Synthesis, characterization, physicochemical, and photophysical studies of redox switchable NIR dye derived from a ruthenium-dioxolene-porphyrin system

Article abstract of DOI:10.1021/ic048805l

Newly synthesized semi-quinone derivatives of the ruthenium polypyridyl, covalently linked to a porphyrin core, show very high e values (59 000-83 500 M^-1cm^-1) for the absorption band in the near infrared (NIR) region of the spectrum. Further, complexes 1-4 show an interesting reversible electrochromic behavior as a function of the redox state of the coordinated dioxolene functionality, and a switching phenomenon between bleaching and the restoration of the NIR peak could be achieved electrochemically. Thus, complexes 1-4 could be ideal candidate materials for NIR-active electrochromic devices. Ultrafast studies on 1 and its mononuclear components, 5-(3,4-dihydroxyphenyl)-10,15,20-triphenyl-21H,23H-porphyrin (H ₂L₁) and Ru(bpy)₂(bsq)⁺, reveal that there is no electron or energy transfer from the porphyrin to the Ru(bpy) ₂sq⁺ (bpy is 2,2′-bipyridine and sq is the deprotonated species of a substituted semi-quinone fragment) fragment or vice versa in 1. The observed decrease in the luminescence quantum yield for 1 compared to that of H₂L₁ can be ascribed to the increased nonradiative pathway due to higher vibronic coupling because of the direct linkage of the metal center to the porphyrin moiety.

Full text of DOI:10.1021/ic048805l

Inorg. Chem. 2005, 44, 2414−2425
Synthesis, Characterization, Physicochemical, and Photophysical
Studies of Redox Switchable NIR Dye Derived from a
Ruthenium Dioxolene Porphyrin System
−
−
D. Amilan Jose, Atindra D. Shukla, D. Krishna Kumar, Bishwajit Ganguly, and Amitava Das*
Central Salt and Marine Chemicals Research Institute, BhaVnagar 364 002, Gujarat, India
G. Ramakrishna, Dipak K. Palit, and Hirendra N. Ghosh*
Radiation Chemistry and Chemical Dynamics DiVision, Bhabha Atomic Research Center,
Mumbai 400 085, India
Received August 29, 2004
Newly synthesized semi-quinone derivatives of the ruthenium polypyridyl, covalently linked to a porphyrin core,
show very high values (59 000
83 500 M-1cm-1) for the absorption band in the near infrared (NIR) region of the
spectrum. Further, complexes 1 4 show an interesting reversible electrochromic behavior as a function of the
redox state of the coordinated dioxolene functionality, and a switching phenomenon between bleaching and the
restoration of the NIR peak could be achieved electrochemically. Thus, complexes 1 4 could be ideal candidate
ꢀ
−
−
−
materials for NIR-active electrochromic devices. Ultrafast studies on 1 and its mononuclear components, 5-(3,4-
dihydroxyphenyl)-10,15,20-triphenyl-21H,23H-porphyrin (H2L1) and Ru(bpy)2(bsq)+, reveal that there is no electron
or energy transfer from the porphyrin to the Ru(bpy)2sq+ (bpy is 2,2
′-bipyridine and sq is the deprotonated species
of a substituted semi-quinone fragment) fragment or vice versa in 1. The observed decrease in the luminescence
quantum yield for 1 compared to that of H2L1 can be ascribed to the increased nonradiative pathway due to higher
vibronic coupling because of the direct linkage of the metal center to the porphyrin moiety.
Introduction
extended conjugated systems, and some conducting poly-
mers.2-6 However, studies in this area mostly involve discrete
NIR-active molecules, and thus, the possibility to achieve
on/off switching of the NIR absorption band as a function
of the redox state of the chromophore is limited.2,7-11 Lever
Molecular systems that absorb strongly in the near infrared
(NIR) region are of considerable interest to both chemists
and material scientists because of their potential in diverse
applications such as optical data processing, photodynamic
therapy, telecommunications, aerospace, military camouflage,
and so forth.1,2 Compounds that generally show strong
absorption in the NIR region include NiII-dithiolene com-
plexes, some mixed valence complexes, molecules with
(3) Mueller-Westerhof, U. T.; Vance, B.; Yoon, D. I. Tetrahedron 1991,
47, 909 and references therein.
(4) (a) Barthram, A. M.; Cleary, R. L.; Kowallick, R.; Ward, M. D. Chem.
Commun. 1998, 2695. (b) Can˜adas, J. G.; Meacham, A. P.; Peter, L.
M.; Ward, M. D. Angew. Chem., Int. Ed. 2003, 42, 3011.
(5) Blake, I. M.; Rees, L. H.; Claridge, T. D. W.; Anderson, H. L. Angew.
Chem., Int. Ed. 2000, 39, 1818 and references therein.
(6) (a) Cho, H. S.; Jeong, D. H.; Cho, S.; Kim, D.; Matsuzaki, Y.; Tanaka,
K.; Tsuda, A.; Osuka, A. J. Am. Chem. Soc. 2002, 124, 14643. (b)
Shukla, A. D.; Bajaj, H. C.; Das, A. Angew. Chem., Int. Ed. 2001, 40,
446.
(7) (a) Mortimer, R. J. Chem. Soc. ReV. 1997, 26, 147. (b) Rosseinsky,
D. R.; Mortimer, R. J. AdV. Mater. (Weinheim, Ger.) 2001, 13, 783.
(8) Haga, M.; Dodsworth, E. S.; Lever, A. B. P. Inorg. Chem. 1986, 25,
447.
* Authors to whom correspondence should be sent. E-mail: amitava@
csmcri.org (A.D.); hnghosh@magnum.barc.ernet.in (H.N.G.).
(1) (a) Emmelius, M.; Pawlowski, G.; Vollmann, H. W. Angew. Chem.,
Int. Ed. Engl. 1989, 28, 1445. (b) Sengupta, S.; Sadhukhan, S. K. J.
Chem. Soc., Perkin Trans. 1 2000, 4332. (c) Fabian, J.; Nakamura, I.
L.; Matsuoka, M. Chem. ReV. 1992, 92, 1197. (d) Tsuda, A.; Osuka,
A. Science 2001, 293, 79. (e) Infrared Optical Circuits and Compo-
nents: Design and Applications; Murphy, E. J., Ed.; Marcel Dekker:
New York, 1999. (f) Chandrasekhar, P.; Zay, B. J.; Birur, G. C.; Rawal,
S.; Pierson, E. A.; Kauder, L.; Swanson, T. AdV. Funct. Mater. 2002,
12, 95.
(9) Lever, A. B. P.; Masui, H.; Metcalfe, R. A.; Stufkens, D. H.;
Dodsworth, E. S.; Auburn, P. R. Coord. Chem. ReV. 1993, 125, 317.
(10) Schwab, P. F. H.; Diegoli, S.; Biancardo, M.; Bignozzi, C. A. Inorg.
Chem. 2003, 42, 6613.
(2) Ward, M. D.; McCleverty, J. A. J. Chem. Soc., Dalton Trans. 2002,
275 and references therein.
2414 Inorganic Chemistry, Vol. 44, No. 7, 2005
10.1021/ic048805l CCC: $30.25
© 2005 American Chemical Society
Published on Web 03/05/2005

Study of Redox Switchable NIR Dye
Chart 1
and co-workers were the first8 among different research
groups2,9-11 who could demonstrate that Ru(bpy)2(bsq)+
complexes (bpy is 2,2′-bipyridine and bsq is the deprotonated
species of the benzo semi-quinone form of 1,2-dihydroxy-
benzene) display an absorption band for the dπRu f π*bsq
metal-to-ligand charge-transfer (MLCT) transition in the NIR
region (ꢀ890 ) 9000 dm3mol-1cm-1), while their one-electron
reduced [Ru(bpy)2(cat); cat is the completely deprotonated
form of 1,2-dihydroxy benzene] or oxidized [Ru(bpy)2(q)2+;
q is its corresponding benzo quinone form] products do not
have any significant absorption in that region. Since then,
several molecules based on the Ru(bpy)2(sq)+ fragment (sq
is deprotonated species of the substituted semi-quinone
fragment) have been reported by various workers, which
show very high ꢀ values for the redox-active NIR band.
Recently, with an attempt to develop efficient and realistic
electrochromic devices, Ru(bpy)(sq)+ molecules, with high
ꢀ values for the NIR absorption band, were anchored on
nanocrystalline SnO2 and showed reversible electrochromic
behavior.10 In both reports, researchers did not observe any
photoinduced interfacial electron transfer between nano-
crystalline semiconductor (SnO2) particles and Ru(bpy)2(bsq)+
fragments. It is worth mentioning here that, apart from the
reversible electrochromic behavior, a higher molar absorp-
tivity value is also desirable to design an efficient and
sensitive sensor device. There are only a few examples for
inorganic metal complexes2,14a available in the literature1d,2,6,14
that show exceptionally high molar absorptivity values for
the absorption band in the NIR region; though, no redox
(12) The thermodynamic driving force (∆G°) for the electron transfer
process can be evaluated from the expression ∆G° ) -E00 + [Eox(D)
- Ered(A)], while for the TPP, E00 is ∼2.0 eV and the [Eox(D) - Ered(A)
]
value for TPP/TPP+ (D) and Ru(bpy)2sq/Ru(bpy)2cat (A) couples is
-0.16 eV. Thus, ∆G° > 1.8 eV and is thermodynamically highly
favorable.
(13) (a)Lee, S.-M.; Marcaccio, M.; McCleverty, J. A.; Ward, M. D. Chem.
Mater. 1998, 10, 3271. (b) Yoshida, K.; Oga, N.; Kadota, M.;
Ogasahara, Y.; Kubo, Y. Chem. Commun. 1992, 1114. (c) Takahashi,
K.; Gunji, A.; Yanagi, K.; Miki, M. J. Org. Chem. 1996, 61, 4784.
(d) Coe, B. J.; Houbrechts, S.; Asselberghs, I.; Persoons, A. Angew.
Chem., Int. Ed. 1999, 38, 366.
(11) (a) Shukla, A. D.; Das, A. Polyhedron, 2000, 19, 2605. (b) Ghosh,
D.; Shukla, A. D.; Banerjee, R.; Das, A. J. Chem. Soc. Dalton Trans.
2001, 1. (c) Shukla, A. D.; Whittle, B.; Bajaj, H. C. Inorg. Chim.
Acta 1999, 285, 89.
(14) (a) Fleischer, E. B.; Shachter, A. M. Inorg. Chem. 1991, 30, 3763.
(b) Deviprasad, G. R.; Keshavan, B.; D’Souza, F. J. Chem. Soc., Perkin
Trans. 1 1998, 3133. (c) Shukla, A. D.; Ganguly, B.; Dave, P. C.;
Samanta; A.; Das, A. Chem. Commun. 2002, 2648.
Inorganic Chemistry, Vol. 44, No. 7, 2005 2415

