Spectroscopic and computational investigations of the temperature-dependent emission behavior of [Re(CNx)5Cl] and [Re(CNx)6]+ complexes

Article abstract of DOI:10.1021/om049443s

[Re(CNx)₅Cl] and [Re(CNx)₆](PF₆) complexes form with the isocyanide ligand, 2,6-dimethylphenylisocyanide (CNx). [Re(CNx)₅Cl] crystallizes in the space group P(2)1/c with a Re-Cl bond length of 2.5278(16) A, a Re-C bond length trans to Cl of 1.937(6) A, and Re-C bond lengths in the equatorial plane ranging from 1.998(6) to 2.034(6) A. The complexes are highly emissive at 77 K with emission lifetimes of 5.1 and 4.6 μs, respectively, but are nonemissive at room temperature. Density functional theory (DFT) geometry optimizations of the ground and the triplet metal-ligand-to-ligand charge transfer ( ³MLLCT) states of the complexes in the gas phase place the energy of the ³MLLCT state for [Re(CNx)₅Cl] 200 cm^-1 higher than the experimental emission peak at 77 K The energies of the singlet excited states of the two complexes calculated in ethanol using time-dependent density functional theory (TDDFT) and conductor-like polarizable continuum model (CPCM) deviate by less than 600 cm^-1 from the corresponding UV-vis peaks in the same solvent. The TDDFT/CPCM calculation confirms the existence of ³d-d states that provide a nonradiative relaxation pathway accounting for loss of emission from the emitting state at room temperature. Calculations also reveal a singlet to triplet excitation at 35 700 cm^-1 near the absorption of the CNx ligand at 35 200 cm^-1 and a ³π→π* state of CNx 200 cm^-1 higher than the experimental phosphorescence energy.

Full text of DOI:10.1021/om049443s

Organometallics 2005, 24, 395-404
395
Spectroscopic and Computational Investigations of the
Temperature-Dependent Emission Behavior of
[Re(CNx)5Cl] and [Re(CNx)6]+ Complexes
John M. Villegas, Stanislav R. Stoyanov, Joseph H. Reibenspies,† and
D. Paul Rillema*
Department of Chemistry, Wichita State University, 1845 N. Fairmount Street,
Wichita, Kansas 67260-0051
Received July 22, 2004
[Re(CNx)5Cl] and [Re(CNx)6](PF6) complexes form with the isocyanide ligand, 2,6-
dimethylphenylisocyanide (CNx). [Re(CNx)5Cl] crystallizes in the space group P(2)1/c with
a Re-Cl bond length of 2.5278(16) Å, a Re-C bond length trans to Cl of 1.937(6) Å, and
Re-C bond lengths in the equatorial plane ranging from 1.998(6) to 2.034(6) Å. The complexes
are highly emissive at 77 K with emission lifetimes of 5.1 and 4.6 µs, respectively, but are
nonemissive at room temperature. Density functional theory (DFT) geometry optimizations
of the ground and the triplet metal-ligand-to-ligand charge transfer (3MLLCT) states of the
complexes in the gas phase place the energy of the 3MLLCT state for [Re(CNx)5Cl] 200 cm-1
higher than the experimental emission peak at 77 K. The energies of the singlet excited
states of the two complexes calculated in ethanol using time-dependent density functional
theory (TDDFT) and conductor-like polarizable continuum model (CPCM) deviate by less
than 600 cm-1 from the corresponding UV-vis peaks in the same solvent. The TDDFT/
CPCM calculation confirms the existence of 3d-d states that provide a nonradiative relaxation
pathway accounting for loss of emission from the emitting state at room temperature.
Calculations also reveal a singlet to triplet excitation at 35 700 cm-1 near the absorption of
the CNx ligand at 35 200 cm-1 and a 3πfπ* state of CNx 200 cm-1 higher than the
experimental phosphorescence energy.
Introduction
sorption spectrum and the photochemistry4 are domi-
nated by the d-d states, while low-lying MLCT states
are dominant in aryl isocyanide complexes of low-valent
metal centers1,2 such as hexakis(aryl isocyanides) of
W(0), W(CNAr)6. The MLCT state occurs as the lowest-
lying excited state, and its population appears to cause
associative ligand substitution reactions.3 Computa-
tional studies of Ru(II) heterocyclic complexes contain-
ing the 2,6-dimethylphenylisocyanide ligand synthe-
sized in our laboratories revealed the presence of d-d
Several studies have revealed that d6 isocyanide
complexes exhibit ligand photosubstitution reactions
analogous to those of metal carbonyls.1-4 However, due
to their more extended electronic structure, complexes
of isocyanide ligands are expected to exhibit more varied
behavior. Isocyanide ligands are known to act like
carbon monoxide by serving as σ donors and/or π
acceptors.1,5-15 In [Fe(CNMe)6]2+ for example, the ab-
3
states a few hundred wavenumbers above the MLCT
state,16 accounting for loss of emission at room temper-
* To whom correspondence should be addressed. E-mail:
paul.rillema@wichita.edu.
3
ature via thermal population of the d-d state.
† Department of Chemistry, Texas A & M University, P.O. Box
30012, College Station, TX 77842-3012.
We have found density functional theory (DFT) to be
a powerful tool for interpretation of electrochemical and
spectroscopic results. Earlier reports from our labora-
tory17 have shown a linear relationship between the
HOMO-LUMO energy gap and the electrochemical
redox potentials for a series of isoelectronic Ru(II)
diimine complexes. DFT calculations on the ground and
MLCT states of a series of Re(I) tricarbonyl complexes
(1) Stacy, N. E.; Conner, K. A.; McMillin, D. R.; Walton, R. A. Inorg.
Chem. 1986, 25, 3649.
(2) Wrighton, M. S.; Geoffroy, G. L. Organometallic Photochemistry;
Academic Press: New York, 1979; Chapter 2.
(3) Gray, H. B.; Mann, K. R.; Lewis, N. S.; Thich, J. A.; Richman,
R. M. Adv. Chem. Ser. 1978, 168, 44.
(4) Costanzo, L. L.; Giuffrida, S.; DeGuidi, G.; Condorelli, G. Inorg.
Chim. Acta 1985, 101, 71.
(5) Malatesta, L.; Bonati, F. Isocyanide Complexes of Metals;
Wiley: New York, 1969.
(6) Bonati, F.; Minghetti, G. Inorg. Chim. Acta 1974, 9, 95.
(7) Barybin, M. V.; Young, V. G.; Ellis, J. E. J. Am. Chem. Soc. 1998,
120, 429.
(13) Szalda, D. J.; Dewan, J. C.; Lippard, S. J. Inorg. Chem. 1981,
20, 3851.
(8) Barybin, M. V.; Young, V. G.; Ellis, J. E. J. Am. Chem. Soc. 1999,
121, 9237.
(14) Shi, Q. Z.; Richmond, T. G.; Trogler, W. C.; Basolo, F. Inorg.
Chem. 1984, 23, 957.
(9) Mialki, W. S.; Wigley, D. E.; Wood, T. E.; Walton, R. A. Inorg.
