Azadipyrromethene dye derivatives in coordination chemistry: The structure-property relationship in homoleptic metal(II) complexes

Article abstract of DOI:10.1021/ic300872m

As a chromophore closely related to dipyrromethene (DPM), the azadipyrromethene (ADPM) family has attracted much interest in the life sciences and optoelectronic fields. A high-yielding microwave-assisted synthesis is reported for new homoleptic complexes of cobalt(II), nickel(II), copper(II) and zinc(II) based on the tetrakis(p-methoxyphenyl)azadipyrromethene ligand 1b. These complexes are compared with other homoleptic complexes of the same metal(II) series based on the tetraphenylazadipyrromethene 1a and also with related BF₂⁺ chelates (Aza-BODIPYs 6a and 6b) for a better understanding of trends arising from substitution of the chelate and/or the electron-donating effect of the p-methoxy substituents. The electrochemical behavior of the new compounds 2b, 3b, and 5b in dichloromethane revealed two pseudoreversible reductions (2b, -1.09 and -1.25 V vs SCE; 3b, -1.05 and -1.29 V; 5b, -1.13 and -1.25 V) followed by a third irreversible process (2b, -1.78 V; 3b, -1.80 V; 5b, -1.77 V) along with two pseudoreversible oxidations (2b, 0.55 and 0.80 V; 3b, 0.56 and 0.80 V; 5b, 0.55 and 0.80 V) followed by two closely spaced irreversible processes (2b, 1.21 and 1.27 V; 3b, 1.21 and 1.28 V; 5b, 1.22 and 1.25 V). On its side, copper(II) homoleptic complex 4b revealed only one pseudoreversible reduction at -0.59 V followed by three irreversible processes at -0.95, -1.54, and -1.74 V, respectively. The oxidation behavior of this complex exhibited two pseudoreversible processes (0.55 and 0.82 V) and two irreversible processes (1.19 and 1.25 V). The redox processes are assigned and discussed in relation to their photophysical properties. X-ray structures for 1b and related copper(II) complex 2b are also discussed.

Full text of DOI:10.1021/ic300872m

Article
pubs.acs.org/IC
Azadipyrromethene Dye Derivatives in Coordination Chemistry: the
Structure−Property Relationship in Homoleptic Metal(II) Complexes
Andre Bessette,†,‡ Janaina G. Ferreira,† Martin Giguere,‡ Francis Belanger,‡ Denis Desilets,‡
́
̀
́
́
and Garry S. Hanan*,†
†Dep
Canada
‡Saint-Jean Photochimie Inc., 725 Trotter Street, Saint-Jean-sur-Richelieu, Queb
́ ́ ́ ́ ́
artement de Chimie, Universite de Montreal, Pavillon J.-A. Bombardier, 5155 Decelles Avenue, Montreal, Quebec H3T-2B1,
́
ec J3B 8J8, Canada
S
* Supporting Information
ABSTRACT: As a chromophore closely related to dipyrromethene (DPM),
the azadipyrromethene (ADPM) family has attracted much interest in the life
sciences and optoelectronic fields. A high-yielding microwave-assisted
synthesis is reported for new homoleptic complexes of cobalt(II), nickel(II),
copper(II) and zinc(II) based on the tetrakis(p-methoxyphenyl)-
azadipyrromethene ligand 1b. These complexes are compared with other
homoleptic complexes of the same metal(II) series based on the
+
tetraphenylazadipyrromethene 1a and also with related BF2 chelates (Aza-
BODIPYs 6a and 6b) for a better understanding of trends arising from
substitution of the chelate and/or the electron-donating effect of the p-
methoxy substituents. The electrochemical behavior of the new compounds
2b, 3b, and 5b in dichloromethane revealed two pseudoreversible reductions
(2b, −1.09 and −1.25 V vs SCE; 3b, −1.05 and −1.29 V; 5b, −1.13 and
−1.25 V) followed by a third irreversible process (2b, −1.78 V; 3b, −1.80 V; 5b, −1.77 V) along with two pseudoreversible
oxidations (2b, 0.55 and 0.80 V; 3b, 0.56 and 0.80 V; 5b, 0.55 and 0.80 V) followed by two closely spaced irreversible processes
(2b, 1.21 and 1.27 V; 3b, 1.21 and 1.28 V; 5b, 1.22 and 1.25 V). On its side, copper(II) homoleptic complex 4b revealed only
one pseudoreversible reduction at −0.59 V followed by three irreversible processes at −0.95, −1.54, and −1.74 V, respectively.
The oxidation behavior of this complex exhibited two pseudoreversible processes (0.55 and 0.82 V) and two irreversible
processes (1.19 and 1.25 V). The redox processes are assigned and discussed in relation to their photophysical properties. X-ray
structures for 1b and related copper(II) complex 2b are also discussed.
The synthesis of homoleptic complexes of the d10 metal ions
INTRODUCTION
■
zinc(II) and mercury(II) with various substituted tetraarylaza-
dipyrromethenes was recently reported by Gray and co-
workers.7 Subsequently, O’Shea et al. studied the spectroscopic
properties of cobalt(II), nickel(II), copper(II), and zinc(II)
chelates 2a−5a based on tetraphenylazadipyrromethene 1a
(Figure 2),8 66 years after their initial synthesis by Rogers.9
The synthesis of ADPM-based homoleptic complexes was
ameliorated by O’Shea et al. as the reaction time was decreased
from 1 day at room temperature (rt) in tetrahydrofuran (THF)
to about 1 h in hot n-butanol, although this procedure was not
high-yielding in every case. Herein, we report a new synthetic
approach using microwave-assisted heating and potassium tert-
butoxide (KOtBu) as a strong base to deprotonate the pyrrolic
unit. This optimization of conditions led to a more
straightforward, high-yielding, and reproducible synthesis on
the time scale of minutes. This procedure allowed us to prepare
new homoleptic four-coordinate complexes of cobalt(II),
nickel(II), copper(II), and zinc(II), 2b−5b, featuring the
Dipyrromethene (DPM) organic dyes have attracted attention
because of their practical applications, e.g., biological probes1
and light-harvesting2 system components. As a close relative of
this intense red-absorbing and near-infrared (NIR)-emitting
compound, azadipyrromethene (ADPM) offers an opportunity
to absorb and emit even further in the red and NIR regions,
respectively. Therefore, this family of chromophores is
attracting increased interest, especially in organometallic
chemistry.3 In fact, ADPM exhibits properties similar to those
of the DPM family including versatile methods of preparation,4
even for unsymmetrical derivatives.5 Consequently, fine-tuning
of the optical and steric properties becomes feasible for the
preparation of suitable photoactive materials for life sciences
and solar energy conversion, including photovoltaic applica-
tions.6 Unlike DPM, the ADPM dye has never been isolated in
its unsubstituted form, being mainly formed as tetraaryl
derivatives in the 3 and 5 positions by two distinct synthetic
methods.4 Figure 1 depicts both generic chromophores and
their classical boron adducts (BODIPY and Aza-BODIPY) with
the position numbering and corresponding nomenclature.
Received: April 28, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Synthesis of Complexes 2b−5b from ADPM
Ligand 1b Assisted by Microwave Irradiation
Figure 1. DPM, ADPM, and their corresponding boron adducts.
Figure 2. ADPM ligand 1a and complexes 2a−5a.
tetrakis(p-methoxyphenyl)-substituted ADPM ligand 1b also
reported by Rogers9 and absent from the scientific literature
since then (Scheme 1).
