS (EI):
m/z 284 (M+, 43), 225 (55), 184 (100), 117 (36). FAB-HRMS:
m/e calcd for (M+) C16H21BN2O2 284.1696, found 284.1709.
quantitative measurements, a 1:1 stoichiometry between
analyte and ligand receptor would be preferable. Principally,
a 1:1 complex stoichiometry can be achieved either by
linking two phenanthroline moieties together or by attaching
two additional donor groups (D) to a single phenanthroline
core as shown in structure B. Molecular modeling studies
of type B complexes suggest that the steric requirements of
the 2-tolyl-donor substituents enforce a C2-symmetric com-
plex with a pseudotetrahedral coordination geometry.
The rigid ligand framework imposed by the biaryl-type
linker in complex B is expected to reduce the previously
mentioned nonradiative deactivation processes such as flat-
tening distortion or exciplex quenching by the solvent. The
following study describes the synthesis, structure, and
photophysical characterization of three complexes of type
B with various donor groups. As revealed by steady-state
and time-resolved emission spectroscopy, the structural
differences significantly affect their photophysical properties.
Time-dependent density functional calculations were per-
formed to gain further insights into possible excited-state
deactivation processes.
(c) 3,5-Dimethyl-1-[2-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-
2-yl)-benzyl]-1H-pyrazole (3b) was synthesized as described for
1
3a. Yield: 60%. Rf: 0.58 (2:1 hexane/EtOAc). H NMR (CDCl3,
300 MHz): δ 1.35 (s, 12H), 2.11 (s, 3H), 2.27 (s, 3H), 5.59 (s,
2H), 5.87 (s, 1H), 6.56 (d, J ) 7.7 Hz, 1H), 7.20 (td, J ) 7.1, 1.1
Hz, 1H), 7.31 (td, J ) 7.7, 1.6 Hz, 1H), 7.85 (dd, J ) 7.1, 1.6 Hz,
1H). MS (EI) m/z 313 (M+, 13), 297 (93), 253 (33), 235 (36), 212
(100), 117 (38), 109 (51). FAB-HRMS: m/e calcd for (M+) C18H26-
BN2O2 313.2087, found 313.2104.
(d) 2,9-Bis-(2-ethylsulfanylmethyl-phenyl)-[1,10]phenanthro-
line (5). A solution of 2,9-dichloro-[1,10]phenanthroline (23 mg,
0.089 mmol), pinacol ester 2 (50 mg, 0.180 mmol), and a catalytic
amount of tetrakis-triphenylphosphine palladium(0) (3 mg) in a
mixture of 5 mL of ethanol and 5 mL of toluene was heated for 48
h at 120 °C in a sealed Schlenk tube under argon together with a
solution of potassium carbonate (345 mg) in 5 mL of water. The
reaction mixture was diluted with 10 mL of saturated NaCl and
extracted twice with dichloromethane. The combined organic phase
was dried (MgSO4) and concentrated under reduced pressure. The
residue was purified by flash chromatography on silica gel (1:1
hexane/ethyl acetate) to yield 32 mg (75%) of phenanthroline
derivative 5 as a colorless solid. Mp: 117-118 °C. Rf: 0.47 (1:1
hexane/EtOAc). 1H NMR (CDCl3, 300 MHz): δ 0.98 (t, J ) 14.8
Hz, 6H), 2.32 (q, J ) 7.7 Hz, 4H), 4.40 (s, 2H), 7.33-7.37 (m,
4H), 7.48-7.51 (m, 2H), 7.57-7.60 (m, 2H), 7.83 (s, 2H), 7.91
(d, J ) 8.2 Hz, 2H), 8.33 (d, J ) 8.2 Hz, 2H). MS (EI): m/z 480
(M+, 16), 419 (59), 389 (80), 357 (100), 178 (44). FAB-HRMS:
m/e calcd for (M+) C30H28N2S2 480.1694, found 480.1694.
Experimental Section
Materials and Reagents. 2,9-Dichloro-[1,10]phenanthroline (4)
was prepared according to a published procedure.13 (2-Bromo-
methylphenyl)-boronic acid pinacol ester (Combiblock), pyrazole
(Alfa Aesar, 98%), dimethyl pyrazole (Alfa Aesar, 99%), ethane-
thiol (Acros, 99%), Pd(PPh3)4 (Aldrich, 99%). NMR: δ in ppm vs
1
SiMe4 (0 ppm, H, 300 MHz). MS: selected peaks; m/z. Melting
points are uncorrected. Flash chromatography (FC): Merck silica
(11) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. Inorg. Chem. 1998,
37, 2285.
(12) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. Inorg. Chem. 1999,
38, 3414.
(13) Yamada, M.; Nakamura, Y.; Kuroda, S.; Shimao, I. Bull. Chem. Soc.
Jpn. 1990, 63, 2710.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4919

Cody et al.
(e) 2,9-Bis-(2-pyrazol-1-ylmethyl-phenyl)-[1,10]phenanthro-
line (6a) was synthesized as described for 5: Starting with 185
mg of boron ester 3a, 57 mg of ligand 6a was isolated as a colorless
solid (34% yield). Mp: 188-190 °C. Rf: 0.27 (1:1 hexane/EtOAc).
1H NMR (CDCl3, 300 MHz): δ 5.72 (s, 4H), 5.78 (t, J ) 4.4 Hz,
2H), 7.22-7.28 (m, 2H), 7.38-7.45 (m, 6H), 7.57-7.63 (m, 2H),
7.68 (d, J ) 2.2 Hz, 2H), 7.82 (d, J ) 8.2 Hz, 2H), 7.89 (s, 2H),
8.36 (d, J ) 8.2 Hz, 2H). MS (FAB): m/z 493 (M+, 43), 493 (100),
425 (33). FAB-HRMS: m/e calcd for (M + H)+ C32H24N6
493.2141, found 493.2157.
fluorimeter. Quantum yields were determined using degassed
dichloromethane with [Ru(bpy)3](PF6)2 as standard (Φf (H2O) )
0.042).14
Time-Resolved Emission Spectroscopy. Time-resolved emis-
sion was measured by time-correlated single photon counting. The
excitation source was a Spectra-Physics (SP) Tsunami Ti-sapphire
laser pumped with a SP Millennia diode-pumped Nd-YVO4 laser.
The repetition rate was set at 4 MHz with 2.2 ps pulse width. The
output was frequency doubled by a SP GWU HG flexible harmonic
generator and used as the excitation source. The emission was
detected by a cooled Hamamatsu (Hamamatsu, Japan) R3809U-50
microchannel plate photomultiplier, while a SPC-330-12 PC module
(Becker & Hickl GmbH Intelligent Measurement and Control
Systems, Berlin, Germany) was used for the photon counting
electronics. The excitation wavelength used was 390 nm, and the
emission was monitored at 650 nm. Luminescence decays were
acquired simultaneously with a lamp profile in 1024 channels of
9.8 ps/channel. Decay curves were deconvoluted using the Beechem
global program.15 Goodness of fit was judged by reduced chi-square,
ør2, and the autocorrelation function of the weighted residuals.