Chem. 1982, 21, 480.
(15) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford
University Press: New York, 1992.
(10) Bohling, D. A.; Mann, K. R. Inorg. Chem. 1983, 22, 1561.
(11) Nielson, R. M.; Wherland, S. Inorg. Chem. 1985, 24, 1803.
(12) Brant, P.; Cotton, F. A.; Sekutowski, J. C.; Wood, T. E.; Walton,
R. A. J. Am. Chem. Soc. 1979, 101, 6588.
(16) Villegas, J. M.; Stoyanov, S. R.; Wei, H.; Lockyear, L. L.;
Reibenspies, J.; Rillema, D. P. Inorg. Chem. 2004, 43, 6383-6396.
(17) Stoyanov, S. R.; Villegas, J. M.; Rillema, D. P. Inorg. Chem.
2002, 41, 2941.
10.1021/om049443s CCC: $30.25 © 2005 American Chemical Society
Publication on Web 01/05/2005

396 Organometallics, Vol. 24, No. 3, 2005
Villegas et al.
another 3-4 h. The solvent was evaporated under vacuum
until about 5-10 mL was left. Then 15 mL of a saturated
have been used by others for the investigation of the
excited-state geometries and electronic structures.18a
Correlations were also reported between time-depend-
ent density functional theory (TDDFT) calculated MLCT
states and UV-vis spectra for Re(I) complexes contain-
ing the ligand azophenine.18b The TDDFT method treats
molecules in the gas phase; therefore the method does
not always give the right electronic transition energies
in solution.19a-c
solution
of
NH4PF6 in water was added. The solution was diluted to 50
mL. The tan-colored precipitate was filtered, washed with
water, and dried in vacuo. Yield: 0.18 g (70%). Anal. Calcd
for ReC54H54N6PF6: C, 58.00; H, 4.87; N, 7.52. Found: C,
57.86; H, 5.02; N, 7.36. IR (KBr pellet): 2066, 1590, 1467, 1382,
1261, 1183, 1034, 842, 775, 719, 657, 557, 491 cm-1. 1H NMR
(DMSO): δ ppm 2.42 (s, 36H), 7.23 (m, 18H).
We have reported that combining the conductor-like
polarizable continuum model (CPCM) with the TDDFT
method gives calculated excited-state energies that
correlate well with the experimental absorption energies
for [Ru(bpy)2(CNx)Cl]+ in seven solvents of different
polarity.19d The use of TDDFT and CPCM as reported
by others has produced dramatic changes in the excited-
state energies and assignments for Ru(II) and Os(II)
polypyridyl complexes.20 It is the primary method used
in our study.
The photoinduced ligand substitution of homoleptic
aryl isocyanide complexes of Re(I) in the presence of
halide ions1 as well as spectroscopic studies on the
analogue complexes of Ru(II) had been reported.21 Here
we elaborate on those earlier studies by examining the
excited-state properties in more detail by way of com-
paring experimental excited-state properties with those
computed by the TDDFT-CPCM method.
Instrumentation and Physical Measurements. UV-vis
spectra were obtained using a Hewlett-Packard model 8452A
diode array spectrophotometer. The IR spectra were acquired
using a Nicolet Avatar 360 FT-IR spectrophotometer. Proton
NMR spectra were obtained using a Varian Mercury 300 FT-
NMR spectrometer. An EG&G PAR model 263A potentiostat/
galvanostat was used to obtain the cyclic voltammograms. The
measurements were carried out in a typical H-cell using a
platinum disk working electrode, a platinum wire counter
electrode, and an Ag/AgCl reference electrode in acetonitrile.
The supporting electrolyte used was 0.1 M tetrabutylammo-
nium hexafluorophosphate (TBAH). Ferrocene was added as
a reference.
The sample preparation for emission studies involved dis-
solving a small amount of sample (∼2 mg) in the appropriate
solvent and the absorbance of the solution was measured. The
concentration of the solution was altered in order to achieve
an absorbance of about 0.10 at the lowest energy transition.
Such a concentration provided enough material for data
acquisition but excluded self-quenching processes. A 3-4 mL
aliquot of the solution was then placed in a 10 mm diameter
Suprasil (Heraeus) nonfluorescent quartz tube equipped with
a tip-off manifold. The sample was then freeze-pump-thaw
degassed for at least three cycles (to approximately 75 mTorr),
removing any gases from the sample. The manifold was then
closed, and the sample was allowed to equilibrate at room
temperature. The solvent evaporation was assumed to be
negligible; therefore the concentrations were assumed to
remain constant throughout this procedure. The corrected
emission spectra were collected using a Spex Tau3 fluorometer.
The excited-state lifetimes were determined using an OPOTEK
optical parametric oscillator pumped by a frequency tripled
Continuum Surlite Nd:YAG laser. The Origin 6.1 program by
OriginLab Corporation was used for curve-fitting analysis. The
excited-state lifetime experiments were conducted as previ-
ously published.22
X-ray Crystallography Data Collection. A Bausch and
Lomb 10× microscope was used to identify a suitable colorless
plate from a representative sample of [Re(CNx)5Cl] crystals
grown by slow evaporation of ethanol. The crystal was coated
in a cryogenic protectant (paratone) and was then fixed to a
loop, which in turn was fashioned to a copper mounting pin.
The mounted crystal was then placed in a cold nitrogen stream
(Oxford) maintained at 110 K.
A Bruker SMART 1000 X-ray three-circle diffractometer was
employed for crystal screening, unit cell determination, and
data collection. The goniometer was controlled using the
SMART software suite, version 5.056 (Microsoft NT operating
system). The sample was optically centered with the aid of a
video camera such that no translations were observed as the
crystal was rotated through all positions. The detector was set
at 5.0 cm from the crystal sample (CCD-PXL-KAF2, SMART
1000, 512 × 512 pixel). The X-ray radiation employed was
generated from a Mo sealed X-ray tube (KR ) 0.70173 Å with
a potential of 50 kV and a current of 40 mA) and filtered with
a graphite monochromator in the parallel mode (175 mm
collimator with 0.5 mm pinholes).
Experimental Section
Materials. The ligand 2,4-dimethylphenylisocyanide (CNx)
was purchased from Fluka. [Re(CO)5Cl] was purchased from
Aldrich. Optima grade methanol was purchased from Fisher
Scientific, while acetonitrile was purchased from Sigma-
Aldrich. AAPER Alcohol and Chemical Company was the
source of absolute ethanol. Ethanol and methanol were used
in a 4:1 (v/v) mixture to prepare solutions for the emission and
emission lifetime studies. Elemental analyses were obtained
from Desert Analytics Laboratory, Tucson, AZ.
Preparation of [Re(CNx)5Cl]. A 0.40 g portion of Re-
(CO)5Cl (1.1 mmol) was added with 1.33 g of CNx (10 mmol)
in a 125 mL round-bottomed flask. Then 50 mL of toluene was
added, and the solution was refluxed for 5 days. A yellow
precipitate formed. It was then filtered and washed with ether.