Jose et al.
temperature and at 77 K on a Bruker ESP-300 spectrometer.
Electrochemical experiments were performed on a CH-660A (USA)
electrochemical instrument; a conventional three-electrode cell
assembly was used. A saturated Ag/AgCl solution as a reference
and platinum as the working electrode were used for all of the
measurements. Ferrocene was added at the end of each experiment
as an internal standard, and all of the potentials are quoted versus
the ferrocene/ferrocenium (Fc/Fc+) couple. Spectra of the electro-
reduced/-oxidized species were recorded either using a spectro-
photometer (model: Shimadzu UV-3101 PC) under a dinitrogen
atmosphere or by recording in situ spectra of the electroreduced/
-oxidized species with a fiber-optic deep probe attached to an Ocean
Optics 2000 CCD spectrophotometer. Room-temperature emission
spectra were obtained using a Perkin-Elmer LS 50B luminescence
spectrofluorimeter fitted with a red-sensitive photomultiplier. The
fluorescence quantum yields (φf) were estimated in the appropriate
solvents (as specified), using the integrated emission intensity of
5,10,15,20-tetraphenyl porphyrin (H2TPP; φf ) 0.11 in benzene)
and ZnTPP (φf(S1) ) 0.033) as a reference (eq 1)17-19 for free base
porphyrins and its metalated derivatives, respectively. The emission
spectra recorded are not corrected, and hence, we rely on the relative
differences in the quantum yields only.
switching for the NIR absorption band for these complexes
was reported.1d,2,5,13
In our attempt to utilize the reversible on and off
phenomena of this redox-active NIR band for developing
redox as well as photoactive photochromic dye, we planned
to couple this NIR-active Ru(bpy)(sq)+ fragment to another
redox or photoactive center that can interact with the former
and induce a reversible bleaching of the NIR absorption band.
Preliminary calculations for the thermodynamic feasibility
of the photoinduced electron transfer from the donor por-
phyrin fragment to the acceptor Ru(bpy)2(bsq)+ unit (based
on the ground-state redox potential values of the porphyrin
and Ru(bpy)2(bsq)+ centers and the steady-state luminescence
property for the porphyrin center12) reveal that it is possible
to achieve photoinduced electron injection from the photo-
active porphyrin center to the covalently linked Ru(bpy)2(sq)+
core. This, coupled with the possibility of achieving a fast
charge recombination process in such a system, may make
Ru(bpy)2(bsq)+ and porphyrin fragments ideal candidates for
use in developing an efficient redox or photoactive photo-
chromic dye. On the basis of available information, we have
synthesized mono- and bis-dioxolene derivatives of porphyrin
(Chart 1) for the syntheses of new complexes (1-4; Chart
1) derived from the Ru(bpy)2(sq) core with an attempt to
develop a reversible redox/photoactive chromophoric dye.
The dioxolene derivatives of porphyrin used in this study
are either previously reported (H2L1) or newly synthesized
(H2L2-H2L4) (Chart 1)14a,b and are used for the synthesis of
complexes 1-4 (Chart 1). Part of the preliminary results were
published in one of our earlier communications.14c
φf ) φ′ (Isample/Istd)(Astd/Asample)(η2sample/η2
)
std
(1)
f
In eq 1, φf′ is the absolute quantum yield for the porphyrin
molecule used as a reference, Isample and Istd are the integrated
emission intensities, Asample and Astd are the absorbances at the
excitation wavelength, and η2
and η2 are the respective
sample
std
refractive indices.
Femtosecond transient absorption measurements were carried out
in a femtosecond tunable visible spectrometer, which has been
developed on the basis of a multipass amplified femtosecond Ti:
sapphire laser system from Avesta, Russia (1 kHz repetition rate
at 800 nm, 50 fs, 800 µJ/pulse), and described earlier.20 The 800
nm output pulse from the multipass amplifier is split into two parts
to generate pump and probe pulses. One part, with 200 µJ/pulse, is
frequency-doubled in BBO crystals to generate pump pulses at 400
nm. To generate visible probe pulses, about 3 µJ of the 800 nm
beam is focused onto a 1.5-mm-thick sapphire window. The
intensity of the 800 nm beam is adjusted by the iris size and neutral
density filters to obtain a stable white-light continuum in the 400
nm to over 1000 nm region. The probe pulses are split into the
signal and reference beams and are detected by two matched
photodiodes with variable gains. We have kept the spot sizes of
the pump and probe beams at the crossing point around 500 and
300 µm, respectively. The excitation energy density (at 400 nm)
was adjusted to ∼2500 µJ/cm2. The noise level of the white light
is about ∼0.5%, with occasional spikes that are due to oscillator
fluctuation. We have noticed that most of the laser noise is low-
frequency noise and can be eliminated by comparing the adjacent
probe laser pulses (pump blocked vs unblocked, using a mechanical
chopper). The typical noise in the measured absorbance change is
about <0.3%. The instrument response function was obtained by
fitting the rise time of the bleach of the sodium salt of meso-tetrakis-
(4-sulfonatophenyl)porphyrin at 710 nm, which has an instantaneous
response.
Experimental Section
I. Materials. H2L1,14a,b Ru(bpy)2Cl2‚2H2O,15 Ru(bpy)2(bsq)(PF6),8
and ferrocenium hexafluorophosphate (FcPF6)16 were prepared
following known methods. [But]4NPF6 was used as a background
electrolyte for the electrochemical studies and was recrystallized
from an ethanolic solution before use. Diisopropylamine, pyridine,
and acetonitrile were dried and distilled over CaH2. Pyrrole was
distilled under reduced pressure prior to use. All of the reactions
were performed under an argon atmosphere unless stated otherwise.
The water that was used was doubly distilled. All of the other
chemicals and solvents were obtained locally and were used as such
without further purification.
II. Physical Measurements. Microanalyses (C, H, and N) were
performed using a Perkin-Elmer 4100 elemental analyzer. Electron-
impact mass spectra were obtained on a Kratos MS9 instrument,
and fast-atom bombardment (FAB) measurements were carried out
on a VG-ZAB instrument, using 3-nitrobezyl alcohol as the matrix.
Fourier transform infrared spectra were recorded as KBr pellets
using a Perkin-Elmer Spectra GX 2000 spectrometer. Absorption
spectra were recorded using a Shimadzu UV-3101 PC spectropho-
tometer. 1H NMR spectra were recorded on a Bruker 200 MHz FT
NMR (model: Avance-DPX 200), using CDCl3/CD3CN as the
solvent and tetramethylsilane as an internal standard. Electron
paramagnetic resonance (EPR) spectra were recorded at room
(17) Hill, R. L.; Gouterman, M.; Ulman, A. Inorg. Chem. 1982, 21, 1450.
(18) Ravikanth, M.; Chandrashekhar, T. K. J. Photochem. Photobiol., A
1993, 74, 181.
(19) Beeston, R. F.; Aldridge, W. S.; Treadway, J. A.; Fitzgerald, M. C.;
DeGraff, B. A. E.; Stitzel, S. Inorg. Chem. 1998, 37, 4368.
(20) Ramakrishna, G.; Singh, A. K.; Palit, D. K.; Ghosh, H. N. J. Phys.
Chem. B. 2004, 108, 1701.
(15) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17,
3334.
(16) (a) Decurtins, S.; Felix, F.; Ferguson, J.; Gu¨del, H. U.; Lud, A. J.
Am. Chem. Soc. 1980, 102, 4102. (b) Connelly, N. G.; Geiger, W. E.
Chem. ReV. 1996, 96, 877.
2416 Inorganic Chemistry, Vol. 44, No. 7, 2005

