Photoactivity of mono- and dicarbonyl complexes of ruthenium(II) bearing an N,N,S-donor ligand: Role of ancillary ligands on the capacity of CO photorelease

Article abstract of DOI:10.1021/ic4016004

One monocarbonyl and one dicarbonyl complex of ruthenium(II), namely, [Ru(Cl)(CO)(qmtpm)(PPh₃)]BF₄ (2) and [Ru(Cl)(CO) ₂(qmtpm)]ClO₄ (3), derived from the tridentate ligand 2-quinoline-N-(2′-methylthiophenyl)methyleneimine (qmtpm) have been synthesized and structurally characterized. The qmtpm ligand binds in a meridional fashion in these carbonyl complexes, and in 3, the two carbon monoxide (CO) ligands are cis to each other. Solutions of 2 in ethanol, chloroform, or acetonitrile rapidly release CO upon illumination with low-power (3-15 mW) light in the 300-450 nm range. Loss of CO from 2 brings about a dramatic color change from yellow to magenta because of the formation of [Ru(Cl)(MeCN)(qmtpm)(PPh₃)]BF₄ (4). In acetonitrile, photorelease of CO from 3 under 360 nm light occurs in two steps, and the violet photoproduct [Ru(Cl)(MeCN)₂(qmtpm)]⁺ upon reaction with Ag⁺ and PPh₃ affords red [Ru(MeCN)₂(qmtpm) (PPh₃)](ClO₄)₂ (5). The structure of 5 has also been determined by X-ray crystallography. Reduced myoglobin assay confirms that 2 and 3 act as photoactive CO-releasing molecules (photoCORMs) that deliver 1 and 2 equiv of CO, respectively. The results of density functional theory (DFT) and time-dependent DFT studies confirm that electronic transitions from molecular orbitals with predominantly Ru-CO character to ligand-based π* orbitals facilitate CO release from these two photoCORMs. Complexes 2-5 have provided an additional opportunity to analyze the roles of the ancillary ligands, namely, PPh₃, Cl^-, and MeCN, in shifting the positions of the metal-to-ligand charge-transfer bands and the associated sensitivity of the two photoCORMs to different wavelengths of light. Collectively, the results provide helpful hints toward the future design of photoCORMs that release CO upon exposure to visible light.

Full text of DOI:10.1021/ic4016004

Article
pubs.acs.org/IC
Photoactivity of Mono- and Dicarbonyl Complexes of Ruthenium(II)
Bearing an N,N,S-Donor Ligand: Role of Ancillary Ligands on the
Capacity of CO Photorelease
Margarita A. Gonzalez,† Samantha J. Carrington,† Indranil Chakraborty,† Marilyn M. Olmstead,‡
and Pradip K. Mascharak*,†
†Department of Chemistry and Biochemistry, University of California, Santa Cruz, California 95064, United States
‡Department of Chemistry, University of California, Davis, California 95616, United States
S
* Supporting Information
ABSTRACT: One monocarbonyl and one dicarbonyl complex of
ruthenium(II), namely, [Ru(Cl)(CO)(qmtpm)(PPh3)]BF4 (2)
and [Ru(Cl)(CO)2(qmtpm)]ClO4 (3), derived from the
tridentate ligand 2-quinoline-N-(2′-methylthiophenyl)-
methyleneimine (qmtpm) have been synthesized and structurally
characterized. The qmtpm ligand binds in a meridional fashion in
these carbonyl complexes, and in 3, the two carbon monoxide
(CO) ligands are cis to each other. Solutions of 2 in ethanol,
chloroform, or acetonitrile rapidly release CO upon illumination
with low-power (3−15 mW) light in the 300−450 nm range. Loss
of CO from 2 brings about a dramatic color change from yellow
to magenta because of the formation of [Ru(Cl)(MeCN)-
(qmtpm)(PPh3)]BF4 (4). In acetonitrile, photorelease of CO
from 3 under 360 nm light occurs in two steps, and the violet photoproduct [Ru(Cl)(MeCN)2(qmtpm)]+ upon reaction with
Ag+ and PPh3 affords red [Ru(MeCN)2(qmtpm)(PPh3)](ClO4)2 (5). The structure of 5 has also been determined by X-ray
crystallography. Reduced myoglobin assay confirms that 2 and 3 act as photoactive CO-releasing molecules (photoCORMs) that
deliver 1 and 2 equiv of CO, respectively. The results of density functional theory (DFT) and time-dependent DFT studies
confirm that electronic transitions from molecular orbitals with predominantly Ru−CO character to ligand-based π* orbitals
facilitate CO release from these two photoCORMs. Complexes 2−5 have provided an additional opportunity to analyze the roles
of the ancillary ligands, namely, PPh3, Cl−, and MeCN, in shifting the positions of the metal-to-ligand charge-transfer bands and
the associated sensitivity of the two photoCORMs to different wavelengths of light. Collectively, the results provide helpful hints
toward the future design of photoCORMs that release CO upon exposure to visible light.
upregulation has been noted in cases of ischemia/reperfusion
injury,7 aid of organ graft survival,8 modulation of inflammatory
conditions such as lung injury,9 and myocardial infarction.10
While exogenous CO gas application has shown promise in
animal models of transplantation,11 site-specific delivery of this
toxic gas is imperative to effectively induce the benefits of CO-
mediated protection and to circumvent toxicity associated with
asphyxiation. In recent years, CO complexes bearing low-valent
metal centers have been studied as photoactive CO-releasing
molecules (photoCORMs) because of the inherent photo-
sensitivity of metal carbonyl complexes.12,13 The initial
photoCORMs reported underwent UV-light-triggered CO
release because metal−CO bonds typically require high
energies for dissociation owing to strong back-bonding
interactions. For example, near-UV photolysis of the water-
soluble salt Na3[W(CO)5(TPPTS)] [TPPTS = tris-
INTRODUCTION
■
Carbon monoxide (CO) is endogenously produced during the
breakdown of heme proteins by either inducible or constitutive
forms of the enzyme heme oxygenase (HO).1 Although the
details of HO action has been known for some time, CO was
initially dismissed as a metabolic waste product. Focus on the
anti-inflammatory properties of the bile pigments biliverdin and
bilirubin2 (generated in the later stages of heme catabolism)
has, however, led to important insights regarding HO activity in
anti-inflammatory responses. It was not until the past decade
that the role of endogenously produced CO was reassessed and
implicated in physiological processes such as vasodilation,3
antiapoptotic activity,4 and cell signaling.5 CO mediates cell
protection by interacting with heme-containing metalloproteins
in stress response pathways, thus earning its reputation as a
cytoprotective agent. For instance, inhibition of mitochondrial
cytochrome c by both exogenously and endogenously produced
CO has been found to precondition neuronal cells against
apoptosis.6 Indeed, increased CO levels associated with HO
Received: June 24, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(sulfonatophenyl)phosphine trianion] resulted in 1.2−1.6 mol
of CO per molecule of CORM released.14 The presence of CO
groups in the autoxidation product suggests that not all metal−
CO bonds are labilized. Similarly, UV light was required for CO
delivery from the iron carbonyl complex [Fe(CO)(N4Py)]-
visible region.20 The results of density functional theory (DFT)
studies reveal that inclusion of an −SMe group on the qmtpm
frame and a σ-donating Br− ligand leads to enhanced CO
photolability by promoting MLCT interactions that weaken the
affinity of the metal center toward CO in these photoCORMs.
Although photoCORMs like fac-[Mn(pqa)(CO)3]ClO4 are
quite stable in biologically relevant media, complexes like fac-
[Mn(Br)(qmtpm)(CO)3] exhibit lower stability because of the
bidentate mode of coordination of the qmtpm ligand (Figure
2). The rigidity arising from inclusion of the imine functionality
in the ligand frame does not allow the thioether group to
participate in coordination in the fac-manganese(I) tricarbonyl
complexes derived from qmtpm. As part of our effort to
establish ligand design strategies for the isolation of photo-
CORMs, we now report the syntheses, structures, and
photochemical properties of the two stable ruthenium(II)
carbonyl complexes derived from qmtpm in which the ligand
coordinates in a tridentate fashion. Carbonylation of [Ru-
(Cl)2(qmtpm)(PPh3)] (1) in ethanol in the presence of Ag+
ion yields the monocarbonyl complex [Ru(Cl)(CO)(qmtpm)-
(PPh3)]BF4 (2), while reaction of [Ru(Cl)2(CO)3)]2 with
qmptm in boiling methanol affords the dicarbonyl complex
[Ru(Cl)(CO)2(qmtpm)]ClO4 (3) in high yield. In both of
these complexes, the qmtpm ligand binds in a meridional
fashion, and in 3, the two CO ligands are cis to each other.