Yield: 0.34 g (35%). Anal. Calcd for ReC45H45N5Cl: C, 61.59;
H, 5.17; N, 7.98. Found: C, 61.54; H, 4.97; N, 7.85. IR (KBr
pellet): 2044, 1588, 1464, 1380, 1188, 1166, 1033, 843, 771,
721, 656, 573, 489 cm-1
30H), 7.21 (m, 15H).
.
1H NMR (DMSO): δ ppm 2.42 (s,
Preparation of [Re(CNx)6](PF6). A 0.20 g sample of [Re-
(CNx)5Cl] (0.23 mmol) was added with 0.06 g of AgCF3SO3
(0.24 mmol) in a 125 mL round-bottomed flask. The mixture
was dissolved in 50 mL of ethanol and was refluxed in the
dark for 5-6 h. It was allowed to cool, and the AgCl precipitate
was removed by filtration. To the filtrate was added 0.035 g
of CNx (0.27 mmol), and the solution was again refluxed for
(18) (a) Dattelbaum, D. M.; Martin, R. L.; Schoonover, J. R.; Meyer,
T. J. J. Phys. Chem. A 2004, 108, 3518-3526. (b) Frantz, S.; Rall, J.;
Hartenbach, I.; Schleid, T.; Zalis, S.; Kaim, W. Chem.-A Eur. J. 2004,
10, 149-154.
(19) (a) Monat, J. E.; Rodriguez, J. H.; McCusker, J. K. J. Phys.
Chem. A 2002, 106, 7399. (b) Rodrigues, J. H.; Wheeler, D. E.;
McCusker, J. K. J. Am. Chem. Soc. 1998, 120, 12051. (c) Stoyanov, S.
R.; Villegas, J. M.; Rillema, D. P. Inorg. Chem. 2003, 42, 7852. (d)
Stoyanov, S. R.; Villegas, J. M.; Rillema, D. P. Inorg. Chem. Commun.
2004, 7, 838-841.
(20) Guillemoles, J.-F.; Barone, V.; Joubert, L.; Adamo, C. J. Phys.
Chem. A 2002, 106, 11345.
(21) Larowe, J. E.; McMillin, D. R.; Stacy, N. E.; Tetrick, S. M.;
Walton, R. A. Inorg. Chem. 1987, 26, 966.
Dark currents were obtained for the appropriate exposure
time of 30 s. A rotation exposure was taken to determine
(22) Villegas, J. M.; Stoyanov, S. R.; Rillema, D. P. Inorg. Chem.
2002, 41, 6688.

[Re(CNx)5Cl] and [Re(CNx)6]+ Complexes
Organometallics, Vol. 24, No. 3, 2005 397
Figure 1. Schematic diagram of the synthesis of the complexes.
Table 1. Crystal Data for [Re(CNx)5Cl]
30 s dark current exposure. The total data collection was
performed for a duration of approximately 13 h at 110 K. No
significant intensity fluctuations of equivalent reflections were
observed.
After data collection, the crystal was measured carefully for
size, morphology, and color. These measurements are reported
in Table 1.
formula
C52H53ClN5Re
fw
969.64
cryst size, mm
0.30 × 0.05 × 0.05
cryst syst
monoclinic
space group
P2(1)/c
a, Å
10.159(4)
b, Å
21.548(4)
c, Å
21.411(4)
Results
R, deg
90
â, deg
92.286(3)
Synthesis. The complexes were prepared according
to the scheme presented in Figure 1. [Re(CNx)5Cl]
formed as a yellow precipitate upon reaction of [Re-
(CO)5Cl] with an excess amount of CNx in refluxing
toluene for 5 days. This complex served as a starting
material for the preparation of the hexakis CNx com-
plex. The latter was obtained by first refluxing the
pentakis complex with an equimolar amount of AgCF3-
SO3 and then removing the precipitated AgCl by filtra-
tion. Then an equimolar amount of CNx (plus slight
excess) was added to the filtrate and the solution was
again refluxed. The volume of the solvent was reduced,
and the desired product was precipitated by adding
NH4PF6. The synthesized complexes were characterized
γ, deg
90
V, Å3
4683(2)
Z
4
density(calcd), g cm-1
absorp coeff, mm-1
F(000)
θ range for data collection, deg
index ranges
1.375
2.691
1968
1.34 to 25.00
-9 e h e 12, -20 e k e 25,
-6 e l e 25
no. of reflns collected
no. of indep reflns
completeness to θ
17 576
17 700 [(R(int) ) 0.0000]
84.1%
max. and min. transmn
goodness-of-fit on F2
final R indices [I>2σ(I)]
R indices (all data)
0.8772 and 0.4990
1.025
R1 ) 0.0538, wR2 ) 0.1370
R1 ) 0.0699, wR2 ) 0.1506
1.717 and -1.917
largest diff peak and hole, e Å-3
1
by IR, H NMR, and elemental analysis.
crystal quality and the X-ray beam intersection with the
detector. The beam intersection coordinates were compared
to the configured coordinates, and changes were made accord-
ingly. The rotation exposure indicated acceptable crystal
quality, and the unit cell determination was undertaken. Sixty
data frames were taken at widths of 0.3° with an exposure
time of 10 s. Over 200 reflections were centered, and their
positions were determined. These reflections were used in the
autoindexing procedure to determine the unit cell. A suitable
cell was found and refined by nonlinear least squares and
Bravais lattice procedures and reported here in Table 1. The
unit cell was verified by examination of the hkl overlays on
several frames of data, including zone photographs. No
supercell or erroneous reflections were observed.
After careful examination of the unit cell, a standard data
collection procedure was initiated. This procedure consists of
collection of one hemisphere of data using omega scans,
involving the collection of 1201 0.3° frames at fixed angles for
φ, 2θ, and ø (2θ ) -28°, ø ) 54.73°), while varying omega.
Each frame was exposed for 30 s and contrasted against a
X-ray Crystal Structure Determination of [Re-
(CNx)5Cl]. The structure of [Re(CNx)5Cl] was deter-
mined by X-ray crystallography as shown in the ORTEP
diagram (Figure 2). The crystal data are presented in
Table 1. Selected bond distances and angles for the
complex are listed in Table 2.
The complex adopted a distorted octahedral geometry.
The angles of the trans ligands at the metal center were
176.1(2)° for C(1)-Re(1)-C(19), 174.98(15)° for Cl(1)-
Re(1)-(C28), and 173.7(2)° for C(10)-Re(1)-C(37). The
Re-C on the axial position (trans to Cl) was ∼0.1 Å
shorter compared to the bond distances of the Re-C in
the equatorial plane. The Re-Cl bond length was
2.5278(16) Å. The C-N-C angle of the CNx ligands in
the equatorial plane ranged from 163.6(6)° to 175.8(6)°.
The distortion of the angles from the 180° value could
be attributed to steric effects. However, the C-N-C
angle of the CNx ligand on the axial position was

398 Organometallics, Vol. 24, No. 3, 2005
Villegas et al.
Table 3. Experimentala Electronic Transitions and
Calculatedb Excited States of the Compounds at
Room Temperature.