1
In addition to characterization by H and 13C NMR, high-
resolution mass spectrometry (HRMS), and elemental analysis,
electrochemistry has been used to establish the exact energy
associated with the highest occupied molecular orbital
(HOMO) and lowest unoccupied molecular orbital (LUMO)
levels of ligands 1a and 1b, complexes 2−5b, and Aza-
BODIPYs 6a and 6b (Figure 3). Because of an emergent area
of ADPM coordination chemistry, the effect of metal
complexation and also of substitution with electron-rich
methoxy groups on the electronic properties of this dye family
was still be lacking from an electrochemical point of view. In
order to establish a general trend, the photophysical character-
ization of those new electron-rich tetra-p-methoxy compounds
is also reported and compared to existing tetraphenylazadi-
pyrromethene derivatives in both dichloromethane (DCM,
CDCl3) and THF solution. Finally, X-ray structures of 1b and
2b are discussed in relation to existing results observed for
similar homoleptic metal(II) complexes in order to have insight
on the effect of the electron-rich tetrakis(p-methoxyphenyl)-
azadipyrromethene. All together, information presented herein
should permit an empirical evaluation of such chromophores
toward their incorporation in photovoltaic applications as
electron-donor or photosensitizer materials.1f,6k,l,10
Figure 3. Aza-BODIPYs 6a and 6b.
carried out under an ambient atmosphere. Solvents were removed
under reduced pressure using a rotary evaporator unless otherwise
stated.
Microwave reactions were performed on a Biotage Initiator 8
equipped with an autosampler in 20 mL flasks from the same
distributor. The degree of absorption was set to normal, with 2 min of
prestirring and a stirring rate of 600 rpm.
Nuclear magnetic resonance (NMR) spectra were recorded in
CDCl3 at rt on a Bruker AV400 spectrometer at 400 MHz for H
1
NMR and at 100 MHz for 13C NMR, otherwise stated. Chemical shifts
are reported in part per million (ppm) relative to residual solvent
protons (7.26 ppm for chloroform-d) and the carbon resonance of the
solvent (77.16 ppm for chloroform-d). Absorption and emission
spectra were measured in CH2Cl2 and THF at concentrations obeying
Beer−Lambert’s law at rt on a Cary 500i UV−vis−NIR spectropho-
tometer and a Cary Eclipse fluorescence spectrophotometer,
respectively.
EXPERIMENTAL SECTION
■
1. Materials and Instrumentation. Literature procedures were
used for the synthesis of compounds 1a, 1b, 2−5a, and 6a.4,8 Aza-
BODIPY 6b was obtained from Saint-Jean Photochemicals Inc. and
used as received. M(OAc)2·xH2O (M = Co, Ni, Cu, Zn; OAc =
acetate), potassium tert-butoxide (KOtBu), and solvents were obtained
commercially and used without further purification. Reactions were
High-resolution electrospray ionization mass spectrometry (HR-
ESIMS) was performed by liquid chromatography/mass spectrometry
with a time-of-flight detector (LC/MS TOF) from Agilent by infusion
of a solution in CH2Cl2 of compounds 2−5b directly into the source
by vacuum aspiration.
B
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Electrochemical measurements were carried out in argon-purged
CH2Cl2 at rt with a BAS CV50W multipurpose equipment interfaced
to a PC. The working electrode used was a glassy carbon electrode for
every compound. The counter electrode was a platinum wire, and the
pseudoreference electrode was a silver wire. The reference was set
using an internal 1 mM ferrocene/ferrocenium sample at 0.46 V vs
SCE in CH2Cl2. The concentration of the compounds was about 1
mM. Tetrabutylammonium hexafluorophosphate (TBAP) was used as
the supporting electrolyte, and its concentration was 0.10 M. Cyclic
voltammetry (CV) scans were obtained at scan rates of 50, 100, 200,
and 500 mV/s. For reversible processes, half-wave potentials (vs SCE)
from CV were used. To establish the potential of irreversible
processes, differential pulse voltammetry (DPV) experiments were
performed with a step rate of 4 mV, a pulse height of 50 mV, and a
frequency of 5 Hz. Criteria for reversibility were the separation of 60
mV between the cathodic and anodic peaks, the close-to-unity ratio of
the intensities of the cathodic and anodic currents, and the constancy
of the peak potential with changing scan rate.
Diffraction data were collected on a Bruker Microstar diffractometer
equipped with a Platinum 135 CCD detector, a Helios optics, a Kappa
goniometer, and Cu Kα radiation. The diffraction quality of the
crystals were checked several times, revealing in all cases poor
diffraction with a large amount of diffuse scattering, signaling extensive
crystal disorder. Cell refinement and data reduction were done using
APEX2.11 Absorption corrections were applied using SADABS.12
Structures were solved by direct methods using SHELXS97 and
refined on F2 by full-matrix least squares using SHELXL97.13 All non-
hydrogen atoms were refined anisotropically. Hydrogen atoms were
refined isotropically on calculated positions using a riding model. For
compound 1b, the highest residual difference peak with an electronic
density of 1.4 e/Å3 is essentially due to a minor spatial disorder of the
N2 atom (evaluated to 18%) and was not modeled in the final
structure. For the nickel(II), cobalt(II), and copper(II) complexes,
crystals showed high mosaicity and low diffraction, as well as high
decay of standard intensities during data collection (around 20%).
Because of the undesirable characteristics described above, the crystal
structures of the nickel, copper, and zinc complexes were not
completed. However, it was possible to determine details of the
connectivity and conformation from those complexes, which were
similar to those obtained for the distorted tetrahedral cobalt complex
2b. For this last complex, the structure exhibited high disorder, mainly
over the methoxyphenyl rings, which was modeled as two
components, with an average ratio of 0.7/0.3, and refined anisotropi-
cally.
concentrated in THF and recrystallized by the slow diffusion of
pentane. The solid was isolated by filtration and washed with pentane
(3 × 5 mL), which afforded a red-purple powder. Yield: 109 mg
1
(89%). H NMR (CDCl3, 400 MHz): δ −1.78 (d, J = 7.46 Hz, 8 H),
2.79 (s, 12 H), 3.44 (s, 12 H), 6.40 (d, J = 7.67 Hz, 8 H), 8.15 (br s, 8
H), 34.53 (br s, 8 H), 61.52 (br s, 4 H). HR-ESIMS. Calcd for
C72H60N6O8Ni ([M+]): m/z 1194.3821. Found: m/z 1194.3828. Elem
anal. Calcd: C, 72.31; H, 5.06; N, 7.03. Found: C, 72.30; H, 4.97; N,
7.08.
Bis[3,5-bis(4-methoxyphenyl)-1H-pyrrol-2-yl-3,5-bis(4-methoxy-
phenyl)-2-ylideneamine]copper(II) (4b). A suspension of compound
1b (118 mg, 0.207 mmol), copper(II) acetate monohydrate
[Cu(OAc)2·H2O; 21 mg, 0.10 mmol], and KOtBu 95% (25 mg,
0.21 mmol) in n-butanol (10 mL) was reacted in a microwave reactor
for 5 min under magnetic stirring. The reaction mixture was
evaporated to dryness, dissolved in CH2Cl2, and filtered through
Celite. The solvent was evaporated, and the resulting blue-purple
residue was concentrated in THF and recrystallized by the slow
diffusion of pentane. The solid was isolated by filtration and washed
with pentane (3 × 5 mL), which afforded a blue-purple powder. Yield:
114 mg (92%). HR-ESIMS. Calcd for C72H60N6O8Cu ([M+]): m/z
1199.3763. Found: m/z 1199.3734. Elem anal. Calcd: C, 72.01; H,
5.04; N, 7.00. Found: C, 71.79; H, 5.23; N, 6.83.