Complete photolysis of 2 in coordinating solvents such as
acetonitrile yields one CO per molecule and forms the solvato
species [Ru(Cl)(MeCN)(qmtpm)(PPh3)]BF4 (4), which upon
reaction with AgClO4 in acetonitrile affords [Ru-
(MeCN)2(qmtpm)(PPh3)](ClO4)2 (5). The structures of 2,
3, and 5 are reported in this account. In addition, DFT and
time-dependent DFT (TDDFT) calculations have been
performed to examine the role of MLCT transitions in
metal−CO bond labilization in 2 and 3. The results described
below strongly suggest that electronic transitions from
molecular orbitals (MOs) with predominantly Ru−CO
character to ligand-based π* orbitals facilitate CO release
from these two photoCORMs.
(ClO4)2 [N4Py = pentadentate tetrakis(pyridine) ligand].15
A
particular subset of photoCORMs consists of fac-manganese(I)
tricarbonyl complexes, wherein a low-valent manganese center
is stabilized by a tripodal ligand and three CO groups.
Schatzschneider and co-workers reported [Mn(CO)3(tpm)]-
PF6, the first in this series,16 and have since expanded their set
of tripodal ligands to include imidazole moieties and their
derivatives.17 For this subgroup of photoCORMs, exhaustive
UV photolysis affords three CO ligands per molecule and a
bridged μ-OMnIII compound as the final photoproduct.18
We have been interested in the nature of electronic
transition(s) in metal carbonyl complexes that promotes CO
photolability in the visible and near-infrared (NIR) region. As
part of such effort, we have recently demonstrated that, for
manganese(I) tricarbonyl complexes of the composition fac-
[Mn(L)(CO)3] (L = tridentate ligands), increased conjugation
in the ligand frame L results in a systematic increase in the
absorptivity of the corresponding complexes in the longer-
wavelength region.19 For example, the carbonyl complex fac-
[Mn(pqa)(CO)3]ClO4 [where pqa = (2-pyridylmethyl)(2-
quinolylmethyl)amine], which features both pyridine and
quinoline moieties (Figure 2) in the ligand frame, exhibits a
strong absorption band with λmax at 360 nm, while fac-
[Mn(dpa)(CO)3]ClO4 [where dpa = N,N-bis(2-
pyridylmethyl)amine, the same ligand frame with two pyridine
moieties] displays a weaker band at 350 nm. In another study,
we have utilized the ligand 2-quinoline-N-(2′-
methylthiophenyl)methyleneimine (qmtpm; Figure 1) to
Figure 1. Structures of the ligands.
EXPERIMENTAL SECTION
■
Materials and Reagents. [Ru(Cl)2(CO)3]2 was purchased from
Sigma Aldrich and used as received. The other starting salt
[Ru(Cl)2(PPh3)3]21 and the ligand 2-quinoline-N-(2′-
methylthiophenyl)methyleneimine (qmtpm)22 were synthesized fol-
lowing literature procedures. Solvents were purified and/or dried by
standard techniques prior to use.
Caution! Transition-metal perchlorates should be prepared in small
quantities and handled with great caution as metal perchlorates may
explode upon heating.
[Ru(Cl)2(qmtpm)(PPh3)] (1). A batch of 0.741 g (0.77 mmol) of
[Ru(Cl)2(PPh3)3] and 0.221 g (0.77 mmol) of qmtpm were added to
20 mL of degassed toluene, and the red-brown slurry was heated to
reflux for 4 h. The solution turned blue, and a deep-blue solid
separated from the solution. The solid was filtered and washed several
times with diethyl ether. After thorough drying, this procedure
afforded 1 in analytically pure form (also confirmed by its IR, NMR,
and electronic absorption spectra (EAS) spectra). Yield: 510 mg
(93%). Anal. Calcd for C35H29N2RuCl2SP: C, 58.98; H, 4.10; N, 3.93.
Found: C, 58.83; H, 4.20; N, 3.91. Selected IR frequencies (KBr disk,
cm−1): 1615 (w), 1593 (w), 1519 (m), 1481 (m), 1434 (m), 1090
investigate the influence of both a thioether moiety and an
imine functionality on the absorption characteristics of the
resulting fac-[Mn(L)(CO)3] complexes. The low affinity of the
MnI center toward thioether donors, however, resulted in the
bidentate mode of coordination of qmtpm to the MnI center in
the resulting carbonyl complexes. Nevertheless, the two
carbonyl complexes fac-[Mn(qmtpm)(MeCN)(CO)3]ClO4
(Figure 2) and fac-[Mn(Br)(qmtpm)(CO)3] exhibit strong
metal-to-ligand charge-transfer (MLCT) bands at 435 and 535
nm, respectively, and photorelease CO upon illumination in the
(m), 827 (w), 787 (w), 753 (m), 697 (s), 525 (s). EAS [CH2Cl2; λmax
,
nm (ε, M−1 cm−1)]: 280 (19226), 355 (17630), 615 (4235). 1H NMR
(CDCl3, 500 MHz, TMS): δ 10.50 (d, 1H), 9.37 (s, 1H), 7.80 (t, 1H),
7.61 (t, 1H), 7.51 (d, 1H), 7.44 (t, 3H), 7.15−7.24 (m, 8H), 7.02 (t,
3H), 6.88 (t, 6H), 6.54 (t, 1H), 3.11 (s, 3H).
Figure 2. Structures of the MnI photoCORMs.
B
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Summary of Crystal Data, Intensity Collection, and Refinement Parameters for 2·H2O, 3, and 5
2
3
5
empirical formula
fw
C36H31BClF4N2O2PRuS
C19H14Cl2N2O6RuS
570.35
C39H35Cl2N4O8PRuS
922.71
809.99
cryst color
cryst size (mm3)
temperature (K)
wavelength (Å)
cryst syst
space group
a (Å)
brown plates
yellow plates
0.40 × 0.30 × 0.08
296(2)
red blocks
0.44 × 0.26 × 0.07
90(2)
0.12 × 0.11 × 0.04
296(2)
0.71073
0.71073
0.71073
orthorhombic
monoclinic
P21/c
triclinic
P212121
P1
̅
9.824(2)
13.8529(6)
9.3341(4)
16.2461(7)
90.00
11.4409(18)
11.936(2)
16.993(3)
79.84(2)
84.16(2)
70.16(2)
2146.4(6)
2
b (Å)
15.179(3)
c (Å)
23.333(6)
α (deg)
90.00
β (deg)
90.00
95.6850(10)
90.00
γ (deg)
V (Å3)
90.00
3479.4(13)
2090.36(16)
4
Z
4
dcalcd (g/cm3)
1.546
1.812
1.428
μ (mm−1
)
0.691
1.146
0.629
a
GOF on F2
1.149
1.041
1.114
b
final R indices [I > 2σ(I)]
R1 = 0.0291, wR2 = 0.0689
R1 = 0.0321, wR2 = 0.0711
R1 = 0.0326, wR2 = 0.0697
R1 = 0.0641, wR2 = 0.1889
c
R indices (all data)
R1 = 0.0510, wR2 = 0.0779
R1 = 0.0940, wR2 = 0.2072
a
b
c
GOF = [∑w(|Fo|2 − |Fc1|2/)22/(No − Nv)]1/2 (No = number of observations; Nv = number of variables). R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = [(∑w(|
Fo|2 − |Fc|2)2/∑w|Fo|2)]
.
