Photodelivery of CO by designed photoCORMs: Correlation between absorption in the visible region and metal-CO bond labilization in carbonyl complexes

Article abstract of DOI:10.1002/cmdc.201402007

The therapeutic potential of photoactive CO-releasing molecules (photoCORMs) have called for close examination of the roles of the ligand(s) and the central metal atoms on the overall photochemical labilization of the metal-CO bonds. Along this line, we have synthesized four metal complexes, namely, [MnBr(azpy)(CO)₃] (1), [Mn(azpy)(CO)₃(PPh ₃)]ClO₄ (2), [ReBr(azpy)(CO)₃] (3), and [Re(azpy)(CO)₃(PPh₃)]ClO₄ (4), derived from 2-phenylazopyridine. These complexes were characterized by spectroscopic and crystallographic studies. Although both 1 and 3 exhibit strong metal-to-ligand charge-transfer bands in the 500-600 nm region, only 1 photoreleases CO upon illumination with visible light. Results of theoretical studies were used to gain insight into this surprising difference. Strong spin-orbit coupling (prominent in heavy metals) appears to promote intersystem crossing to a triplet state in 3, a step that discourages CO release upon illumination with visible light. Slow release of CO from 2 and 4 also indicates that strong σ-donating ligands, such as Br^-, accelerate the rate of CO photorelease relative to π-acid ligands, such as PPh₃. Break it up! Despite strong absorptions in the visible range, [MnBr(azpy)(CO) ₃] and [ReBr(azpy)(CO)₃], two structurally and electronically similar metal-carbonyl complexes, exhibit very different CO photolability upon exposure to visible light. Our theoretical study results that explain these differences may be useful to future studies of photoactive CO-donating drugs triggered by visible/near-IR light.

Full text of DOI:10.1002/cmdc.201402007

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201402007
Photodelivery of CO by Designed PhotoCORMs:
Correlation between Absorption in the Visible Region and
Metal–CO Bond Labilization in Carbonyl Complexes
Indranil Chakraborty, Samantha J. Carrington, and Pradip K. Mascharak*[a]
The therapeutic potential of photoactive CO-releasing mole-
cules (photoCORMs) have called for close examination of the
roles of the ligand(s) and the central metal atoms on the over-
all photochemical labilization of the metal–CO bonds. Along
this line, we have synthesized four metal complexes, namely,
[MnBr(azpy)(CO)3] (1), [Mn(azpy)(CO)3(PPh3)]ClO4 (2), [ReBr-
(azpy)(CO)3] (3), and [Re(azpy)(CO)3(PPh3)]ClO4 (4), derived from
2-phenylazopyridine. These complexes were characterized by
spectroscopic and crystallographic studies. Although both
1 and 3 exhibit strong metal-to-ligand charge-transfer bands in
the 500–600 nm region, only 1 photoreleases CO upon illumi-
nation with visible light. Results of theoretical studies were
used to gain insight into this surprising difference. Strong spin-
orbit coupling (prominent in heavy metals) appears to pro-
mote intersystem crossing to a triplet state in 3, a step that
discourages CO release upon illumination with visible light.
Slow release of CO from 2 and 4 also indicates that strong s-
donating ligands, such as Brꢀ, accelerate the rate of CO photo-
release relative to p-acid ligands, such as PPh3.
Introduction
The deleterious effects of carbon monoxide (CO) observed in
mammalian physiology arise from its strong affinity to heme
centers in proteins. Binding of CO to hemoglobin leads to as-
phyxia, an effect that has earned this diatomic molecule the
moniker of “silent killer”. However, CO is produced endoge-
nously through heme degradation by the heme oxygenase
(HO) enzyme[1] and in low doses, CO has recently been shown
to impart beneficial effects in various physiological pathways,
including vasorelaxation. More surprisingly, low doses of CO
exhibit anti-inflammatory and anti-apoptotic properties.[2,3] As
a consequence, CO provides protection to oxidatively dam-
aged tissues, such as in ischemic reperfusion injury and endo-
thelial impairment during balloon angioplasty. It is therefore
expected that CO could play a crucial role as a therapeutic
agent in cardiovascular disease and organ transplantation pro-
tocols.[4] In addition, Motterlini et al. have shown that, although
CO induces an anti-apoptotic effect in endothelial cells, it can
impart considerable pro-apoptotic effects in hyperproliferative
tissues.[2] Collectively, these findings have prompted considera-
ble research effort in recent times to use CO as a chemothera-
peutic in various settings.[2,3] However, administration of CO in
gaseous form raises serious issues in terms of controlled and
safe delivery to biological targets. To circumvent these obsta-
cles, various research groups have directed their efforts to de-
velop suitably designed metal–carbonyl complexes[5–7] as CO-
releasing molecules (CORMs) to deliver CO in a more con-
trolled fashion. The major drawbacks of such first-generation
CORMs are associated with their solubility and stability under
ambient conditions.[2,8] Air-stable and water-soluble metal–car-
bonyls, such as [Re(CO)3(H2O)3]+ and [Tc(CO)3(H2O)3]+, show no
reactivity in terms of CO release,[9] and in cases of amino acid-
derived carbonyl complexes such as [RuCl(gly)(CO)3] (CORM-3),
solvent-assisted CO release is triggered through hydrolysis, and
a significantly shortened half-life under specific physiological
conditions inhibits sustained delivery of CO to desired tar-
gets.[8b,c]
During the past few years, the photo-induced CO-releasing
molecules (photoCORMs) have emerged as credible alterna-
tives.[10–12] Here, the CO release from the metal–carbonyl com-
plexes (which are otherwise stable under dark conditions) can
be achieved through exposure to light. In earlier attempts, typ-
ical carbonyls such as [Mn2(CO)10] and [Fe(CO)5] were used for
photodelivery of CO.[13] Unfortunately, high toxicity and lack of
chemical amenability restricted their applicability in biological
systems. In recent years, several research groups have devel-
oped suitably designed photoCORMs based on transition
metal–carbonyls to alleviate such limitations.[11,12] For example,
the rhenium-based water-soluble photoCORM [Re(bpy)(CO)3-
(thp)]+ (thp=tris(hydroxymethyl)phosphine), developed by
Ford and co-workers, is readily internalized by human prostatic
carcinoma cells with no apparent cytotoxicity.[14] When the
loaded cells are irradiated with UV light (405 nm), CO release
can be visualized by a change in fluorescence. Schatzschneider
and co-workers have synthesized the cationic [Mn(CO)3(tpm)]+
(tpm=tris(pyrazolyl)methane) complex that initiates photode-
livery of CO upon illumination with UV light (365 nm).[15] This
photoCORM has been shown to eradicate human colon cancer
[a] Dr. I. Chakraborty, S. J. Carrington, Prof. P. K. Mascharak
Department of Chemistry and Biochemistry
University of California Santa Cruz
1156 High Street, Santa Cruz, CA 95064 (USA)
E-mail: pradip@ucsc.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402007.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1266

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Results and Discussion
cell (HT29) through efficient internalization and CO delivery.