Transient Absorption Spectroscopy. Femtosecond time-
resolved transient absorption measurements were recorded with a
femtosecond laser system that has been previously described.16
Samples were excited at 390 nm with a laser pulse of 120 fs duration
(fwhm) and 0.8 mJ/cm2 output energy per pulse. The excitation
beam is focused to a spot diameter of about 500 µm and the probe
beam to 100 µm. The samples were measured in a quartz cuvette
of 2 mm path length and stirred by a cell stirrer to avoid permanent
bleaching of the pump-probe volume element in the solution. All
pump-probe experiments were carried out under ambient condi-
tions.
Cyclic Voltammetry. The cyclic voltammograms were acquired
in acetonitrile (freshly distilled from calcium hydride) containing
0.1 M Bu4NPF6 as electrolyte using a CH-Instruments potentiostat
(model 600A). The samples were measured under inert gas at a
concentration of 3 mM in a single compartment cell with Pt working
and counter electrode and Ag/AgCl reference electrode. The half-
wave potentials were referenced to ferrocene as internal standard,
and the measurements were typically performed with a 100 mV
s-1 scan rate.
Computational Methods. All computations were carried out
using the Q-Chem electronic structure package.17 Geometries of
the electronic ground states were determined using density func-
tional theory (DFT), applying the B3LYP method18 which includes
Becke’s three-parameter hybrid exchange functional19 and the
gradient-corrected correlation functional of Lee, Yang, and Parr.20
Computations employed the 6-31G** basis,21,22 which is a valence
double-ú basis with d polarization functions on first-row atoms, p
polarization on hydrogens, and f polarization on copper. Estimates
(f) 2,9-Bis-[2-(3,5-dimethyl-pyrazol-1-ylmethyl)-phenyl][1,10]-
phenanthroline (6b) was synthesized as described for 5: Starting
with 153 mg of boron ester 3b, 80 mg of ligand 6b was isolated as
1
a colorless oil (30% yield). H NMR (CDCl3, 300 MHz): δ 1.88
(s, 6H), 2.17 (s, 6H), 5.68 (s, 4H), 5.73 (s, 2H), 6.73-6.76 (m,
2H), 7.30-7.38 (m, 4H), 7.59-7.65 (m, 2H), 7.86 (d, J ) 8.2 Hz,
2H), 7.86 (s, 2H), 8.34 (d, J ) 8.2 Hz, 2H). ΜS (FAB): 548 (M+,
3), 452 (55), 356 (100). FAB-HRMS: m/e calcd for (M+) C36H32N6
548.2675, found 548.2689.
(g) 2,9-Bis-(2-ethylsulfanylmethyl-phenyl)-[1,10]phenanthro-
line copper(I) hexafluorophosphate (7). A solution of phenan-
throline derivative 5 (20 mg, 41 µmol) in 1 mL of acetonitrile was
added dropwise to a solution of tetrakis(acetonitrile) copper(I)
hexafluorophosphate (15.5 mg, 41 µmol). After stirring for 30 min
at room temperature, the solvent was evaporated under reduced
pressure and the residue recrystallized from dichloromethane/hexane
affording 22.5 mg (32.6 µmol, 78%) of yellow crystals. Single
crystals suitable for X-ray analysis were obtained from an aceto-
nitrile solution of the complex by vapor diffusion with toluene.
1
Mp: >300 °C. H NMR (CDCl3, 300 MHz): δ 0.74 (t, J ) 7.4
Hz, 6H), 2.39 (q, J ) 7.4 Hz, 4H), 3.32 (d, J ) 11.5 Hz, 2H), 3.80
(d, J ) 11.5 Hz, 2H), 7.39-7.41 (m, 2H), 7.51-7.61 (m, 6H),
8.07 (d, J ) 8.2 Hz, 2H), 8.21 (s, 2H), 8.78 (d, J ) 8.2 Hz, 2H).
MS (FAB): m/z 543 (M+, 100). FAB-HRMS: m/e calcd for (M+)
C30H28CuN2S2 543.0989, found 543.0988.
(h) 2,9-Bis-(2-pyrazol-1-ylmethyl-phenyl)-[1,10]phenanthro-
line copper(I) hexafluorophosphate (8a) was synthesized from
2,9-bis-(2-pyrazol-1-ylmethyl-phenyl)-[1,10]phenanthroline 6a as
described for complex 7. Yield: 78%, orange crystals. Single
crystals suitable for X-ray analysis were obtained from an aceto-
nitrile solution of the complex by vapor diffusion with toluene.
1
Mp: >300 °C. H NMR (CDCl3, 300 MHz): δ 5.02-5.06 (AB
multiplet, 4H), 5.54 (d, J ) 2.2 Hz, 1H), 6.00 (t, J ) 2.2 Hz, 2H),
7.58-7.67 (m, 8H), 8.13 (s, 2H), 8.16 (d, J ) 8.2 Hz, 2H), 8.72
(d, J ) 8.2 Hz, 2H). MS (FAB): m/z 555 (M+, 100). FAB-
HRMS: m/e calcd for (M+) C32H24N6Cu 555.1358, found 555.1332.
(i) 2,9-Bis-[2-(3,5-dimethyl-pyrazol-1-ylmethyl)-phenyl][1,10]-
phenanthroline copper(I) hexafluorophosphate (8b) was syn-
thesized from 2,9-bis-[2-(3,5-dimethyl-pyrazol-1-ylmethyl)-phenyl]-
[1,10]phenanthroline 6b as described for complex 7. Yield: 77%,
red crystals. Single crystals suitable for X-ray analysis were obtained
by vapor diffusion with toluene of an acetonitrile solution of the
complex. Mp: >300 °C. 1H NMR (CDCl3, 300 MHz): δ 0.75 (s,
6H), 2.39 (s, 6H), 4.82-4.86 (AB multiplet, 4H), 5.68 (s, 2H),
7.17 (d, J ) 7.14 Hz, 2H), 7.45-7.56 (m, 4H), 7.70 (d, J ) 8.2
Hz, 2H), 8.05 (d, J ) 8.2 Hz, 2H), 8.15 (s, 2H), 8.72 (d, J ) 8.2
Hz, 2H). MS (FAB): m/z 611 (M+, 100). FAB-HRMS: m/e calcd
for (M+) C36H32N6Cu 611.1984, found 611.1982.
(14) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583.
(15) Beechem, J. M. Chem. Phys. Lipids 1989, 50, 237.
(16) Burda, C.; Samia, A. C. S.; Hathcock, D. J.; Huang, H. J.; Yang, S.
J. Am. Chem. Soc. 2002, 124, 12400.
(17) Kong, J.; White, C. A.; Krylov, A. I.; Sherrill, D.; Adamson, R. D.;
Furlani, T. R.; Lee, M. S.; Lee, A. M.; Gwaltney, S. R.; Adams, T.
R.; Ochsenfeld, C.; Gilbert, A. T. B.; Kedziora, G. S.; Rassolov, V.
A.; Maurice, D. R.; Nair, N.; Shao, Y. H.; Besley, N. A.; Maslen, P.
E.; Dombroski, J. P.; Daschel, H.; Zhang, W. M.; Korambath, P. P.;
Baker, J.; Byrd, E. F. C.; Van Voorhis, T.; Oumi, M.; Hirata, S.; Hsu,
C. P.; Ishikawa, N.; Florian, J.; Warshel, A.; Johnson, B. G.; Gill, P.
M. W.; Head-Gordon, M.; Pople, J. A. J. Comput. Chem. 2000, 21,
1532.