Eexp, × 103 cm-1 Ecalc, × 103 cm-1
compound
CNx
(M-1 cm-1
)
(M-1 cm-1
)
assignment
48.5 (13 100)
42.0 (6400)
50.2
42.9
35.7
1πfπ*
1πfπ*
35.2 (780)c
3πfπ*
[Re(CNx)5Cl]
42.4 (44 600)
32.2 (78 300)
26.0 (22 700)
22.5 (6000) sd
43.1 (72 400)
33.3 (88 600)
28.4 (32 600) s
26.2 (10 100) s
22.5 (2700) s
LC (πfπ*)
MLLCT
MLLCT
MLLCT
LC (πfπ*)
MLLCT
MLLCT
MLLCT
MLLCT
31.9
26.6
[Re(CNx)6](PF6)
33.4
28.4
a In 4:1 (v/v) ethanol/methanol. b In ethanol. d s ) shoulder.
c Data from ref 21.
Figure 2. ORTEP diagram of [Re(CNx)5Cl].
Table 2. Selected Bond Length (Å) and Angles
(deg) for [Re(CNx)5Cl].
Re(1)-Cl(1)
Re(1)-C(1)
Re(1)-C(10)
Re(1)-C(19)
Re(1)-C(28)
Re(1)-C(37)
N(1)-C(1)
2.5278(16)
2.019(7)
1.998(6)
2.034(6)
1.937(6)
2.021(6)
1.185(7)
1.387(8)
1.189(7)
1.413(7)
1.162(7)
1.395(7)
1.176(7)
1.408(8)
1.189(7)
1.376(8)
C(1)-Re(1)-C(19)
Cl(1)-Re(1)-C(28)
C(10)-Re(1)-C(37)
N(1)-C(1)- Re(1)
N(2)-C(10)-Re(1)
N(3)-C(19)-Re(1)
N(4)-C(28)-Re(1)
N(5)-C(37)-Re(1)
C(1)-N(1)-C(2)
C(10)-N(2)-C(11)
C(19)-N(3)-C(20)
C(28)-N(4)-C(29)
C(37)-N(5)-C(38)
176.1(2)
174.98(15)
173.7(2)
177.7(5)
176.2(5)
176.8(5)
174.8(5)
174.3(5)
170.2(6)
175.8(6)
163.6(6)
154.6(6)
171.5(6)
N(1)-C(2)
N(2)-C(10)
N(2)-C(11)
N(3)-C(19)
N(3)-C(20)
N(4)-C(28)
N(4)-C(29)
N(5)-C(37)
N(5)-C(38)
Figure 3. Experimental absorption spectrum of [Re-
(CNx)6]+ and calculated singlet excited states. The excited
states are shown as vertical bars with height equal to the
extinction coefficient.19a
154.6(6)°, significantly smaller than those in the equa-
torial position.
cm-1 increased as the number of CNx ligands increased
from five to six. The positions of the lowest energy
transition for both complexes were the same at 22 500
cm-1. However, the molar absorption coefficient of the
pentakis complex was about twice as large as that of
the hexakis complex.
Electrochemical Studies. The redox potentials of
the two complexes were determined using cyclic volta-
mmetry. The electrochemical properties of the com-
plexes are tabulated in Table 4. No observable reduc-
tions were noted for the two complexes. The pentakis
complex showed a quasi-reversible oxidation at 0.41 V
(∆Ep ) 0.074 V) assigned to Re oxidation and an
irreversible oxidation at 1.11 V assigned to oxidation
of the chloro ligand. The hexakis complex underwent a
quasi-reversible oxidation at 1.08 V (∆Ep ) 0.080 V)
assigned to Re oxidation and an irreversible oxidation
at 1.99 V attributed to the second oxidation of
Re(I).
Electronic Absorption Studies. The absorption
spectrum of the CNx ligand was determined in 4:1
ethanol/methanol as solvent at room temperature for
comparison to electronic absorption spectra of the
complexes. The intense absorption at ∼43 500 cm-1 was
1
previously assigned as a πfπ* transition, while the
absorption at ∼35 700 cm-1 was assigned as a weak
3πfπ* absorption transition.21
The electronic absorption properties of the complexes
were also studied at room temperature in 4:1 ethanol/
methanol as solvent. Molar absorption coefficients (ꢀ)
were determined using Beer’s law studies consisting of
at least five dilution points. The probable assignments
of these bands were made on the basis of the docu-
mented optical transitions of similar types of Re(I)
complexes20,23-26 and the computational results (Table
3). The electronic spectrum of [Re(CNx)6]+ is shown in
Figure 3.
The lowest energy transition of the complexes was
assigned as metal-ligand-to-ligand charge transfer
(MLLCT), while those at higher energies were assigned
as CNx ligand πfπ* transitions. The molar absorptivi-
ties of the electronic transitions at 33 300 and ∼43 500
Emission Properties and Excited-State Life-
times. The emission spectra of the ligand at 77 K and
room temperature are presented in Figure 4A. Both
complexes were nonemissive at room temperature but
showed high-intensity emission at 77 K. As shown in

[Re(CNx)5Cl] and [Re(CNx)6]+ Complexes
Organometallics, Vol. 24, No. 3, 2005 399
Table 4. Electrochemical Properties of the Two
Complexes and [Ru(bpy)3]2+ in CH3CN at Room
Temperature
Table 5. Experimentala Emission Properties and
Calculatedb Lowest-Lying Triplet-State Energies of
the Compounds at 77 K
complex
E1/2(ox), Va
E1/2(red), Va
Eem, × 103 cm-1
τem, µs
compound
Ecalc, × 103 cm-1
77 K
77 K
[Re(CNx)5Cl]
0.41b
1.11c
1.08b
1.99c
1.27d
CNx
26.4
20.4
22.8
26.2
20.2
20.3
7 × 106c
[Re(CNx)6](PF6)
[Ru(bpy)3]2+ 32
[Re(CNx)5Cl]
5.1
[Re(CNx)6](PF6)
4.6
-1.31
-1.50
-1.77
a In 4:1 (v/v) ethanol/methanol. b In ethanol. c Data from ref 21.
Computational Technique. The geometries of the
ligand CNx, [Re(CNx)5Cl], and [Re(CNx)6]+ were opti-
mized in the singlet ground state in the gas phase using
the B3LYP27 functional of the Gaussian ’0328 program
package. The Stuttgart-Dresden (SDD) ECP29 was used
for the Re core potentials. The (8s7p6d)/[6s5p3d]-GTO
was applied for the valence shell of Re together with
the all-electron 6-311G* basis set30 for N, C, Cl, and H
atoms.
a Potential in V vs SSCE (scan rate ) 250 mV/s). b Quasi-
reversible oxidation wave. c Irreversible oxidation wave. d RuIII/
II redox couple.