[Ru(Cl)(CO)(qmtpm)(PPh3)]BF4 (2). A slurry of 0.122 g (0.17
mmol) of 1 in 15 mL of ethanol was treated with 1.1 equiv of AgBF4
(0.037g, 0.19 mmol) to obtain a magenta solution, which was degassed
by a freeze−pump−thaw technique. The reaction mixture was allowed
to stir for 12 h at room temperature and subsequently heated to reflux
under nitrogen for 1 h. The resulting purple reaction mixture was
filtered through a Celite pad. A steady stream of CO gas was bubbled
through the filtrate for a period of 15 min, during which the color
changed from purple to brown and a solid precipitated from the
solution. The brown solid was collected by filtration and washed
several times with diethyl ether. Yield: 80 mg (59%). Anal. Calcd for
C36H29N2RuClBF4OSP: C, 53.51; H, 3.62; N, 3.47. Found: C, 53.57;
H, 3.60; N, 3.39. Selected IR frequencies (KBr disk, cm−1): 2028 (vs,
[Ru(MeCN)2(qmtpm)(PPh3)](ClO4)2 (5). To a batch of 0.051 g (0.71
mmol) of 1 in 8 mL of acetonitrile was added 2 equiv of AgClO4
(0.031 g, 0.150 mmol) dissolved in 7 mL of acetonitrile. The resulting
magenta solution was degassed and stirred at room temperature for 12
h, during which a deep-red color developed. The solution was filtered
through a Celite pad, and the filtrate was allowed to evaporate slowly
to obtain red blocks of 5 suitable for X-ray diffraction. Yield: 40 mg
(61%). Anal. Calcd for C39H35N4RuCl2O8SP: C, 50.76; H, 3.82; N,
6.07. Found: C, 50.72; H, 3.91; N, 6.06. Selected IR frequencies (KBr
disk, cm−1): 1618 (w), 1592 (w), 1435 (w), 1094 (s, νClO ), 757 (w),
4
698 (w), 623 (m), 530 (m). EAS [CH3CN; λmax, nm (ε, M−1 cm−1)]:
270 (17125), 310 (14240), 368 (15500), 495 (5400). 1H NMR
(CD3CN, 500 MHz, TMS): δ 9.56 (s, 1H), 8.60 (d, 1H), 8.55 (d,
1H), 8.13 (d, 1H), 8.01 (d, 1H), 7.95 (t, 1H), 7.85 (m, 2H), 7.61 (d,
1H), 7.47−7.52 (m, 2H), 7.22 (m, 3H), 7.06 (m, 15H), 2.90 (s, 3H),
2.69 (s, 3H).
νCO), 1637 (w), 1516 (w), 1480 (w), 1436 (m), 1084 (s, νBF ), 839
4
(w), 759 (m), 748 (m), 698 (m), 520 (m). EAS [CH2Cl2; λmax, nm (ε,
M−1 cm−1)]: 292 (15900), 370 (14350), 465 (5310). 1H NMR
(CDCl3, 500 MHz, TMS): δ 9.92 (d, 1H), 9.61 (s, 1H), 8.52 (d, 1H),
8.34 (d, 1H), 8.14 (t, 1H) 7.96 (t, 1H), 7.80 (m, 2H), 7.47 (s, 1H),
7.28 (t, 7H), 7.20 (t, 3H), 7.04 (t, 5H), 3.08 (s, 3H).
1
Physical Measurements. H NMR spectra were recorded at 298
K on a Varian Unity Inova 500 MHz instrument. A Perkin-Elmer
Spectrum One Fourier transform infrared sepctrometer was utilized to
monitor the IR spectra of the compounds. Electronic absorption
spectrometry (EAS) spectra were obtained on a Varian Cary 50
spectrophotometer. Room temperature magnetic susceptibility meas-
urements were performed with a Johnson Matthey magnetic
susceptibility balance.
Photolysis Experiments. For continuous-wave photolysis experi-
ments, the light source was an IL-410 illuminator (Electro FiberOptics
Corp.) equipped with various light filters. The power of the incident
light was measured with a Field MaxII-TO laser power meter (from
Coherent, Portland, OR). The rates of CO photorelease were
measured by recording the EAS spectra of 2 and 3 in acetonitrile
and ethanol (both 0.25 mM, 1.4 mL) and monitoring either the loss of
absorbance at 460 nm (for 2) or the rise in absorbance of the 565-nm
charge-transfer band (for 3) of samples exposed to light at definite
time intervals. The plots were fitted to the three-parameter exponential
equation A(t) = A∞ + (Ao − A∞) exp(−kCOt), where Ao and A∞ are
the initial and final absorbance values, respectively. The apparent rates
of CO loss (kCO) were calculated from the ln(C) versus time (T) plot
for each compound.
[Ru(Cl)(CO)2(qmtpm)]ClO4 (3). A batch of 0.167 g (0.326 mmol) of
[Ru(Cl)2(CO)3]2 was heated to reflux in 7 mL of methanol to
generate a pale-yellow solution. Separately, a slurry of 0.183 g (0.656
mmol) of the qmtpm ligand in 10 mL of methanol was added to a hot
solution of [Ru(Cl)2(CO)3]2. The resulting red-orange solution was
heated to reflux under a N2 atmosphere for 5 h. Next the volume was
reduced to 5 mL under reduced pressure, and the orange solid that
separated from the solution was filtered. To the filtrate was added 1
equiv of NaClO4 (0.043 g, 0.348 mmol), and the mixture was stirred at
room temperature for 30 min, during which complex 3 precipitated
from the solution as an orange-yellow solid. Yield: 112 mg (60%).
Anal. Calcd for C19H14N2RuCl2O6S: C, 40.01; H, 2.47; N, 4.91.
Found: C, 40.10; H, 2.48; N, 4.88. Selected IR frequencies (KBr disk,
cm−1): 2086 (s, νCO), 2038 (s, νCO), 1634 (w), 1514 (w), 1434 (w),
1296 (w), 1121 (m), 1083 (s, νClO ), 825 (w), 759 (w), 623 (w), 505
4
(w). EAS [acetonitrile; λmax, nm (ε, M−1 cm−1)]: 252 (25350), 275
(25700), 385 (25100), 405 (sh, 20770). 1H NMR (CD3CN, 500
MHz, TMS): δ 9.76 (d, 1H), 8.92 (d, 1H), 8.42−8.47 (q, 1H), 8.40 (t,
1H), 8.26 (d, 2H), 8.22 (t, 1H), 7.96−8.01 (m, 2H), 7.82 (t, 1H), 3.13
(s, 2H), 3.02 (s, 1H).
The quantum yield (ϕ) values of CO release were determined by
using a Newport Oriel Apex Illuminator (150 W xenon lamp)
C
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
equipped with an Oriel 1/8m Cornerstone monochromator. Standard
ferrioxalate actinometry23 was used to determine the quantum yields at
436 nm (for 2) and 392 nm (for 3). Solutions of 2 and 3 were
prepared under dim-light conditions and placed in a 2 mm × 10 mm
quartz cuvette, 2 cm away from the light source. Solutions were
prepared to ensure sufficient absorbance (>90%) at the irradiation
wavelength, and changes in the EAS spectra in the 460 and 565 nm
regions (<10% photolysis) were used to determine the extent of CO
release.
Table 2. Selected Bond Distances (Å) and Angles (deg) for
2, 3, and 5
2
3
5
Ru−N1
Ru−N2
Ru−S1
Ru−C1
Ru−C2
Ru−P1
Ru−Cl1
Ru−N4
Ru−N3
C1−O1
C2−O2
2.1633(18)
2.0190(19)
2.3166(6)
1.923(2)
2.1305(19)
2.060(2)
2.3357(7)
1.903(3)
1.890(3)
2.155(5)
2.017(2)
2.3137(18)
Myoglobin Assay. Horse heart myoglobin was dissolved in
phosphate-buffered saline (100 mM, pH 7.4) and reduced by adding
sodium dithionite. The concentration of the resulting deoxymyoglobin
(Mb) was calculated from the absorbance of the Soret band at 435 nm
(extinction coefficient = 121 mM−1 cm−1). Because sodium dithionite
is known to facilitate the release of CO,24 an apparatus was
constructed using two quartz cuvettes. In the first cuvette under
anaerobic conditions, the photoactive complexes were exposed to light
evolving CO into the headspace. The photogenerated CO was then
transferred into the second cuvette containing the reduced Mb
solution via a cannula by a positive pressure of N2(g), and the
absorbance was taken after 5 min to ensure complete capture of the
photoreleased CO. Conversion of Mb to carbonyl myoglobin (Mb-
CO) was monitored at defined time intervals. A shift in λmax from 435
to 424 nm was noted in each case because of the formation of Mb-CO.