Study of Redox Switchable NIR Dye
Table 1. UV-Vis and Electrochemical Data of Ligands and Their Zn Derivatives
dioxolene
porphyrin-based
E1/2 (∆E mV)a
λ
max (CHCl3)
E1/2 (∆E mV)a
E1/2 (∆E mV)a
compound
Me2L1
(∈ × 104 dm3 mol-1 cm-1
)
sq/cat
sq/q
419 (298), 515 (15.5), 551(6.9),
591 (4.6), 647 (4.0)
418 (301), 511 (2.6), 547 (19.6),
585 (3.4)
419 (283), 515 (13.8), 550 (5.7),
589 (4.0)
417 (308), 509 (5.5), 547 (37.5),
585 (6.1)
421 (252), 516 (10.9), 552 (6.0),
591 (3.4)
0.81 (142), -1.73 (150)
-2.06 (152)
0.5 (qr)
0.34 (qr)
0.47 (ir)
Me2L2
H2L1
0.65 (100), -1.86 (100)
0.74 (100), -1.59 (100)
H2L2
0.75 (100), -2.14 (120),
-2.36 (120)
0.74, -1.72 (181),
-2.05 (286)
0.7 (100), -1.85 (100)
0.29b
0.47 (qr)
0.47 (qr)
Me4L3
Me4L4
422 (283), 509 (2.2), 548 (16.2),
586 (3.2)
420 (280), 516 (13.9), 552 (6.2),
591 (4.2)
0.36 (qr)
b
H4L3
1.2, 0.62, -1.23, -2.26
-0.34 (qr)
-0.86(qr)
0.19,
0.35
c
H4L4
a Cyclic voltammograms were recorded in CH2Cl2 in 0.1 M [But]4NPF6. The potential values are reported (scan rate of 100 mV s-1) versus those of the
ferrocene/ferrocenium couple. Electrochemical measurements were made with Pt bead as the working electrode, Pt wire as the auxiliary electrode, and
Ag-AgCl as the reference electrode. b Values are quoted from the square wave voltammogram. c Insoluble in common organic solvents.
The geometries of the model biradical (singlet and triplet forms)
and quinonoid species were optimized following the unrestricted
Hartree-Fock UHF method without any symmetry constraint by
using the UPM3 Hamiltonian.21,22
flask. 3,4-dihydroxybenzaldehyde (3.45 g, 25 mM) was added to
this, the mixture was heated to ∼60 °C with vigorous stirring, and,
then, freshly distilled and dry pyrrole (7.0 mL 100 mM) was added
in a dropwise manner. The resulting mixture was allowed to reflux
for 1 h. Then, the heat was removed, and the mixture was allowed
to cool and was left overnight at room temperature. The solid was
separated and removed by filtration, and it was found to be
5,10,15,20-tetraphenyl-21H,23H-porphyrin (H2TPP). The filtrate
was collected in a 500 mL conical flask and evaporated slowly on
an oil bath, which yielded a dark-colored, sticky mass. This was
washed with hot water to remove the remaining propionic acid and
other undesired polymeric tars. Solid lumps obtained thereby were
dried and subjected to gravity chromatography for purification; an
ethyl acetate/methanol solvent mixture (97:3 v/v) as the eluent and
silica gel as the stationary phase were used. This resulted in the
desired product with a minor amount of H2TPP as an impurity and
was further purified by column chromatography on SiO2, using ethyl
acetate/n-hexane (80:20 v/v) as the eluent. The purified yield of
the product was 1.1 g (∼6.8%). Analytical and various spectro-
scopic data are similar to those observed for H2L1 obtained by
method 1.
III. Syntheses. A. H2L1. The synthesis of H2L1 was achieved
following two different methodologies.
III.A.1. Method 1. The bis O-methylated product (Me2L1) was
synthesized, as an intermediate, following a published procedure14a,b
with some modifications in the purification process. Pyrrole (7 mL,
100mM), benzaldehyde (7.6 mL, 75 mM), and 3,4-dimethoxybenz-
aldehyde (4.1 g, 25 mM) were used for the reaction. The crude
product was purified by column chromatography, using silica gel
(60-120 mesh) as the stationary phase and CHCl3/n-hexane (60:
40 v/v) as the eluent. The desired product was isolated as a second
fraction and was characterized by various analytical and spectro-
scopic techniques. Elem Anal. Calcd: C, 81.88; H, 5.04; N, 8.3.
1
Found: C, 81.1; H, 5.1; N, 8.0. FAB-MS: 674 (M+). H NMR
(CDCl3, ppm): δH 8.91 (d, 2H, J ) 4 Hz, Hpyrrole), 8.85 (d, 6H, J
) 4 Hz, Hpyrrole), 8.22 (d, 6H, J ) 8 Hz, Hphenyl), 7.78-7.71 (m,
11H, Hphenyl), 7.26 (d, 1H, J ) 8.0 Hz, Hsubstituted phenyl), 4.18 (s,
3H, Hpara-methoxy), 3.99 (s, 3H, Hmeta-methoxy), -2.77 (s, 2H, HNH).
IR (KBr; ν/cm-1): 3434 (-NH), 1026 (-OCH3), 798-702
(pyrrole, HCdC). The desired product, H2L1, was obtained through
demethylation from Me2L1 following the methodology reported
earlier,14a,b except the reaction temperature was maintained at ∼160
°C. The final yield for H2L1 is 1.48 g (9.2%) with respect to the
reactant pyrrole and different aldehydes used. Elem Anal. Calcd:
C, 81.71; H, 4.68; N, 8.66. Found: C, 79.9; H, 4.3; N, 8.7. FAB-
MS: 645 (M+). 1H NMR (CDCl3, ppm): δH 8.82 (d, 8H, J ) 4.8
Hz, Hpyrrole), 8.21 (d, 6H, J ) 8.0 Hz, Hphenyl), 7.76-7.73 (m, 11H,
III.B.1. Me4L3. Method 1. A total of 22.4 mL (0.32 M) of
pyrrole was mixed with 0.8 mL (7.9 mM) of benzaldehyde. The
mixture was degassed with argon for 30 min. Then, 0.068 mL of
trifluoroacetic acid (TFA) was added slowly with a syringe strictly
under an argon atmosphere while the color of the reaction mixture
changed to dark orange-brown. About 15 mL of CH2Cl2 was added
to this and was allowed to stir at room temperature for ∼45 min
while the progress of the reaction was monitored by thin-layer
chromatography (TLC). Then, the reaction mixture was diluted with
∼100 mL of CH2Cl2. The organic layer was washed with 0.1 N
NaOH solution followed by distilled water. This organic layer was
collected and dried over Na2SO4. After the removal of CH2Cl2, the
unreacted pyrrole was removed from the concentrated mass at a
pressure of ∼0 mbar and a temperature of 50-55 °C. The product,
meso-phenyldipyrromethane, was identified as having a pale yellow
color with Rf ) 0.46 [on a silica gel TLC plate with CHCl3/n-
hexane/triethylamine (68:30:2 v/v/v) as the eluent]. For purification,
this dark-yellow-colored, sticky mass was subjected to flash
chromatography with 230-400 mesh-size silica gel and a cyclo-
hexanes/ethyl acetate/triethylamine (79:20:1 v/v/v) mixture as the
eluent. Yield: 1.25 g (∼70%). 1H NMR (CDCl3, ppm): δH 7.9 (br
H
phenyl), 7.58 (d, 1H, 8.0 Hz, Hsubstituted phenyl), -2.77 (s, 2H, HNH).
IR (KBr pellet; ν/cm-1): 3448 (br, -NH and -OH), 1260 (C-
O), 798 (pyrrole HCdC). UV-vis and E1/2 values are summarized
in Table 1.
III.A.2. Method 2. Benzaldehyde (7.6 mL, 75 mM) was
dissolved in ∼200 mL of propionic acid in a two-neck round-bottom
(21) Reports on the calculations of porphyrin and its related isomers suggest
that the UHF method seems to be a better choice than RHF for
porphyrin systems. (Reynolds, C. H. J. Org. Chem. 1988, 53, 6061).
Therefore, all of the geometries have been completely optimized
without any symmetry constraints by using the UPM3 Hamiltonian.
(22) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209.
Inorganic Chemistry, Vol. 44, No. 7, 2005 2417