The development of designed photoCORMs, however, faces
the major challenge of synthesizing suitable metal–carbonyl
complexes that can liberate CO upon irradiation with light in
the biocompatible range (500–900 nm).[12] To date, the vast
majority of photoCORMs have shown sensitivity toward UV
light (300–450 nm),[11,12,16–22] an untenable range of wave-
lengths in terms of developing phototherapeutics, with very
few exceptions.[23–25]
Synthesis
Reaction of [MnBr(CO)5] with one equivalent of azpy ligand in
dichloromethane at room temperature afforded the fac-[MnBr-
(azpy)(CO)3] (1) complex. During the synthesis, the entire reac-
tion setup was properly covered with aluminum foil to protect
from exposure to ambient light source. The Re analogue fac-
[ReBr(azpy)(CO)3] (3) was synthesized by reacting [ReBr(CO)5]
with one equivalent of azpy ligand in boiling benzene. Previ-
ously, Ishitani and co-workers showed that incorporation of
a p-acceptor ligand like PPh3 in metal–carbonyl complexes en-
hances the rate of CO release.[26] In such species, competition
between PPh3 and the trans CO group for the same p-symme-
try orbital causes CO labilization. We therefore undertook the
task of synthesizing the phosphine complexes through re-
placement of the bromide group of 1 and 3. The fac-[Mn-
(azpy)(CO)3(PPh3)]ClO4 (2) complex was synthesized in two
steps. Complex 1 was first stirred with one equivalent of
AgClO4 in tetrahydrofuran (THF), and the resulting AgBr was re-
moved by filtration. Following removal of THF, one equivalent
of PPh3 in dichloromethane was added to the residue (presum-
ably the [Mn(azpy)(CO)3(THF)]ClO4 complex), and the mixture
was stirred at room temperature for an extended period to iso-
late 2. In the case of fac-[Re(azpy)(CO)3(PPh3)]ClO4 (4), both
steps were carried out at reflux in THF and chloroform, respec-
tively. The general structures of 1–4 are depicted in Scheme 1.
In our attempt to correlate the light absorption parameters
of designed metal–carbonyl complexes with their ability to
photorelease CO, we looked at various ligands that give rise to
carbonyl complexes with varying numbers of CO ligands.[16,25]
In our recent effort, we selected the ligand 2-phenyazopyridine
(azpy), which resembles a,a’-diimine ligands such as bipyridine
(bpy) and ortho-phenanthroline (o-phen). Metal–carbonyl com-
plexes derived from these a,a’-diimine ligands have recently
been employed as photoCORMs by different groups.[11,12,14] In
a previous communication, we reported a very efficient Mn-
based photoCORM, namely, fac-[MnBr(azpy)(CO)3] (1), which
rapidly releases CO (quantum yield f=0.48) upon illumination
with low-power visible light.[23] Moreover, CO liberated from
this complex has been used to inflict severe damage to HeLa
and MDA-MB-231 cancer cells. The phosphine-substituted
complex fac-[Mn(azpy)(CO)3(PPh3)]ClO4 (2) also exhibits sensi-
tivity to visible light. To further increase the stability of this
type of photoCORMs in biological media, we have now synthe-
sized the corresponding rhenium complexes, fac-[ReBr-
(azpy)(CO)3] (3) and fac-[Re(azpy)(CO)3(PPh3)]ClO4 (4), and ex-
amined their CO-releasing properties. As both of these com-
plexes display strong absorption bands in the visible (~
500 nm) region, analogous to their Mn progenitors, we expect-
ed that the associated metal-to-ligand charge transfer (MLCT)
transitions would augment CO release. However, both 3 and 4
release CO only upon illumination with UV light. In addition,
the rates of CO release from 3 and 4 are significantly slower
than from the corresponding Mn complexes 1 and 2. Clearly,
these findings raise the critical question of why the structurally
and electronically similar rhenium carbonyl complexes fail to
photorelease CO upon exposure to visible light, despite strong
absorption in the visible region. To explore the cause of dis-
crepancy in light sensitivity between 1 and 3 (derived from
metal centers with same low-spin d6 configuration), we per-
formed density functional theory (DFT) and time-dependent
density functional theory (TDDFT) calculations on 1–4. The re-
sults, as described in this article, reveal for the first time the
role of transition metal centers (within the same group) on the
CO-releasing capacities of structurally identical complexes. It is
now evident that judicious choice of the metal center, along
with proper combination of ligand/co-ligand, are critical to
achieve the objective of CO delivery under the control of visi-
ble/near-IR light.
Scheme 1. Manganese and rhenium carbonyl complexes reported herein.
To analyze the effect of the azpy ligand, we also synthesized
fac-[MnBr(bpy)(CO)3] and fac-[ReBr(bpy)(CO)3] by following pro-
cedures developed in our laboratory (see below).
X-ray structures
Dark, block-shaped crystals of 1 and 3 and orange needles of
2, 4, and fac-[MnBr(bpy)(CO)3] were obtained by layering hex-
anes over their solution in dichloromethane. We reported the
structures of 1 and 2 in a previous communication.[23] The
structures of complexes 3 and 4 with atom labeling are shown
in Figures 1 and 2. Selected metric parameters of 3 and 4 are
listed in Table 1. Complete crystal structure determination and
refinement parameters for 1–4 (Table S1), metric parameters
for 1 and 2 (Table S2), and crystal and metric parameters for
fac-[MnBr(bpy)(CO)3] (Table S3) are available in the Supporting
Information.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1267

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(bpy)(CO)3] reveals certain differences. The Mn–C2 distance in
1 is 1.8139(18) ꢁ, which is noticeably longer than the average
of Mn–C2 distances in fac-[MnBr(bpy)(CO)3] (1.803(4) ꢁ). This
weakening can be attributed to the superior p-acidity of the
azo group[28] trans to the Mn–C2 bond in 1. In contrast, the
Re–C2 distance in 3 (1.920(4) ꢁ) is slightly shorter than the
average of the corresponding distances in the structurally simi-
lar fac-[ReBr(Me2bpy)(CO)3] (1.925(3) ꢁ).[29] Similarly, the Re–C2
distance in 4 (1.931(2) ꢁ) is slightly shorter than the average of
the corresponding distances in fac-[Re(bpy)(CO)3(PPh3)]+
(1.939(16) ꢁ) (Figure 2).[30] These structural characteristics sug-
Figure 1. Molecular structure of fac-[ReBr(azpy)(CO)3] (3). Thermal ellipsoids
are shown at 50% probability level, with the hydrogen atoms omitted for
clarity.
Table 1. Selected bond distances [ꢁ] and angles [8].