Final concentrations of Mb-CO were assessed at 424 nm (extinction
coefficient = 207 mM−1 cm−1) and compared to the initial Mb present
in solution to quantify the amount of CO released by each compound.
X-ray Crystallography. X-ray-quality crystals of brown plates of 2
were grown by diffusion of pentane into a dilute solution of the
complex in dichloromethane. Yellow plates of 3 were obtained by
diffusion of toluene into an acetonitrile solution. Diffraction data were
collected on a Bruker APEX-II instrument using monochromated Mo
Kα radiation (λ = 0.71073 A). The data for 2 were collected at 90 K,
while for 3 and 5, the data collection temperature was 296 K. All data
were corrected for absorption, and the structures were solved by direct
methods using the SHELXTL (1995−1999) software package (Bruker
Analytical X-ray Systems Inc.). Additional refinement details are
contained in the CIF files (Supporting Information). Crystal data,
instrument, and data collection parameters are summarized in Table 1.
Selected bond distances and angles are listed in Table 2.
2.4719(7)
2.4352(6)
2.3485(19)
2.4176(7)
2.056(6)
2.087(6)
1.127(3)
1.137(3)
1.130(3)
C1−Ru−P1
Cl1−Ru−P1
Cl1−Ru−N2
N1−Ru−S1
C1−Ru−S1
C1−Ru−N2
C2−Ru−S1
C1−Ru−C2
N2−Ru−S1
N2−Ru−N1
Cl1−Ru−C2
Cl1−Ru−C1
Cl1−Ru−N1
Cl1−Ru−S1
P1−Ru−N3
N4−Ru−N2
P1−Ru−N2
P1−Ru−N1
P1−Ru−S1
N1−Ru−N4
C1−Ru−N1
N1−Ru−C2
N2−Ru−C2
N1−Ru−N3
N2−Ru−N3
N3−Ru−S1
N4−Ru−S1
N4−Ru−P1
N4−Ru−N3
Ru−C1−O1
Ru−C2−O2
175.70(6)
90.27(2)
176.11(5)
163.22(5)
92.31(7)
93.44(8)
86.21(6)
161.32(6)
93.45(8)
171.24(10)
95.99(8)
94.71(11)
83.96(6)
78.13(8)
179.13(8)
85.22(8)
88.64(6)
84.88(2)
162.87(15)
84.75(5)
78.48(7)
85.29(15)
77.9(2)
85.58(6)
105.25(5)
91.53(2)
176.21(16)
176.4(2)
91.92(15)
94.12(14)
89.89(7)
104.7(2)
90.63(5)
92.88(5)
86.67(2)
89.30(8)
103.47(10)
90.53(9)
DFT and TDDFT Calculations. DFT and TDDFT studies were
executed with the PC-GAMESS program25 using the hybrid functional
PBE0. DFT calculations were carried out using the double-ζ basis set
6-31G* for all atoms except ruthenium, for which the quasi-relativistic
93.89(10)
84.5(2)
Stuttgart−Dresden effective core potential was implemented. The X-
̈
91.2(2)
ray crystal structure coordinates of complexes 2, 3, and 5 were used as
a starting point for the gas-phase geometry optimization. Electronic
transition energies and oscillator strengths were then calculated for 2−
5 at their PBE0-optimized geometries using TDDFT. For each
compound, the 20 lowest-energy electronic excitations were calculated
and the solvent effect was added using the polarized continuum
model26 and ethanol as the solvent. The calculated MOs were
visualized with the aid of MacMolPlt.27
92.46(16)
91.89(17)
90.24(16)
86.7(2)
173.7(2)
173.4(32)
175.6(2)
In the present work, complex 1 was synthesized for its
potential use as the starting material for mono- and dicarbonyl
species. However, attempts to replace the Cl− ligands of 1 with
CO led to very different results depending on the solvent. For
example, in acetonitrile, heating with 1 equiv of AgBF4
generated the magenta complex 4 (Scheme 1). The further
addition of 1 equiv of AgBF4 afforded the red complex
[Ru(MeCN)2(qmtpm)(PPh3)]2+ (Scheme 1), which has been
structurally characterized as the perchlorate salt 5. Bubbling of
CO gas to such solutions did not yield either the monocarbonyl
complex 2 or [Ru(CO)2(qmtpm)(PPh3)]2+, the expected
dicarbonyl species. In contrast, the treatment of 1 with Ag+
salts in boiling ethanol followed by passage of a steady stream
of CO gas at room temperature led to the replacement of one
Cl−, as evidenced by formation of the brown monocarbonyl
SYNTHESES AND INTERCONVERSIONS
■
The reaction of [Ru(Cl)2(PPh3)3] with qmtpm in toluene at
refluxing temperature afforded 1 in good yield. The
composition of this complex was established with the aid of
1
microanalytical data and its H NMR spectrum in CDCl3. The
singlet observed at 3.11 ppm confirms S-ligation of qmtpm to
the RuII center in 1. In free qmtpm, the −SMe signal is noted at
2.52 ppm. Integration of the proton resonances indicates the
presence of one PPh3 ligand, while the other two coordination
positions are occupied by the two Cl− ligands. We have
tentatively assigned a cis disposition to the two Cl− ligands in 1
on the basis of the crystal structure of the related complex
[Ru(Cl)2(pmtpm)(PPh3)] [where pmtpm = 2-pyridyl-N-(2′-
methylthiophenyl)methyleneimine].28
D
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1
complex 2 (Scheme 1). It is important to note that even in
ethanol only one Cl− can be exchanged with CO. Collectively,
these reactions indicate that acetonitrile is a better donor for
the RuII center, and any replacement of the Cl− ligand of 1 with
CO is possible only in hot ethanol. The affinity of the RuII
center toward acetonitrile is further indicated by the
quantitative formation of 5 upon reaction of 2 with 1 equiv
of Ag+ salts in hot acetonitrile (Scheme 1).
Success in the isolation of the dicarbonyl complex cis-3 was
achieved when [Ru(Cl)2(CO)3]2 was employed as the starting
salt. In a boiling methanolic solution, the chloro-bridged
dimeric carbonyl was allowed to react with qmtpm to afford 3
(as a perchlorate salt) in good yield (eq 1). The reaction
presumably proceeds through the solvent-assisted bridge
cleavage of [Ru(Cl)2(CO)3]2,29 followed by coordination of
qmtpm to the RuII center. Interestingly, the reaction of the
dicarbonyl precursor cis-[Ru(Cl)2(CO)2(PPh3)2]21 with
qmtpm in solvents such as methanol, toluene, chloroform,
and acetonitrile did not afford 3.