Bis[3,5-bis(4-methoxyphenyl)-1H-pyrrol-2-yl-3,5-bis(4-methoxy-
phenyl)-2-ylideneamine]zinc(II) (5b). A suspension of compound 1b
(142 mg, 0.249 mmol), zinc(II) acetate dihydrate [Zn(OAc)2·2H2O;
27 mg, 0.13 mmol], and 95% KOtBu (29 mg, 0.25 mmol) in n-butanol
(10 mL) was reacted in a microwave reactor for 5 min under magnetic
stirring. The reaction mixture was evaporated to dryness, dissolved in
CH2Cl2, and filtered through Celite. The solvent was evaporated, and
the resulting red-purple residue was concentrated in THF and
recrystallized by the slow diffusion of pentane. The solid was isolated
by filtration and washed with pentane (3 × 5 mL), which afforded a
1
blue-purple powder. Yield: 134 mg (89%). H NMR (CDCl3, 400
MHz): δ 3.40 (s, 12 H), 3.92 (s, 12 H), 6.57−6.66 (m, 12 H), 6.97 (d,
J = 8.78 Hz, 8 H), 7.44 (d, J = 8.78 Hz, 8 H), 7.89 (d, J = 8.78 Hz, 8
H). 13C NMR (CDCl3, 100 MHz): δ 55.01, 55.36, 72.21, 113.41,
113.42, 115.28, 126.05, 127.50, 128.31, 130.87, 143.93, 147.74, 159.28,
159.50, 160.45. HR-ESIMS. Calcd for C72H60N6O8Zn ([M+]): m/z
1200.3759. Found: m/z 1200.3724. Elem anal. Calcd: C, 71.90; H,
5.03; N, 6.99. Found: C, 71.73; H, 5.22; N, 6.76.
RESULTS AND DISCUSSION
■
Synthesis. Complexes 2a−5a were obtained following
literature procedures using ligand 1a.8 The new methodology
developed to access complexes 2b−5b consist of irradiating 2
equiv of 1b with 1 equiv of the corresponding M(OAc)2·xH2O
(M = Co, Ni, Cu, Zn; OAc = acetate) in the presence of 2
equiv of KOtBu at 140 °C in a microwave reactor for 5−15 min,
using n-butanol as the solvent. Microwave irradiation
accelerated the reaction while giving yields ranging from 89%
to quantitative. The solvent, n-butanol, is suitable for such
complexation because of its high boiling point, its stability
toward side reactions and decomposition, and the ease of
purification resulting from the insolubility of the product upon
formation. Recrystallization was performed for each complex by
the slow diffusion of pentane in a saturated solution of THF to
remove residual traces of ligand, after filtration of salts on Celite
in CH2Cl2. Dark-purple powders with tints of blue to red were
Experimental uncertainties are as follows: absorption maxima,
2
nm; molar absorption coefficient, 10%; emission maxima, 5 nm;
redox potentials, 10 mV.
2. Experimental Procedures and Data. Bis[3,5-bis(4-methox-
yphenyl)-1H-pyrrol-2-yl-3,5-bis(4-methoxyphenyl)-2-ylideneamine]-
cobalt(II) (2b). A suspension of compound 1b (116 mg, 0.203 mmol),
cobalt(II) acetate tetrahydrate [Co(OAc)2·4H2O; 25 mg, 0.10 mmol],
and KOtBu 95% (24 mg, 0.20 mmol) in n-butanol (10 mL) was
reacted in a microwave reactor for 15 min under magnetic stirring. The
reaction mixture was evaporated to dryness, dissolved in CH2Cl2, and
filtered through Celite. The solvent was evaporated to afford a blue-
purple powder. Yield: 123 mg (>99%). 1H NMR (CDCl3, 500 MHz):
δ 7.22 (br s), 7.84 (br s), 8.73 (br s), 15.02 (br s), 15.33 (br s), 63.66
(br s). HR-ESIMS. Calcd for C72H60N6O8Co ([M+]): m/z 1195.3799.
Found: m/z 1195.3781. Elem anal. Calcd: C, 72.29; H, 5.06; N, 7.03.
Found: C, 72.23; H, 5.07; N, 7.00. Crystals suitable for X-ray analysis
were obtained from double-layer crystallization in CH2Cl2 and
hexanes.
1
isolated. Elemental analysis, H and 13C NMR [except for the
Bis[3,5-bis(4-methoxyphenyl)-1H-pyrrol-2-yl-3,5-bis(4-methoxy-
phenyl)-2-ylideneamine]nickel(II) (3b). A suspension of compound
1b (117 mg, 0.206 mmol), nickel(II) acetate tetrahydrate [Ni-
(OAc)2·4H2O; 26 mg, 0.10 mmol], and KOtBu 95% (24 mg, 0.20
mmol) in n-butanol (10 mL) was reacted in a microwave reactor for
10 min under magnetic stirring. The reaction mixture was evaporated
to dryness, dissolved in CH2Cl2, and filtered through Celite. The
solvent was evaporated, and the resulting red-purple residue was
paramagnetic copper(II) complex 4b], and HRMS all
supported conversion to the corresponding homoleptic
complexes. The use of an aprotic solvent for MS gives the
M+ cation, while a proton source such as formic acid gives
mainly the MH+ species. Single crystals of the ligand 1b and
cobalt(II) complex 2b suitable for X-ray diffraction were
obtained from a double layer of DCM and hexanes.
C
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Electrochemical Data for ADPM Ligands 1a and 1b, Complexes 2−5b, and Aza-BODIPYs 6a and 6b
a
a
compound
chelate
E1/2(Ox)
E1/2(Red)
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
none
none
CoII
CoII
NiII
1.36 (137)
1.01 (91)
0.71 (61)
0.85 (97)
0.55 (115)
0.85 (106)
0.56 (77)
0.87 (86)
0.55 (82)
0.86 (98)
0.55 (86)
1.32 (122)
−0.78 (72)
−0.92 (106)
−0.93 (108)
−1.09 (109)
−0.89 (115)
−1.05 (86)
−0.46 (150)
−0.59 (106)
−0.97 (111)
−1.13 (82)
−0.40 (101)
−0.58 (107)
−1.45 (16)
−1.52 (193)
−1.10 (104)
−1.25 (114)
−1.15 (113)
−1.29 (91)
−0.80 (88)
b,
b,
b,
b,
b,
b,
c
c
c
c
c
c
b,c
1.30
1.21
1.21
1.19
1.22
1.52
1.10
1.15 (102)
0.80 (136)
1.15 (111)
b
b
b
b
,
,
,
,
c
c
c
c
b
b
,
,
c
c
1.27
1.28
1.25
1.25
−1.78
−1.80
c
d
NiII
0.80 (111)
e
CuII
CuII
ZnII
ZnII
1.18 (95)
0.82 (107)
1.15 (104)
0.80 (98)
−1.41 (112)
c
f
−0.95
−1.54
−1.11 (114)
−1.25 (88)
−1.19 (96)
−1.37 (92)
b,c
−1.77
c
d
+
BF2
1.45 (62)
b
,
c
b,c
+
BF2
1.34
0.95
a
Potentials are in volts vs SCE for CH2Cl2 deaerated solutions, 0.1 M TBAP, recorded at 25 1 °C at a sweep rate of 50 mV/s. The differences
b
c
d
between the cathodic and anodic peak potentials (mV) are given in parentheses. Irreversible. Determined by DPV. Obtained from CV before the
addition of ferrocene. Another pseudoreversible reduction process was observed at −1.60 (133) V (see the SI). An irreversible reduction at −1.74
V was also observed by DPV (see the SI).
e
f
Electrochemistry. In the perspective of incorporating
ADPM complexes in various photoactive systems, access to
information concerning the energy levels of HOMO and
LUMO becomes crucial in order to further fine-tune their
properties. For instance, we can easily consider that a too low-
lying LUMO level would not permit efficient electron transfer
in potential photosensitizers for dye-sensitized solar cells or
organic photovoltaic devices.2c,14 Similarly, the photophysical
mechanisms implied in fluorescent sensors based on such
materials might also be affected by improper energy levels; i.e.,
high-lying HOMO levels might oxidize in biological environ-
ments.1a While optical properties such as absorption and
emission can indicate the differences in the energy levels
between the HOMO and LUMO, electrochemistry is able to
further provide their exact position in the ground state. The
effect of metal substitution from the d7 CoII to the d10 ZnII in
ADPM homoleptic complexes 2−5b was compared to the
corresponding ligands 1a and 1b by CV and DPV techniques
(Table 1 and the Supporting Information, SI). These
techniques afford critical information concerning the electronic
processes occurring and exact energies associated with their
HOMO/LUMO states. To complete the study, Aza-BODIPYs
6a and 6b are also presented.