Nonequilibrium TDDFT31/CPCM32,33 calculations were
employed to produce a number of singlet excited states
of CNx, [Re(CNx)5Cl], and [Re(CNx)6]+ in ethanol based
on the singlet ground-state geometry optimized in the
gas phase.34 The output contained information for the
excited-state energies and oscillator strengths (f) and a
list of the transitions that give rise to each excited state,
the orbitals involved, and the orbital contribution coef-
ficients of the transitions. Selected properties of the
singlet excited states of the CNx ligand with f > 0.01
and the [Re(CNx)5Cl] complex with f > 0.05 are listed
(23) Wrighton, M.; Morse, D. L. J. Am. Chem. Soc. 1974, 96, 998.
(24) Giordano, P. J.; Wrighton, M. S. J. Am. Chem. Soc. 1979, 101,
2888.
(25) Luong, J. C.; Nadjo, L.; Wrighton, M. S. J. Am. Chem. Soc. 1978,
100, 5790.
(26) Wallace, L.; Rillema, D. P. Inorg. Chem. 1993, 32, 3836-3843.
(27) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. (c) Vosko, S. H.; Wilk, L.;
Nusair, M. Can. J. Phys. 1980, 58, 1200.
(28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma,
K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.03; Gaussian, Inc.:
Pittsburgh, PA, 2003.
Figure 4. Emission spectra of the CNx ligand at room
temperature (- - -) and 77 K (s) (A) and [Re(CNx)5Cl]
(- - -) and [Re(CNx)6]+ (s) at 77 K (B) in 4:1 (v/v) ethanol/
methanol.
(29) Andrae, D.; Hauessermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
(30) (a) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72,
5639. (b) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650.
(31) (a) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. J. Chem.
Phys. 1998, 109, 8218. (b) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys.
Lett. 1996, 256, 454. (c) Casida, M. E.; Jamorski, C.; Casida, K. C.;
Salahub, D. R. J. Chem. Phys. 1998, 108, 4439.
(32) Cossi, M.; Barone, V. J. Chem. Phys. 2001, 115, 4708.
(33) (a) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995. (b)
Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem. 2003,
24, 669.
(34) Geometry optimization in solvents was not achieved. Partial
optimizations (change in distance of less than 0.001 Å and change in
angles of less than 0.01°) followed by TDDFT/CPCM calculation
produced excited-state energies that were not in better agreement with
the experimental excited-state energies than the excited-state energies
based on the gas phase optimized geometry.
Figure 4B, there was no vibronic structure in the
emission profile for both complexes. Both complexes
emit at a higher energy than the standard complex,
[Ru(bpy)3]2+, at 77 K under the same experimental
conditions. The emission lifetimes at 77 K of the two
complexes decreased slightly from 5.1 to 4.6 µs when
Cl was replaced with a CNx moiety in the coordination
sphere. The emission properties and the excited-state
lifetimes of the complexes studied at 77 K are listed in
Table 5.

400 Organometallics, Vol. 24, No. 3, 2005
Villegas et al.
Table 6. Calculated Singlet and Triplet Excited
States of CNxa
Table 8. Calculated Triplet Excited States of
[Re(CNx)5Cl] (A) and [Re(CNx)6]+ (B) in Ethanol
Based on the Lowest-Lying Triplet-State
state
f
ψofψv
type
EVER
Geometrya
T1
T2
T3
S1
S2
S4
0
0
0
0.04
0.25
0.48
HfL (0.7)
πfπ* (phenyl)
πfπ* (phenyl)
πfπ* (phenyl)
πfπ* (phenyl)
πfπ* (phenyl)
πfπ* (phenyl)
27.8
33.6
35.7
39.4
42.9
50.2
A
H-1fL (0.7)
H-1fL+1 (0.7)
H-1fL (0.6)
HfL (0.6)
state
f
ψfψv
type
EVER
1
2
3
4
0.00
0.01
0.00
0.05
H-1fH (1.0)
LfL+1 (0.9)
H-2fH (1.0)
LfL+2 (0.9)
Red, CNxfRed, CNx
CNxfCNx
Red, CNxfRed, CNx
CNxfCNx
23.4
24.4
25.2
26.0
HfL+1 (0.6)
a The triplet excited states are computed using the ground-state
geometry. EVER is the energy of the vertical transition in ×103
cm-1, f is the oscillator strength, and ψo and ψv are the occupied
and the virtual orbitals that define the transition. The transition
type is determined on the basis of the change in the spatial
distribution from occupied to virtual orbital. The absolute value
of the transition coefficient for each transition is given in paren-
theses. H ) HOMO and L ) LUMO. (See text for calculation
details.)
B
state
f
ψofψv
type
EVER
1
2
3
0.00
0.00
0.03
H-1fH (1.0)
H-2fH (1.0)
LfL+3 (0.6)
H-3fH (0.8)
LfL+1 (1.0)
Red, CNxfRed, CNx
Red, CNxfRed, CNx
CNxfCNx
CNxfRed, CNx
CNxfCNx
28.4
29.1
30.5
4
0.00
31.0
a EVER is the energy of the vertical transition in ×103 cm-1, f is
the oscillator strength, and ψo and ψv are the occupied and the
virtual orbitals that define the transition. The transition type is
determined on the basis of the change in the spatial distribution
from occupied to virtual orbital. The absolute value of the
transition coefficient for each transition is given in parentheses.
H ) HOMO and L ) LUMO. (See text for calculation details.)
Table 7. Calculated Singlet Excited States of
[Re(CNx)5Cl] in Ethanola
state
S7
f
ψofψv
type
EVER
0.13 H-2fL (0.4)
H-2fL+2 (0.4)
0.51 H-2fL (0.4)
H-2fL+2 (0.4)
0.48 H-2fL+1 (0.5)
0.13 H-1fL+3 (0.6)
0.34 HfL+3 (0.6)
0.27 HfL+4 (0.5)
0.11 H-1fL+4 (0.5)
0.08 H-2fL+3 (0.5)
H-1fL+5 (0.4)
0.21 H-1fL+4 (0.4)
HfL+5 (0.4)
0.16 H-1fL+5 (0.5)
HfL+5 (0.4)
0.10 H-2fL+5 (0.5)
0.06 H-3fL (0.6)
0.18 H-4fL (0.5)
0.08 HfL+13 (0.5)
0.20 H-1fL+12 (0.4) Red,CNxfCNx,Rep
H-1fL+13 (0.4) Red,CNxfCNx,Rep
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNxfCNx
Red, CNx f CNx
CNx, Cl f CNx
CNx, Cl f CNx
Red,CNxfCNx,Rep
26.6
S8
28.0
S9
28.7
29.0
29.4
30.3
30.8
31.3
with height equal to the extinction coefficient calculated
from the oscillator strength.19a
S10
S11
S12
S13
S14
The TDDFT/CPCM calculations are nonequilibrium
calculations with respect to the polarization process
between the solvent reaction field and the charge
density of the electronic state indicated in the input.