Jose et al.
s, 2H, HNH), 7.35-7.19 (m, 5H, Hphenyl), 6.7 (d, 2H, Hpyrrole), 6.17
(d, 2H, Hpyrrole), 5.9 (t, 2H, Hpyrrole), 5.48 (s, 1H, Hmeso-H).
redissolved in a minimum amount of CHCl3 and, then, subjected
to column chromatography, using silica gel as the stationary phase
and a CHCl3/methanol (95:5 v/v) mixture as the eluent. The first
major deep-purple band was found to be the desired product.
Yield: 0.130 g (90%). The isolated product was characterized by
various analytical and spectroscopic techniques. Elem Anal.
Calcd: C, 74.85; H, 4.37; N, 7.59. Found: C, 74.3; H, 4.2; N, 7.2.
FAB-MS: 737 (M+). 1H NMR (CDCl3, ppm): δH 9.02-8.93 (m,
8H, Hpyrrole), 8.2 (d, 6H, J ) 8.0 Hz, Hphenyl), 7.76-7.48 (m, 11H,
A total of 1.6 g (7.2 mM) of meso-phenyldipyrromethane and
1.2 g (7.2 mM) of 3,4-dimethoxybenzaldehyde were dissolved in
∼80 mL of degassed dichloromethane. The solution was stirred in
an inert atmosphere for 30 min. To this, 2 to 3 drops of TFA were
added as a catalyst. The mixture was stirred vigorously in the dark
for 15 h, a 4 mol equiv of chloranil (∼7.0 g) was added, and the
mixture was heated to reflux for 1 h. Then, it was cooled to room
temperature and filtered. The filtrate was washed with 0.1N NaOH
solution followed by distilled water to remove TFA. The organic
layer was collected and dried. The desired product was purified by
column chromatography, using silica gel (100-200 mesh) as the
stationary phase and CHCl3/n-hexane (80:20 v/v) as the eluent. The
major purple fraction was found to contain the desired product.
Yield: 0.200 g (∼7.5%). Elem Anal. Calcd: C, 78.47; H, 5.17; N,
H
H
phenyl), 7.23 (d, 1H, J ) 8.0 Hz, Hsubstituted phenyl), 4.09 (s, 3H,
para methoxy), 3.90 (s, 3H, Hmeta methoxy). IR (KBr pellet; ν/cm-1):
1026 (-OCH3), 997 (C-O), 796 (pyrrole, CdC). The UV-vis
spectral data and E1/2 values are given in Table 1.
III.E. H2L2. This metalated porphyrin was synthesized following
a synthetic methodology similar to that adopted for Me2L2. The
only exception is that 1.4 equiv (0.116 g, 0.52 mM) of Zn(CH3-
COO)2‚2H2O was used for the reaction with 0.243 g (0.37 mM) of
H2L1. Yield: 0.244 g (93%). Elem Anal. Calcd: C, 74.44; H, 3.95;
N, 7.89. Found: C, 73.9; H, 4.1; N, 7.8. FAB-MS: 709 (M+). 1H
NMR (CD2Cl2, ppm): δH 9.03-9.01 (2H, Hpyrrole), 8.94-8.92 (m,
6H, Hpyrrole), 8.21 (d, 6H, J ) 7.4 Hz, Hphenyl), 7.78-7.53 (m, 11H,
1
7.62. Found: C, 78.6; H, 5.0; N, 7.5. FAB-MS: 734 (M+). H
NMR (CDCl3, ppm): δH 8.91 (d, 4H, J ) 4.0 Hz, Hpyrrole), 8.85
(d, 4H, J ) 4.0 Hz, Hpyrrole), 8.22 (d, 4H, J ) 8.0 Hz, Hphenyl),
7.78-7.71 (m, 10H, Hphenyl), 7.22 (s, 2H, Hsubstituted phenyl), 4.18 (s,
6H, Hpara-methoxy), 3.99 (s, 6H, Hmeta,methoxy), -2.75 (s, 2H, HNH).
IR (KBr pellet; ν/cm-1): 3443 (br, -NH), 1510 (aromatic CdC),
1259 (-OCH3), 797-703 (pyrrole, HCdC). The UV-vis spectral
data and E1/2 values are presented in Table 1.
Hphenyl), 7.22 (d, 1H, J ) 8.0 Hz, Hsubstituted phenyl). IR (KBr pellet;
ν/cm-1): 3428 (br, -OH), 995 (C-O), 795 (pyrrole, CdC). The
UV-vis spectral data and E1/2 values are given in Table 1.
III.F. Me4L4. This was synthesized by reacting Zn(CH3COO)2‚
2H2O (0.062 g, 0.28 mM) and Me4L3 (0.070 g, 0.095 mM),
following the same procedure as that adopted for the preparation
of Me2L2. Yield: 0.130 g (90%). Elem Anal. Calcd: C, 72.23; H,
4.55; N, 7.02. Found: C, 71.6; H, 4.5; N, 6.8%. FAB-MS: 797
III.B.2. Method 2. In this method, the procedure that was
adopted was the same as that mentioned in method 1 for the
synthesis of Me2L1, except the molar ratio for benzaldehyde (5.07
mL, 100 mM), 3,4-dimethoxybenzaldehyde (8.3 g, 100 mM), and
pyrrole (7.0 mL, 100mM) used for the reaction was different.
During the chromatographic separation, a third fraction was
collected. This fraction was found to be the mixture of the cis {5,10-
(bisphenyl)-15,20-[bis-(3,4-dimethoxy)]-21H,23H-porphyrin} and
desired trans {5,15-(bisphenyl)-10,20-[bis-(3,4-dimethoxy)]-21H,23H-
porphyrin} forms. This mixture was further purified by column
chromatography, using a silica gel column (column dimensions:
length 75 cm and diameter 4 cm) and CHCl3/n-hexane (95:5 v/v)
as the eluent to start with. The polarity of the eluent mixture was
raised by slowly decreasing the proportion for n-hexane. Eventually,
the desired trans derivative was eluted with pure CHCl3. The yield
of the pure product was 0.400 g (∼2.2%). The Rf value and other
analytical data for this were found to be similar to those for Me4L3
obtained by following method 1.
1
(M+). H NMR (CDCl3, ppm): δH 9.02-9.0 (4H, J ) 4.8 Hz,
Hpyrrole), 8.96-8.94 (4H, J ) 4.6 Hz, Hpyrrole), 8.22 (d, 4H, J ) 8.0
Hz, Hphenyl ring), 7.79-7.76 (m, 10H, Hphenyl ring), 7.26 (d, 2H, J )
8.0 Hz, Hsubstituted phenyl), 4.18 (s, 6H, Hpara methoxy), 3.98 (s, 6H,
H
meta methoxy). IR (KBr pellet; ν/cm-1): 1027 (-OCH3), 1001 (C-
O), 797 (pyrrole, CdC). The UV-vis spectral data and E1/2 values
are presented in Table 1.
III.G. H4L4. This compound was synthesized following the
above-mentioned procedure, resulting in a yield of ∼85%, by
reacting H4L3 (0.068 g, 0.100 mM) and Zn(CH3COO)2‚2H2O (0.045
g, 0.2 mM). However, this crude product could not be purified using
column chromatography because of its limited solubility in common
organic solvents. The crude product was washed thoroughly with
CHCl3 to remove any unreacted H4L3 and was used as such for
further reactions. Yield: 85% (0.060 g). Elem Anal. Calcd: C,
71.21; H, 3.77; N, 7.55. Found: C, 70.8; H, 4.0; N, 7.4. FAB-
MS: 741 (M+). IR (KBr pellet; ν/cm-1): 3429 (br, -OH), 995
(C-O), 797 (pyrrole, CdC). The UV-vis and redox potential
values are provided in Table 1.
III.H. Complex 1. To an ethanolic solution of ligand H2L1 (0.300
g, 0.464 mM) was added ∼2.1 equiv of KOH (0.054 g), and the
reaction mixture was allowed to warm to ∼60 °C under an argon
atmosphere. Then, Ru(bpy)2Cl2‚2H2O (0.248 g, 0.48 mM) was
added to it. The resulting mixture was refluxed for 5 h in an inert
atmosphere and was allowed to cool to room temperature. FcPF6
(0.154 g, 0.046 mM) was added to this, and the mixture was left
exposed to the air with stirring at room temperature for 30 min.
After that, an excess aqueous solution of KPF6 was added, and the
volume was reduced in vacuo to give a dark-brown precipitate.
This was filtered off, washed with cold water, diethyl ether, and
air dried. Column chromatography with a 100-200 mesh-size silica
column and CH3CN/aq‚NH4PF6 (98:2 v/v) as the eluent gave the
desired product as the first major band. This was dried under
reduced pressure and redissolved in CH2Cl2, and excess hexfluoro-
III.C. H4L3. The procedure adopted for the demethylation of
Me4L3 (0.250 g, 0.341 mM) was similar to the one adopted for
Me2L1. The yield for H4L3 was about 67% (0.155 g). Elem Anal.
Calcd: C, 77.87; H, 4.42; N, 8.25. Found: C, 77.2; H, 4.6; N,
8.1%. FAB-MS: 678 (M+). 1H NMR (CDCl3, ppm): δH 8.95 (d,
4H, J ) 4.4 Hz, Hpyrrole), 8.84 (d, 4H, J ) 4.4 Hz, Hpyrrole), 8.27-
8.22 (m, 4H, Hphenyl), 7.85-7.75 (m, 8H, Hphenyl), 7.57 (d, 2H, J )
8 Hz, Hsubstituted phenyl), 7.29 (s, 2H, Hsubstituted phenyl), -2.75 (s, 2H,
H
NH). IR (KBr pellet; ν/cm-1): 3396 (br, -NH and -OH), 1272
(C-O), 798 (pyrrole CdC). The UV-vis spectral data and E1/2
values are summarized in Table 1.
III.D. Me2L2. A total of 0.150 g (0.222 mM) of Me2L1 was
dissolved in ∼80 mL of N,N-dimethylformamide in a round-bottom
flask and heated to reflux. Then, an excess of Zn(CH3COO)2‚2H2O
(0.200 g, 0.90 mM) was added to that mixture in three parts in
15-20 min intervals. The color of the solution changed from purple
to deep blue-green. The mixture was allowed to reflux for 1 h, and
then, it was cooled to room temperature. The reaction mixture was
poured into ∼300 mL of ice-cold distilled water and kept refriger-
ated overnight. The precipitates obtained were filtered off, washed
twice with water, and then dried in air. The precipitates were
2418 Inorganic Chemistry, Vol. 44, No. 7, 2005