The molecular orbitals (MOs) implied in redox processes for
ADPM ligands 1a and 1b can be assigned by comparing the
electron density map obtained from molecular modeling
reported in the literature for various related ADPM and Aza-
BODIPY compounds. An in silico study of interest from Russo
et al. presents time-dependent density functional theory (TD-
DFT) calculations for Aza-BODIPY 6a and related −OMe
derivatives in proximal and distal positions (6c and 6d,
respectively; Figure 4).16 Their modelization for 6a showed a
The electrochemistry of ligands 1a and 1b gives an indication
of the difference in stability of the ligands due to the electron-
rich p-methoxyphenyl groups (see the SI). The irreversibility of
the redox processes for 1b may be the result of secondary
reactions of the oxidized electron-rich anisole groups. This is of
specific importance in the context of using derivatives based on
ligand 1b either for life science or light-harvesting system
components because they are more likely to decompose over
time in such environments.1a,14a,15 While ligand 1a presents two
“pseudoreversible” oxidations of one electron each [1.01 (91)
and 1.36 (137) V vs SCE] along with two other reductions
[−0.78 (72) and −1.45 (16) V; Table 1], the ligand 1b bearing
four p-methoxyphenyl groups shows only one pseudoreversible
oxidation at 0.71 (61) V followed by two irreversible peaks
(1.10 and 1.30 V). The presence of a third oxidation process in
ADPM ligand 1b compared to 1a is due to the electron-rich
methoxy substituent groups. The reduction processes of 1b are
“pseudoreversible” [−0.92 (106) and −1.52 (193) V] in a
fashionsimilar to the tetraphenylazadipyrromethene derivative,
while being at a more negative potential.
Figure 4. Related ADPM and Aza-BODIPY derivatives.
HOMO mainly based on the π system, implying the dipyrrolic
core and the four phenyl rings, but with a node on the nitrogen
bridge arising from C2 symmetry in the molecule. Furthermore,
the electron densities calculated for HOMO−1 and HOMO−2
were both mainly centered on the distal phenyl rings. In the
case of the LUMO, delocalization throughout the whole π
system was restored, including the nitrogen bridge, along with a
slightly higher contribution implying the BF2+ fragment. On the
basis of these observations, it appears reasonable to use Aza-
BODIPY 6a as a good model for the assignation of MOs from
ADPM ligands 1a and 1b, especially because the three
derivatives are symmetrical molecules. Our assignment is also
supported by two other in silico studies. One was reported in
D
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2011 by Kobayashi et al., presenting a direct comparison
between the fused-ring ADPM ligand 7 and its corresponding
Aza-BODIPY derivative 8 (Figure 4). Computational studies
revealed that both corresponding HOMOs and LUMOs have
very similar electronic charge distribution.6f Jacquemin et al.
lately reported analysis of the frontier orbitals of the Aza-
BODIPY derivative 6e bearing a dimethylamino electron-
donating group installed in the proximal position through an
extended π-conjugation system (Figure 4).6c Calculations for
HOMO−4 to HOMO and LUMO to LUMO+4 orbitals were
presented, and similarities are notable concerning the HOMO−
2, HOMO−1, HOMO, and LUMO electron density local-
ization previously reported by Russo et al. for 6a. While the
LUMO of 6e is centered on the dipyrrolic moiety like in 6a, the
LUMO+1 to LUMO+4 of 6e are centered on the appended
phenyl rings.
Scheme 2. Energy Diagram (in eV) of Compounds 1−6
Including the HOMO−LUMO Difference (ΔE) As
Calculated by Electrochemistry
Once converted into electronvolts (eV), the HOMO and
LUMO levels lie at −5.95 and −4.17 eV (ΔE = 1.78 eV),
respectively, for 1a, while they are at −5.66 and −4.03 eV (ΔE
= 1.63 eV) for 1b (Table 2). A look at Scheme 2 leads to a
Table 2. HOMO/LUMO Levels (in eV) As Determined by
Electrochemistry along with ΔE Obtained from Redox and
Optical Methods for ADPM Ligands 1a and 1b, Complexes
2−5b, and Aza-BODIPYs 6a and 6b
(62) V (Table 1), with the last one being limited by the
potential window of DCM (see the SI). At negative potential,
two “pseudoreversible” processes are present at −0.40 (101)
and −1.19 (96) V. In general, the same number of redox
processes is present compared to ADPM ligand 1a, although
they are all markedly stabilized by the presence of the Lewis
acid chelate. For Aza-BODIPY 6b bearing four p-methox-
yphenyl groups, the three oxidation processes at 0.95, 1.34, and
1.52 V are irreversible. The reduction processes are, however,
pseudoreversible [−0.58 (107) and −1.37 (92) V]. In a
comparison of Aza-BODIPY 6b to its ADPM ligand 1b, the
overall effect is stabilization of both the oxidation and reduction
a
a
b
c
compound
chelate
HOMO
LUMO
ΔERedox
ΔEOpt
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
none
none
CoII
CoII
NiII
−5.95
−5.66
−5.80
−5.49
−5.80
−5.50
−5.82
−5.49
−5.81
−5.49
−6.26
−4.17
−4.03
−4.02
−3.86
−4.06
−3.90
−4.49
−4.36
−3.97
−3.81
−4.54
−4.36
1.78
1.63
1.78
1.63
1.74
1.60
1.33
1.13
1.84
1.68
1.72
1.54
1.93
1.81
d
d
d
d
d
d
d
d
1.69
1.60
1.69
1.62
1.76
1.65
1.75
1.64
1.83
1.71
NiII
CuII
CuII
ZnII
ZnII
+
processes upon coordination of the BF2 chelate.
As mentioned previously in the assignation of the redox
processes of ADPM ligands 1a and 1b from the molecular
modeling of various Aza-BODIPYs,6c,f,16 the first oxidations
observed for both 6a and 6b are centered on the dipyrrolic core
with charge stabilization throughout the appended phenyls,
while the second and third (for 6b) ones are more likely to be
centered on the peripheral phenyl rings. For the reduction
processes, the first reduction is based on the pyrrolic moiety,
while peripheral phenyls are subsequently reduced.
+
BF2
+
BF2
−5.90
b
a
Reported in electronvolts. Energetic difference between HOMO and
c
LUMO obtained by electrochemistry in a DCM solution. Energetic
difference obtained from conversion of the emission wavelength in a
DCM solution into electronvolts using the formula E = hc/λ,
d
otherwise stated. Energetic difference qualitatively obtained from the
A closer look at the frontier orbitals of those Aza-BODIPY
derivatives shows that tetraphenyl-Aza-BODIPY 6a presents a
ligand-centered HOMO that is harder to oxidize than that of
the p-methoxyphenyl-substituted 6b by about 0.36 eV (−6.26
and −5.90 eV, respectively) and a ligand-centered LUMO
red end of the Gaussian peak of the λmax shoulder in a DCM solution and
converted in electronvolts using the same formula as that in footnote
c).
better understanding of the electron-rich p-methoxyphenyl
groups’ effect on the energy levels of ADPM derivatives. In fact,
the HOMO is easier to oxidize by about 0.29 eV in 1b
compared to 1a. This is an indication of the additional electron
density brought by the methoxy substituents. On the other
hand, the first reduction of 1b is about 0.14 eV more negative
because of destabilization of the antibonding orbital.
Combined, these effects decreased the energetic transition
(ΔE) between the HOMO and LUMO by 0.15 eV in 1b. In
other words, the absorption band of this dye should be
displaced to the red region because of a less energetic
transition, as will be discussed in the absorption spectroscopy
section.