For singlet excited states, the state indicated in the
input is the singlet ground state.36 The CPCM is
designed to account for the bulk physical properties of
the solvent. It does not account for specific solvent-
solute interactions. The TDDFT is known to perform
well for the computing of CT excited states between
closely spaced moieties. The tandem use of CPCM and
TDDFT is currently the most suitable computational
approach for the treatment of the solvent effects on the
excited-state energies of transition metal complexes.19d
For the CNx ligand, a number of triplet excited states
were computed in ethanol based on the ground-state
geometry and are listed in Table 6. The singlet to triplet
transitions from the singlet ground state to the triplet
excited states are multiplicity forbidden (f ) 0) and are
correlated with the low-energy absorption of the ligand.
The lowest-lying triplet-state geometries of the ligand
CNx and of the two complexes were calculated using
unrestricted B3LYP in the gas phase. The spin con-
tamination from states of higher multiplicity was low.
The value of S2 was 2.021 for the ligand CNx, 2.015
for [Re(CNx)5Cl], and 2.020 for [Re(CNx)6]+. The ener-
gies of the lowest-lying triplet states were higher than
those of the corresponding ground states by 26 400 cm-1
for CNx, 20 400 cm-1 for [Re(CNx)5Cl], and 22 800 cm-1
for [Re(CNx)6]+ (Table 5). The lowest-lying triplet states
of the complexes were 3MLLCT states and featured
single HOMO and LUMO occupancy.
S15
S16
31.4
31.9
S18
S38
S39
S40
S41
33.3
38.1
38.3
38.4
38.7
S42
S44
0.08 H-3fL+1 (0.5)
0.05 H-4fL+1 (0.3)
H-3fL+2 (0.4)
CNx, ClfCNx
CNx, ClfCNx
CNx, ClfCNx
CNx, ClfCNx
39.0
39.3
S45
0.06 H-4fL+1 (0.3)
39.4
H-1fL+12 (0.3) Red,CNxfCNx,Rep
S47
S48
0.05 H-5fL (0.5)
0.27 H-4fL+2 (0.3)
πfπ* (CNx)
CNx, ClfCNx
39.8
39.8
H-2fL+12 (0.3) Red,CNxfCNx,Rep
S49
0.11 H-5fL (0.3)
H-9fL (0.4)
πfπ* (CNx)
πfπ* (CNx)
39.8
a EVER is the energy of the vertical transition in ×103 cm-1, f is
the oscillator strength, and ψo and ψv are the occupied and the
virtual orbitals that define the transition. The transition type is
determined on the basis of the change in the spatial distribution
from occupied to virtual orbital. The absolute value of the
transition coefficient for each transition is given in parentheses.
H ) HOMO and L ) LUMO. (See text for calculation details.)
in Tables 6 and 7, respectively. For each excited state
the orbitals involved in the major transition are given
followed by the transition coefficient in parentheses. The
transition coefficients are the absolute values of the
wave function coefficients for each excitation35 and are
directly proportional to the contribution of the given
excitation to the excited state. The singlet excited states
of [Re(CNx)6]+ are presented in Figure 3 as vertical bars
A number of triplet excited states were computed
on the basis of the 3MLLCT state geometry for
[Re(CNx)5Cl] and [Re(CNx)6]+. The lowest-lying triplet
excited states are listed in Table 8, even if the f value
(35) Foresman, J. B.; Frisch, Æ. Exploring Chemistry with Electronic
Structure methods, 2nd ed.; Gaussian Inc.: Pittsburgh, PA, 1996; pp
206, 215.
(36) Frisch, Æ.; Frisch, M. J.; Trucks, G. W. Gaussian 03 User’s
Reference, version 7.0; Gaussian Inc.: Carnegie, PA, 2003; p 206.

[Re(CNx)5Cl] and [Re(CNx)6]+ Complexes
Organometallics, Vol. 24, No. 3, 2005 401
chloro ligand contributions, with the remaining electron
density located on the CNx ligand.
The HOMO of [Re(CNx)5Cl] (shown in Figure 6)
contained ∼53% Red, 36% CNx, and 11% Cl ligand
contributions, whereas the LUMO contained 94% CNx
ligand character. The HOMO of the hexakis complex
contained ∼53% Red and 47% CNx ligand contributions,
whereas the LUMO contained 93% CNx ligand and 6%
Rep contributions (Supporting Information Table S2).
The electrochemical oxidation involves the removal
of an electron from the HOMO. The difference between
the first oxidation potentials of the pentakis and hexakis
complexes of 0.67 V is in agreement with the calculated
energy difference between the HOMOs of the two
complexes of 0.75 eV. As the HOMOs are primarily
located on the Re, the quasi-reversible waves of the first
oxidation were assigned to the ReII/I couple.
Figure 5. Molecular orbital energy diagram for complexes
[Re(CNx)5Cl] (1) and [Re(CNx)6]+ (2) in ethanol.
Geometry Optimization. The geometries of
[Re(CNx)6]+ and [Re(CNx)5Cl] were optimized in the
singlet ground and in the lowest-lying triplet state. For
the hexakis complex the changes in the molecular
geometry in the triplet state relative to the ground state
were insignificant. The average Re-C (CNx) distance
was 2.05 Å in both the singlet ground and lowest-lying
triplet states.
is low. These states are used for the interpretation of
the temperature-dependent emission properties of the
complexes.
Discussion
X-ray Structure. As noted in the Results section, the
pentakis complex adopted a distorted octahedral geom-
etry. The angles of the trans ligands at the metal center
were in the range 173.7(2)-176.1(2)°. The Re-C28
distance (trans to Cl) was found to be ∼0.1 Å shorter
compared to the bond distances of the Re-C in the
equatorial plane. The Re-Cl bond length was 2.5278(16)
Å. The C-N-C angle of the CNx ligands in the
equatorial plane ranged from 163.6(6)° to 175.8(6)°. The
distortion of the angles could be attributed to steric
effects. However, the C-N-C angle of the CNx ligand
in the axial position (trans to Cl) was significantly
smaller than those in the equatorial position at 154.6(6)°.
This could be due to crystal packing effects.
For comparison, the Re-Cl distance of 2.515(2) Å in
[Re(CO)5Cl]37a was 0.01 Å shorter than the value for
[Re(CNx)5Cl]. Similar to our results, the Re-C (CO
trans to Cl) distance was ∼0.1 Å shorter than the Re-C
(CO cis to Cl) distance. However, the Re-C distances
(both cis and trans to Cl) in [Re(CO)5Cl] were shorter
by ∼0.02 Å relative to the corresponding Re-C distances
in [Re(CNx)5Cl].
Molecular Orbital Assignment. According to the
molecular orbital population analysis of the CNx ligand,
the HOMO-2 and LUMO+2 are located on the CtN
moiety. The HOMO-1, HOMO, and LUMO+1 are
located on the phenyl ring.