Bond
fac-[ReBr(azpy)(CO)3] (3)
fac-[Re(azpy)(CO)3(PPh3)]ClO4 (4)
ReꢀN1
ReꢀN3
2.149(3)
2.156(3)
1.922(4)
1.920(4)
1.929(4)
2.6217(5)
–
1.271(4)
87.79(16)
87.82(15)
90.63(15)
97.94(13)
172.37(12)
94.57(13)
170.74(13)
101.27(12)
93.84(13)
72.85(10)
179.33(11)
92.30(12)
90.04(11)
84.75(7)
85.96(7)
–
2.1511(18)
2.1588(19)
1.967(3)
1.931(2)
1.936(2)
–
2.5199(6)
1.277(3)
87.65(10)
90.77(10)
91.25(10)
94.57(9)
174.66(8)
88.82(9)
167.18(9)
101.89(9)
94.05(9)
72.78(7)
–
ReꢀC1
ReꢀC2
ReꢀC3
ReꢀBr
ReꢀP1
N2ꢀN3
C2ꢀReꢀC1
C2ꢀReꢀC3
C1ꢀReꢀC3
C2ꢀReꢀN1
C3ꢀReꢀN1
C1ꢀReꢀN1
C2ꢀReꢀN3
C3ꢀReꢀN3
C1ꢀReꢀN3
N1ꢀReꢀN3
C1ꢀReꢀBr
C2ꢀReꢀBr
C3ꢀReꢀBr
N1ꢀReꢀBr
N3ꢀReꢀBr
C2ꢀReꢀP1
C3ꢀReꢀP1
C1ꢀReꢀP1
N1ꢀReꢀP1
N3ꢀReꢀP1
Figure 2. Molecular structure of the cation of fac-[Re(azpy)(CO)3(PPh3)]ClO4
(4). Thermal ellipsoids are shown at 50% probability level, with the hydro-
gen atoms omitted for clarity.
gest that the trans effect of the azo group is not prominent in
rhenium complexes. Comparison of the metal–N distances of
1–4 with the corresponding bpy complexes also indicates that
the azpy ligand binds the metal centers more strongly than
the bpy/Me2bpy ligand in the corresponding complexes. Once
again, this can be ascribed to the superior p-acceptor charac-
ter of the azpy ligand, which is well-suited for binding the low-
valent low-spin d6 metal centers in 1–4.
–
–
–
–
88.90(8)
88.38(7)
176.53(8)
91.87(5)
89.40(5)
–
–
–
–
Spectroscopic properties
The three facially disposed CO ligands in all of the complexes
gave rise to nCO bands in the expected regions (2030, 1930,
1920 cmꢀ1 for 3; 2050, 1970, 1940 cmꢀ1 for 4). In addition, the
nN stretch was observed near 1370 cmꢀ1 for all of the com-
The coordination geometry of the Mn and Re centers in 1–4
is distorted octahedral, and the three CO ligands are facially
disposed. The two nitrogen atoms of the azpy ligand and the
two carbon atoms from the CO groups constitute the equatori-
al plane, while the axial positions are occupied by a CO group
and Brꢀ or PPh3. In all of the structures, the NꢀN distances of
the azo group are uniformly longer than that of the uncoordi-
nated 2-phenylazopyridine ligand (1.25(3) ꢁ).[27] This lengthen-
ing indicates strong d(Mn/Re)!azo(p*) back-bonding charac-
ter in the present complexes. Careful scrutiny of the metric pa-
rameters of 1[23] with the structurally similar fac-[MnBr-
=
N
1
plexes. All complexes display H NMR spectra consistent with
the diamagnetic ground state of the ReI and MnI centers (ex-
ample shown in Figure 3).
Solutions of the complexes in methanol, chloroform, di-
chloromethane, and acetonitrile are indefinitely stable in the
absence of light. The complexes are also stable under aerobic
conditions, which is important for biological compatibility and
controlled release of CO. The electronic absorption spectra of
the complexes consist of two bands. One of the bands appears
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1268

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Figure 3. 1H NMR spectrum of [ReIBr(azpy)(CO)3] (3) in CDCl3 solution at 298 K.
Photorelease of CO from 1–4
Manganese complexes 1 and 2 exhibit excellent photoactivity
upon exposure to low-power visible light (10–15 mWcmꢀ2).[23]
Such illumination causes rapid changes in the absorption spec-
tra due to loss of CO, and distinct isosbestic points indicate
the clean conversion of the complexes to their corresponding
photo products. Photorelease of CO has been confirmed in
these photolytic processes by standard myoglobin (Mb) assay.
Results of a representative Mb assay are shown in Figure 5.
Figure 4. Electronic absorption spectra of 1–4 in dichloromethane.
in the range 330–390 nm, while the absorption maxima of the
second band span a range of 460–590 nm (Figure 4). The latter
absorption is likely to arise from the MLCT (metal(dp)!
azo(p*)) transition, with considerable XLCT (halide(p)!azo(p*))
contributions for 1 and 3 (see below). A close inspection of the
electronic absorption spectra of 1 and 3 reveals the effect of
replacement of the group 7 d6 metal center. Complex 1 displays
a dark royal blue color in dichloromethane solution (MLCT
band at 586 nm). Replacement of the Mn center with ReI in 3
results in a blue shift of the MLCT band to 530 nm (Figure 4),
and the color of the dichloromethane solution changes to
deep purple. It is important to note that the lower energy
bands are highly blue shifted in 2 and 4 relative to 1 and 3.
The PPh3 ligand (a good p-acceptor) in these complexes draws
more electron density from the metal center and stabilizes the
highest-occupied orbitals. Such stabilization increases the
energy of the MLCT transitions and causes the blue shift ob-
served with 2 and 4.
Figure 5. UV/Vis traces from the myoglobin (Mb) assay for 2. Formation of
the MbꢀCO adduct from reduced Mb is evident by the shift in the Soret
band from 435 to 424 nm.
The rates of CO photorelease (kCO) from all four complexes
were determined spectrophotometrically. In dichloromethane,
1 and 2 exhibited kCO values of 21.94ꢁ0.01 minꢀ1 (conc.:
1.23ꢂ10ꢀ4 m) and 15.28ꢁ0.01 minꢀ1 (conc.: 3.07ꢂ10ꢀ4 m), re-
spectively, upon illumination with visible light. In contrast, pho-
torelease of CO from 3 and 4 could only be initiated by expo-
sure to low power UV light (centered at 305 nm, 5 mWcmꢀ2
)
despite strong MLCT bands in the 430–530 nm region. Expo-
sure to UV light also resulted in distinct isosbestic points in
their absorption spectra, suggesting clean conversions
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1269

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
therefore necessary to elucidate the photobehavior of the
present photoCORMs.
DFT and TDDFT studies
In our previous work, we were able to move the MLCT band(s)
of metal–carbonyl complexes through modification of the
ligand frame.[16b,25] Such ligand alteration allowed us to isolate
photoCORMs that deliver CO upon exposure to visible light. In
the present work, changing the metal center in a set of analo-
gous complexes afforded 1–4, which exhibit strong MLCT
bands in the visible region. The surprising absence of sensitivi-
ty toward visible light in case of 3 and 4, however, indicated
that their strong absorptions might not aid in labilization of
the ReꢀCO bonds. We therefore proceeded to examine the
nature of the MLCT transitions with the aid of DFT and TDDFT
calculations to determine the differences in photosensitivity of
the structurally similar complexes 1–4.