+
including contributions from the BF2 chelate that reduces
more easily by about 0.18 eV (−4.54 and −4.36 eV,
respectively; Table 2). This correlates perfectly with the
observations made for 1a and 1b ligands arising from
stabilization of the bonding orbital and destabilization of the
antibonding orbital induced by the methoxy groups. In
+
addition, the presence of the BF2 electropositive fragment
leads to an overall stabilization of both the HOMO and LUMO
because of the Lewis base−acid interaction provided upon
coordination of the chelate. For the HOMO, this stabilization is
about 0.25 eV (−6.26 vs −5.95 eV for 6a vs 1a; −5.90 vs −5.66
eV for 6b vs 1b), while it is of about 0.35 eV for the LUMO
(−4.54 vs −4.17 eV for 6a vs 1a; −4.36 vs −4.03 eV for 6b vs
1b). As mentioned previously, this implies a contribution from
the BF2+ chelating agent in the first unoccupied MO of the Aza-
+
After coordination of the Lewis acidic BF2 fragment,
tetraphenyl-Aza-BODIPY 6a presents a one-electron “pseudor-
eversible” oxidation process at 1.32 (122) V and another at 1.45
E
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
BODIPYs. The computational study from Jacquemin et al. on
Aza-BODIPY derivatives tends to support this hypothesis
because they reported a contraction of the B−N bonds and a
strong elongation of the central C−N bonds of the central rings
in the S1 excited state compared to his ground state S0.6c
Overall, ΔE from redox measurements decreased to 1.72 eV for
6a and 1.54 eV for 6b, with a relative order that correlates with
the observed emission values (Tables 2 and 3).
Having a HOMO energy level difference ranging from 0.30 eV
for cobalt(II) complexes 2a and 2b to 0.33 eV for copper(II)
complexes 4a and 4b, they all respect the relative difference of
ADPM ligand 1a compared to 1b along with a very similar
difference (0.29 eV for the ligands). These observations point
toward a ligand-based HOMO, notwithstanding the slight
destabilization (about 0.15 eV) by the presence of two ADPM
ligands on the same homoleptic complex. This assignment of
the HOMO is also in accordance with the computational
models of the zinc(II) complex 5a reported by Gray et al.7 The
second oxidation process follows the exact same behavior, with
potentials ranging from 1.15 to 1.18 V for complexes based on
ADPM ligand 1a and between 0.80 and 0.82 V for the ones
based on ADPM ligand 1b. Therefore, this second pseudor-
eversible oxidation also seems to be ligand-based. For
complexes 2b−5b, the third irreversible oxidation process
ranging between 1.19 and 1.21 V and the fourth one (also
irreversible) ranging between 1.25 and 1.28 V are also pointing
toward ligand-based processes because they are very similar
along the metal(II) series.
While oxidation processes were slightly affected by variation
of the metal center throughout the series, the trend is
somewhat different for the LUMO (Scheme 2) and other
reduction processes (Table 2). The presence of cobalt(II)
induced stabilization of the first pseudoreversible reduction
process, representing the LUMO, of 0.15 eV for complex 2a
(−4.02 eV/−0.93 V) and 0.17 eV for complex 2b (−3.86 eV/−
1.09 V) compared to their respective ADPM ligands 1a (−4.17
eV/−0.78) and 1b (−4.03 eV/−0.92 V). For nickel(II)-based
complexes, this stabilization decreased to 0.11 and 0.13 eV for
3a (−4.06 eV/−0.89 V) and 3b (−3.90 eV/−1.05 V),
respectively. On their side, zinc(II) complexes are showing
0.20 and 0.22 eV of LUMO stabilization for 5a (−3.97 eV/−
0.97 V) and 5b (−3.81 eV/−1.13 V), respectively. However, all
complexes based on cobalt(II), nickel(II), and zinc(II)
presented a second pseudoreversible reduction process less
stable than the one of their corresponding ADPM ligand by
about 0.35 V for 1a derivatives and 0.25 V for the ones based
on 1b. All three complexes based on this last ligand present an
additional irreversible oxidation between −1.77 V for the
zinc(II) complex 5b and −1.80 V for the nickel(II) complex 3b.
On the basis of models of complex 5a by Gray et al. mentioned
previously,7 all reduction processes seem to be based on the
core of the ADPM ligand without significant contribution from
the metal(II) present. However, variation of the metal(II)
center seems to have an influence on the exact energy of the
LUMO, without a clear trend going from the d7 CoII to the d10
ZnII. Finally, the combined effect of a relatively constant
HOMO with a fluctuating LUMO leads to ΔERedox = 1.78 eV
for the cobalt(II) complex 2a compared to ΔERedox = 1.63 eV
for 2b. The ΔERedox values observed are exactly the same as
those found for their respective ADPM derivatives. For
nickel(II) complexes, slightly smaller ΔERedox values are
obtained (1.74 and 1.60 eV respectively for 3a and 3b), while
larger ones are obtained for zinc(II) complexes (1.84 and 1.68
eV respectively for 5a and 5b). Overall, the ΔERedox values
observed are following the same relative order as the ΔEOpt
ones obtained from absorption spectroscopy (Table 3).
Table 3. Compiled UV−Vis Absorption and Emission Data
for Ligand 1, Complexes 2−5, and Corresponding Aza-
BODIPYs 6 in DCM and THF Deaerated Solutions
a,
b
a c
−
absorption
emission
λmax near‑UV
nm (ε, ×103
compound chelate M−1 cm−1
, λmax red, nm λshoulder, nm
(ε, ×103
(ε, ×103
)
M−1cm−1
)
M−1cm−1
)
λem, nm
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
none
302 (41)
596 (46)
597 (47)
627 (52)
627 (54)
603 (68)
603 (68)
632 (77)
632 (73)
607 (58)
607 (56)
631 (67)
631 (58)
565 (55)
566 (57)
591 (64)
591 (71)
590 (88)
591 (86)
612 (88)
613 (90)
648 (80)
651 (84)
691 (79)
644
642
684
678
d
302 (42)
322 (40)
322 (41)
305 (64)
305 (64)
325 (66)
325 (63)
306 (62)
306 (60)
325 (63)
325 (54)
303 (57)
303 (59)
322 (59)
323 (63)
304 (66)
303 (65)
326 (63)
326 (64)
310 (25)
310 (26)
335 (19)
335 (19)
none
d
CoII
d
675 (41)
675 (40)
691 (54)
695 (51)
653 (49)
655 (47)
692 (50)
693 (57)
640 (51)
641 (51)
670 (55)
671 (58)
644 (56)
642 (56)
675 (58)
674 (59)
CoII
d
NiII
d
NiII
d
CuII
d
CuII
d
ZnII
d
ZnII
d
+
BF2
d
677
678
723
+
BF2
d
694 (81)
721
a
b
c
In CH2Cl2 unless otherwise stated. Acquisition at 293 K. Value
d
obtained from excitation at λmax red = 10 nm. In THF (for the entire
line).
When the redox behavior of homoleptic complexes 2−5b is
considered, it appears to change significantly compared to
ADPM ligands 1a and 1b and their corresponding boron
adducts 6a and 6b. The four main observations for the
organometallic complexes (Table 2 and Scheme 2) are as
follows: (i) access to one additional oxidation and one
reduction process for complexes 2b−5b; (ii) the presence of
two additional reduction processes upon coordination to
copper(II) for both 4a and 4b complexes; (iii) a general
destabilization of HOMOs to the same extent throughout the
metal(II) series; (iv) general stabilization of LUMOs to various
extents, except for copper(II) derivatives 4a and 4b.
Looking at the effect of variation of the metal center on the
oxidation potentials, only a slight difference is observed among
the complexes for the two first oxidation potentials. Referring
to the HOMO level, complexes based on 1a are all lying
between −5.80 and −5.82 eV (0.85−0.87 V), while the ones
based on 1b are between −5.49 and −5.50 eV (0.55−0.56 V).