triphenylphosphine is axially positioned trans to the CO ligand,
while Cl− occupies the remaining site, in the same plane as
qmtpm. Interestingly, this particular orientation of the ancillary
ligands (PPh3 and Cl) relative to CO in 2 has been noted in
another ruthenium monocarbonyl complex [Ru(Cl)(CO)-
(terpy)(PPh3)]PF6, which features the planar tridentate
2,2′:6′,2″-terpyridine (terpy) ligand.30 The Ru−P bond length
at 2.4719(7) Å in 2 is slightly longer than those noted in other
ruthenium(II) phosphinecarbonyl complexes such as [Ru(Cl)-
(CO)(PPh3)(H-pybox)]BF4 [where H-pybox = 2,6-bis-
(dihydrooxazolin-2′-yl)pyridine; Ru−P = 2.311(3) Å]31 and
[RuCl(CO)(NN′N)(PPh3)]OTf [where NN′N = 2,6-bis-
[(dimethylamino)methyl]pyridine; Ru−P = 2.2384(2) Å].32
This lengthening of the Ru−P bond is expected because, unlike
these latter carbonyl complexes, the PPh3 ligand is trans to CO
in 2. The short Ru−C(O) bond distance [1.923(2) Å] and the
almost linear Ru−C(O) bond angle are both typical of
ruthenium(II) phosphinecarbonyl complexes.30,31
3. The structure of the cation of 3 is shown in Figure 4, and
the selected structural parameters are listed in Table 2. Here
[Ru(Cl)2(CO)3]2 + qmtpm
(a) MeOH, heat
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ [Ru(Cl)(CO)2(qmtpm)]ClO (3)
4
(b) NaClO
(1)
4
Structures of the Complexes. 2. The structure of the
cation of 2 is shown in Figure 3, and selected metric parameters
are listed in Table 2. The ruthenium center in this carbonyl
complex resides in an octahedral environment, with the qmtpm
ligand coordinated in a meridional fashion. The bulky
Figure 4. Thermal ellipsoid (50% probability level) plot of
[Ru(Cl)(CO)2(qmtpm)]+ (cation of 3) with the atom-labeling
scheme. Hydrogen atoms have been omitted for the sake of clarity.
again the qmtpm ligand is bound to the RuII center in a
meridional fashion. The two bound CO molecules are cis to
each other, while one Cl− ligand occupies an axial site. The
Ru−Cl bond distances in 2 and 3 are similar [2.4352(6) Å vs
2.4176(7) Å] regardless of their coordination with respect to
the imine nitrogen atom of qmtpm (in 2) versus CO (in 3).
The Ru−C(O) bond distances in 3 are almost identical
[1.903(3) and 1.890(3) Å], suggesting a uniform extent of
back-bonding between the CO ligands and the metal center.
Such Ru−C(O) bond lengths and the linearity of the
corresponding bond angles (∠O2−C2−Ru1: 175.6°) have
been observed in other cis-ruthenium(II) dicarbonyl complexes
such as [Ru(CO)2(mnt-κ2SS′)(terpy-κ2NN′)] (where mnt-
Figure 3. Thermal ellipsoid (50% probability level) plot of
[Ru(Cl)(CO)(qmtpm)(PPh3)]+ (cation of 2) with the atom-labeling
scheme. Hydrogen atoms have been omitted for the sake of clarity.
E
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
κ2SS′ = maleonitriledithiolate and terpy-κ2NN′ = 2,2′:6′,2″-
length at 2.3485(19) Å (Table 2) falls in the range of known
ruthenium(II) triphenylphosphine complexes.31,32 The Ru−
Nacn (acn = acetonitrile) bond lengths of 4 [2.056(6) and
2.087(6) Å] are slightly longer than those observed in other
ruthenium bis(acetonitrile) complexes such as [Ru(Cl)-
(MeCN)2)(L)(PPh3)]Cl [where L = 2-(2-pyridyl)-
benzothiazole; Ru−Nacn = 1.97(13) and 2.01(12) Å].34
terpyridine).33
5. The structure of the cation of 5 (shown in Figure 5)
reveals meridional binding of the qmtpm ligand to the RuII
Spectroscopic Properties. All of the ruthenium complexes
reported herein are diamagnetic in the solid state and in
1
solution, as evidenced by their clean H NMR spectra. For
1
example, the H NMR spectrum of 2 in CDCl3 is shown in
Figure 6. Ligation of the thioether end of the qmtpm ligand to
the RuII center is confirmed by the downfield shift of the SCH3
signal from 2.52 ppm in the free ligand to 3.08 and 2.90 ppm in
2 and 5, respectively. Interestingly, 3 exhibits two resonances
for the SCH3 unit (3.02 and 3.14 ppm) presumably because of
two distinct orientations of this group with respect to the planar
qmtpm ligand in solution. In its Fourier transform infrared
spectrum, 2 exhibits one strong νCO band at 2028 cm−1, similar
to [Ru(Cl)(CO)(terpy)(PPh3)]PF6.30 As expected, the dicar-
bonyl complex 3 displays two strong νCO bands at 2086 and
2038 cm−1.31,33
The EAS spectra of 2, 4, and 5 all exhibit one strong MLCT
band in the visible region (Figure 7). These signature spectra
allow one to readily identify the products of interconversions
outlined in Scheme 1. For example, when 2 is heated in
acetonitrile in the presence of AgClO4, formation of 5 is readily
evidenced by a shift of the absorption maximum of the 460 nm
band to 495 nm. Similarly, loss of CO from 2 in acetonitrile
Figure 5. Thermal ellipsoid (50% probability level) plot of
[Ru(MeCN)2(qmtpm)(PPh3)]+ (cation of 5) with the atom-labeling
scheme. Hydrogen atoms have been omitted for the sake of clarity.
center of this complex. Two MeCN molecules (cis to each
other) and a PPh3 ligand complete the octahedral coordination
around the RuII center. Much like in 2, the triphenylphosphine
ligand is perpendicular to the qmtpm plane. The Ru−P bond
Figure 6. Partial (7−10 ppm) 1H NMR (500 MHz) spectrum of 2 in CDCl3 at 298 K. The resonances of the 11 protons of the qmtpm ligand are
spread out in the 7.4−10 ppm region, while peaks for the PPh3 protons appear between 7 and 7.3 ppm. Inset: Resonance of the −SMe at 3.08 ppm.
F
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 7. EAS spectra of 2 (black trace), 4 (blue trace), and 5 (red
trace) in acetonitrile.
Figure 8. Changes in the absorption spectrum of 2 upon exposure to
365 nm light (power = 10 mW/cm2).
upon illumination (vide infra) leads to a shift of λmax from 465
to 535 nm, indicating formation of 4. It is interesting to note
that the typical blue (for 1), magenta (for 4), or red (for 5)
colors of the RuIIqmtpm complexes are lost upon coordination
of one or two CO ligand(s). For example, coordination of one
CO to 1 changes the deep-blue color (λmax = 610 nm) to
orange-yellow (λmax = 460 nm; Figure 5). Likewise, 3 with two
bound CO ligands exhibits its λmax at 385 nm (yellow). As
described in a forthcoming section, DFT studies have provided
insight into such changes in the MLCT bands of these
complexes.
Photorelease of CO in Solution. When a solution of 2 in
ethanol or chloroform is kept in the dark, no change is noted
with time. However, solutions of 2 in acetonitrile show small
changes in its absorption spectrum over hours because of the
slow replacement of CO with acetonitrile. The strong affinity of
acetonitrile toward RuII centers as well as the strong trans-
labilizing effect of the PPh3 ligand promotes such CO loss. This
behavior has also been reported for [Ru(Cl)(CO)(terpy)-
(PPh3)]PF6 under similar conditions.30 As expected, 3 with no
PPh3 ligand is indefinitely stable in acetonitrile. Quite in
contrast to these behaviors, exposure to light brings about rapid
changes in the absorption spectra of 2 and 3. When solutions of
2 or 3 in ethanol or acetonitrile are exposed to 350−500 nm
light, the absorption spectrum undergoes rapid changes because
of photorelease of CO, as evidenced by reduced myoglobin
assay (Figure S2, Supporting Information). Changes in the
absorption spectrum of 2 in acetonitrile upon exposure to 365
nm light are shown in Figure 8. Careful inspection of the
spectra reveals that 2 is quantitatively converted to 4 (eq 2)
(power = 15 mW/cm2) with cutoff filters at λ ≥ 380 nm and λ
≥ 440 nm, respectively. In ethanol, the kCO values drop to
0.031 0.001 and 0.0032 0.0001 min−1 when illuminated
with the same light sources. Notable acceleration in CO
photorelease from 2 is observed with low-power (5−10 mW/
cm2) UV light. For example, in ethanol, a kCO value of 0.81
0.01 min−1 is obtained under 310 nm illumination.