Study of Redox Switchable NIR Dye
Table 2. Electronic Spectral Data and Redox Potential Values for Complexes 1-4
dioxolene based
porphyrin-/bpy-based
λmax
E1/2 (∆E mV)b
E1/2 (∆E mV)b
a
complex
(∈ × 104 dm3 mol-1 cm-1
)
E1/2 (∆E mV)b
sq/cat
sq/q
RuII/III
1
292 (34), 419 (229), 515 (15.2),
549 (7.8), 588 (5.4), 927 (6.4)
291 (51.1), 419 (276), 546 (18.1),
584 (4.5), 933 (8.5)
0.46 (110), 0.78 (120),
-1.79 (140), -2.17 (145)
0.45 (80), 0.71 (90), -1.82 (120),
-2.19 (150)
-0.68 (75)
0.227 (80)
1.07 (140)
1.05 (100)
2
-0.66 (80)
0.14 (70)
[2]red
[2]ox
3
420 (364), 546 (45.2)
415 (308), 547 (35.1)
286 (107), 418 (298), 514 (17),
548 (9.1), 910 (79.2)
286 (142), 420 (302), 545 (18.2),
945 (83.5)
424 (333), 555 (61.3)
0.4 (110), 0.74 (80), 1.1 (100),
-1.89 (120), -2.14 (140)
0.46 (110), 0.8 (85), -1.67 (110),
-2.0 (115), -2.29 (125)
-0.66 (120)
-0.66 (120)
0.14 (125)
0.30 (135)
1.02 (140)
1.06 (126)
4
[4]red
[4]ox
408 (289), 558 (43)
a Electronic spectra were recorded in CH2Cl2. Spectra of the electroreduced and -oxidized species were recorded in situ in the presence of 0.1 M [But]N4PF6
in CH3CN. b Cyclic voltammograms were recorded in CH3CN in 0.1 M [But]4NPF6. The potential values are reported (scan rate of 100 mVs-1) versus those
of the ferrocene/ferrocenium couple. Electrochemical measurements were made with Pt bead as the working electrode, Pt wire as an auxiliary electrode, and
Ag-AgCl as the reference electrode.
phosphate salt was removed in the aqueous layer by solvent
extraction. The CH2Cl2 layer was dried, and the compound thus
obtained was further purified by additional gravity chromatography,
using grade III Al2O3 as the stationary phase and a freshly distilled
CH3CN/toluene (7:3 v/v) solvent mixture as the eluent. The first
major fraction was found to contain the pure desired product.
Yield: 0.390 g (∼70%). Elem Anal. Calcd: C, 63.84; H, 3.66; N,
9.31. Found: C, 62.9; H, 3.9; N, 8.9. FAB-MS: 1202 (M+, ∼2%),
1057 (M+-PF6, 25%). IR (KBr pellet; ν/cm-1): 1443 (semi-
quinone stretching), 841 (PF6). EPR spectral data (77 K; CH2Cl2-
toluene glass): g1, 2.058; g2, 1.9885; g3,1.9585; gav, 2.0016. The
UV-vis/NIR spectral data and redox potential values are presented
in Table 2.
III.I. {(bpy)2Ru(H2L3)}sq[PF6] (H2L3 is the doubly deprotonated
form of H4L3). The synthetic procedure for this compound is similar
to that of complex 1, except H4L3 was used as the catechol
derivative. H4L3 (0.28 g, 0.412 mM), KOH (0.047 g, 0.84 mM),
and Ru(bpy)2Cl2‚2H2O (0.214g, 0.412 mM) were used for the
reaction. Yield: 0.345 g (∼65%). FAB-MS: 1090 (M+, <10%),
945 (M+-PF6, ∼20%). IR (KBr pellet; ν/cm-1): 1447 (semi-
quinone stretching), 840 (PF6).
III.J. Complex 2. The synthetic methodology followed for this
complex is similar to that for complex 1, except H2L2 (0.100 g,
0.141 mM), KOH (0.016 g, 0.29 mM), and Ru(bpy)2Cl2‚2H2O
(0.074 g, 0.14 mM) were used for the reaction. Yield: 0.130 g
(∼72%). This was characterized by standard analytical and
spectroscopic techniques. Elem Anal. Calcd: C, 60.65; H, 3.42;
N, 8.85. Found: C, 60.4; H, 3.4; N, 8.6. FAB-MS: 1120 (M+-
PF6, 10%). IR (KBr pellet; ν/cm-1): 1443 (semi-quinone stretch-
ing), 841 (PF6). EPR spectral data (77 K; CH2Cl2-toluene glass):
g1, 2.058; g2, 1.99; g3, 1.958; gav, 2.0015. The UV-vis/NIR spectral
data and redox potential values are provided in Table 2.
III.K.1. Complex 3. Method 1. This synthetic procedure is
similar to that for complex 1, except H4L3 was used as the ligand.
H4L3 (0.300 g, 0.442 mM), KOH (0.105 g, 1.2 mM), and
Ru(bpy)2Cl2‚2H2O (0.336 g, 0.647 mM) were used for the reaction.
During chromatographic purification, the desired product was eluted
as the first fraction. Yield: 0.26 g (50%). Elem Anal. Calcd: C,
56.31; H, 3.24; N, 9.39. Found: C, 56.0; H, 3.2; N, 9.2. FAB-
MS: 1789 (M+, ∼2%): 1644 (M+-PF6, ∼15%). IR (KBr pellet;
ν/cm-1): 1447 (semi-quinone stretching), 840 (PF6). EPR spectral
data (77 K; CH2Cl2-toluene glass): g1, 2.0593; g2, 1.9965; g3,
1.945; gav, 2.001. The UV-vis/NIR spectral data and redox potential
values are presented in Table 2.
III.K.2. Method 2. This synthetic procedure is similar to that
adopted for complex 1. A total of 0.124 g (0.1 mM) of {(bpy)2Ru-
(H2L3)}sq[PF6], dissolved in ethanolic KOH (0.012g, 0.21 mM),
was reacted with 0.057 g (0.11 mM) of Ru(bpy)2Cl2‚2H2O. Yield:
0.143 g (80%). The purification procedure is similar to that followed
for method 1. The analytical and various spectroscopic data are
similar to those of the desired product synthesized following method
1.
III.L.1. Complex 4. Method 1. Complex 3 (0.110 g, 0.061 mM)
was dissolved in 50 mL of CH3OH and allowed to warm to ∼60
°C; to this, Zn(CH3COO)2‚2H2O (0.054 g, 0.246 mM) was added,
and the resulting mixture was refluxed for 1.5 h under an argon
atmosphere. Then, the solution was dried in vacuo. The solid residue
was redissolved in a minimum amount of CH3CN and purified by
column chromatography on a silica column (100-200 mesh size),
using a CH3CN/aq‚NH4PF6 (98:2 v/v) mixture as eluent. The first
major fraction was found to be the desired product and was
collected. This was dried in vacuo and redissolved in CH2Cl2; excess
NH4PF6 was removed in the aqueous layer through solvent
extraction. Yield: 0.082 g (71%). Elem Anal. Calcd: C, 54.38; H,
3.02; N, 9.06. Found: C, 54.3; H, 3.3; N, 9.0. FAB-MS: 1708
(M+-PF6, ∼25%). IR (KBr pellet; ν/cm-1): 1447 (semi-quinone
stretching), 840 (PF6). EPR spectral data (77 K; CH2Cl2-toluene
glass): g1, 2.0596; g2, 1.998; g3, 1.949; gav, 2.0011. Various
spectroscopic data and redox potential values are summarized in
Table 2.
III.L.2. Method 2. This synthetic procedure is similar to that
adopted in method 1 for complex 3, except H4L4 was used as the
ligand instead of H4L3. H4L4 (0.125 g, 0.168 mM), KOH (0.017 g,
2.1 mM), and Ru(bpy)2Cl2‚2H2O (0.175 g, 0.336 mM) were used
for the reaction. Yield: 0.195 g (∼65%). Analytical and spectro-
scopic data and redox potential values are similar to those observed
for complex 4 synthesized following method 1.
Results and Discussion
Ligand 5-(3,4-dimethoxyphenyl)-10,15,20-triphenyl-21H,
23H-porphyrin (Me2L1) was prepared following the standard
“one-pot synthesis” procedure, resulting in a reasonable yield.
Attempts to improve the yield for Me2L1 by repeating the
reaction in CH2Cl2 in the presence of a catalytic amount of
TFA were not successful. The selective synthesis of trans
bis-substituted porphyrin, 5,15-bis(3,4-dimethoxyphenyl)-
10,20-bisphenyl-21H,23H-porphyrin (Me4L3), was prepared
Inorganic Chemistry, Vol. 44, No. 7, 2005 2419