The case of copper(II) complexes 4a and 4b is special
because the first reduction process leads to strong destabiliza-
tion (0.32 and 0.33 eV, respectively) compared to their
corresponding ADPM ligands. The LUMOs are therefore low-
lying at −4.49 eV (−0.46 V) and −4.36 eV (−0.59 V) for 4a
F
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
and 4b, respectively. De facto, the transitions between the
HOMO and LUMO levels of the two complexes are the lowest
observed with ΔERedox = 1.33 eV for 4a and ΔERedox = 1.13 eV
for 4b. Such low-lying LUMOs are attributable to the CuII/CuI
reduction process in a fashion similar to that of metal-
loporphyrin and copper(I) phenanthroline [Cu(NN)2]+
derivatives.17 In addition, this metal-based reduction leads to
the loss of one ADPM ligand from the formally anionic
complex. This explains the small pseudoreversible reduction
peak observed for complexes 4a and 4b in CV and DPV (see
the SI) around the same potentials as their corresponding
ADPM ligand, i.e., at −0.80 and −0.95 V, respectively. Two
other pseudoreversible processes of one electron each were
observed at −1.41 (112) and −1.60 (133) V for 4a. In the case
of 4b, these two reduction processes were irreversible (−1.54
and 1.74 V). On the basis of the related cobalt(II), nickel(II),
and zinc(II) complexes, these last two reductions are attributed
to the ligand.
Optical Properties. To complete our study of ADPM dye
derivatives in coordination chemistry, absorption and emission
spectra of the newly synthetized complexes 2b−5b and Aza-
BODIPY 6b based on ligand 1b were recorded (Figures 5 and
Figure 6. Emission spectra of 1b and 6b in DCM.
bridge of the ADPM. Typical of ADPM dyes, a second
absorption band is observed in the 300 nm region (λmax near‑UV
)
arising from variable orbital contributions, mainly π−π*
transitions, depending of the substituents present in the
chromophore.6c As mentioned by Jacquemin and co-workers,
these orbital contributions are hard to formally establish
without in silico simulations.6c,19 Also of interest, 1b offers a
molar absorptivity (ε) of 54000 M−1 cm−1 in THF, about 20%
more than any of the other p-methoxy-substituted or
tetraphenyldipyrromethenes. They are also on the same order
of magnitude as what was observed by Gresser and co-workers
for benzannulated ADPM derivatives that exhibit better π
conjugation.6g In fluorescence spectroscopy, 1a presents an
emission at 644 nm in DCM, which shifts negligibly in more
polar THF (642 nm). ADPM ligand 1b exhibited an emission
centered at 684 nm in DCM (Figure 6), with a larger
hypsochromic effect of 6 nm compared to 1a, bringing the
emission up to 678 nm in THF (see the SI). Emission peaks are
tailing out to approximately 850 nm in both solvents, which
make ADPM ligand 1b an interesting candidate for NIR
applications. Once converted into electronvolts, the optical
band gap (ΔEOpt) can be compared to the one obtained by
electrochemistry (ΔERedox) in order to evaluate whether the
trends are respected. While the values of the band gap do not
correspond perfectly between redox and optical measurements
(1.78 vs 1.93 eV, respectively, for 1a and 1.63 vs 1.81 eV for
1b), the relative order within each method is respected (Table
2).
The presence of two ligands brought into proximity by a
metallic center in homoleptic complexes 2−5 gives rise to
interesting behavior. As explained by O’Shea et al. for the
copper(II) complex 4a, the clear splitting of the longest
wavelength band with λ values of 565 and 640 nm in DCM
(slight bathochromic shift of 1 nm in THF for both λ values) is
indicative of the strain within the ligand.8 The same behavior is
observed for the copper(II) complex 4b bearing the four p-
methoxy groups, with λ values red-shifted to 591 and 670 nm in
DCM (slight bathochromic shift of 1 nm in THF for the
second λ value only). For complexes 2, 3, 5a, and 5b, only a
distinct red-shifted shoulder (λshoulder) of the main absorption
band λmax red is observed. In that last series of complexes, only a
slight shift is discernible between the λshoulder recorded in DCM
and THF, in the margin of error of the measurements.
Figure 5. Absorption spectra of ligand 1b, complexes 2b−5b, and Aza-
BODIPY 6b in DCM.
6, respectively). In a desire to standardize and complete the
various data available in the existing literature, data were
collected for ligand 1a, complexes 2a−5a, and Aza-BODIPY 6a
in both DCM and THF (Table 3). All together, this
information should lead to a better understanding of trends
arising from variation of the electronic density of the ligand
and/or coordinated center.
The presence of four electron-rich methoxy groups on 1b led
to a bathochromic shift of 37 nm in DCM and 31 nm in THF
compared to 1a. In fact, λmax in the red region for 1b was of 627
nm in both solvents, compared to 590 nm in DCM and 596 nm
in THF for 1a. Such a bathochromic shift follows the tendency
observed in the literature as the presence of two p-methoxy
substituents in the distal position gave a λmax at 607 nm and at
620 nm in the proximal position (in THF).7 A further
comparison can be made between ADPM ligand 1a and its
tetraphenyldipyrromethene derivative reported by Guy and
Jones.18 In this case, a bathochromic shift of 50 nm is observed
simply based on the presence of the nitrogen atom on the
G
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 7. (a) Thermal ellipsoid projection of the ligand 1b (ellipsoids are shown at 30% probability). (b) Space-filling model of the view of L along
the central plane, showing the tilt of the rings. The hydrogen atoms were omitted for clarity. Selected bond lengths (Å) and angles (deg): N1−C1
1.407(5), N1−C10 1.348(4), N2−C1 1.327(4), N2−C17 1.319(4), N3−C17 1.374(5), N3−C26 1.361(4); C1−N2−C17 126.5(3), N1−C1−N2
121.2(3), N2−C17−N3 122.3(3).
Complexes 2b−5b presented a bathochromic shift in DCM of
λmax red compared to their tetraphenyl counterpart ranging from
22 nm between zinc(II) complexes to 29 nm between
cobalt(II) ones. As mentioned previously, the molar
absorptivity of ligand 1b was higher than that for 1a, and this
trend transferred to the homoleptic complexes. Through the
series, metal(II) complexes based on ligand 1b exhibit a higher
molar absorptivity by roughly 9000 M−1 cm−1 compared to the
ones based on ADPM ligand 1a in a DCM solution, except for
zinc(II) complexes, where no difference is observed. The more
polar THF solution seems to yield the same overall trend but
with lower values, in general, than those in a DCM solution.
Both copper(II) complexes present an absorption spectrum in
the red region similar to their respective ligand and related
metal(II) complexes, which means that electronic transitions
observed in that absorption region possess mainly a cyanine-
like character, which implies π−π* transitions.6c This
observation further supports the idea that the first reduction
process observed by electrochemistry is metal-centered on the
CuII/CuI process. Similar to what was reported by Gray and
O’Shea about homoleptic metal(II) complexes bearing ADPM
derivatives, no emission was observed for the newly synthesized
complexes 2b−5b. Therefore, a comparison of the band-gap
values with redox measurements needs to be done based on an
optical band gap evaluated using the intersection of the end of
λshoulder with the baseline (edge absorption; Table 2). Again, a
direct comparison of values obtained by redox versus optical
measurements for a given compound shows that they do not
match perfectly. A systematic underestimation from the latter
method is observed, still giving values that are closer compared
to those obtained from the emissive compounds. However, the
relative order between given metal(II) complexes based on
ADPM ligand 1a vs 1b is consistent.
6b (81000 M−1 cm−1) are significantly increased compared to
their ADPM precursors 1a and 1b, a typical behavior in this
family of chromophore.20 In addition, the presence of four
electron-rich −OMe groups in 6b gave a 46 cm−1 bathochromic
shift in DCM to the sharp emission at 723 nm compared to
tetraphenyl 6a (677 nm). Similar emission values were
observed in THF and are consistent with the shift observed
in absorption and the tailing of the emission band out to 850
nm, in a fashion similar to that of ADPM ligand 1b, which
shows potential for NIR applications if photo- and chemo-
stability can be ensured in the given system. Concerning the
band gaps, ΔEOpt is systematically overestimated, as was found
for ADPM ligands 1a and 1b; however, the relative order
obtained is consistent for the two methods.