Yellow solutions of 3 in ethanol and acetonitrile release CO
upon exposure to light in the 360−300 nm range. For example,
when exposed to a narrow-width UV-light source with λmax at
310 nm (power = 7 mW/cm2), 3 exhibits rapid changes in its
EAS spectrum in acetonitrile with isosbestic points at 480, 365,
and 300 nm and affords a kCO value of 0.22
0.01 min−1.
Because exhaustive photolysis of 3 followed by replacement of
the Cl− ligand with PPh3 affords 5 quantitatively (eq 3), it is
evident that 3 delivers 2 equiv of CO under exposure to 310
nm light. When λmax of the light source is shifted to 325 nm, the
value of kCO drops to 0.08 0.01 min−1, again showing the
dependence of kCO on the wavelength of light.
DFT and TDDFT Studies. DFT and TDDFT calculations
were performed to obtain the optimized geometries, MO
electron densities, and calculated electronic transitions for 2−5.
The primary objective of this study was to identify the
electronic transitions that give rise to CO release from
photoCORMs like 2 and 3 under illumination. Another
major objective was to determine how the auxiliary ligands
such as PPh3, MeCN, and Cl− affect the MLCT transitions and
facilitate Ru−CO bond scission. The DFT-optimized structures
of 2, 3, and 5 show very good agreement with the bond lengths
and bond angles observed in the corresponding crystal
structure of each complex (Table S1, Supporting Information).
The structure of 4 was obtained by exchanging an acetonitrile
of 5 with Cl− followed by optimization.
within 60 min with no apparent decomposition, as evidenced
by clean isosbestic points at 490, 400, 380, and 355 nm.Data
from myoglobin assay confirm that 2 delivers 1 equiv of CO
under such illumination (eq 2). The apparent first-order rate
constant of CO release kCO from 2 is faster in acetonitrile than
that in ethanol, and the kCO values change with the wavelength
of light. For example, in acetonitrile, 2 exhibits kCO values of
0.065
0.001 and 0.0063
0.0002 min−1 with visible light
G
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 4. Energies (E, nm) and Oscillator Strengths (f) of the Calculated (TDDFT) Electronic Transitions with the MOs
a
Involved with Each Transition
energy (nm)
oscillator strength
[Ru(Cl)(CO)(qmtpm)(PPh3)]BF4 (2)
transition
467
431
387
381
374
365
323
0.0489867
0.0566891
0.0302262
0.0529700
0.0289463
0.2146560
0.0339253
π(Ru−CO)−p(Cl)−π(Ru−PPh3) → π*(Q−PhS−SB)
π(PPh3)−π(Ru−PPh3)−π(Ru−CO) → π*(Q−PhS−SB)
π(Ru−CO−SMe) → π*(Q−PhS−SB)
π(PPh3)−π(Ru−CO−SMe) → π*(Q−PhS−SB)
π(PPh3)−π(PhS) → π*(Q−PhS−SB)
π(Ru−CO−SMe)−π(PPh3)−π(PhS) → π*(Q−PhS−SB)
π(Ru−Q−SMe−SB)−p(Cl) → π*(Q−PhS−SB)
[Ru(Cl)(CO)2(qmtpm)]ClO4 (3)
445
408
396
387
370
0.0275726
0.0165650
0.1794866
0.2110407
0.0729107
π(Q−PhS−SB) → π*(Q−SB−PhS)
π(Q−PhS−SB)−π(Ru−CO−Cl) → π*(Q−SB−PhS)
π(Ru−CO)−p(Cl)−π(PhS) → π*(Q−SB−PhS)
π(Ru−CO)−p(Cl)−π(PhS) → π*(Q−SB−PhS)
π(Ru−SMe−CO)−π(PhS)−p(Cl) → π*(Q−SB−PhS)
[Ru(Cl)(MeCN)(qmtpm)(PPh3)]BF4 (4)
589
539
414
391
382
370
365
357
353
350
342
0.0139661
0.0442363
0.0384803
0.0905007
0.0434632
0.0384393
0.0147778
0.0790384
0.0131655
0.0223908
0.0602548
π(Ru−SMe−MeCN−Cl) → π*(Q−PhS−SB)
π(Ru−Cl−SMe−MeCN)−π(Q) → π*(Q−PhS−SB)
π(Ru−PPh3) → π*(Q−PhS−SB)
π(Ru−Cl−PPh3)−π(Q) → π*(Q−PhS−SB)
π(PPh3)−π(Ru−SB) → π*(Q−PhS−SB)
π(Ru−PPh3) → π*(Q−PhS−SB)
π(Q)−π(Ru−PPh3) → π*(Q−PhS−SB)
π(Ru−PPh3−SMe)−p(Cl) → π*(Q−PhS−SB)
π(Ru−PPh3−SMe)−p(Cl) → π*(Q−PhS−SB)
π(PPh3) → π*(Q−PhS−SB)
π(PPh3)−π(Sme) → π*(Q−PhS−SB)
[Ru(MeCN)2(qmtpm)(PPh3)]ClO4 (5)
449
378
375
357
347
329
0.0787027
0.0210149
0.0685953
0.0148138
0.2105311
0.0473331
π(Ru−MeCN−PPh3) → π*(Q−SB−PhS)
π(Ru−Q−PPh3) → π*(Q−SB−PhS)
π(Ru−PPh3)−π(PhS) → π*(Q−SB−PhS)
π(Ru−PPh3) → π*(Q−SB−PhS)
π(PPh3−PhS)−π(Ru−SB) → π*(Q−SB−PhS)
π(Ru−PhS−SMe) → π*(Q−SB−PhS)
a
Orbitals with greater contributions are listed first.
As stated above, the transformations of Scheme 1 are
accompanied by sharp changes in color. For example, the
replacement of one Cl− (a strong σ donor) of 1 with CO (an
excellent π acceptor) results in a loss of the deep-blue color and
formation of 2 (pale orange), while the replacement of CO of 2
with acetonitrile (a weak π acceptor) changes the color to deep
magenta, indicating formation of 4 (Scheme 1). Similarly, loss
of the two CO ligands from 3 (pale yellow) with acetonitrile
affords a purple solution of [Ru(Cl)(MeCN)2(qmtpm)]+ (λmax
= 565 nm), which upon replacement of the Cl− with PPh3 (a
good π acceptor) turns deep red because of formation of 5.
TDDFT studies were utilized to gain insight into the nature of
transitions that lead to these drastic color changes observed
upon CO addition or loss to the [Ru(qmtpm)] frame along
with variation in the auxiliary ligands PPh3, MeCN, and Cl−.
The calculated electronic transitions of 2−5 with oscillator
strengths above 0.0139 were collected, and those with energies
falling within the range of 700−300 nm for each complex are
presented in Table 4. As shown in Figure 9, the theoretical
electronic spectrum of 2 is in good agreement with the
experimental spectrum obtained in acetonitrile. Complex 2
exhibits two major bands in the 300−500 nm region, namely,
the MLCT band at 460 nm and a stronger absorption at 370
nm. TDDFT results indicate that, for 2, two theoretical
Figure 9. Experimental (solid line) and calculated (dashed line) EAS
spectra of 2.
transitions that occur in the ∼450 nm region are HOMO−1 →
LUMO (462 nm) and HOMO−3 → LUMO (431 nm). The
370 nm band corresponds to the theoretical transition with the
highest oscillation strength at 355 nm along with a few weaker
transitions (Figure 9). Close inspection of the MO energy
diagram (Figure 10) displaying the electron densities of the
H
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 10. Calculated HOMO/LUMO energy diagrams of 2, 4, and 5 (left to right). The most prominent MOs involving transitions under the low-
energy band and their diagrams are shown.