Jose et al.
Figure 1. EPR spectra recorded for 1 (A) and 3 (B) at 77 K in CH2Cl2/toluene glass.
by employing a building block approach23 through the
condensation of presynthesized meso-phenyldipyrromethane
with 1 mol equiv of 3,4-dimethoxybenzaldehyde. This was
followed by the oxidation of the intermediate product with
p-chloranil. The demethylations of Me2L1 and Me4L3 were
achieved in molten pyrridinium hydrochloride salt at ∼160-
165 °C in an inert atmosphere. An increase in the reaction
temperature to ∼170-180 °C resulted in the desired por-
phyrin derivative with a considerably lower yield (<50%).
Attempts to achieve a better yield for the demethylation
Figure 2. Cyclic voltammogram (200 mVs-1) and differential-pulse
voltammetry of 2 in CH3CN. The potential scale shown is for the Fc/Fc+
couple.
reaction in a CH2Cl2 solvent and with BBr3 as a catalyst,
following standard procedure,24 were not successful either
(yield < 35%). These free-base porphyrin ligands are
metalated with ZnII following standard procedures. Ruthe-
nium-dioxolene complexes, 1-4, are synthesized from reac-
tions of the completely deprotonated form of 5-(3,4-dihy-
droxyphenyl)-10,15,20-triphenyl-21H,23H-porphyrin (H2L1),
5,10-bis(3,4-dihydroxyphenyl)-10,20-bisphenyl-21H,23H-
porphyrin (H4L3), the corresponding Zn derivatives (H2L2
and H4L4, respectively), and an appropriate mol equiv of
Ru(bpy)2Cl2‚2H2O in absolute ethanol. After refluxing for
4 h, the reaction mixture was cooled and exposed to the air
to ensure the oxidation of the initially formed catecholate
complex to the corresponding semi-quinone form. Part of
the oxidative conversion from the Ru(bpy)2(cat*) fragment
to the Ru(bpy)2(sq)+ fragment (cat* is a substituted catechol
derivative) may also take place during the workup procedure
or during column chromatography as the complexes get
exposure to the air. SiO2 and Al2O3, used as column material,
are also known to act as the oxidant.25 In a few cases, to
avoid any possibility of an incomplete conversion to the
semi-quinone form, a mild oxidant (FcPF6) was added to
the crude reaction mixture at room temperature. Analytical
data (elemental analysis and electrospray mass spectral data)
and the results of the various spectroscopic data [viz.,
Complexes 1-4 were found to be paramagnetic and ESR-
active. The paramagnetic nature of these dioxolene deriva-
tives excludes the possibility of these complexes being
present as either catecholate or quinone derivatives.4,9-11
These complexes show a featureless room-temperature ESR
spectrum consisting of a broadened signal centered at gav
≈
2.003 in CH2Cl2. However, spectra recorded at 77 K are
anisotropic (Figure 1) because the molecular orbital contain-
ing the unpaired electrons of the semi-quinone fragments is
partially localized on the ruthenium center.4,9-11 The room-
temperature spectral features of the mono- (complexes 1 and
2) and biradical (complexes 3 and 4) species are similar,
except that the spectra for monoradical species are less broad
than those for biradical ones, presumably because of a weak
interaction between the unpaired electrons of the two semi-
quinone fragments.
A weak electronic interaction between the two redox-active
Ru(bpy)2(sq)+ fragments is also evident in the results of
cyclic and square wave voltammetric studies for complexes
1-4 (Table 2). The porphyrin ligands and their Zn deriva-
tives showed characteristic features because the porphyrin-
based oxidation couple was observed within 0.6-0.8 V
versus that of Fc/Fc+, while the porphyrin-based reduction
couple was observed within -1.5 to -1.85 V versus that of
Fc/Fc+. As expected, 3,4-dimethoxybenzene or 3,4-dihy-
droxybenzene fragments attached to the porphyrin ring and
showed either quasi-reversible or irreversible dioxolene-based
redox processes (Table 1).11 For complexes 1-4, the
Ru(bpy)2(sq) cores undergo reversible redox processes at
potentials consistent with that of the individual component
along with the expected redox processes associated with the
porphyrin core (Table 2). The representative cyclic voltam-
1
vibrational; electronic; H NMR spectral data for Me2L1,
H2L1, H2L2, Me4L3, H4L3, and H4L4; and electron spin
resonance (ESR) spectra data for 1, 2, 3, and 4] agree well
with the formulations proposed for the respective ligands
and the corresponding complexes.
(23) (a) Lee, C.-H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427. (b) Manka,
J. S.; Lawrence, D. S. Tetrahedron Lett. 1989, 30, 6989.
(24) McOmie, J. F.; Watts, M. L.; West, E. Tetrahedron 1968, 24, 2289.
(25) Ernst, S.; Hanel, P.; Jordanov, J.; Kaim, W.; Kasack, V.; Roth, E. J.
Am Chem. Soc. 1989, 111, 1733.
2420 Inorganic Chemistry, Vol. 44, No. 7, 2005

Study of Redox Switchable NIR Dye
Figure 3. UPM3 optimized structures of model systems X1 and X2.
mogram for complex 2 is shown in Figure 2. However, the
cat/sq and sq/q redox waves for 3 and 4 are much broader
[(Epc - Epa) g 120-135 mV; Table 2) than those observed
for complexes 1 and 2, which is understandable if one
considers the possibility of overlapping redox waves for the
two weakly interacting Ru(bpy)2(sq)+ fragments in 3 and 4.
Such behavior is not surprising if one considers that two
appended ruthenium-semiquinone fragments are separated
by ∼14.2 Å with an orthogonal orientation, with respect to
the porphyrin plane (an analogous model system is repre-
sented as X1, vide infra).20,21
To corroborate the presumption, we have employed UHF
PM3 calculations to optimize the geometry of two possible
isomeric model compounds, X1 and X2, at the UPM3 level
(Figure 3).21,22 The possibility of X1 being present as
biradical (singlet and triplet) has been considered. The
energy-optimized geometry for model system X1 (singlet and
triplet) predicts a planar porphyrin ring and orthogonal sq-
fragment orientations, with π electron delocalization in the
porphyrin unit (Figure 3), while the planar orientation of the
semi-quinone ring in X2 causes the porphyrin ring to be away
from a planar structure, with disruption of the π electron
delocalization. The triplet form of the biradical X1 is more
stable than X2 by 18.7 kcal/mol. Calculations also predict
that the singlet form of biradical X1 is less stable than the
triplet form by 1.9 kcal/mol. Thus, the energy-minimized
structure of the biradical species (X1) suggests an orthogonal
orientation of the semi-quinone fragment with respect to the
porphyrin plane in complexes 1-4 and thereby negates the
possibility of extended conjugation and the delocalization
of unpaired electrons across the bridging porphyrin core in
complexes 2 and 4.
∼940 nm for the dπRu f π*bsq transition (ꢀ ) 59 000-
83 500 dm3mol-1cm-1). To the best of our knowledge,
barring two examples available in the literature to date,1d,5
where the ꢀ values reported for the absorption band in the
NIR region are ∼100 000 dm3mol-1cm-1, the ꢀ values for
the NIR absorption band observed for these complexes are
among the highest known for any inorganic metal complexes.
A red shift of ∼50 nm for λmax of the MLCT (dπRu f π*bsq
)
transition band of complexes 1-4, compared to that for
Ru(bpy)2(bsq)+ species),8 is observed. An even higher red
shift of ∼100 nm was observed by us for this MLCT band,
where the coordinated sq functionality is part of an extended
conjugated system.11 Presumably, this higher red shift
observed in the RuII-dioxolene complexes reported earlier
is due to the more effective lowering of the energy of semi-
quinone singly occupied molecular orbitals arising from the
extended conjugation.8,9,11 This also indicates either the
absence of any or the presence of little extended conjugation
in complexes 1-4 and agrees well with the proposed
conformation. A similar shift of ∼75 nm is reported by Ward
and co-workers for the bi-/trinuclear complexes, where two
Ru(bpy)2(sq) units are linked by a Ru(tpy)2 center (tpy is a
2,2′:6′,2”-terpyridyl derivative).26 For complexes 1-4, sq-
porphyrin or sq-porphyrin-sqRu(bpy)2 is expected to be a
better electron acceptor than the bsq alone, and this,
presumably, is responsible for the observed red shift.
Spectroelectrochemical studies with complexes 1-4 show
that either oxidation or reduction of the semi-quinone
fragment causes a complete bleaching of the MLCT band
in the NIR region (Figure 4). Spectral changes are found to
be completely reversible, and a quantitative recovery of the
MLCT band at the NIR region occurs upon the reversal of
the redox processes. This observation is in agreement with
the reversible nature of the dioxolene-based redox processes,
which is observed in cyclic voltammograms for complexes
Electronic spectral data for the ruthenium-dioxolene
complexes along with those of the reduced and oxidized
forms of 2 and 4 are presented in Table 2. Electronic spectral
data reveal that bi-/trichromophoric systems, 1-4, show
expected spectral features of the individual mononuclear
components along with an unusually strong absorbance at
(26) Whittle, B; Everest, N. S.; Howard, C.; Ward, M. D. Inorg Chem.
1995, 34, 2025.
Inorganic Chemistry, Vol. 44, No. 7, 2005 2421