X-ray Diffraction. The slow diffusion of hexanes into a
CH2Cl2 solution at rt of 1b and 2b afforded single crystals for
X-ray crystallography. Both compounds crystallized in the
monoclinic space group P21/c (see the SI). The molecular
structure of the ligand 1b confirms the conjugated nature of the
azadipyrrole unit, with bond lengths for the bridging nitrogen
to both pyrrole rings of 1.330(4) and 1.319(4) Å for N2−C1
and N2−C17, respectively (Figure 7a). All other bond
distances are in the expected range for those found in
conjugated systems of this type of chromophore. As previously
observed for other ADPM derivatives,8 these structures show
two very planar central pyrrolic rings [the angle of intersection
between the two is negligible at 1.29(3)°] and four twisted p-
methoxyphenyl groups. Of interest, ADPM ligand 1b features
internal asymmetry relative to the pyrrolic rings in the solid
state, which is due to tautomerism for the pyrrole hydrogen.
The structure was therefore refined with a disordered hydrogen
atom on N1 and N3 atoms, leading to a ratio of about 1:2 in
favor of N3. In addition, the distance between the centroids of
the phenyl rings in position 3 of the pyrrolic ring is slightly
shorter compared to the ones in position 5 [6.29(5) and
7.12(5) Å, respectively]. Also, the tilt between the rings formed
by the azadipyrrole moiety and the phenyl rings varies from
24.62(9)° for distal positions to only 18.47(8)° for proximal
ones.
Aza-BODIPYs 6a and 6b absorb in the red region (λmax red
)
at 648 and 691 nm, respectively, in a DCM solution.
Measurements in a THF solution give rise to a red shift of 3
nm in both cases. The bathochromic shift induced by the
presence of the tetrakis(p-methoxyphenyl) substituents reached
43 nm, slightly more than that for the parent ADPM ligands.
The molar absorptivities of 6a (84000 M−1 cm−1 in THF) and
H
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
+
The crystal structure of the homoleptic cobalt(II) complex
BF2 chelates (Aza-BODIPY) for a better understanding of
trends arising from substitution of the chelate and/or electron-
donating effect of the p-methoxy substituents. Electrochemical
characterization revealed four main observations for the
organometallic complexes: (i) access to one additional
oxidation and one reduction process for complexes 2b−5b
compared to related ADPM ligand 1b; (ii) the presence of two
additional reduction processes upon coordination to copper(II)
for both 4a and 4b complexes; (iii) a general destabilization of
HOMOs to the same extent throughout the metal(II) series;
(iv) a general stabilization of LUMOs to various extents, except
for copper(II) derivatives 4a and 4b. These two copper(II)
complexes presented a metal-centered first reduction process
(CuII to CuI) that engenders decoordination of one ADPM
ligand and also a lower ΔERedox compared to related complexes.
Optical characterization further supports this hypothesis by
exhibiting similar absorption properties for copper(II) com-
plexes compared to cobalt(II), nickel(II), and zinc(II) ones and
their free ligands. Such behavior is indicative of a cyanine-like
character, implying mainly π−π* transitions. As might be
expected from electron-rich −OMe substituents, emission of1b
and Aza-BODIPY derivative 6b gave a bathochromic shift
compared to related tetraarylazadipyrromethene compounds
previously reported. X-ray structures for 1b and related
cobalt(II) complex 2b were also presented.
2b is shown in Figure 8. The two ADPM bidentate ligands 1b
ASSOCIATED CONTENT
■
S
* Supporting Information
HR-ESIMS for newly synthesized complexes 2b−5b, cyclic and
differential pulse voltammograms for ADPM 1a and 1b,
complexes 2−5b, and Aza-BODIPYs 6a and 6b, absorption
spectra in DCM for ADPM 1a, complexes 2a−5a, and Aza-
BODIPY 6a, absorption spectra in THF for ADPM 1a and 1b,
complexes 2−5b, and Aza-BODIPYs 6a and 6b, emission
spectra in DCM for 1a and 6a, emission spectra in THF for 1a,
1b, 6a, and 6b, crystal data and details of the structure
determination for 1b (CCDC 876955) and 2b (CCDC
876954) along with the ORTEP representation of 2b at the
50% probability level, and X-ray crystallographic data in CIF
format. This material is available free of charge via the Internet
at http://pubs.acs.org.
Figure 8. Thermal ellipsoid projection of cobalt(II) complex 2b (the
disorder over the methoxyphenyl rings, corresponding to 30% of the
occupancy factor, was omitted for clarity). (a) Ellipsoids are shown at
30% probability (the labeling and hydrogen atoms were omitted for
clarity). Selected bond lengths (Å) and angles (deg): Co−N1
1.975(3), Co−N3 1.987(3), Co−N4 1.992(3), Co−N6 1.975(2);
N1−Co−N4 106.55(12), N3−Co−N6 108.08(11), N1−Co−N6
128.68(11), N3 −Co-N4 128.22(11). (b) Space-filling model of the
view of Co along the N2−Co−N5 plane and N1−Co−N6, showing
the tilt of the rings. The hydrogen and disordered atoms were omitted
for clarity.
force a distorted tetrahedral geometry at the metal center, with
Co−N bond distances ranging from 1.975(2) Å for Co−N6 to
1.992(3) Å for Co−N4 and chelate angles of N1−Co−N3
[94.7(1)°] and N4−Co−N6 [94.5(1)°] being slightly different
from each other. Other bond distances and angles are generally
very similar to those found by Gray and O’Shea for related
complexes.7,8 As expected, the chelating ligand is planar, with
an angle between best-fit planes of 67°. The crystal structure of
2b also revealed that, upon coordination of the ligands, the
phenyl rings arrange in a clockwise fashion (Figure 8b),
although no significant π interaction was found in the crystal
packing. Again, this behavior compares with similar complexes
reported by Gray and O’Shea.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: garry.hanan@umontreal.ca. Tel: +1 514 340 5156.
Fax: +1 514 343 7586.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
G.S.H. thanks the Natural Sciences and Engineering Research
Council of Canada, the Centre for Self-Assembled Chemical
Structures, and the Universite de Montreal (UdeM) for
́ ́
financial support. A.B. thanks NSERC, FQRNT, and Saint-
Jean Photochemicals Inc. (sjpc.com) for a BMP-Innovation
grant. J.G.F. is thankful to the Canadian Postdoctoral Research
Fellowship Programme. The authors also thank Anne-
Catherine Bedard, Daniel Chartrand, and the Combinational
́
Chemistry, NMR, Mass Spectrometry, Elemental Analysis, and
X-ray laboratories of UdeM for their help.
CONCLUSIONS
■
In summary, we reported herein a high-yielding microwave-
assisted synthesis of four new homoleptic complexes of
cobalt(II), nickel(II), copper(II), and zinc(II) based on 1b.
These complexes were compared with already known ADPM
homoleptic complexes of the same metal(II) series and related
I
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(c) Jørgensen, M.; Norrman, K.; Gevorgyan, S. A.; Tromholt, T.;
Andreasen, B.; Krebs, F. C. Adv. Mater. 2012, 24, 580−612.
(16) Quartarolo, A. D.; Russo, N.; Sicilia, E. Chem.Eur. J. 2006, 12,
6797−6803.
REFERENCES
■
(1) (a) Boens, N.; Leen, V.; Dehaen, W. Chem. Soc. Rev. 2012, 41,
1130−1172. (b) Ulrich, G.; Ziessel, R.; Harriman, A. Angew. Chem.,
Int. Ed. 2008, 47, 1184−1201. (c) Smith, N. W.; Alonso, A.; Brown, C.