MOs that comprise these levels reveals that HOMO−1 is
mostly π(Ru−CO) bonding in character mixed with a
moderate π(Ru−PPh3) contribution while HOMO−3 is mainly
π(Ru−PPh3) with modest π(Ru−CO) participation. The
LUMO, on the other hand, is mostly composed of the ligand
frame consisting of the π* orbital of the quinoline, the imine
functionality, and the π* orbital of the Ph(SMe) unit. It is
important to note that exposure to visible light (λ ≥ 440 nm)
causes moderate CO photorelease from 2. The TDDFT results
strongly suggest that such Ru−CO bond labilization arises from
a shift of the electron density from the Ru−CO moiety to the
ligand frame. Partial reduction in Ru−PPh3 back-bonding also
occurs during the 431 nm transition. The much faster
photorelease of CO from 2 upon exposure to light of λ ≥
380 nm stems from the predicted HOMO−9 → LUMO
transition at 365 nm (Figure 9). Interestingly, HOMO−9 also
has a strong π(Ru−CO) character in addition to a moderate
contribution from the quinoline π orbital (not shown in Figure
10). Transfer of the electron density from this orbital to the
LUMO therefore results in accelerated CO release. Taken
together, these results suggest that transitions shifting the
electron density from the π(Ru−CO) moiety to the ligand
frame promote CO release from photoCORMs like 2.
The photoproduct 4 exhibits its MLCT band at 535 nm
(Figure 7). TDDFT results predict a theoretical HOMO−2 →
LUMO transition at 588 nm and a comparatively stronger
HOMO−3 → LUMO transition at 538 nm. As is evident from
Figure 10, the replacement of CO with MeCN (a moderate π
acceptor) raises the energies of the occupied MOs while the
LUMO (mostly comprised of the π* orbitals spread over the
qmtpm ligand frame) is minimally affected. This effect is clearly
reflected in the change of color from orange-yellow (for 2) to
magenta (in 4) upon CO photorelease. Replacement of Cl− of
4 with MeCN (a moderate π acceptor), on the other hand,
stabilizes the HOMO−1 in addition to slightly lowering the
LUMO in the case of 5. The combined effect of such an
alteration leads to a net increase in the transition energy (from
3.401 to 3.592 eV; Figure 10), as reflected in the color change
from magenta (in 4) to deep red (in 5).
As described above, the dicarbonyl complex 3 exhibits CO
photorelease with UV light. The calculated MOs involved in
the electronic transitions of 3 are displayed in Figure 11
I
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. Calculated HOMO/LUMO energy diagram of 2 and 3 (left to right). The most prominent MOs involved with transitions under the low-
energy band and their diagrams are shown.
alongside those of 2. TDDFT studies predict that, for 3,
transitions originating from both HOMO−2 and HOMO−3
into the LUMO give rise to the experimental absorption
maximum at ∼380 nm (Figure S1, Supporting Information).
This HOMO−2/HOMO−3 transition into the antibonding
orbitals of the ligand frame (LUMO) again corresponds to a
shift in the electron density from the RuII center to the qmtpm
ligand frame, consistent with a MLCT process. Scrutiny of the
nature of HOMO−2 and HOMO−3 reveals a combined Ru−
Cl and Ru−CO π-bonding character in these occupied orbitals
(Figure 11). Placement of Cl− (a σ donor) trans to CO clearly
fosters strong back-bonding interactions between the metal and
the CO group. Such favorable interactions must be overcome
to promote a shift in the electron density toward the LUMO
(mostly ligand π* orbitals). As a consequence, 3 requires UV
light for the release of CO.
(b) Although stable in the dark, solutions of 2 and 3 in
ethanol, acetonitrile, and dichloromethane are sensitive to light.
Both complexes exhibit excellent CO photolability under low-
power (5−10 mW) UV light (300−350 nm). In the case of 2,
moderate photorelease of CO is also noted upon exposure to
low-power (15−30 mW) visible light in the range 350−480
nm. The results of myoglobin assay have confirmed quantitative
CO release in both cases. Loss of CO from 2 followed by
replacement of Cl− (by Ag+) in acetonitrile gives rise to 5 as the
photoproduct, which has been characterized by crystallography.
Exhaustive photolysis of 3 followed by replacement of the Cl−
ligand with PPh3 also affords 5 quantitatively.
(c) The Results of DFT and TDDFT studies demonstrate
that, with both photoCORMs, exposure to light promotes
transfer of the electronic charge density from MOs with strong
π(Ru−CO) bonding in character to a LUMO comprised of
mostly the qmtpm ligand π frame. Such loss of π-back-bonding
leads to rapid CO release. An increase in the number of CO
ligands in the photoCORMs (going from 2 to 3) shifts the
MLCT bands to higher energy, and as a consequence, 3
requires UV light for CO release. The theoretical results also
indicate that the ancillary ligand PPh3 (trans to CO) plays a
significant role in the Ru−CO bond labilization process in 2.
(d) The low-lying LUMOs in 2 and 3, a direct consequence
of the extended conjugation across the imine function and the
quinoline moiety (plus the −SMe appendage) of the qmtpm
ligand, bring the major electronic transitions to the 350−450
nm range. This fact underscores the importance of designed
SUMMARY AND CONCLUSIONS
■
The following are the summary and conclusions of this
investigation:
(a) Mono- and dicarbonyl complexes, namely, 2 and 3, have
been synthesized through the careful choice of starting
ruthenium(II) sources and the tridentate ligand qmtpm. The
crystal structures of these two carbonyl complexes reveal that,
unlike the case with the MnI center,20 the ligand qmtpm binds
the RuII center in 2 and 3 in a tridentate (and meridional)
fashion. In 3, the two CO ligands are cis to each other.
J
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(9) (a) Stein, A. B.; Bolli, R.; Dawn, B.; Sanganalmath, S. K.; Zhu, Y.;
Wang, O. L.; Guo, Y.; Motterlini, R.; Xuan, Y. T. J. Mol. Cell. Cardiol.
2012, 52, 228−236. (b) Stein, A. B.; Guo, Y.; Tan, W.; Wu, W. J.; Zhu,
X.; Li, Q.; Luo, C.; Dawn, B.; Johnson, T. R.; Motterlini, R.; Bolli, R. J.
Mol. Cell. Cardiol. 2005, 38, 127−134. (c) Otterbein, L. E.; Mantell, L.
L.; Choi, A. M. K. Am. J. Physiol. 1999, 276, L688−L694.
(10) (a) Stein, A. B.; Bolli, R.; Dawn, B.; Sanganalmath, S. K.; Zhu,
Y.; Wang, O. L.; Guo, Y.; Motterlini, R.; Xuan, Y. T. J. Mol. Cell.
Cardiol. 2012, 52, 228−236. (b) Stein, A. B.; Guo, Y.; Tan, W.; Wu,
W. J.; Zhu, X.; Li, Q.; Luo, C.; Dawn, B.; Johnson, T. R.; Motterlini,
R.; Bolli, R. J. Mol. Cell. Cardiol. 2005, 38, 127−134.
ligands for the successful isolation of photoCORMs that are
sensitive to visible and near-IR light. Although metal carbonyl
complexes derived from typical ligands such as bipyridine and
water-soluble phosphines14 have been shown to release CO
upon exposure to UV light, novel ligands that would initiate
specific MLCT interactions will be required the isolation of
photoCORMs suitable for use in the photodelivery of CO to
biological targets.
ASSOCIATED CONTENT
■
(11) (a) Nakao, A.; Choi, A. M. K.; Murase, N. J. Cell. Mol. Med.
2006, 10, 650−671. (b) Sato, K.; Balla, J.; Otterbin, L.; Smith, R. N.;
Brouard, S.; Lin, Y.; Csizmadia, E.; Sevigny, J.; Robson, S. C.;
Vercellotti, G.; Choi, A. M. K.; Bach, F. H.; Soares, M. P. J. Immunol.
2001, 166, 4185−4194.
S
* Supporting Information
X-ray crystallographic data (in CIF format) for 2·H2O, 3, and 5,
experimental and calculated elctronic absorption spectra of 3
(Figure S1), traces of spectra showing conversion of reduced
Mb to Mb-CO in a typical myoglobin assay experiment (Figure
S2), and selected bond distances (Å) and angles (deg) for 2−5
with DFT-optimized bond distances and angles (Table S1).
This material is available free of charge via the Internet at
http://pubs.acs.org.
(12) (a) Zobi, F. Future Med. Chem. 2013, 5, 175−188. (b) Romao,
̃
C. C.; Blattler, W. A.; Seixas, J. D.; Bernardes, G. J. L. Chem. Soc. Rev.
̈
2012, 41, 3571−3583. (c) Motterlini, R.; Otterbein, L. E. Nat. Rev.