Jose et al.
Figure 4. Electronic spectral changes when 4 is subjected to bulk electrolysis (A) -0.4 V (vs SCE) and (B) 0.7 V (vs SCE).
Table 3. Emission Data of Free-Base Porphyrine Derivatives and Their
tum yield (φ) for the free ligand is ∼0.1. On the basis of the
definitions of φ [φ ) kr/(kr + ∑knr)] and the lifetime [τ; τ )
1/(kr + ∑knr)], in terms of the radiative (kr) rate constant
and the sum of all of the nonradiative (∑knr) rate constants,
one may conclude that ∑knr > kr. Therefore, if the value for
φ has to decrease, either kr should decrease or (kr + ∑knr)
must increase. However, if the latter were true, the value
for τ would have decreased. In the present investigation, the
observed value for τ is almost the same for both of the
species (H2L1 and 1). ∑knr being the dominant term in the
expression for φ, it is difficult to see how an additional term
in ∑knr (that accounts for a new, fast nonradiative decay)
could be the cause of the quantum yield decrease without
any change in the lifetime as well. Thus, our data tends to
suggest that the presence of the Ru(bpy)2(sq)+ fragment
somehow reduces the value for kr. Because τ*/τ ) kr and
the τ values are the same, then the change in φ reflects a
change in kr. Because H2L1 and 1 are two different
compounds, it is quite possible that their natural radiative
lifetimes (τ*) will be different. (This interpretation was
suggested by one of the reviewers and is consistent with the
observed results.)
Dioxolene Complexes (1-4)a
Q (0,0)
Q (0,1)
compound
λ
max (nm)
λ
max (nm)
φf
H2L1
H4L3
complex 1
complex 3
651 (239)
653 (275)
652 (91)
652 (23)
708 (16)
708 (17)
707 (6.3)
0.0977
0.1117
0.03379
0.0093
S2 emission
S1 emission
B (0,0)
Q (0,0)
max (nm)
Q (0,1)
max (nm)
compound
λ
max (nm) φf × 104
λ
λ
φf
H2L2
H4L4
436 (84)
1.37
596 (514) 640 (262) 0.03612
b
complex 2
complex 4
433 (21)
431 (37)
0.32
0.77
595 (66)
595 (35)
640 (31)
640 (15)
0.0054
0.00393
a All of the spectra were recorded in CH3CN. b Data could not be obtained
because of its limited solubility in the desired solvent.
1-4. Thus, these complexes fulfill the basic criteria for
possible use as redox switchable photochromic dyes.27
However, to understand the system better and to investi-
gate the possibility of electron or energy transfer processes
being involved, we looked into the steady-state emission
properties of all the bi-/trichromophoric systems and the
transient absorption spectra of the excited state of the simple
bichromophoric system (1) and its individual mononuclear
components on an ultrafast time scale. The data obtained
from the steady-state emission measurements for the bichro-
mophoric complexes of interest (1-4) reveal to us that the
fluorescence quantum yield of the complex is at least 3 times
less (Table 3) than that of the precursor molecule.
The emission lifetime measurements carried out for H2L1
and 1 have shown that there is no change in the emission
lifetime of the porphyrin core in these two species when
excited at 408 nm. The excited state lifetime for the porphyrin
moiety was found to be 8.8 ( 0.2 ns. This suggests that the
process that is responsible for the reduction in the fluores-
cence quantum yield is expected to be much faster than the
time resolution of the time-correlated single photon counting
(TCSPC) technique used for lifetime measurements (∼200
ps). Thus, the observed decrement in the emission quantum
yield may be due to intramolecular electrons or an energy
transfer between Ru(bpy)2sq+ and porphyrin moieties, or it
may occur through other nonradiative pathways. The quan-
The possibility of an intramolecular energy transfer process
can be ruled out because the emission spectra of the
porphyrin core and the absorption spectra of the Ru(bpy)2sq+
fragment do not overlap. However, the possibility of electron
transfer from photoexcited porphyrin to the Ru(bpy)2sq+
moiety may not be ruled out because the process is also
favored thermodynamically. The other factor, which may also
cause the decrease in the fluorescence quantum yield, is
basically a consequence of the aggregation of the porphyrin
molecules.28 This possibility of aggregate formation can be
clarified with the aid of concentration-dependent optical
absorption and emission measurements. These did not show
any red shift in the absorption spectrum with respect to the
Soret band, which is the signature of aggregate formation.28
These measurements have unambiguously nullified the
plausible involvement of aggregate formation for the decrease
in the quantum yield. To elucidate the actual mechanism for
the decrement in the emission quantum yield of the complex
molecule, excited-state dynamics of H2L1 and 1 have been
studied using transient absorption spectroscopy in shorter
time scales.
(27) DeSilva, A.P.; Gunaratne, H. Q. N.; Gunnaugsson, T.; Huxley, A. J.
M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997,
97, 1515 and references therein.
(28) Akins, D. L.; Ozcelik, S.; Zhu, H.-Ru.; Guo, C. J. Phys. Chem. 1996,
100, 14390.
2422 Inorganic Chemistry, Vol. 44, No. 7, 2005

Study of Redox Switchable NIR Dye
Table 4. Transient Decay Times of H2L1 and [Ru(bpy)2bsq]+ and Their Complex (1) at Representative Wavelengths Measured after Exciting with a
100 fs Laser Light at 400 nm
H2L1/ACN
τ in ps (a%)
1/ACN
τ in ps (a%)
Ru-sq/ACN
τ in ps (a%)
wavelength (nm)
470
0.22 ( 0.05 (43.5)
1.5 ( 0.2 (5.3)
13 ( 2 (6.8)
0.28 ( 0.05 (32.3)
1.5 ( 0.2 (24.7)
10 ( 2 (3.1)
>400 (44.4)
>400 (39.9)
550
670
710
900
0.22 ( 0.05 (33.8)
1.4 ( 0.2 (16.2)
10 ( 2 (8.1)
0.20 ( 0.05 (52.6)
1.5 ( 0.2 (15.8)
26 ( 2 (8.4)
>400 (41.9)
>400 (23.2)
0.22 ( 0.05 (61.3)
1.5 ( 0.2 (12.5)
10 ( 2 (1.3)
0.23 ( 0.05 (72.2)
1.4 ( 0.2 (8.3)
28 ( 2 (8.2)
0.27 ( 0.05 (62.7)
1.5 ( 0.2 (7.4)
10 ( 2 (7.4)
>400 (24.9)
>400 (11.3)
>400 (22.5)
0.20 ( 0.05 (51.1)
1.5 ( 0.2 (17.8)
13 ( 2 (12.9)
>400 (18.2)
0.23 ( 0.05 (68.9)
1.4 ( 0.2 (12.1)
25 ( 2 (10.6)
>400 (8.4%)
0.21 ( 0.05 (50)
1.5 ( 0.2 (10.3)
10 ( 2 (3.1)
0.27 ( 0.05 (-83)
1.6 ( 0.2 (-12.3)
23 ( 3 (-8.9)
>400 (4.1)
0.27 ( 0.05 (-65.7)
1.5 ( 0.2 (-15.7)
19 ( 3 (-22.4)
>400 (3.8)
>400 (46.6)
To understand the relaxation dynamics of the complex
molecule 1, it is crucial to study the excited-state dynamics
of the porphyrin moiety separately. We have recorded
transient absorption spectra for H2L1 in acetonitrile after
exciting it with a 400 nm laser pulse (full width at half
maximum ) 100 fs), and we have shown them in Figure 5.
The transient spectrum consists of absorption peaks centered
at 490, 530, 570, 610, and 690 nm, with a dip at 510 nm.
The observed transients for H2L1 are assigned to the S1-Sn
absorption bands. The dip at 510 nm is observed because of
the overlapping ground-state absorption of the Q band of
the porphyrin moiety. In addition, all of the sharp structures
observed in the transient spectra correspond to the overlap-
at 657 nm can be fitted multiexponentially with time
constants of 200 fs, 1.4 ps, and 10-20 ps and a very long
lifetime component. They have attributed three fast compo-
nents, 200 fs, 1.4 ps, and 10-20 ps, to intramolecular
vibrational energy redistribution, vibrational redistribution
caused by elastic collision with solvent molecules, and
thermal equilibration by energy exchange with the sol-
vent, respectively. Similarly, we have also attributed the
300 fs component to intramolecular vibrational energy
redistribution, the 2 ps component to vibration cooling, and
the 13 ps component to thermal equilibration with the
solvent. To understand and find out the decay time constant
of the longer component, which is the excited-state lifetime
of H2L1, we have measured the fluorescence lifetime using
TCSPC measurements, and it has been found to be 8.8 (
0.2 ns.
30
ping Q-band only. Kong and co-workers have estimated
the internal conversion rate ∼ 10 fs from S2 to S1 for H2TPP
by the femtosecond time-resolved fluorescence depletion
method, following excitation at 398 nm in chloroform.
Similar ultrafast or instrument-limited increases of the S1
state have also been observed by Zewail and co-workers29
for H2TPP, and experimental data of the fluorescence up-
conversion measurements suggest that the time duration of
internal conversion observed for the transition from the S2
to S1 state can be as fast as <50 fs for H2TPP. Kinetic decay
traces (after excitation by a 400 nm laser pulse) of the
acetonitrile solution of H2L1 at 470 nm, 550 nm, and 710
nm, measured with femtosecond pump-probe spectroscopy,
were obtained. Observed decay traces, at these respective
wavelengths, can be best fitted with a multiexponential
function with typical time constants of 300 fs, 2 ps, and 13
ps and a very long lifetime component for all of the
wavelengths (Table 4). Zewail and co-workers29 also had a
similar observation while carrying out transient absorption
measurements for H2TPP, following excitation with a 397
nm femtosecond laser pulse; the transient absorption kinetics
The transient absorption spectrum of 1 at various time
delays, after excitation at 400 nm, is shown in Figure 6. The
transient spectrum at each time delay consists of positive
absorption peaks centered on 490, 550, 610, and 710 nm
(29) Baskin, J. S.; Yu, H. Z.; Zewail, A. H. J. Phys. Chem. A 2002, 106,
Figure 5. Transient absorption spectra of H2L1 in acetonitrile at different
delay times after excitation at 400 nm. The transient absorption has been
attributed to the S1-Sn absorption.
9837.
(30) Zhong, Q.; Wang, Z.; Liu, Y.; Zhu, Q. H.; Kong, F. A. J. Chem. Phys.
1996, 105, 5377.
Inorganic Chemistry, Vol. 44, No. 7, 2005 2423