The molecular orbital population analysis of the
complexes indicated that the orbitals involved in forma-
tion of excited states of interest were located primarily
on the CNx ligand with the following exceptions: The
three highest in energy occupied molecular orbitals
(HOMO-2, HOMO-1, and HOMO) of the two com-
plexes contained more than 50% rhenium d orbital
character, with the remaining population located pri-
marily on the CNx ligand (Figure 5). For [Re(CNx)5Cl]
the HOMO-3 and HOMO-4 contained more than 44%
Selected parameters of the optimized geometry of [Re-
(CNx)5Cl] in the singlet ground state presented in Table
9 were in a very good agreement with the corresponding
X-ray values. The optimized Re-C (CNx) distances of
2.05 Å were 0.03 Å longer than the X-ray result for the
four equatorial CNx ligands. For the Re-C28 distance,
the geometry optimization produced 1.96 Å, which was
∼0.02-0.03 Å longer than the experimental result. The
calculated Re-C28 distance was ∼0.09 Å shorter than
the Re-C (equatorial ligands) compared to the experi-
mental difference of ∼0.1 Å. These results obtained with
one of the largest basis sets appear satisfactory despite
the shortcomings associated with the calculation of the
metal-ligand bond lengths using B3LYP theory.19 In
the triplet state the Re-C (CNx) distance did not change
significantly for the equatorial ligands, but for Re-C28
it increased by 0.03 Å. The Re-Cl distance decreased
from 2.61 Å in the ground state to 2.55 Å in the lowest-
lying triplet state. The geometry changes in the triplet
state can be interpreted using the frontier orbital spatial
distributions shown in Figure 6. The HOMO was Re-
Cl antibonding, and its single electron occupation
decreased the antibonding character, thus decreasing
the Re-Cl bond length in the triplet state. The bonding
and antibonding character was determined by visual
examination of the phases of the molecular orbital for
each diagram. The phases are related to the spatial
distributions of alpha (R) and beta (â) electron densi-
ties35 shown in red and blue, respectively. The elonga-
tion of the Re-C28 distances in the LUMO is a result of
the reduced electron density on the Re atom and on the
CNx ligand trans to Cl. The computed C28-N4-C29
angle was not significantly different from the corre-
sponding angles for the other ligands. Hence, the
smaller C28-N4-C29 angle reported in the crystal
structure (vide supra) was attributed to crystal packing.
CNx Ligand Excited States. A number of singlet
and triplet excited states of the CNx ligand were
computed on the basis of the singlet ground-state
(37) (a) Cotton, F. A.; Daniels, L. M. Acta Crystallogr., Sect. C 1983,
C39, 1495-1496. (b) Drago, R. S. Physical Methods for Chemist, 2nd
ed.; Saunders College Publishing: Orlando, FL; 1992; p 123.

402 Organometallics, Vol. 24, No. 3, 2005
Villegas et al.
Figure 6. Schematic diagrams of selected molecular orbitals of Re(CNx)5Cl (1) and [Re(CNx)6]+ (2) in ethanol.
Table 9. Selected Geometry Parameters of
[Re(CNx)5Cl] from X-ray Crystallography and
Calculated Singlet Ground-State and
were based on the CNx, ClfCNx, where the CNx
contributions of the occupied and virtual orbitals were
located on different CNx ligands. These states were
assigned as ligand-to-ligand charge transfer (LLCT).
Excited states 41 and 44 were based primarily on two
excitations that involved orbitals of similar spatial
distributions. Excited states 45 and 48 were based on
excitations that involved LLCT transitions and also Red,
CNxfCNx, Rep transitions. States 47 and 49 were
assigned as CNx ligand πfπ* states. The energies of
the MLLCT excited states 7 (26 600 cm-1) and 16
(31 900 cm-1) were 600 cm-1 higher and 300 cm-1 lower
than the respective UV-vis peaks as shown in Table 3.
For the hexakis complex, [Re(CNx)6]+, all singlet
excited states were assigned as Red, CNxfCNx or
MLLCT states and originated from excitations from
HOMO, HOMO-1, and HOMO-2 to virtual orbitals.
Several of the excited states were highly mixed. In
Figure 3 the singlet excited states of the hexakis
complex are presented as vertical bars with height equal
to the extinction coefficient.19a Excited states 13 and 15
at 32 600 and 33 400 cm-1, respectively, were close in
energy to the experimental peak at 33 300 cm-1. Excited
states 13 and 15 had the highest ꢀ (Supporting Informa-
tion Table S2). The results from the simulation of the
excited states to Gaussian curves and integration,
following a procedure previously reported19a (not shown),
did not correlate well with the UV-vis spectrum
curvature. The use of the TDDFT/CPCM method is
known to produce optical energies in good agreement
with the experimental absorption spectra. However, the
ꢀ values produced from this method are found to deviate
from the experiment.19d
Lowest-Lying Triplet-State Geometries
Re-C1, Re-C28
,
C1-N1,
Re-Cl,
C28-N4-C29,
source
Å
Å
Å
Å
deg
X-ray 2.019(7) 1.937(6) 1.185(7) 2.5278(16)
154.6(6)
175
179
singlet 2.05
triplet 2.05
1.96
1.99
1.18
1.18
2.61
2.55
geometry (Table 6) and correlated with the absorption
spectrum peaks (Table 3). In Table 6, singlet excited
state 1 (S1) at 39 400 cm-1 resulted from the HOMO-1
to LUMO excitation, S2 at 42 900 cm-1 from the HOMO
to LUMO excitation, and S4 at 50 200 cm-1 from the
HOMO to LUMO+1 excitation. All of the above singlet
excited states were phenyl πfπ* states. (State S3 had
f lower than 0.01.) States S2 and S4 were close in energy
to the experimental absorption peaks at 42 000 and
48 500 cm-1, respectively, shown in Table 3. In the
experimental absorption spectrum there were peaks
with very low molar absorptivity coefficients (ꢀ) at
∼35 000 cm-1. These were correlated with the triplet
excited states shown in Table 6. Triplet excited state 1
(T1) resulted from the HOMO to LUMO transition, T2
from HOMO-1 to LUMO, and T3 from HOMO-1 to
LUMO+1. These three triplet excited states were also
phenyl πfπ* states. State T3 was 500 cm-1 higher in
energy than the experimental absorption peak at 35 200
cm-1 (Table 3). These assignments are consistent with
multiplicity forbidden excited-state transitions having
low absorption coefficients.
Singlet Excited States of the Complexes. The
singlet excited states of [Re(CNx)5Cl] with f > 0.05 are
listed in Table 7. The lowest-lying 35 singlet excited
states were assigned as Red, CNxfCNx or MLLCT
states. These were associated with excitations from the
HOMO, HOMO-1, and HOMO-2 (Figure 5). For ex-
cited states 7, 8, 14, 15, and 16, there were two
excitations per state as shown in Table 7 that were
similar in type and provided almost equal contributions.
The majority of the higher energy states (S38 and higher)
Triplet Excited States of the Complexes. The
lowest-lying triplet excited states of [Re(CNx)5Cl] and
3
[Re(CNx)6]+ complexes were MLLCT states. The cal-
3
culated energy (20 400 cm-1) of the MLLCT state for
[Re(CNx)5Cl] was only 200 cm-1 higher than the ex-
perimental emission energy at 77 K (Table 5). The
geometries of the 3MLLCT states of the two complexes
were optimized and a number of triplet excited states

[Re(CNx)5Cl] and [Re(CNx)6]+ Complexes
Organometallics, Vol. 24, No. 3, 2005 403
Figure 7. Triplet excited-state energy diagram for com-
plexes Re(CNx)5Cl (1) and [Re(CNx)6]+ (2) in ethanol.