Drug Discovery 2010, 9, 728−743. (d) Mann, B. E. Top. Organomet.
Chem. 2010, 32, 247−285.
(13) (a) Rimmer, R. D.; Pierri, A. E.; Ford, P. C. Coord. Chem. Rev.
2012, 256, 1509−1519. (b) Schatzschneider, U. Inorg. Chim. Acta
2011, 374, 19−23. (c) Schatzschneider, U. Eur. J. Inorg. Chem. 2010,
1451−1467.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: pradip@ucsc.edu.
Notes
(14) Rimmer, R. D.; Richter, H.; Ford, P. C. Inorg. Chem. 2010, 49,
1180−1185.
(15) Jackson, C. S.; Schmitt, S.; Dou, Q. P.; Kodanko, J. J. Inorg.
Chem. 2011, 50, 5336−5338.
The authors declare no competing financial interest.
(16) Niesel, J.; Pinto, A.; Peindy N’Dongo, H. W.; Merz, K.; Ott, I.;
Gust, R.; Schatzschneider, U. Chem. Commun. 2008, 1798.
(17) (a) Huber, W.; Linder, R.; Niesel, J.; Schatzschneider, U.;
Spingler, B.; Kunz, P. C. Eur. J. Inorg. Chem. 2012, 3140. (b) Kunz, P.
C.; Huber, W.; Rojas, A.; Schatzschneider, U.; Spingler, B. Eur. J. Inorg.
Chem. 2009, 5358. (c) Pfeiffer, H.; Rojas, A.; Niesel, J.;
Schatzschneider, U. Dalton Trans. 2009, 4292−4298.
(18) Berends, H. M.; Kurz, P. Inorg. Chim. Acta 2012, 380, 141−147.
(19) Gonzalez, M. A.; Yim, M. A.; Cheng, S.; Moyes, A.; Hobbs, A. J.;
Mascharak, P. K. Inorg. Chem. 2012, 51, 601−608.
(20) Gonzalez, M. A.; Carrington, S. J.; Fry, N. L.; Martinez, J. L.;
Mascharak, P. K. Inorg. Chem. 2012, 51, 11930−11940.
(21) Stephenson, T. A.; Wilkinson, G. J. Inorg. Nucl. Chem. 1966, 28,
945−956.
ACKNOWLEDGMENTS
■
This research was supported by a grant from the National
Science Foundation (Grant CHE-0957251). M.A.G. was
supported by IMSD Grant GM-58903.
REFERENCES
■
(1) (a) Kikuchi, G.; Yoshida, T.; Noguchi, M. Biochem. Biophys. Res.
Commun. 2005, 338, 558−567. (b) Wu, L.; Wang, R. Pharmacol. Rev.
2005, 57, 585−630.
̈
(2) (a) Ollinger, R.; Wang, H.; Yamashita, K.; Wegiel, B.; Thomas,
M.; Margreiter, R.; Bach, F. H. Antioxid. Redox Signal. 2007, 9, 2175−
̈
2185. (b) Ollinger, R.; Yamashita, K.; Bilban, M.; Erat, A.; Kogler, P.;
Thomas, M.; Csizmadia, E.; Usheva, A.; Margreiter, R.; Bach, F. H.
Cell Cycle 2007, 6, 39−43. (c) Fondevila, C.; Shen, X. D.; Tsuchiyashi,
S.; Yamashita, K.; Csizmadia, E.; Lassman, C.; Busuttil, R. W.; Kupiec-
Weglinski, J. W.; Bach, F. H. Hepatology 2004, 40, 1333−1341.
(3) (a) Komuro, T.; Borsody, M. K.; Ono, S.; Marton, L. S.; Weir, B.
K.; Zhang, Z. D.; Paik, E.; Macdonald, R. L. Expt. Biol. Med. 2001, 226,
860−865. (b) Zhang, F.; Kaide, J.-I.; Rodriguez-Mulero, F.; Abraham,
N. G.; Nasjletti, A. Am. J. Hypertens. 2001, 14, 62S−67S.
(4) Brouard, S.; Otterbin, L. E.; Anrather, J.; Tobiasch, E.; Bach, F.
H.; Choi, A. M. K.; Soares, M. P. J. Exp. Med. 2000, 192, 1015−1025.
(5) Kim, H. P.; Ryter, S. W.; Choi, A. M. K. Annu. Rev. Pharmocol.
Toxicol. 2006, 46, 411−449.
(22) Castro, A. G.; Costa, J. S.; Pievo, R.; Massera, C.; Mutikainen, I.;
Turpeinen, U.; Gamez, P.; Reedijik, J. Z. Anorg. Allg. Chem. 2008, 634,
2477−2482.
(23) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M. T. Handbook of
Photochemistry, 3rd ed.; CRC Press: Boca Raton, FL, 2006.
(24) McLean, S.; Mann, B. E.; Poole, R. K. Anal. Biochem. 2012, 427,
36−40.
(25) Nemukhin, A. V.; Grigorenko, B. L.; Granovsky, A. A. Moscow
Univ. Chem. Bull. 2004, 45, 75−102.
(26) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117−
120.
(27) Waller, M. P.; Braun, H.; Hojdis, N.; Buhl, M. J. Chem. Theory
̈
(6) Vieira, H. L. A.; Queiroga, C. S. F.; Alves, P. M. J. Neurochem.
2008, 107, 375−384.
Comput. 2007, 3, 2234−2242.
(28) Hossain, M.; Maji, M.; Chattopadhyay, S. K.; Ghosh, S.; Blake,
A. J. Polyhedron 1998, 17, 1897−1906.
(7) (a) Faller, S.; Hoetzel, A. Curr. Pharmaceut. Biotechnol. 2012, 13,
777−786. (b) Ghosh, S.; Wilson, M. R.; Choudhury, S.; Yamamoto,
H.; Goddard, M. E.; Falusi, B.; Marczin, N.; Takata, M. Am. J. Physiol.
Lung Cell Mol. Physiol. 2005, 288, L1003−L1009. (c) Zhang, X.; Shan,
P.; Alam, J.; Davis, R. J.; Flavell, R. A.; Lee, P. J. J. Biol. Chem. 2003,
278, 22061−22070.
(29) Moreno, M. A.; Haukka, M.; Kallinen, M.; Pakkanen, T. A. Appl.
Organomet. Chem. 2006, 20, 51−69.
(30) Ooyama, D.; Saito, M. Inorg. Chim. Acta 2006, 359, 800−806.
́
(31) Cuervo, D.; Menendez-Pedregal, E.; Díez, J.; Gamasa, M. P. J.
Organomet. Chem. 2011, 696, 1861−1867.
(32) Abbenhuis, R. A. T. M.; del Rio, I.; Bergshoef, M. M.; Boersma,
J.; Veldman, N.; Spek, A. L.; van Koten, G. Inorg. Chem. 1998, 37,
1749−1758.
(33) Sugimoto, H.; Tsuge, K.; Tanaka, K. J. Chem. Soc., Dalton Trans.
2001, 57−63.
(8) (a) Kohmoto, J.; Nakao, A.; Stolz, D. B.; Kaizu, T.; Tsung, A.;
Ikeda, A.; Shimizu, H.; Takahashi, T.; Tomiyama, K.; Sugimoto, R.;
Choi, A. M. K.; Billiar, T. R.; Murase, N.; McCurry, K. R. Am. J.
Transplant. 2007, 7, 2279−2290. (b) Fujita, T.; Toda, K.; Karimova,
A.; Yan, S. F.; Naka, Y.; Yet, S. F.; Pinsky, D. J. Nat. Med. 2001, 7,
598−604.
K
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(34) Maji, M.; Sengupta, P.; Chattopadhyay, S. K.; Mostafa, G.;
Schwalbe, C. H.; Ghosh, S. J. Coord. Chem. 2001, 54, 13−24.
L
dx.doi.org/10.1021/ic4016004 | Inorg. Chem. XXXX, XXX, XXX−XXX