Jose et al.
Figure 7. Transient absorption spectrum of Ru(bpy)2(bsq)+ in acetonitrile
at a 100 fs time delay after excitation with a 400 nm laser pulse. The
transient absorptions at 690 and 570 nm have been attributed to the excited-
and triplet-state absorptions of Ru(bpy)2(bsq)+, respectively, and the negative
absorption at 900 nm is due to the bleach of the ground-state absorption.
(The inset shows the bleach recovery kinetics monitored at 900 nm.)
Figure 6. Transient absorption spectra of 1 in acetonitrile at different delay
times after excitation at 400 nm. (The inset shows the bleach recovery
kinetics monitored at 950 nm.)
along with a dip at 510 nm and strong bleaching in the 935
nm region. The positions of the transient absorption peaks
matched pretty well with those of the transients in H2L1
(Figure 5). It is worth mentioning here that, even though
the peak positions are similar, the relative intensities and
width of the bands are markedly different. This is due to the
fact that complex 1 is basically a different molecule, with
the Ru(bpy)2(sq) fragment attached to the porphyrin moiety,
compared to the individual fragment H2L1; the attached
Ru(bpy)2(sq) fragment is expected to have an influence on
the ꢀ values of different bands. The additional big bleaching
at 950 nm can be attributed to the disappearance of the
absorption peak due to the dπRuf π*sq-based MLCT
transition of the Ru(bpy)2sq moiety of 1. The dynamics of 1
have been monitored at various wavelengths (e.g., at 470
550, 670, and 710 nm), and the respective decay trace has
been fitted with a multiexponential function with different
time constants; these data are given in Table 4. It is
interesting to note that the decay kinetics at 470 nm for 1
matched pretty well with those for H2L1 at the same
wavelength and can be attributed to the excited-state dynam-
ics of the photoexcited porphyrin moiety of 1 (vide infra).
However, the decay kinetics for 1 at 670 and 710 nm are
relatively faster than those observed for H2L1 at these
respective wavelengths. We have also monitored bleaching
recovery kinetics at 950 nm (inset of Figure 6), and they
have also been fitted to a multiexponential function; the time
constants that were obtained are provided in Table 4. In our
earlier studies,14c we carried out a time-resolved transient
absorption of 1 in acetonitrile following excitation at 532
nm with a picosecond laser pulse (full width at half
maximum ) 35 ps). We observed bleaching around 900 nm
in the time-resolved spectra, which was attributed to the
forward electron transfer from photoexcited porphyrin to the
Ru(bpy)2sq moiety, while the recovery kinetics of the
bleaching were attributed to the corresponding back electron-
transfer process. Thus, our present in-depth study seems to
contradict our earlier proposition based on experiments on
a picosecond time domain. To resolve this further, we have
also studied the excited-state dynamics of [Ru(bpy)2bsq]+
separately, after excitation at 400 nm.
The transient absorption spectrum of photoexcited
[Ru(bpy)2bsq]+ following excitation at 400 nm is shown in
Figure 7. The transient spectrum consists of absorption peaks
at 580 and 690 nm with a bleach at 900 nm; a similar bleach
was observed for complex 1 and has been described earlier.
The transient decay kinetics of the absorption have been fitted
multiexponentially, and the respective rate constants are
provided in Table 4. The transient peaks have been attributed
to the excited-state absorption of the dπRu f π*bsq-based
MLCT transition of the Ru(bpy)2(bsq)+ moiety. The bleach
at 900 nm (inset of Figure 7) can be attributed to the electron
transfer from the photoexcited RuII center to the semi-quinone
radical, while the recovery kinetics can be attributed to the
back electron transfer from the reduced semi-quinone radical
to the RuIII center. Spectroelectrochemical studies on 1 and
[Ru(bpy)2(bsq)]+ 11b have already established that, for both
complexes, the NIR absorption bands around 930 and 890
nm, respectively, bleach reversibly upon the reduction of the
semi-quinone fragment to the corresponding catechol one
and reappear with their original intensity upon the reversal
of the redox process. As we have observed, in the present
investigation, both porphyrin and Ru(bpy)2bsq moieties can
be excited by a 400 nm laser pulse. Thus, it is expected that,
in 1, both porphyrin and Ru(bpy)2(sq) centers will get excited
when they are exposed to a 400 nm laser pulse. One can
expect that the transient absorption spectrum of 1 will be a
mixture of the excited states of both porphyrin and Ru(bpy)2sq
fragments, unless there is any interaction between them in
the excited states. Although, Figure 6 shows that the transient
spectrum for 1 is roughly an addition of the transient spectra
of two individual components, H2L1 (Figure 5) and Ru(bpy)2-
bsq+ (Figure 7); slight variation is observed as a result of
the differences in the energy and ꢀ values associated with
the MLCT transition for the dπRu f π*sq-based transitions
in 1 and Ru(bpy)2(bsq)+. Further, in the transient spectrum
of 1, we have observed neither any special transient due to
the porphyrin cation radical nor any transient growth in the
absorption spectrum in the wavelength region studied (460-
1000 nm; Figure 6) when it was excited at 400 nm. Thus,
we can unambiguously rule out the possibility of electron
2424 Inorganic Chemistry, Vol. 44, No. 7, 2005

Study of Redox Switchable NIR Dye
transfer from the photoexcited porphyrin moiety to the
Ru(bpy)2sq fragment. In the present investigation, the bleach
at 950 nm is instantaneous, which is due to a direct
photoexcitation of the Ru(bpy)2(sq) moiety instead of an
excited energy transfer from the porphyrin moiety. Hence,
we can rule out the possibility of an intramolecular energy
transfer process within the photoexcited complex.
The above results suggest that the excited-state dynamics
of 1 are a combination of the deactivations of the two
individual moieties. Thus, the dynamics of the complex
molecule, 1, on photoexcitation is explained in the following
manner. The fast component of 220 fs has been assigned to
an intramolecular vibrational relaxation similar to that of
H2L1; only, the amplitude of the decay is more in that of 1.
The 1.6 and 11 ps components can be assigned to the
vibrational cooling. The long-lived transient is the singlet
state of the molecule, which decays with a long lifetime.
In conclusion, the excited-state deactivation of the por-
phyrin center, linked with the Ru(bpy)2(sq) moiety, reveals
that the observed dynamics are a mixed set, having inde-
pendent contributions from the porphyrin and Ru(bpy)2(sq)
components. The results unambiguously rule out the pos-
sibility of either an electron or energy transfer from the
porphyrin to the Ru(bpy)2(sq)+ fragment or vice versa. The
observed decrement in the luminescence quantum yield can
be ascribed to the increased nonradiative pathway due to
large vibronic coupling because of the direct linkage of the
metal center to the porphyrin moiety.
Acknowledgment. The Department of Science and
Technology (DST) and the Board of Research in Nuclear
Science (BRNS), government of India, support this work.
Authors D.A.J., A.D.S., and D.K.K. acknowledge CSIR for
the Sr. Research Fellowship. We also thank all of the
reviewers for their suggestions and comments. A.D. and
H.N.G. acknowledge Dr. P. K. Ghosh (CSMCRI, Bhavnagar,
India) and Dr. T. Mukherjee (BARC, Mumbai, India) for
their keen interest in this work.
Supporting Information Available: Schemes for the redox
processes involving Ru(bpy)2(bsq)+, synthetic methodologies for
the syntheses of Me2L1 and H4L3, room-temperature EPR spectra
for complexes 1 (A) and 3 (B), relative energy diagrams for X1
(both singlet and triplet forms) and X2, and kinetic decay traces of
the different transients of H2L1 and L1 in acetonitrile monitored at
different wavelengths. This material is available free of charge via
the Internet at http://pubs.acs.org.
IC048805L
Inorganic Chemistry, Vol. 44, No. 7, 2005 2425