Figure 6. Changes in the electronic absorption spectrum of 4 in dichlorome-
thane solution upon exposure to UV light (5 mWcmꢀ2). The inset displays
the kCO rate plot for complex 4.
In the initial step, DFT optimization of the structures of the
complexes was performed, starting from the X-ray coordinates.
The optimized structures of 1–4 agree well with respect to
bond lengths and angles as listed in Tables S4 and S5 (Sup-
porting Information). Next, TDDFT calculations were performed
to obtain the MO electron densities and the calculated elec-
tronic transitions (Table 2). The theoretical spectra of 1–4 agree
considerably well with experimental data. The MO contribu-
tions that make up the MLCT bands were closely examined,
along with the associated UV bands that were experimentally
observed to release CO in the case of the Re complexes. Close
scrutiny of Table 2 reveals that, in all cases, the lowest-energy
band corresponds to the transition from MOs with a strong
metal–CO bonding interaction to the LUMO, primarily consist-
ing the ligand–p* orbital. However, in all cases, the soft auxili-
ary ligand (Brꢀ or PPh3) also makes a significant contributions.
Examination of the MO electron densities of 1 (Table 2)
shows that the absorption at 586 nm (responsible for rapid CO
release and calculated as 616 nm) arises from a transition
(HOMOꢀ1 to LUMO) that is comprised of both MLCT and
halide!p* (XLCT) character. The HOMOꢀ1 level has 41%
p(MnꢀCO) bonding character, along with 37% Brꢀ bonding
contribution with the rest of the orbital densities in the p-
bonding ligand frame. Upon illumination, electron density is
transferred to the LUMO consisting of 70% p*MO of azpy
along with 8% Brꢀ and 18% p(MnꢀCO) bonding character
(Figure 7). Reduction in MnꢀCO p-backbonding in such a trans-
fer promotes rapid CO release. In case of 3, the lowest-energy
transition at 530 nm is again a HOMOꢀ1 to LUMO transition.
Here, the HOMOꢀ1 consists of 48% p(ReꢀCO) and 22% Brꢀ
bonding character, while the LUMO is composed of 63%
p*MO of azpy, along with 8% Brꢀ and 27% p(ReꢀCO) bonding
character. Despite such similarity in the nature of electronic
transition, CO is surprisingly not released from 3 upon illumina-
tion at 530 nm (calculated: 517 nm). Clearly, there exists anoth-
er pathway by which this energy is released by 3 without scis-
sion of the ReꢀCO bond. In rhenium–carbonyl complexes of a-
diimine and related ligands, stronger spin-orbit interactions
(than those with manganese) are known to lead to better
(Figure 6), and the rates of CO photorelease were similar. For
example, in dichloromethane, 3 and 4 exhibited kCO rates of
0.25ꢁ0.01 minꢀ1 (conc.: 1.26ꢂ10ꢀ4 m) and 0.21ꢁ0.01 minꢀ1
(conc.: 6.94ꢂ10ꢀ5 m), respectively.
Light-induced loss of CO from structurally similar bpy com-
plexes has been reported by several groups.[31] Close scrutiny
of the results reveals that the structurally similar fac-[MnBr-
(bpy)(CO)3] displays its most red-shifted MLCT band at 420 nm.
When we exposed this complex to 420 nm light, spectral
changes due to CO photorelease were also observed. The rate
of CO photorelease, however, was slower (1.21ꢁ0.01 minꢀ1
,
conc.: 1.15ꢂ10ꢀ4 m) than that noted with 1 (Figure S9, Sup-
porting Information). Faster CO release was observed when
this complex was irradiated with UV light (~300–325 nm). In di-
chloromethane, the kCO rate of fac-[MnBr(bpy)(CO)3] was found
to be 22.11ꢁ0.01 minꢀ1 (conc.: 1.17ꢂ10ꢀ4 m). It is therefore
evident that, much like 1, the bpy complex fac-[MnBr-
(bpy)(CO)3] also releases CO when irradiated with light corre-
sponding to the lmax of the low-energy MLCT band. In contrast,
the corresponding rhenium complex fac-[ReBr(bpy)(CO)3], with
an MLCT band at 391 nm, releases CO only upon exposure to
UV light (lmax ~300 nm), with kCO =0.23ꢁ0.02 minꢀ1 (conc.:
7.49ꢂ10ꢀ5 m) in dichloromethane.
Together, these CO photorelease studies indicate that 1–4
are efficient photoCORMs that release CO upon exposure to
light. All four complexes display relatively strong absorption
bands in the visible region of the spectrum (Figure 4). Howev-
er, Mn complexes 1 and 2 release CO upon exposure to low-
power visible light (lꢂ500 nm), while the corresponding Re
complexes (3 and 4) release CO only when exposed to UV
light (~300 nm). It is therefore evident that the presence of
strong MLCT bands in the visible region does not translate to
sensitivity of a designed photoCORM toward visible light. Labi-
lization of the CO ligand is expected only when significant
electron density is transferred from a molecular orbital (MO)
with significant metal–CO bonding contribution to a mostly
ligand-based MO.[25] Assessment of such contributions was
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1270

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Further evidence of CO release
Table 2. Calculated (TDDFT) energies (E), oscillator strengths (f), and nature of transitions[a] in the complexes.
1
from the low-lying MLCT transi-
E [nm]
f
Transition
tion, in the case of manganese–
carbonyl complexes, comes from
1
the loss of CO from fac-[MnBr-
616
396
354
350
348
336
331
326
324
320
0.0099492
0.0720846
0.0797801
0.1818711
0.0346971
0.0267677
0.0489395
0.0207783
0.0239514
0.0283438
p(Mn-CO)-p(Br)!p*(Py-Azo-Ph) (HOMOꢀ1!LUMO)
p(Mn-CO)-p(Br)!p*(Py-Azo-Ph)
p(Ph)!p*(Py-Azo-Ph)
(bpy)(CO)3] under 420 nm illumi-
nation. The HOMOꢀ1 to LUMO
p(Ph)-d(Mn)!p*(Py-Azo-Ph)
p(Br)-p(Mn-CO)!p*(Ph)-p(CO)-d(Mn)
p(Br)-p(Mn-CO)!p*(Py-Ph-Azo)
p(Pyr)!p*(Py-Azo-Ph)
transition of this complex is
comprised of both MLCT and
XLCT character. The HOMOꢀ1
p(Mn-CO)-p(Br)!p*(Py-Ph-Azo)
p(Mn-CO)!p*(Py-Ph)
level has 52% p(MnꢀCO) bond-
ing character, along with 28%
p(Br)-p(Mn-CO)!p*(Py-Ph-Azo)
Brꢀ bonding contribution, while
2
the LUMO consists of 74%
472
430
416
403
392
364
360
328
322
0.0134903
0.0410682
0.0233788
0.0198349
0.0143996
0.1836637
0.0870086
0.0122870
0.0217908
p(Mn-PPh3-CO)!p*(Py-Azo-Ph) (HOMOꢀ4!LUMO)
p*MO of bpy, 8% Brꢀ, and 17%
p(Mn-PPh3-CO)!p*(Py-Azo-Ph)
p(MnꢀCO) bonding character.
p(Mn-PPh3)-p(Ph)!p*(Py-Azo-Ph)
Transfer of electrons during this
p(Mn-PPh3)!p*(Py-Azo-Ph)
p(Mn-PPh3-CO)-p(Py)-p(Ph)!p*(Py-Azo-Ph)
p(Mn-PPh3-CO)-p(Ph)-p(Py)!p*(Py-Azo-Ph)
p(Mn-PPh3-CO)-p(Ph)-p(Mn-CO)!p*(Py-Azo-Ph)
p(Mn-PPh3-CO)-p(Ph)!d(Mn)-p*(PPh3)-p(CO)-p*(Py)
p(Mn-CO-PPh3)-p(Ph)!d(Mn)-p*(PPh3)-p(CO)-p*(Py)
transition also causes labilization
of the MnꢀCO bond, much like
1.