Steady-State Absorption and Emission Spectroscopy. UV-
vis absorption spectra were recorded at 25 °C using a Varian Cary
Bio50 spectrometer with constant-temperature accessory. Steady-
state emission and excitation spectra were recorded with a PTI
(18) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(19) Becke, A. D. J. Chem. Phys. 1993, 98, 1372.
(20) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
4920 Inorganic Chemistry, Vol. 42, No. 16, 2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
Table 1. Crystallographic Data for Complexes 7, 8a, and 8b
[Cu(bempp)]PF6 (7)
[Cu(bpmpp)]PF6 (8a)
33.5H27Cl3CuF6N6P
dark red prisms
0.24 × 0.17 × 0.07
triclinic
[Cu(bdmpp)]PF6 (8b)
formula
color, habit
cryst size (mm)
cryst syst
C30H28CuF6N2PS2
yellow prisms
0.25 × 0.14 × 0.03
monoclinic
P2(1)/n
C
C30H32CuF6N6P
dark red prisms
0.44 × 0.37 × 0.27
monoclinic
P2(1)/n
space group
P1h
a (Å)
b (Å)
c (Å)
12.740(2)
14.140(2)
16.198(3)
90
95.061(3)
90
2906.7(8)
4
1.575
10.6375(13)
12.7868(16)
14.6402(18)
109.599(2)
111.207(2)
93.59(2)
1710.1(4)
2
12.0457(18)
9.3115(14)
29.488(5)
90
91.961(3)
90
3305.6(9)
4
1.521
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
calcd density (g/cm3)
F(000)
1.609
838
1408
1552
temp (K)
188(2)
198(2)
198(2)
2θ range (deg)
3.82-50.0
3.24-50.0
2.76-50.30
no. of reflns collected
no. of indep reflns
final R indices (obsd data) (%): R, Rw
R indices (all data) (%): R, Rw
GOF on F2
15158
12921
23985
5118
5989
5917
5.16, 7.72
12.24, 9.27
0.910
5.50, 14.11
8.38, 15.86
1.044
3.73. 9.38
4.77, 10.29
1.064
of vertical excitation energies were obtained using time-dependent
density functional theory (TD-DFT)23,24 in the Tamm-Dancoff
approximation24 with the B3LYP functional and the 6-31G** basis
at the ground-state optimized geometries. Molecular orbitals and
electron attach/detach densities25 were visualized with the MOLE-
KEL software26 using the Q-Chem plot data. Details for all
computations, including the coordinates of optimized geometries,
are provided in the Supporting Information.
X-ray Structure Analysis. Crystals of complexes 7, 8a, and 8b
suitable for X-ray structural analysis were grown from acetonitrile-
toluene by slow evaporation of the solvent over the period of several
days. The X-ray data were collected on a Siemens SMART 1K
CCD diffractometer with graphite-monochromated Mo KR radiation
(λ ) 0.71073 Å). The program SADABS (Sheldrick)27 and SAINT
6.22 (Bruker)28 were used for absorption corrections. All structures
were solved by direct methods and refined by least-squares
calculations with the SHELXTL 5.10 software package.28 The
hydrogen atoms were added using ideal geometries with a fixed
C-H bond distance of 0.96 Å. A summary of the crystallographic
parameters and data is given in Table 1.
bromomethyl-phenyl)-[1,10]phenanthroline results only in
monosubstitution, and is accompanied by intramolecular
cyclization of the second benzylic site to give a cationic
N-alkylated product. Nucleophilic substitution of the bro-
momethyl group must therefore be carried out prior to the
cross coupling reaction. The strategy is expected to be
generally applicable to the synthesis of a wide variety of
donor-substituted 2,9-diphenyl-[1,10]phenanthroline deriva-
tives. Reaction of ligands 5, 6a, and 6b with 1 molar equiv
of Cu(CH3CN)4 provided the corresponding 1:1 complexes
with good yield (78%).
2. Structural Studies. Crystal Structure of Cu(bempp)-
PF6 (7). Vapor diffusion of toluene into an acetonitrile
solution of complex 7 yielded yellow crystals that were
suitable for diffraction studies. An ORTEP view with atom-
numbering scheme of the cation is given in Figure 1, and
selected interatomic distances and angles are compiled in
Table 2. As anticipated from initial molecular modeling
studies, the copper center adopts a twisted tetrahedral
coordination geometry that results in a pseudo-C2 symmetry
of the complex cation. With a dihedral angle of 74.1°
between the N(1)-Cu-N(2) and S(1)-Cu-S(2) planes the
coordination geometry deviates substantially from an ideal
tetrahedral arrangement. The copper-sulfur bond distances
Cu-S(1) and Cu-S(2) are 224 and 225 pm, respectively,
and compare well with the distances of similar monovalent
copper complexes of aliphatic sulfur donor ligands. For
example, the tetrahedral complex bis(2,5-dithiahexane-S,S′′)-
copper(I) perchlorate30 exhibits a Cu-S distance of 226 pm,
and an almost identical bond length of 225 pm was
determined for the mixed sulfur-nitrogen donor ligand 1,5-
dithia-9,13-diazacyclohexadecane.31 Similarly, the copper-
nitrogen bond lengths Cu-N(1) (206 pm) and Cu-N(2) (204
pm) compare well with the distances found for bis-phenan-
Results and Discussion
1. Synthesis. The three disubstituted phenanthroline
ligands 5, 6a, and 6b were synthesized via Suzuki cross
coupling of 2,9-dichloro-[1,10]phenanthroline13 4 with the
corresponding boron-picolinate as outlined in Scheme 1.
Nucleophilic substitution of 2-bromomethylphenyl boron
ester 1 with ethanethiol, pyrazole, or 2,4-dimethyl-pyrazole
yielded the coupling partners 2, 3a, and 3b. As reported by
Bates and Parker29 nucleophilic substitution of 2,9-bis-(2-
(21) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(22) Rassolov, V. A.; Pople, J. A.; Ratner, M. A.; Windus, T. L. J. Chem.
Phys. 1998, 109, 1223.
(23) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454.
(24) Hirata, S.; Head-Gordon, M. Chem. Phys. Lett. 1999, 302, 375.
(25) Head-Gordon, M.; Grana, A. M.; Maurice, D.; White, C. A. J. Phys.
Chem. 1995, 99, 14261.
(26) Flu¨kiger, P.; Lu¨thi, H. P.; Portmann, S.; Weber, J. MOLEKEL, v. 4.1;
Swiss Center for Scientific Computing: Manno, 2000-2001.
(27) Blessing, R. H. Acta Crystallogr., Sect. A 1995, 51, 33.
(28) Bruker AXS: Madison, WI, 1998.
(30) Olmstead, M. M.; Musker, W. K.; Kessler, R. M. Inorg. Chem. 1981,
20, 151.
(31) Balakrishnan, K. P.; Riesen, A.; Zuberbu¨hler, A. D.; Kaden, T. A.
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1990, 46, 1236.
(29) Bates, G. B.; Parker, D. Tetrahedron Lett. 1996, 37, 267.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4921

Cody et al.