Figure 8. Schematic diagram of the photophysical pro-
2+ 40
cesses of Ru(bpy)3
.
were computed using the TDDFT/CPCM method (Table
3
8 and Figure 7). Two d-d states of [Re(CNx)6]+ were
The emission behavior of the two complexes was
different from that of other Re(I) complexes. For ex-
ample, the [Re(CO)3(L-L)X] system, where L-L is a
diimine ligand and X is the ancillary ligand, emitted at
both room temperature and 77 K. The lowest excited
state in these systems usually has been assigned as the
3MLCT state.38b
3
located 5600 and 6300 cm-1 above the MLLCT state.
The excited state at 30 500 cm-1 was mixed with large
3LLCT and triplet ligand-metal-to-ligand charge trans-
fer (3LMLCT) contributions. The 3d-d state in the
pentakis complex at 23 400 cm-1 was 3000 cm-1 above
the 3MLCT states, followed by a 3LLCT state at 24 400
3
3
cm-1. Additional d-d and LLCT states were located
at 25 200 and 26 000 cm-1, respectively. Above these
states were additional 3LLCT states (not shown) for the
two complexes.
The loss of room-temperature emission for the pen-
takis complex could be readily explained since the
presence of chloride in the coordination sphere provides
a mechanism for nonradiative decay as a result of the
heavy atom effect. An alternative explanation could
attribute the loss of emission via thermal population of
The 3d-d and 3LLCT states were obtained via single-
electron vertical excitations from the 3MLCT states, and
the excited-state energies reported were not the minima.
3
3
the d-d state from the MLLCT state. The TDDFT
3
The d-d transitions are symmetry forbidden37b and
3
calculations for the d-d states presented in Figure 7
have very low oscillator strengths (Table 8). The role of
were determined relative to the 3MLLCT state and did
not account for spin-orbit coupling. For third-row
transition metal complexes the treatment with spin-
orbit coupling could lower the predicted triplet-state
energies through interactions with higher singlet and
triplet states by 1700-2500 cm-1, making the 3d-d
states thermally more accessible.39 Because the energy
differences between the states given in Table 8 are for
vertical transitions, the actual thermal activation ener-
gies could be lower. Unfortunately, there is no reliable
method available yet for structural optimization of
higher-lying triplet excited-state geometries to obtain
more accurate values.
3
the d-d states in the lack of room-temperature emis-
sion is discussed in the next section.
Emission Properties and Excited-State Life-
times. The emission spectrum of the CNx ligand was
obtained both at room temperature and at 77 K (Figure
4A). The emission maximum was slightly blue-shifted
when the sample was cooled to 77 K from room tem-
perature. The emission peak shifted from ∼33 600 cm-1
at room temperature to 34 000 cm-1 at 77 K. This
1
indicates that the π*fπ emission is not very temper-
ature dependent. A broad phosphorescence emission
band was observed for the ligand at 77 K from 21 700
to 26 200 cm-1. The unrestricted B3LYP calculated
lowest-lying triplet state of CNx at 26 400 cm-1 is a
phenyl πfπ* state. The energy of this state was only
200 cm-1 higher than the highest-energy experimental
phosphorescence emission peak (Figure 4 and Table 5).
The observation of phosphorescence emission at rigid
glass was not unusual since in fluid solution the species
is susceptible to quenching collisions with adventitious
impurities, while at 77 K the diffusion of quenchers is
strongly inhibited.38a
A general scheme used for the observed excited-state
dynamic behavior of the standard complex [Ru(bpy)3]2+
is presented in Figure 8. The excitation to a higher
energy state upon absorption of light is followed by the
relaxation to the ground state via competing mecha-
nisms. In this model, the 3MLCT is populated with unit
3
quantum efficiency (f ) 1). Once formed, MLCT un-
dergoes three decay processes: radiative decay (kr),
nonradiative decay (knr), and thermal population of a
low-lying excited state, which undergoes efficient ra-
diationless decay and results in a low-yield photosub-
(38) (a) Barltrop, J. A.; Coyle, J. D. Excited States in Organic
Chemistry; John Wiley & Sons: Bristol, UK, 1975; p 81. (b) Juris, A.;
Campagna, S.; Bidd, I.; Lehn, J.-M.; Ziessel, R. Inorg. Chem. 1988,
27, 4007-4011.
(39) Makedonas, C.; Mitsopoulou, C. A.; Lahoz, F. J.; Balana, A. I.
Inorg. Chem. 2003, 42, 8853-8865.

404 Organometallics, Vol. 24, No. 3, 2005
Villegas et al.
stitution reaction.40 This model can also be applied to
complexes correlated quite well with the results from
TDDFT/CPCM calculations; (2) the nonradiative decay
can be attributed to the thermal population of the 3d-d
state from the 3MLCT, resulting in radiationless decay
to the ground state; (3) at low temperature, the emitting
states for most complexes were found to be MLLCT,
as indicated by their broad, structureless bands and (4)
assignment of the first oxidation to the Re center.
n-5
the excited-state behavior of the [Re(CNx)n(Cl)6-n
]
system where n ) 5 or 6.
The presence of the CNx ligand resulted in an
enhancement of nonradiative decay pathway(s) at room
temperature compared to rhenium tricarbonyl diimine
complexes. One possible explanation is thermal activa-
tion from the 3MLCT to the 3d-d state (∆E in Figure 8)
is greatly enhanced, facilitating radiationless decay to
the ground state.41
3
Acknowledgment. We thank the support of
National Science Foundation under Grant No. EIA-
0216178 and Grant No. EPS-0236913, matching support
from the State of Kansas and the Wichita State Uni-
versity High Performance Computing Center, the Wichita
State University Office of Research Administration, the
Department of Energy, and Parker Fellowships (J.M.V.
and S.R.S.).
Conclusion
Complexes of Re(I) with the general formula
[Re(CNx)n(Cl)6-n]
n-5, where n ) 5 or 6, were synthesized
and characterized. The structure of the pentakis com-
plex was determined by X-ray crystallography, provid-
ing useful information for performing B3LYP calcula-
tions related to the singlet ground-state geometry
optimization. The geometry of the emitting 3MLLCT
Supporting Information Available: Crystallographic
data for [Re(CNx)5Cl] in CIF format, the optimized geometries
(Table S1), the percent orbital contributions (Table S2), and
the calculated singlet excited-state energies of [Re(CNx)6]+
(Table S3) in text format. This material is available free of
charge via the Internet at http://pubs.acs.org.
3
state was also optimized using B3LYP, and the d-d
states were located above the emitting state.
The calculations account for the following: (1) The
absorption spectra of the CNx ligand and the two
(40) Ollino, M.; Cherry, W. R. Inorg. Chem. 1985, 24, 1417.
(41) Caspar, J. V.; Meyer, T. J. Inorg. Chem. 1983, 22, 2444.
OM049443S