The significant role of the soft
auxiliary ligand is evident in the
3
relatively slower photorelease of
CO from 2. Replacement of Brꢀ
with PPh3 (a strong p acceptor)
517
462
380
358
346
325
291
289
282
0.0380878
0.0166441
0.2157011
0.1375852
0.0578752
0.0286452
0.0129495
0.0376803
0.0247467
p(Re-CO)-p(Br)!p*(Py-Azo-Ph) (HOMOꢀ1!LUMO)
p(Re-CO)-p(Ph)!p*(Py-Azo-Ph)
p(Re-CO)-p(Ph)-p(Py)!p*(Py-Azo-Ph)
p(Re-CO)-p(Br)-p(Py)!p*(Py-Azo-Ph)
p(Re-CO)-p(Br)-p(Ph)!p*(Py-Azo-Ph)
p(Py-Azo-Ph)!p*(Py-Azo-Ph)
shifts the lowest-energy transi-
tion (HOMOꢀ4!LUMO) of 2 to
520 nm. Upon illumination, elec-
p(Re-CO)-p(Br)!p*(Py-Azo-Ph)
tron density is transferred from
p(Re-CO)-p(Br)!p*(Py-Azo)
p(Re-CO)-p(Br)!p*(Py-Azo)-p*(Re-CO) (HOMOꢀ1!LUMO+1/LUMO+2)
the HOMOꢀ4 orbital, consisting
of 28% p(MnꢀCO) and 38%
4
PPh3 bonding character, to
462
448
424
410
404
390
374
372
350
275
0.0233700
0.0176888
0.0167423
0.0139487
0.0280426
0.0498241
0.0103896
0.2393981
0.0137366
0.0176668
p(Re-PPh3-CO)!p*(Py-Azo-Ph) (HOMOꢀ4!LUMO)
p(Re-CO-PPh3)-p(Ph)!p*(Py-Azo-Ph)
p(Re-PPh3-CO)!p*(Py-Azo-Ph)
LUMO, which has 61% p*MO of
azpy along with 18% p(MnꢀCO)
p(PPh3)!p*(Py-Azo-Ph)
and 18% PPh3 bonding charac-
p(Re-CO-PPh3)-p(Ph)!p*(Py-Azo-Ph)
p(Re-PPh3-CO)!p*(Py-Azo-Ph)
ter. Overall, this transition trans-
fers more electron density from
p(Re-CO-PPh3)-p(Ph)!p*(Py-Azo-Ph)
p(Re-CO-PPh3)-p(Ph)!p*(Py-Azo-Ph)
p(Re-PPh3-CO)!p*(Py-Azo-Ph)
the PPh3 ligand, which competes
with CO in terms of p-backbond-
p(Py-Azo-Ph)-p(Re-CO-PPh3)!p*(Py-Azo-Ph)
ing. As a consequence, weaken-
[a] Orbitals with greater contributions are listed first.
ing of the MnꢀCO bond is not
as severe as in the case of 1, in
which Brꢀ acts more as a donor
ligand.
1
3
mixing of the low-lying MLCT with MLCT, resulting in metal–
halide bond homolysis (instead of CO dissociation via a spin-
singlet process) upon illumination with light in the visible
region.[32,33] Time-resolved spectroscopic studies also indicated
that the halide ligand stays within the solvent cage and rapidly
recombines in most cases. Bond restoration depends on the
Finally, in the case of 3, absorption of ~300 nm light trans-
fers electron density from HOMOꢀ1 to LUMO+1 (consisting
of 69% p*MO of azpy, 26% p(ReꢀCO), and 5% Brꢀ) and
LUMO+2 (consisting of 31% p*MO of azpy, 63% p*(ReꢀCO)
and 4% Brꢀ, Figure 7). Despite intersystem crossing to their re-
spective non-dissociative triplet states, the potential energy
surfaces are somewhat dissociative, presumably due to avoid-
1
3
admixture of MLCT with the MLCT state of the highly coupled
radical pair which, in turn, depends on spin-orbit coupling
(prominent in 3). As a consequence, the energy absorbed
through this transition in the visible range is dissipated, with
no net CO dissociation from the excited species in these rheni-
um complexes.[34] In the present study, 3 therefore exhibits no
photorelease of CO when illuminated with 530 nm light.
1
ing crossing with a higher LF state along the reaction coordi-
nate.[35] As a consequence, minor ReꢀCO labilization is ob-
served when 3 is exposed to ~300 nm light, and the rate of
CO photorelease from 3 is ~500-fold slower than that noted
with 1 under similar conditions.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1271

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Experimental Section
[Mn(CO)5Br] and AgClO4·H2O were purchased from Alpha Aesar,
and [Re(CO)5Br] was procured from Strem Chemical, Inc. The lattice
water molecules of AgClO4·H2O was removed via trituration with
CH3CN several times prior to use. The ligand 2-phenylazopyridine
(azpy) was synthesized by following a reported procedure.[20] Sol-
vents were purified and/or dried by standard techniques prior to
use. 1H NMR spectra were recorded at 298 K on a Varian Unity
Inova 500 MHz instrument. A PerkinElmer Spectrum-One FT-IR was
employed to monitor IR spectra, while the UV/Vis spectra were ob-
tained with a Varian Cary 5000 UV/Vis-NIR spectrophotometer. Mi-
croanalyses (C, H, N) were performed using a PerkinElmer 2400 Ser-
ies II elemental analyzer. Horse heart myoglobin (Mb) was pur-
chased from Sigma–Aldrich and used as received.
Caution! Transition metal perchlorates should be prepared in small
quantities and handled with great caution, as metal perchlorates
may explode upon heating.
Synthesis of complexes
[MnBr(azpy)(CO)3] (1):
A batch of 100 mg (0.36 mmol) of
Figure 7. Calculated energy diagram of 1 and 3 (left to right). The most
prominent MOs involved with transitions under the band associated with
CO release and their compositions are shown.