M.; Dzyuba, S. V. Biochem. Biophys. Res. Commun. 2010, 391, 1455−
1458. (d) Nierth, A.; Kobitski, A. Y.; Nienhaus, G. U.; Jaschke, A. J.
Am. Chem. Soc. 2010, 132, 2646−2654. (e) Buyukcakir, O.; Bozdemir,
O. A.; Kolemen, S.; Erbas, S.; Akkaya, E. U. Org. Lett. 2009, 11, 4644−
4647. (f) Nepomnyashchii, A. B.; Bard, A. J. Acc. Chem. Res. 2012,
DOI: 10.1021/ar200278b.
(2) (a) Ziessel, R.; Harriman, A. Chem. Commun. 2011, 47, 611−631.
(b) Iehl, J.; Nierengarten, J.-F.; Harriman, A.; Bura, T.; Ziessel, R. J.
Am. Chem. Soc. 2012, 134, 988−998. (c) Kolemen, S.; Bozdemir, O.
A.; Cakmak, Y.; Barin, G.; Erten-Ela, S.; Marszalek, M.; Yum, J.-H.;
Zakeeruddin, S. M.; Nazeeruddin, M. K.; Gratzel, M.; Akkaya, E. U.
Chem. Sci. 2011, 2, 949−954. (d) Bozdemir, O. A.; Yilmaz, M. D.;
Buyukcakir, O.; Siemiarczuk, A.; Tutas, M.; Akkaya, E. U. New J. Chem.
2010, 34, 151−155. (e) Rio, Y.; Seitz, W.; Gouloumis, A.; Vazquez, P.;
Sessler, J. L.; Guldi, D. M.; Torres, T. Chem.Eur. J. 2010, 16, 1929−
1940.
(17) (a) Smith, K. M. Porphyrins and Metalloporphyrins; Elsevier:
New York, 1975; p 910. (b) Armaroli, N. Chem. Soc. Rev. 2001, 30,
113−124. (c) Accorsi, G.; Armaroli, N.; Duhayon, C.; Saquet, A.;
Delavaux-Nicot, B.; Welter, R.; Moudam, O.; Holler, M.;
Nierengarten, J.-F. Eur. J. Inorg. Chem. 2010, 2010, 164−173.
(18) Guy, R.; Jones, R. Aust. J. Chem. 1966, 19, 1871−1885.
(19) (a) Jacquemin, D.; Planchat, A.; Adamo, C.; Mennucci, B. J.
Chem. Theory Comput. 2012, 8, 2359−2372. (b) Chibani, S.; Le
Guennic, B.; Charaf-Eddin, A.; Maury, O.; Andraud, C.; Jacquemin, D.
J. Chem. Theory Comput. 2012, 3303−3313.
(20) Gorman, A.; Killoran, J.; O’Shea, C.; Kenna, T.; Gallagher, W.
M.; O’Shea, D. F. J. Am. Chem. Soc. 2004, 126, 10619−10631.
(3) (a) Teets, T. S.; Partyka, D. V.; Esswein, A. J.; Updegraff, J. B.,
III; Zeller, M.; Hunter, A. D.; Gray, T. G. Inorg. Chem. 2007, 46,
6218−6220. (b) Partyka, D. V.; Deligonul, N.; Washington, M. P.;
Gray, T. G. Organometallics 2009, 28, 5837−5840. (c) Teets, T. S.;
Updegraff, J. B., III; Esswein, A. J.; Gray, T. G. Inorg. Chem. 2009, 48,
8134−8144.
(4) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891−4932.
(5) (a) Hall, M. J.; McDonnell, S. O.; Killoran, J.; O’Shea, D. F. J.
Org. Chem. 2005, 70, 5571−5578. (b) Yakubovskyi, V. P.; Shandura,
M. P.; Kovtun, Y. P. Synth. Commun. 2010, 40, 944−950. (c) Li, Y.;
Dolphin, D.; Patrick, B. O. Tetrahedron Lett. 2010, 51, 811−814.
(6) (a) Bellier, Q.; Dalier, F.; Jeanneau, E.; Maury, O.; Andraud, C.
New J. Chem. 2012, 36, 768−773. (b) Amin, A. N.; El-Khouly, M. E.;
Subbaiyan, N. K.; Zandler, M. E.; Fukuzumi, S.; D’Souza, F. Chem.
Commun. 2012, 48, 206−208. (c) Le Guennic, B.; Maury, O.;
Jacquemin, D. Phys. Chem. Chem. Phys. 2012, 14, 157−164.
(d) D’Souza, F.; Amin, A. N.; El-Khouly, M. E.; Subbaiyan, N. K.;
Zandler, M. E.; Fukuzumi, S. J. Am. Chem. Soc. 2011, 134, 654−664.
(e) Batat, P.; Cantuel, M.; Jonusauskas, G.; Scarpantonio, L.; Palma,
A.; O’Shea, D. F.; McClenaghan, N. D. J. Phys. Chem. A 2011, 115,
14034−14039. (f) Lu, H.; Shimizu, S.; Mack, J.; Shen, Z.; Kobayashi,
N. Chem.Asian J. 2011, 6, 1026−1037. (g) Gresser, R.; Hummert,
M.; Hartmann, H.; Leo, K.; Riede, M. Chem.Eur. J. 2011, 17, 2939−
́
2947. (h) Gao, L.; Senevirathna, W.; Sauve, G. v. Org. Lett. 2011, 13,
5354−5357. (i) Tasior, M.; Murtagh, J.; Frimannsson, D. O.;
McDonnell, S. O.; O’Shea, D. F. Org. Biomol. Chem. 2010, 8, 522−
525. (j) Yuan, M.; Yin, X.; Zheng, H.; Ouyang, C.; Zuo, Z.; Liu, H.; Li,
Y. Chem.Asian J. 2009, 4, 707−713. (k) Mueller, T.; Gresser, R.;
Leo, K.; Riede, M. Sol. Energy Mater. Sol. Cells 2012, 99, 176−181.
(l) Leblebici, S. Y.; Catane, L.; Barclay, D. E.; Olson, T.; Chen, T. L.;
Ma, B. ACS Appl. Mater. Interfaces 2011, 3, 4469−4474.
(7) Teets, T. S.; Partyka, D. V.; Updegraff, J. B.; Gray, T. G. Inorg.
Chem. 2008, 47, 2338−2346.
(8) Palma, A.; Gallagher, J. F.; Mueller-Bunz, H.; Wolowska, J.;
McInnes, E. J. L.; O’Shea, D. F. Dalton Trans. 2009, 273−279.
(9) Rogers, M. A. T. J. Chem. Soc. 1943, 590−596.
(10) (a) Rousseau, T.; Cravino, A.; Ripaud, E.; Leriche, P.; Rihn, S.;
De Nicola, A.; Ziessel, R.; Roncali, J. Chem. Commun. 2010, 46, 5082−
5084. (b) Kubo, Y.; Watanabe, K.; Nishiyabu, R.; Hata, R.; Murakami,
A.; Shoda, T.; Ota, H. Org. Lett. 2011, 13, 4574−4577.
(11) APEX2; Bruker AXS Inc.: Madison, WI, 2009.
(12) Sheldrick, G. M. SADABS, Bruker AXS Inc.: Madison, WI, 1996.
(13) SHELXTL, version 6.12; Bruker AXS Inc.: Madison, WI, 2001.
(14) (a) Mishra, A.; Bauerle, P. Angew. Chem., Int. Ed. 2012, 51,
̈
2020−2067. (b) Li, Y. Acc. Chem. Res. 2012, 45, 723−733.
(15) (a) Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.;
̂
Neagu-Plesu, R.; Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M. J.
Am. Chem. Soc. 2007, 130, 732−742. (b) Zeng, L.; Jiao, C.; Huang, X.;
Huang, K.-W.; Chin, W.-S.; Wu, J. Org. Lett. 2011, 13, 6026−6029.
J
dx.doi.org/10.1021/ic300872m | Inorg. Chem. XXXX, XXX, XXX−XXX