Scheme 1
Table 2. Selected Experimental Structural Parameters for Cu(I) Complexes 7, 8a, and 8ba
[Cu(bempp)]PF6 (7)
[Cu(bpmpp)]PF6 (8a)
[Cu(bdmpp)]PF6 (8b)
Cu(1)-N(1) (Å)
Cu(1)-N(2) (Å)
Cu(1)-S(1) (Å)
Cu(1)-S(2) (Å)
2.062(3)
2.044(4)
2.2397(13)
2.2544(13)
Cu(1)-N(1) (Å)
Cu(1)-N(2) (Å)
Cu(1)-N(3) (Å)
Cu(1)-N(5) (Å)
2.046(4)
2.060(4)
1.983(4)
1.985(4)
2.169(2)
2.091(2)
2.042(2)
2.061(2)
N(1)-Cu(1)-N(2) (deg)
N(1)-Cu(1)-S(1) (deg)
N(1)-Cu(1)-S(2) (deg)
N(2)-Cu(1)-S(1) (deg)
N(2)-Cu(1)-S(2) (deg)
S(1)-Cu(1)-S(2) (deg)
81.90(14)
107.45(10)
121.59(10)
127.06(10)
102.83(10)
113.71(5)
N(1)-Cu(1)-N(2) (deg)
N(1)-Cu(1)-N(3) (deg)
N(1)-Cu(1)-N(5) (deg)
N(2)-Cu(1)-N(3) (deg)
N(2)-Cu(1)-N(5) (deg)
N(3)-Cu(1)-N(5) (deg)
81.74(14)
79.13(8)
116.84(15)
118.62(15)
117.95(15)
114.49(14)
106.27(15)
113.20(8)
120.76(8)
124.16(8)
110.86(8)
107.46(8)
θb (deg)
74.06
θb (deg)
87.74
80.94
a Numbering scheme given in Figure 1. b θ is the angle between the two intersecting planes of the coordination tetrahedron: N(1)-Cu-N(2) and S(1)-
Cu-S(2) for complex 7, and N(1)-Cu-N(2) and N(3)-Cu-N(5) for complexes 8a and 8b, respectively.
throline copper(I) complexes, such as [Cu(dmp)2]BPh4 (202
pm)32 or [Cu(dpdmp)2]PF6 (202-211 pm).11 Despite the
considerably rigid architecture of 5, the rotational freedom
of the aryl substituents in the 2- and 9-positions provides
sufficient flexibility to adapt to the natural bonding distances
of the copper center.
1H NMR of Cu(bempp)PF6 (7). The 1H NMR data of 7
in CDCl3 suggest that, in agreement with the solid-state
structure, the complex adopts also a monomeric C2-sym-
metric geometry in solution. The ligand architecture ef-
fectively prevents formation of oligo- or polymeric coordi-
nation complexes, which are characteristic for the coordination
chemistry of monovalent copper with thioethers. Consistent
with coordination of the thioether to the copper center, the
benzylic CH2 proton resonances are shifted to higher field
by 0.6 and 1.08 ppm compared to the free ligand. Copper
coordination renders the two benzylic protons chemically
nonequivalent, resulting in an AM spin system with two
doublets and significantly different chemical shifts (3.32 and
3.80 ppm). This observation indicates that dynamic exchange
of the diastereotopic proton pair is slow on the NMR time
scale at room temperature.
Crystal Structure of Cu(bpmpp)PF6 (8a). Compared to
complex 7, the dihedral angle between the intersecting planes
of N(1)-Cu-N(2) and N(3)-Cu-N(5) is substantially less
twisted and approaches 87.7°, a nearly ideal tetrahedral
geometry. The pyrazole substituent extends the chelate ring
by an additional atom, which naturally leads to a different
coordination geometry. The extension of the chelate ring to
an eight-membered structure is also reflected in a slightly
larger dihedral angle N(1)-C(1)-C(13)-C(18) and N(2)-
C(10)-C(20)-C(25) between the phenanthroline ring plane
and the aryl substituent. In complex 7 the two angles were
measured to be 52.4° and 56.7° whereas in structure 8a the
analogous dihedral angles are 57.6° and 57.7°. The bond
distances between copper and the phenanthroline nitrogens
N(1) and N(2) are 205 and 206 pm, respectively, and are
essentially identical compared to complex 7. The copper
pyrazole-nitrogen distances are slightly shorter with Cu-
N(3) and Cu-N(5) equal 198 pm.
Crystal Structure of Cu(bdmpp)PF6 (8b). Due to the
steric requirements of the added methyl groups the coordina-
(32) Blake, A. J.; Hill, S. J.; Hubberstey, P.; Li, W. S. J. Chem. Soc., Dalton
Trans. 1998, 909.
4922 Inorganic Chemistry, Vol. 42, No. 16, 2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
Figure 2. Space-filling representation for complex 7 (left), 8a (middle),
and 8b (right).
coordination space around the copper center differs between
the three complexes.
3. Photophysical Studies. 3.1. Steady-State Absorption
and Emission. The spectroscopic properties of copper(I)
phenanthroline complexes are dominated by low-lying metal-
to-ligand charge-transfer excited states (MLCT), which give
rise to characteristic charge-transfer absorption bands in the
visible range.7 In [1,10]phenanthroline two energetically
close-lying π* orbitals a2(ø) and b1(ψ) act as electron
acceptors upon excitation into the CT band (Figure 3).33,34
The labels ψ (antisymmetric) and ø (symmetric) were
introduced by Orgel and refer to the orbital symmetry with
respect to the C2 axis that passes through the ligand.35
As illustrated in Figure 3a the D2d-symmetric coordination
environment of copper(I) bis-phenanthroline complexes
results in energetically degenerate dxz and dyz orbitals.
Replacement of one of the phenanthroline ligands with a
donor pair D leads to symmetry reduction and renders the
two orbitals nondegenerate. In the resulting C2V-symmetric
environment almost all of the possible d f π* electronic
transitions are symmetry allowed. The only two exceptions
are excitation into the LUMO (ø) from a d-orbital with a1
2
2
2
symmetry (dx -y and dz ) and excitation into LUMO + 1
(ψ) from a d-orbital with b2 symmetry (dyz). Nevertheless,
according to Day and Sanders the MLCT absorption intensity
of C2V symmetric [1,10]phenanthroline complexes is derived
mainly from transitions that are polarized along the C2 axis
of the complex.36 As shown in Figure 3 for an idealized C2V
symmetry, there are two such transitions possible, b1(ψ) r
dxz and a2(ø) r dxy. In a twisted C2-symmetric environment
the previously mentioned symmetry-forbidden transitions are
now all polarized along the C2 axis. Given the higher energy
expected for these transitions, a twisted coordination geom-
etry will presumably alter only the blue edge of the MLCT
absorption band. Apart from symmetry properties, the
oscillator strength varies also with the degree of orbital
overlap between the metal d and antibonding π* ligand
orbitals. The coefficients of the LUMO orbitals, in particular
at the coordinating nitrogen atoms, are therefore of critical
importance. Using both symmetry and orbital overlap as
qualitative arguments, it is apparent that the b1(ψ) r dxz
transition is favored over the a2(ø) r dxy absorption (Figure
3a). The b1(ψ) orbital has indeed been identified by EPR
spectroscopy as the predominant electron acceptor for a
Figure 1. ORTEP plots and atom-numbering scheme for Cu(I) complexes
7 (top), 8a (middle), and 8b (bottom). Hydrogen atoms are omitted for
clarity.
tion geometry of complex 8b differs significantly from that
of 8a, despite the overall structural similarity. The dihedral
angle between the intersecting coordination planes N(1)-
Cu-N(2) and N(3)-Cu-N(5) of 80.9° is significantly
flattened compared to 8a. The steric demand of the methyl
substituents is also reflected in longer bond distances of the
coordination tetrahedron. The copper-nitrogen bond dis-
tances are generally increased by 3 to 12 pm. Interestingly,
the deviation from an ideal C2-symmetry is more pronounced
in 8b compared to the other two structures. The Cu-N(1)
bond distance was determined to be 217 pm, whereas the
symmetrical Cu-N(2) distance is only 209 pm. The space-
filling representation depicted in Figure 2 illustrates how the
(33) Phifer, C. C.; McMillin, D. R. Inorg. Chem. 1986, 25, 1329.