[MnBr(CO)5] was mixed with 80 mg (0.44 mmol) azpy in 20 mL
CH2Cl2, and the reaction mixture was stirred for 24 h at room tem-
perature. During this time, the dark blue solution was covered with
aluminum foil to prevent exposure from ambient light. After 24 h,
the solvent was removed to obtain a dark blue solid, which was
washed thoroughly with hexanes. Dark blocks of 1 in good yield
(90 mg, 63%) were obtained through recrystallization by layering
hexanes over its CH2Cl2 solution. 1H NMR (500 MHz, CD3CN): d=
9.23 (d, 1H), 8.66 (d, 1H), 8.33 (t, 1H), 7.87 (d, 2H), 7.74 (t, 1H),
Conclusions
Together, results of the present work demonstrate that even
though two very structurally similar metal–carbonyl complexes
(1 and 3) exhibit strong MLCT bands in the visible region, the
rhenium congener fails to release CO upon illumination with
visible light. As a consequence, despite higher stability in bio-
logical media, 3 can hardly be employed as a photoCORM
under the control of visible/near-IR light. To date, most of the
photochemical studies on rhenium–carbonyl complexes with
a,a’-diimine ligands have been performed with UV light.[34] The
recent report on the use of [Re(bpy)(CO)3(thp)]+ as a photo-
CORM also employed UV light for CO release.[14] In general,
these rhenium(I) complexes are pale in color (yellow to light
orange) and exhibit no strong absorption in the visible
range.[36] We now show that although proper choice of ligands
in complexes like 3 could lead to strong absorption in the visi-
ble region (deep purple in color), CO release is not observed
when the complex is exposed to light of similar wavelengths.
The energy absorbed by 3 in this region is dissipated through
pathways that do not initiate CO release. It therefore appears
that the potential of rhenium–carbonyl complexes as photo-
CORMs in the visible region is rather limited relative to that of
the manganese congeners. Indeed, photoCORMs that exhibit
CO release under visible light have so far been centered
around elements of the first transition row (Mn and Fe).[10,12]
Our results also indicate that soft halide ligands such as Brꢀ aid
in red-shifting the MLCT bands of such complexes[23,25] and in
promoting faster CO release through mixing of the XLCT and
MLCT transitions. We anticipate that these findings will provide
helpful tips in the future quest for photoactive CO-donating
drugs that could be triggered by visible/near-IR light.
7.68 ppm (d, 3H); IR (KBr): n˜CO =2040, 1960, and 1940, n˜N
=
=
N
1370 cmꢀ1; UV/Vis (CH2Cl2), lmax(e)=330 (13000), 586 (3900); Anal.
calcd for C14H9N3O3BrMn: C 41.83, H 2.26, N 10.46, found: C 41.60,
H 2.41, N 10.52.
[Mn(azpy)(CO)3(PPh3)](ClO4) (2): A batch of 33 mg (0.16 mmol) of
AgClO4 was added to a solution of 50 mg (0.12 mmol) of [MnBr-
(azpy)(CO)3] in 10 mL of THF, and the resulting blue reaction mix-
ture was stirred for 3 h in the dark, resulting in a purple solution.
The precipitate of AgBr was then filtered with a wet Celite pad,
and the filtrate was evaporated to dryness. The solid residue was
washed with hexanes. Next, a solution of 66 mg (0.25 mmol) of
PPh3 in 10 mL of CH2Cl2 was added to the residue, and the reaction
mixture was stirred for 24 h. The reaction flask in all steps was cov-
ered with aluminum foil. Finally, the solvent was removed, and the
residue was washed thoroughly with benzene to obtain 2 as an
orange–red solid in moderate yield (38 mg, 45%): 1H NMR
(500 MHz, CDCl3): d=8.59 (m), 8.43 (m), 7.67 (m), 7.52 (m), 7.42
(m), 7.26 (m), 7.06 (m), and 5.30 ppm (m); IR (KBr): n˜CO =2045,
1980, and 1950, n˜N N =1370, n˜ClO4 =1090 cmꢀ1; UV/Vis (CH2Cl2):
=
lmax(e)=370 (11500), 520 (4050); Anal. calcd for C32H24N3O7PClMn:
C 56.20, H 3.54, N 6.15, found: C 55.95, H 3.62, N 6.10.
[ReBr(azpy)(CO)3] (3):
A mixture of 74 mg (0.18 mmol) of
[ReBr(CO)5] and 40 mg (0.22 mmol) of azpy in 30 mL of benzene
was stirred at reflux for 3 h. The volume of the deep-purple solu-
tion was then decreased to ~5 mL, and it was stored at 48C for
6 h. The resulting purple solid was collected by filtration and
washed thoroughly with hexanes. Block-shaped crystals were ob-
tained through recrystallization by layering hexanes over a CH2Cl2
1
solution to afford 3 in good yield (70 mg, 72%): H NMR (500 MHz,
CDCl3): d=9.08 (d, 1H), 8.63 (d, 1H), 8.25 (t, 1H), 7.97 (d, 2H), 7.64
(t, 2H), 7.59 ppm (t, 2H); IR (KBr): n˜CO =2030, 1930, and 1920, n˜N
=
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1272

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
N =1370 cmꢀ1; UV/Vis (CH2Cl2): lmax(e)=370 (12400), 530 (4600);
Anal. calcd for C14H9N3O3BrRe: C 31.55, H 1.70, N 7.89, found: C
31.60, H 1.57, N 7.82.
Photolysis experiment: For visible light irradiation, an IL410 illumi-
nation system (Electro FiberOptics Corporation; power: 10–
15 mWcmꢀ2) was used. The UV light source employed in this study
was a UV-Transilluminator (UVP Inc.) with peak intensity at 305 nm
(power: 5 mWcmꢀ2). Apparent rates of CO release (kCO) were fol-
lowed at an appropriate wavelength for each complexes, and
ln[concentration] versus time (T) plots were generated. The myo-
globin (Mb) assay was carried out following standard protocols.[15]
[Re(azpy)(CO)3(PPh3)](ClO4) (4): A mixture of 51 mg (0.25 mmol) of
AgClO4 and 100 mg (0.19 mmol) of [ReBr(azpy)(CO)3] was dissolved
in 10 mL of THF, and the purple solution was stirred at reflux for
3 h, at which point the color changed to dark orange. The solid
AgBr was then filtered on a wet Celite pad, and the filtrate was
evaporated to dryness. Next, the orange residue was dissolved in
10 mL of chloroform along with 65 mg (0.25 mmol) of PPh3, and
the reaction mixture was stirred at reflux for 8 h. The solvent was
then evaporated under reduced pressure, and the resulting solid
was dissolved in 5 mL of CH2Cl2 and subjected to column chroma-
tography (silica gel, 60–100 mesh). Initially, the column was eluted
with benzene (10 mLꢂ3) to remove any trace of the parent com-
pound. Finally, the deep orange–red band was eluted using a mix-
ture of benzene and CH3CN (25:2, v/v) in which a small amount of
(Et4N)ClO4 was dissolved. The eluate was evaporated to dryness
and recrystallized by layering hexanes over its solution in CH2Cl2.