(34) Kaim, W. J. Am. Chem. Soc. 1982, 104, 3833.
(35) Orgel, L. E. J. Chem. Soc. 1961, 3683.
(36) Day, P.; Sanders, N. J. Chem. Soc. A 1967, 1530.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4923

Cody et al.
Figure 3. (a) Qualitative orbital energy diagram for Cu(I)-phenanthroline complexes with D2d and C2V symmetry. The arrows indicate transitions which
are polarized along the C2 axis (z coordinate). (b) Coefficients for the LUMO a2(ø) and LUMO + 1 b1(ψ) of phenanthroline.
Table 3. Electrochemical and Photophysical Data for Complexes 7, 8a,
and 8b in Dichloromethanea
data
1/2 (V)b
7
8a
8b
E
0.59 (85)
0.33 (65)
0.48 (69)
absorption λmax (nm), 440, shoulder (780) 428 (3010)
418 (3070)
c
ꢀ (cm-1 M-1
)
emissiond λmax (nm)
quantum yielde
lifetime (ns)
585, 618 (shoulder) 700
685
0.13%
0.15%
0.004%
15, 155
5.2, 1.0, 0.09 110
a Measured at 25 °C. b Scan speed 100 mV s-1, 0.1 M TBAP as
electrolyte in nitrogen-saturated acetonitrile. The difference between anodic
and cathodic current peak potential is given in parentheses (in mV). c Molar
extinction coefficient in parentheses. d Excitation at 390 nm. e Ru(bpy)3(PF6)2
as reference (Φf(H2O) ) 0.042).14
Figure 4. Charge-transfer (left) and normalized emission spectra (right)
of complexes 7, 8a, and 8b in dichloromethane at 25 °C. The (very weak)
emission spectrum of 8a has been omitted.
by the above symmetry analysis, the MLCT band is
constituted of several possible transitions; however, flattening
distortion to a C2-symmetric environment is expected to
induce changes in the blue edge of the MLCT band. Hence,
the observed low-energy shoulders are consistent with both
a C2V- or C2-symmetric coordination environment, and should
not be mistaken as an indicator for dihedral distortion. The
UV-vis absorption spectrum of 7 differs significantly from
the spectra of the pyrazole-substituted complexes 8a and 8b.
The MLCT absorption band appears only as a shoulder
around 440 nm and with significantly smaller oscillator
strength. The reason for this difference is not immediately
apparent, but might be attributed to reduced orbital overlap
due to the flattened coordination environment of 7. Alter-
natively, the MLCT absorption might be shifted to higher
energy and overlap with the strongly absorbing ligand-
centered transitions.
number of phenanthroline complexes with low-lying MLCT
transitions.34,37
UV-Vis Spectra. The room-temperature charge-transfer
absorption spectra for complexes 7, 8a, and 8b in dichlo-
romethane are shown in Figure 4. All relevant photophysical
parameters including the ground-state oxidation potentials
are compiled in Table 3. The two complexes 8a and 8b with
the pyrazole-substituted ligands exhibit both strong CT
absorptions over a broad range of the visible spectrum with
a maximum centered around 428 and 418 nm, respectively.
The molar absorptivity is very similar to the data reported
for the 2,9-diphenyl-substituted phenanthroline copper(I)
complex [Cu(dpp)2]+ and suggests a similar degree of
electronic delocalization.12 In the case of the structurally
related bis-phenanthroline Cu(I) complexes the low-energy
shoulder in the 550-600 nm range has been attributed to a
geometrical distortion from D2d symmetry.38,39 As illustrated
Emission Spectra. Upon excitation at 390 nm complexes
8a and 8b exhibit a single broad emission band at 700 and
685 nm, respectively. The thioether derivative 7 emits at
(37) Farrell, I. R.; Hartl, F.; Zalis, S.; Mahabiersing, T.; Vlcek, A. J. Chem.
Soc., Dalton Trans. 2000, 4323.
(38) Klemens, F. K.; Palmer, C. E. A.; Rolland, S. M.; Fanwick, P. E.;
McMillin, D. R.; Sauvage, J. P. New J. Chem. 1990, 14, 129.
(39) Ichinaga, A. K.; Kirchhoff, J. R.; McMillin, D. R.; Dietrich-Buchecker,
C. O.; Marnot, P. A.; Sauvage, J. P. Inorg. Chem. 1987, 26, 4290.
4924 Inorganic Chemistry, Vol. 42, No. 16, 2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
nential decay law with short and long lifetime components
of 15 and 155 ns, respectively. The emission of complex 8b
exhibits a slightly smaller lifetime of 110 ns, but follows a
monoexponential decay law. The decay kinetics of complex
8a is much faster and was fitted with a triexponential rate
law yielding the lifetime components 5.2, 1.0, and 0.09 ns.
In the discussion of photophysical properties it is often
instructive to extract the radiative (kr) and nonradiative (knr)
decay rate from the quantum yield and the observed rate
constant kobs. The second- and third-order decay laws for 7
and 8a imply that emission occurs from more than one
excited state and therefore kr and knr cannot be simply
calculated. On a qualitative level the measured lifetimes
parallel the trend in quantum yields.
According to detailed studies of copper(I) bis(phenanthro-
line) complexes, the shortening of the MLCT excited-state
lifetime can be rationalized on the basis of energy gap law
considerations or exciplex quenching with solvent mole-
cules.41-43 As previously noted, conformational locking by
sterically demanding substituents restricts structural relax-
ation, induces a blue shift of the emission energy, and should
increase the quantum yield as a consequence of the energy
gap law. This prediction is based on the assumption that the
vibrational overlap between the electronic ground and excited
states is decreasing with increasing energy gap. This model
is also helpful to rationalize the observed differences in the
excited-state dynamics of 8a and 8b. The steric requirements
of the methyl groups in 8b not only reduce the conforma-
tional flexibility, but also restrict solvent access for exciplex
quenching. As a result, the quantum yield and luminescence
lifetime of 8b are greatly improved compared to 8a. The
molecular architecture of complex 7 allows better solvent
access, but is inherently less flexible due to the smaller size
of the chelate ring. The resulting blue shift of the emission
energy is in agreement with an increase in quantum yield
and luminescence lifetime as predicted by the energy gap
law.
Figure 5. Luminescence decay data of complexes 7, 8a, and 8b in
degassed dichloromethane at 25 °C (time-correlated single photon counting).
considerably higher energy with a maximum at 585 nm and
a shoulder around 618 nm (Figure 4). A detailed analysis of
the emission spectra of copper(I) bis(phenanthroline) com-
plexes by Kirchhoff et al. revealed that these complexes emit
from close-lying, thermally equilibrated singlet and triplet
excited states.40 Quantum chemical calculations suggest a
similar situation for the mono-phenanthroline complexes 7,
8a, and 8b (section 4). The appearance of a single band does
therefore not necessarily imply that emission occurs from a
single excited state. Unlike compound 7 the emission maxima
for 8a and 8b compare well with the peak emission of 690
nm reported for copper(I) bis(2,9-diphenylphenanthroline).12
As shown for a series of structurally related copper(I) bis-
(phenanthroline) complexes, the emission energy reflects the
extent of flattening distortion in the excited state. Upon
increase of the steric requirements of the substituents in the
2- and 9-positions of the phenanthroline ring, the emission
maxima are blue-shifted.9,12,41 Differences in the flexibility
of the ligand backbone might also be the key factor that leads
to the higher emission energy of thioether derivative 7. In
fact, compound 7 contains a seven-membered chelate ring,
whereas 8a and 8b comprise a larger eight-membered ring.
The increased flexibility of 8a and 8b possibly allows a
higher degree of flattening distortion in the Cu(II)-like excited
state. This assumption is corroborated by the decrease of
the emission quantum yield and excited-state lifetime (see
below).
Transient Absorption Spectra. Figure 6 depicts time-
resolved absorbance-difference spectra for complexes 7, 8a,
and 8b. The samples were excited at 390 nm with a laser
pulse of 120 fs duration and the spectra recorded with a
“white-light” probe pulse at the indicated delay times. All
three compounds exhibit a characteristic increase in absorp-
tion between 450 and 700 nm within 600 fs upon excitation.
In contrast to previously reported transient absorption data
+ 44,45
for Cu(dmp)2 ,
the shape and peak position of the
spectrum change as a function of time. The data are therefore
not consistent with a simple excited-state dynamics and
suggest the consecutive formation of at least three different
species. Since vibrational relaxation occurs on the picosecond
time scale, the spectra observed during evolution of the signal
3.2. Time-Resolved Spectroscopy. As shown in Figure
5, the decay kinetics of complexes 7, 8a, and 8b vary
substantially. The thioether derivative 7 follows a biexpo-
(42) Everly, R. M.; Ziessel, R.; Suffert, J.; McMillin, D. R. Inorg. Chem.
1991, 30, 559.
(43) Felder, D.; Nierengarten, J. F.; Barigelletti, F.; Ventura, B.; Armaroli,
N. J. Am. Chem. Soc. 2001, 123, 6291.
(40) Kirchhoff, J. R.; Gamache, R. E., Jr.; Blaskie, M. W.; Del Paggio, A.
A.; Lengel, R. K.; McMillin, D. R. Inorg. Chem. 1983, 22, 2380.
(41) Cunningham, C. T.; Cunningham, K. L. H.; Michalec, J. F.; McMillin,
D. R. Inorg. Chem. 1999, 38, 4388.
(44) Palmer, C. E. A.; McMillin, D. R.; Kirmaier, C.; Holten, D. Inorg.
Chem. 1987, 26, 3167.
(45) Chen, L. X.; Jennings, G.; Liu, T.; Gosztola, D. J.; Hessler, J. P.;
Scaltrito, D. V.; Meyer, G. J. J. Am. Chem. Soc. 2002, 124, 10861.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4925

Cody et al.
Figure 7. Cyclovoltammogram for complexes 7 (a), 8a (c), and 8b (b) in
acetonitrile with 0.1 M Bu4NPF6 as supporting electrolyte, glassy carbon
electrode, and Ag/Ag+ reference electrode (potential plotted against
Fc/Fc+ couple).
ation during the flattening process followed by structural
adjustments for the reduced ligand and possibly formation
of a five-coordinate species.45
3.3. Redox Properties. Cyclic voltammetry studies of
complexes 7, 8a, and 8b in acetonitrile reveal for all
compounds a quasi-reversible one-electron oxidation-reduc-
tion wave associated with the Cu(I)/Cu(II) couple (Figure
7). The measured half-wave potentials range between 0.33
and 0.59 V (vs Fc/Fc+) and indicate significant stabilization
of the monovalent oxidation state (Table 3). A comparison
with published half-wave potentials of substituted bis-
phenanthroline complexes suggests that the rigid molecular
architecture of 7, 8a, and 8b yields a comparable stabiliza-
tion. For example, Cu(dmp)2+ (dmp ) 2,9-dimethylphenan-
throline) is oxidized at E ) 0.27 V (vs Fc/Fc+)9 whereas
the potential of the sterically more crowded complex
Cu(dpp)2+ (dpp ) 2,9-diphenylphenanthroline) was measured
to be E ) 0.32 V (vs Fc/Fc+).12 The significantly greater
potential observed for complex 7 can be attributed to the
different coordination environment. The replacement of a
nitrogen donor atom with a sulfur has been shown to
significantly reduce the affinity toward Cu(II),46 and therefore
the Cu(I)/Cu(II) redox couple is shifted to a more positive
potential. In contrast, complex 8a and 8b have the same set
of donor atoms, and the potential difference of 0.15 V is
primarily due to differences in the ligand architecture.
Copper(II) complexes with diimine ligands have a high
tendency to form five-coordinate complexes in solution,
typically via coordination of a solvent molecule as the fifth
ligand.11 Steric shielding by the four methyl substituents in
8b reduces solvent accessibility and therefore destabilizes
the Cu(II) oxidation state. Solvent coordination in the
Cu(II) state is additionally supported by the larger than
expected potential difference between the oxidation and
reduction peaks. For all complexes the differences are
significantly higher than the 56.5 mV expected for a perfectly
reversible redox process (Table 3). Structural rearrangements
followed by coordination of a solvent molecule between the
oxidation and reduction steps may well account for this
observation.
Figure 6. Room-temperature transient absorption spectra obtained after
pulsed light excitation at 390 nm (0.8 mJ/cm2, 120 fs fwhm) of (a)
Cu(bempp)PF6 (7), (b) Cu(bpmpp)PF6 (8a), and (c) Cu(bdmpp)PF6 (8b).
The dotted spectra show the initial evolution of the signal with 0, 120,
240, 360, and 480 fs delay times.
(<500 fs) reflect only changes in the electronic structure.
The initial spectra for compound 8a and 8b exhibit each a
prominent signature with maxima at 540 and 590 nm (Figure
6b,c). The band shape resembles the ground state absorption
spectrum of the phenanthroline radical anion with maxima
at 585 and 636 nm, though with a significant blue shift.