Orange needles of 4 were obtained in moderate yield (85 mg,
DFT and TDDFT studies: Density functional theory (DFT) and time-
dependent density functional theory (TDDFT) studies were per-
formed with the aid of the PC-GAMESS program[39] using the
hybrid functionals PBE0 and PBE1PW91 for Mn and Re complexes,
respectively. Optimizations for the Mn atom were performed by
employing the LANL2DZ basis set in conjunction with effective
core potential (ECP). For the Re atom, a valence double zeta (cc-
pVDZ-PP) basis set was used. The Pople 6-311G* split-valence
triple-z basis set with polarization was used for Br, while for all
other atoms, the 6-31G* basis set was employed with valence
double-z polarization (VDZP). The X-ray crystal structure coordi-
nates of complexes 1–4 were used as a starting point for the gas-
phase geometry optimization of the low spin (S=0) ground states.
TDDFT was used to calculate the electronic transitions and associ-
ated energies. Transitions with oscillator strengths above 0.0099
were then taken for analysis. For calculations on 1–4, the 40
lowest-energy electronic excitations were calculated. For each Re
compound, solvent effects were added using the polarized contin-
uum model (PCM)[40] using EtOH as the solvent. The calculated mo-
lecular orbitals were visualized using MacMolPlt.[41]
1
55%): H NMR (500 MHz, CDCl3): d=8.65 (d, 1H), 8.57 (d, 1H), 8.49
(t, 1H), 7.68 (m, 6H), 7.55 (m, 4H), 7.40 (t, 3H), 7.28 (ir, 3H),
7.04 ppm (t, 5H); IR (KBr): n˜CO =2050, 1970, and 1940, n˜N N =1370,
=
n˜ClO4 =1090 cmꢀ1
; UV/Vis (CH2Cl2): lmax (e)=390 (12400), 460
(6100); Anal. calcd for C32H24N3O7PClRe: C 47.15, H 2.97, N 5.16,
found: C 47.25, H 2.78, N 5.12.
[MnBr(bpy)(CO)3]: A mixture of 100 mg (0.36 mmol) of [MnBr(CO)5]
and 56 mg (0.36 mmol) of bpy in 20 mL of benzene was stirred at
reflux for 2 h. The yellow-orange solution was then evaporated to
dryness, and the residue was washed thoroughly with hexanes.
The yellow solid was finally recrystallized by layering the hexanes
over its CH2Cl2 solution. After 4 days, block-shaped orange crystals
of [MnBr(bpy)(CO)3] were obtained in good yield (95 mg, 70%):
1H NMR (500 MHz, CDCl3): d=9.23 (1H), 8.34 (1H), 8.12 (1H),
7.62 ppm (2H); IR (KBr): n˜CO =2023, 1945, and 1925 cmꢀ1; UV/Vis
(CH2Cl2), lmax (e)=300 (10500), 420 (1100); Anal. calcd for
C13H8N2O3BrMn: C 41.59, H 2.13, N 7.47, found: C 41.83, H 2.12, N
7.44.
CCDC 977174, 977175, and 977176 contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from the Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
Financial support from the NSF (grant DMR-1105296) is gratefully
acknowledged.
[ReBr(bpy)(CO)3]: This complex was obtained in good yield
(93 mg, 75%) by following a similar procedure as for [MnBr-
(bpy)(CO)3] except with
Keywords: carbonyl ligands · drug design · manganese ·
photolysis · rhenium
a
longer reflux time (7 h): 1H NMR
(500 MHz, CDCl3): d=9.11 (d, 1H), 8.20 (d, 1H), 8.08 (t, 1H),
7.56 ppm (t, 2H); IR (KBr): n˜CO =2012, 1905, 1882 cmꢀ1; UV/Vis
(CH2Cl2): lmax(e)=300 (19400), 400 (3600); Anal. calcd for
C13H8N2O3BrRe: C 30.81, H 1.58, N 5.53, found: C 31.03, H 1.49, N
5.41.
[1] R. Tenhunen, H. S. Marver, R. Schmid, Proc. Natl. Acad. Sci. USA 1968, 61,
748–755.
[2] a) R. Motterlini, L. E. Otterbein, Nat. Rev. Drug Discovery 2010, 9, 728–
743; b) J. McDaid, K. Yamashita, A. Chora, R. ꢃllinger, T. B. Strom, X. C. Li,
F. H. Bach, M. P. Soares, FASEB J. 2005, 19, 458–460; c) R. Song, Z. Zhou,
P. K. M. Kim, R. A. Shapiro, F. Liu, C. Ferran, A. M. K. Choi, L. E. Otterbein,
J. Biol. Chem. 2004, 279, 44327–44334.
[3] S. W. Ryter, A. M. K. Choi, Korean J. Intern. Med. 2013, 28, 123–140.
[4] K. Sato, J. Balla, L. Otterbein, R. N. Smith, S. Brouard, Y. Lin, E. Csizmadia,
J. Sevigny, S. C. Robson, G. Vercellotti, A. M. Choi, F. H. Bach, M. P.
Soares, J. Immunol. 2001, 166, 4185–4194.
[5] B. E. Mann, R. Motterlini, Chem. Commun. 2007, 4197–4208.
[6] R. Alberto, R. Motterlini, Dalton Trans. 2007, 1651–1660.
[7] T. R. Johnson, B. E. Mann, J. E. Clark, R. Foresti, C. J. Green, R. Motterlini,
Angew. Chem. Int. Ed. 2003, 42, 3722–3729; Angew. Chem. 2003, 115,
3850–3858.
[8] a) C. C. Rom¼o, W. A. Blꢄtter, J. D. Seixas, G. J. L. Bernardes, Chem. Soc.
Rev. 2012, 41, 3571–3583; b) T. Santos-Silva, A. Mukhopadhyay, J. D.
Seixas, G. J. L. Bernardes, C. C. Rom¼o, M. J. Rom¼o, J. Am. Chem. Soc.
2011, 133, 1192–1195; c) T. R. Johnson, B. E. Mann, I. P. Teasdale, H.
Adams, R. Foresti, C. J. Green, R. Motterlini, Dalton Trans. 2007, 1500–
1508.
Crystallography: Single crystals of 1–4 were obtained by layering
hexanes over their solutions in CH2Cl2. Data were collected on
a Bruker APEX II single crystal X-ray diffractometer with graphite
monochromated MoKa radiation (l=0.71073 ꢁ) by w-scan tech-
nique in the range of 3ꢂ2qꢂ558 for complex 1, 3ꢂ2qꢂ578 for
complex 2, and 3ꢂ2qꢂ568 for complexes 3 and 4. All data were
corrected for Lorentz polarization and absorption.[37] The metal
atoms were located from the Patterson maps, and the rest of the
non-hydrogen atoms emerged from successive Fourier syntheses.
The structures were refined by the full-matrix least-squares proce-
dure on F2. All non-hydrogen atoms were refined anisotropically.
All hydrogen atoms were included in calculated positions. The ab-
sorption corrections were done using SADABS. Calculations were
performed using the SHELXTL ver. 6.14 software package.[38] Crys-
tallographic data are presented in Table S1 (Supporting Informa-
tion).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1273

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[9] R. Alberto, Eur. J. Inorg. Chem. 2009, 21–31.
[24] F. Zobi, L. Quaroni, G. Santoro, T. Zlateva, O. Blacque, B. Sarafimov, M. C.
Schaub, A. Y. Bogdanova, J. Med. Chem. 2013, 56, 6719–6731.
[25] M. A. Gonzalez, S. J. Carrington, N. L. Fry, J. L. Martinez, P. K. Mascharak,
Inorg. Chem. 2012, 51, 11930–11940.
[10] M. A. Gonzales, P. K. Mascharak, J. Inorg. Biochem. 2014, 133, 127-135.
[11] R. D. Rimmer, A. E. Pierri, P. C. Ford, Coord. Chem. Rev. 2012, 256, 1509–
1519.
[26] K. Koike, N. Okoshi, H. Hori, K. Takeuchi, O. Ishitani, H. Tsubaki, I. P. Clark,
M. W. George, F. P. A. Johnson, J. J. Turner, J. Am. Chem. Soc. 2002, 124,
11448–11455.
[27] A. Mostad, C. Romming, Acta Chem. Scand. 1971, 25, 3561–3568.
[28] S. J. Dougan, M. Melchart, A. Habtemariam, S. Parsons, P. J. Sadler, Inorg.
Chem. 2006, 45, 10882–10894.
[12] U. Schatzschneider, Inorg. Chim. Acta 2011, 374, 19–23.
[13] a) B. Arregui, B. Lꢅpez, M. G. Salom, F. Valero, C. Navarro, F. J. Fenoy,
Kidney Int. 2004, 65, 564–574; b) P. Koneru, C. W. Leffler, Am. J. Physiol
Heart Circ. Physiol. 2004, 286, H304–H309; c) R. Motterlini, J. E. Clark, R.
Foresti, P. Sarathchandra, B. E. Mann, C. J. Green, Circ. Res. 2002, 90,
E17–E24.
[29] B. J. Liddle, S. V. Lindeman, D. L. Reger, J. R. Gardinier, Inorg. Chem.
2007, 46, 8484–8486.
[14] A. E. Pierri, A. Pallaoro, G. Wu, P. C. Ford, J. Am. Chem. Soc. 2012, 134,
18197–18200.
[30] H. Tsubaki, S. Tohyama, K. Koike, H. Saitoh, O. Ishitani, Dalton Trans.
2005, 385–395.
[15] J. Niesel, A. Pinto, H. W. P. N’Dongo, K. Merz, I. Ott, R. Gust, U. Schatzsch-
neider, Chem. Commun. 2008, 1798–1800.
ˇ
[31] A. Vlcek, Jr., Coord. Chem. Rev. 2002, 230, 225–242.
[16] a) S. J. Carrington, I. Chakraborty, J. R. Alvarado, P. K. Mascharak, Inorg.
Chim. Acta 2013, 407, 121–125; b) M. A. Gonzalez, S. J. Carrington, I.
Chakraborty, M. M. Olmstead, P. K. Mascharak, Inorg. Chem. 2013, 52,
11320–11331.
[32] C. Daniel, D. Guillaumont, C. Ribbing, B. Minaev, J. Phys. Chem. A 1999,
103, 5766–5772.
[33] a) A. Vogler, H. Kunkely, Coord. Chem. Rev. 2000, 200–202, 991–1008;
b) D. J. Stufkens, Coord. Chem. Rev. 1990, 104, 39–112.
[17] a) C. Bischof, T. Joshi, A. Dimri, L. Spiccia, U. Schatzschneider, Inorg.
Chem. 2013, 52, 9297–9308; b) P. Govender, S. Pai, U. Schatzschneider,
G. S. Smith, Inorg. Chem. 2013, 52, 5470–5478; c) G. Dçrdelmann, T.
Meinhardt, T. Sowik, A. Krueger, U. Schatzschneider, Chem. Commun.
2012, 48, 11528–11530; d) P. C. Kunz, W. Huber, A. Rojas, U. Schatzsch-
neider, B. Spingler, Eur. J. Inorg. Chem. 2009, 5358–5366.
[18] J. D. Seixas, A. Mukhopadhyay, T. Santos-Silva, L. E. Otterbein, D. J. Gallo,
S. S. Rodrigues, B. H. Guerreiro, A. M. L. GonÅalves, N. Penacho, A. R.
Marques, A. C. Coelho, P. M. Reis, M. J. Rom¼o, C. C. Rom¼o, Dalton
Trans. 2013, 42, 5985–5998.
[34] a) S. Sato, A. Sekine, Y. Ohashi, O. Ishitani, A. M. Blanco-Rodriguez, A.
ˇ
Vlcek, Jr., T. Unno, K. Koike, Inorg. Chem. 2007, 46, 3531–3540; b) M.
Wrighton, D. L. Morse, J. Am. Chem. Soc. 1974, 96, 998–1003.
[35] I. R. Farrell, P. Matousek, M. Towrie, A. W. Parker, D. C. Grills, M. W.
George, A. Vlcek, Jr., Inorg. Chem. 2002, 41, 4318–4323.
[36] R. A. Kirgan, B. P. Sullivan, D. P. Rillema, Top. Curr. Chem. 2007, 281, 45–
100.
[37] A. C. T. North, D. C. Philips, F. S. Mathews, Acta Crystallogr. 1968, 24,
351–359.
ˇ
[38] G. M. Sheldrick, SHELXTL ver. 6.14, Bruker Analytical X-ray Systems, Mad-
ison, WI (USA), 2000.
[39] A. V. Nemukhin, B. L. Grigorenko, A. A. Granovsky, Moscow Univ. Chem.
Bull. 2004, 45, 75–102.
[40] S. Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 1981, 55, 117–120.
[41] M. P. Waller, H. Braun, N. Hojdis, M. Buhl, J. Chem. Theory Comput. 2007,
3, 2234–2242.
[19] W. Q. Zhang, A. C. Whitwood, I. J. S. Fairlamb, J. M. Lynam, Inorg. Chem.
2010, 49, 8941–8952.
[20] R. D. Rimmer, H. Richter, P. C. Ford, Inorg. Chem. 2010, 49, 1180–1185.
[21] D. E. Bikiel, E. G. Solveyra, F. D. Salvo, H. M. S. Milagre, M. N. Eberlin, R. S.
CorrÞa, J. Ellena, D. A. Estrin, F. Doctorovich, Inorg. Chem. 2011, 50,
2334–2345.
[22] F. Zobi, A. Degonda, M. C. Schaub, A. Y. Bogdanova, Inorg. Chem. 2010,
49, 7313–7322.
[23] S. J. Carrington, I. Chakraborty, P. K. Mascharak, Chem. Commun. 2013,
49, 11254–11256.
Received: February 3, 2014
Published online on April 23, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1266 – 1274 1274