Time-resolved absorption studies with copper(I) bis(2,9-
dimethylphenanthroline) showed qualitatively similar blue-
shifted spectra, which were interpreted by the formation of
the ligand radical anion in the excited state.44,45 Interestingly,
after 40 ps the two maxima coalesce to a single broad band,
which again disappears to form a qualitatively similar
spectrum compared to the one observed within the first 500
fs. The transient absorption spectra series for complex 7
shows qualitatively the same pattern (Figure 6a). The initial
spectra also exhibit two maxima at 520 and 570 nm, which
are slightly more blue-shifted compared to 8a and 8b. As
observed previously for the pyrazole derivatives 8a and 8b,
the two maxima become gradually broader and again yield
the original band shape after about 1 ns. In summary, the
data strongly support the formation of the ligand radical anion
in the excited state and show qualitatively a very similar
trend for all three complexes 7, 8a, and 8b. The temporary
spectral broadening is presumably due to vibrational relax-
(46) Ambundo, E. A.; Deydier, M.-V.; Grall, A. J.; Aguera-Vega, N.;
Dressel, L. T.; Cooper, T. H.; Heeg, M. J.; Ochrymowycz, L. A.;
Rorabacher, D. B. Inorg. Chem. 1999, 38, 4233.
4926 Inorganic Chemistry, Vol. 42, No. 16, 2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
Table 4. Calculated Structural Parameters for Cu(I) Complexes 7, 8a, and 8ba
[Cu(bempp)]+ (7)
[Cu(bpmpp)]+ (8a)
[Cu(bdmpp)]+ (8b)
Cu(1)-N(1) (Å)
Cu(1)-N(2) (Å)
Cu(1)-S(1) (Å)
Cu(1)-S(2) (Å)
2.014
2.000
2.270
2.298
Cu(1)-N(1) (Å)
Cu(1)-N(2) (Å)
Cu(1)-N(3) (Å)
Cu(1)-N(5) (Å)
2.040
2.044
1.985
1.987
2.081
2.089
2.018
2.022
N(1)-Cu(1)-N(2) (deg)
N(1)-Cu(1)-S(1) (deg)
N(1)-Cu(1)-S(2) (deg)
N(2)-Cu(1)-S(1) (deg)
N(2)-Cu(1)-S(2) (deg)
S(1)-Cu(1)-S(2) (deg)
84.6
N(1)-Cu(1)-N(2) (deg)
N(1)-Cu(1)-N(3) (deg)
N(1)-Cu(1)-N(5) (deg)
N(2)-Cu(1)-N(3) (deg)
N(2)-Cu(1)-N(5) (deg)
N(3)-Cu(1)-N(5) (deg)
82.2
81.0
107.0
122.0
128.3
103.1
111.1
116.0
119.4
119.5
115.7
104.0
113.4
121.7
122.1
113.1
105.2
θb (deg)
73.5
θb (deg)
87.0
82.6
a B3LYP/6-31G** optimized geometries, with numbering scheme given in Figure 1. b θ is the angle between the two intersecting planes of the coordination
tetrahedron: N(1)-Cu-N(2) and S(1)-Cu-S(2) for complex 7, and N(1)-Cu-N(2) and N(3)-Cu-N(5) for complexes 8a and 8b, respectively.
4. Quantum Chemical Calculations. 4.1. Ground-State
Geometries. Geometries obtained by optimization at the
B3LYP/6-31G** level of theory are presented in Table 4.
In general, there is good agreement with the observed X-ray
structures in Table 2. Distances between Cu and the
coordinating atoms are generally within 0.04 Å of experi-
ment; this compares to an average error of 0.013 Å at this
level of theory47 for bonds in the G2 data set,48 which
contains primarily covalent bonds between first-row atoms.
The somewhat larger discrepancies in this case seem reason-
able given that the distances are larger and correspond to
weaker binding than in the G2 molecules, and they are
smaller than errors (0.09 Å) in predicted distances for
tetracoordinated Cu(I) in [Cu2(pip)2] [pip ) (2-picolylimi-
nomethyl)pyrrole anion] using B3LYP/LANL2DZ.49 Bond
angles are within 3° (compared to an average error of 0.6°
for the G2 molecules), and the angle between coordinating
planes is predicted within 2° of the experimental value. These
results indicate that there are no major distortions in the
coordination geometry due to crystal packing forces or the
influence of the counterion.
The experimental results indicate a slight distortion from
local C2 symmetry, as seen by the differences in the distance
to Cu of N(1) and N(2) [or N(3) and N(5), or S(1) and S(2)].
The theoretical results are also consistent with this distortion,
although the difference in bond lengths is generally reduced.
The most significant distortion is a 0.078 Å difference
between the distances to N(1) and N(2) in 8b, while this
difference is only 0.008 Å theoretically. The crystal structures
were used as starting geometries during optimization, and it
is possible that the potential energy surface is sufficiently
flat that the optimization criteria (maximum gradient 0.0003
hartree/bohr, energy change 10-6 hartree) may be satisfied
before the precise minimum geometry, which might be
symmetric, has been reached. This seems likely for 8a and
8b, where the distortion in the theoretical geometry is only
a few thousandths of an angstrom. The difference is larger
in 7, probably because of the different orientations of the
two pendant ethyl groups observed in the crystal structure
and retained in the theoretical geometry.
4.2. TD-DFT Calculations of Excited States. To under-
stand the electronic spectrum of complexes 7, 8a, and 8b in
more detail, vertical excitation energies were computed using
time-dependent density functional theory (TD-DFT), again
with the B3LYP functional and 6-31G** basis set. A large
number of singlet and triplet states are predicted between
1.7 and 4 eV, and the calculated excitation energies, oscillator
strengths, and approximate composition of these states are
tabulated in the Supporting Information. In general, TD-DFT
is a fairly reliable and inexpensive technique, yielding
excitation energies for valence states with a typical accuracy
of about 0.4 eV.23,24 Previous applications of TD-DFT to
vertical spectra of transition metal complexes indicate that
energies of charge-transfer states are typically underestimated
by several tenths of an electronvolt,37,50 while Tozer et al.51
have shown that these errors can grow to 1 eV or more for
certain cases. One very recent study indicates that B3LYP
TD-DFT succeeds in accurately modeling the photophysics
of [Cu(I)(dmp)2]+.52 Our goals in the present study are to
obtain a qualitative picture of the excited states and to
evaluate the performance of TD-DFT for this case.
The observed λmax values for absorption of 7, 8a, and 8b
(Table 3) are 440 nm (2.82 eV), 428 nm (2.90 eV), and 418
nm (2.97 eV). The observed excitation energies are reason-
ably well matched by the TD-DFT predictions, which are
depicted schematically in Figure 8 for vertical singlet and
triplet transition energies. The excited singlet states with the
largest oscillator strengths are at 2.57 (8a, strength 0.065),
2.48 (8b, strength 0.047), and two near-degenerate states for
7 at 2.79 and 2.83 eV (strength 0.032). For 8a and 8b, the
excited-state energy is underestimated as expected for charge-
transfer states at this level of theory. For 7, the agreement
with experiment appears fortuitously good. However, it is
possible that this state is also theoretically underestimated
if the MLCT is not the shoulder assigned in Figure 4 but
instead is buried within the higher-energy transitions below
400 nm.
(47) Bauschlicher, C. W. Chem. Phys. Lett. 1995, 246, 40.
(48) Curtiss, L. A.; Raghavachari, K.; Trucks, G. W.; Pople, J. A. J. Chem.
Phys. 1991, 94, 7221.
(49) Liao, Y.; Novoa, J. J.; Arif, A.; Miller, J. S. Chem. Commun. 2002,
3008.
(50) Full, J.; Gonzalez, L.; Daniel, C. J. Phys. Chem. A 2001, 105, 184.
(51) Tozer, D. J.; Amos, R. D.; Handy, N. C.; Roos, B. O.; Serrano-Andres,
L. Mol. Phys. 1999, 97, 859.
(52) Zgierski, M. Z. J. Chem. Phys. 2003, 118, 4045.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4927

Cody et al.
Figure 8. Energy level diagram of the calculated singlet (left) and triplet
(right) excited states for 7 (a), 8a (b), and 8b (c) (TD-DFT B3LYP/6-
31G**).
The nature of the low-lying excited states can be elucidated
by the electron attachment and detachment analysis of Head-
Gordon and co-workers.25 This procedure decomposes the
difference density between ground and excited states into
the negative of a “detachment density” corresponding to the
removal of charge from the ground state plus an “attachment
density” corresponding to the new arrangement of this charge
in the excited state. The attachment/detachment density plots
for the lowest seven excited states of 7, 8a, and 8b are
presented in Figure 9. The plots are very similar among the
molecules considered, and they clearly indicate that all of
the lowest-lying states are MLCT states in which electron
density is detached from Cu and attached onto the phenan-
throline moiety. In some cases (particularly for 7) a minor
amount of density is also removed from some of the atoms
in closest proximity to Cu. In nearly all of the lowest-lying
singlet states, the electron attachment density is consistent
with either the b1(ψ) or a2(ø) phenanthroline π* orbitals,
which are very close energetically (within 0.001 hartree for
the three molecules considered). The rest of the ligand serves
primarily as a scaffold to arrange approximate tetrahedral
coordination in the ground state and does not participate
directly in the vertical electronic excitation process.
For the states predicted to be the strongest absorbers in
8a and 8b (S6 and S7, respectively), the attachment/
detachment density analysis suggests the promotion of an
electron from a dxz orbital to the phenanthroline b1(ψ) π*
orbital. As described in section 3.1, this is expected to be
the strongest MLCT transition because it should feature the
best orbital overlap between the two orbitals involved in the
transition. Moreover, the computed transition moment lies
primarily along the approximate C2 axis, which is again
consistent with the general rule that the primary intensity of
Figure 9. Attachment/detachment densities for the seven lowest excited
states of 7 (left), 8a (middle), and 8b (right). Plots are arranged with
increasing energy from bottom to top (TD-DFT B3LYP/6-31G**).
MLCT absorption in such complexes arises from transitions
polarized in the same direction as the charge transfer.36 The
second most intense absorbing state in 8a and 8b is the S4
state, which looks qualitatively like a dxz to phenanthroline
a2(ø) π* transition and is polarized primarily along the y
axis. Results for 7 are similar, although the electron attach-
ment density for the two most intensely absorbing states (S4
and S6) mixes the near-degenerate unoccupied b1(ψ) and
a2(ø) phenanthroline orbitals. The detachment density like-
wise shows some d orbital mixing. The reduced intensity of
the strongest transition in 7 compared to 8a and 8b seems
consistent with a reduction in orbital overlap caused by the
larger deviation from a tetrahedral coordinating geometry.
The TD-DFT computations reveal a dense manifold of
low-lying triplet states, similar to the singlet manifold (see
Figure 8). It is therefore challenging to assign the emission
4928 Inorganic Chemistry, Vol. 42, No. 16, 2003

Geometrically Constrained Cu(I)-Phenanthroline Complexes
spectrum, since any of these states are potentially emissive;
indeed, the time-resolved spectra discussed above already
indicate emission from multiple states for 7 and 8a. At the
ground-state geometry, the lowest-lying triplet states are at
1.78 (7, expt 2.12), 1.59 (8a, expt 1.77), and 1.42 eV (8a,
expt 1.81). These energies are roughly consistent with the
observed λmax values for emission for the given level of
theory, but such a comparison neglects geometry relaxation
effects in the excited state which will reduce the energy gap.
To attempt to estimate the effect of geometry relaxation,
B3LYP/6-31G** optimizations were performed for the
lowest-lying triplets of each complex. The vertical emission
energies from this lowest triplet, computed as the energy
difference between the B3LYP energies for the triplet and
singlet at the triplet geometries, took on much lower values
of 1.1-1.3 eV. This suggests that either there is a funda-
mental difficulty of B3LYP for computing vertical singlet-
triplet gaps for charge-transfer states or most emission is
taking place from higher-lying triplets and/or before geom-
etry relaxation is completed.
are both key elements to enhance the emission energy, extend
the emission lifetime, and improve the quantum yield.41
McMillin and co-workers have recently demonstrated that
this strategy leads to highly emissive Cu(I) complexes with
very impressive quantum yields of up to 16%.53,54 By
increasing the size of the aryl substituent similar enhance-
ments might be possible for the investigated complexes with
1:1 binding stoichiometry.
Acknowledgment. Financial support from the Georgia
Institute of Technology is gratefully acknowledged. C.D.S.
acknowledges support by the National Science Foundation
through Grant CHE-0091380, and A.G. was supported by
the Georgia Institute of Technology’s Faculty Development
Program and NSF Grant 9974841. Computations were
supported by the Center for Computational Molecular
Science and Technology at the Georgia Institute of Techno-
logy and partially funded through a Shared University
Research (SUR) grant from IBM and the Georgia Institute
of Technology. We also thank Sarah Shealy and David
Bostwick for mass spectral data, and Dr. Mira Jasowicz for
her support with the electrochemical measurements.
Conclusions
Supporting Information Available: X-ray structural acquisition
and analysis (CIF) and additional information as mentioned in the
text. This material is available free of charge via the Internet at
http://pubs.acs.org. The crystallographic data for the structural
analyses have been deposited with the Cambridge Crystallo-
graphic Data Centre (deposition numbers CCDC 213571-213573).
Copies of this information may be also obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge CB21EZ,
U.K. (fax, +44-1223-336033; e-mail, deposit@ccdc.cam.ac.uk or
http://www.ccdc.cam.ac.uk).
The detailed structural and spectroscopic characterization
of a series of geometrically constrained Cu(I) phenanthroline
complexes revealed a picture that strongly resembles the
photophysics of related and well-investigated copper(I) bis-
phenathroline complexes. The rigid ligand architecture of
the studied complexes provides significant stabilization to
restrict structural relaxation in the excited state. Thus, the
observed emission lifetimes and quantum yields compare
well with the structurally related bis-phenanthroline com-
plexes. However, transient absorption data and electrochemi-
cal measurements both indicate that coordinative rearrange-
ments occur in the excited state leading to nonradiative
deactivation via exciplex quenching. Conformational con-
straints combined with suppression of exciplex quenching
IC034529J
(53) Cuttell, D. G.; Kuang, S. M.; Fanwick, P. E.; McMillin, D. R.; Walton,
R. A. J. Am. Chem. Soc. 2002, 124, 6.
(54) Kuang, S. M.; Cuttell, D. G.; McMillin, D. R.; Fanwick, P. E.; Walton,
R. A. Inorg. Chem. 2002, 41, 3313.
Inorganic Chemistry, Vol. 42, No. 16, 2003 4929