Synthesis and ultrafast excited-state dynamics of zinc and palladium triply fused diporphyrins

Article abstract of DOI:10.1021/jp3043895

We report the synthesis and ultrafast excited-state dynamics of two new meso-meso, β-β, β-β triply fused diporphyrins, Zn-3DP and Pd-3DP. Both compounds were found to have short excited-state lifetimes: Zn-3DP possessed an average S₁ lifetime of 14 ps before nonradiative deactivation to the ground state, whereas Pd-3DP displayed a longer average S₁ lifetime of 18 ps before crossing to the T₁ state, which itself possessed a very short triplet lifetime of 1.7 ns. The excited-state dynamics of Zn-3DP, compared to similar zinc(II) diporphyrins reported in the literature, suggests that a conical intersection of the S ₁ and S₀ potential energy surfaces plays a major role as a deactivation pathway of these molecules. Furthermore, the short triplet lifetime of Pd-3DP, compared to other diporphyrins that also exploit the intramolecular heavy atom effect, reveals that the position of the heavy atom within the diporphyrin framework influences the strength of spin-orbit coupling. The implications for employing triply fused diporphyrins as NIR-absorbing triplet sensitizers are discussed.

Full text of DOI:10.1021/jp3043895

Article
pubs.acs.org/JPCA
Synthesis and Ultrafast Excited-State Dynamics of Zinc and
Palladium Triply Fused Diporphyrins
Derrick A. Roberts, Burkhard Fuckel, Raphael G. C. R. Clady, Yuen Yap Cheng, Maxwell J. Crossley,*
̈
̈
and Timothy W. Schmidt*
School of Chemistry, The University of Sydney, Sydney, NSW 2006, Australia
S
* Supporting Information
ABSTRACT: We report the synthesis and ultrafast excited-state dynamics
of two new meso−meso, β−β, β−β triply fused diporphyrins, Zn-3DP and
Pd-3DP. Both compounds were found to have short excited-state lifetimes:
Zn-3DP possessed an average S1 lifetime of 14 ps before nonradiative
deactivation to the ground state, whereas Pd-3DP displayed a longer
average S1 lifetime of 18 ps before crossing to the T1 state, which itself
possessed a very short triplet lifetime of 1.7 ns. The excited-state dynamics
of Zn-3DP, compared to similar zinc(II) diporphyrins reported in the
literature, suggests that a conical intersection of the S1 and S0 potential
energy surfaces plays a major role as a deactivation pathway of these
molecules. Furthermore, the short triplet lifetime of Pd-3DP, compared to
other diporphyrins that also exploit the intramolecular heavy atom effect,
reveals that the position of the heavy atom within the diporphyrin
framework influences the strength of spin−orbit coupling. The implications
for employing triply fused diporphyrins as NIR-absorbing triplet sensitizers are discussed.
the dynamics of the involved excited electronic states. Organic
INTRODUCTION
■
molecules with electronic absorption bands in the NIR pose a
particular challenge in this respect because the rate of excited-
state deactivation typically increases exponentially as the
electronic energy gaps decrease.21 Therefore, it is crucial to
understand the excited-state dynamics of triply fused
porphyrins and their metal chelates in order to assess and
tailor their photophysical properties for the aforementioned
applications. The simplest representative of this class of
molecule is the meso−meso, β−β, β−β triply fused dimer
(hereafter, triply fused diporphyrin), which serves as a useful
model for understanding the excited-state dynamics of large
aromatic heterocycles with extensive π-conjugation.
There are two key synthetic approaches to triply fused
diporphyrins: (1) direct oxidative coupling of two monomeric
porphyrins, each with at least one free meso position,22−26 and
(2) planarization of a meso−meso singly linked diporphyr-
in.1,3,9,27−30 The first approach has become increasingly
preferred in the recent literature; however, this method
requires the use of 5,10,15-trisubstituted porphyrin precursors
to prevent the formation of fused and linked oligomers, rather
than dimers.6,31,32 Unfortunately, the common routes to
trisubstituted porphyrins are synthetically laborious when
approached through the tripyrrane and inefficient via statistical
condensation. The addition−dimerization methodology devel-
Since their first reported synthesis over a decade ago,1 meso−
meso, β−β, β−β triply fused porphyrin arrays have been the
subject of intense research due to their extremely red-shifted
absorption bands, high electrical conductivity, and nonlinear
optical properties,1−5 which render this class of molecule
potentially useful for applications in molecular electronics,6
reverse saturable absorption,7 near-infrared sensing,8 artificial
photosynthesis,9 and photodynamic therapy.10 The latter
applications are based on near-infrared (NIR) absorption
features; hence, triply fused porphyrins can also be envisaged as
NIR sensitizers for photochemical upconversion (PUC). In this
process, two low-energy photons are fused incoherently to
produce one photon of higher energy. Usually, PUC proceeds
via sensitized triplet−triplet annihilation and thus makes use of
metalloporphyrin sensitizers,11−13 which possess large absorp-
tion cross sections and high intersystem crossing (ISC) rates. In
previous work, we have employed a π-expanded tetrakisquinox-
alinoporphyrin, synthesized using a stepwise ring-annulation
strategy,14,15 as an efficient sensitizer for PUC of light in the
650−700 nm range.16,17 Extending upconversion sensitization
further into the NIR has been the focus of recent work in this
field;13,18−20 the wavelength regime above 900 nm, however,
remains a barrier for PUC. The extremely red-shifted
absorption bands of triply fused porphyrins have the potential
to overcome this barrier.
While PUC mainly requires a high ISC yield and reasonably
long triplet-state lifetime, the suitability of triply fused
porphyrins for all of the applications listed above depends on
Received: May 7, 2012
Revised: July 5, 2012
Published: July 9, 2012
© 2012 American Chemical Society
7898
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
a
Scheme 1
a
Reagents and conditions: (i) n-butyllithium (1.4 M in hexanes), −78 °C, THF, 15 min and then DDQ, 30 min; (ii) Zn(OAc)2·2H2O in CHCl3−
CH3OH, 1 h; (iii) Sc(OTf)3−DDQ, 50 °C, toluene, 1 h; (iv) H2SO4−CHCl3, rt, 5 min; (v) Pd(OAc)2 and NaOAc in CHCl3−CH3OH, protected
from light, 2 days. Ar = 3,5-di-tert-butylphenyl.
oped by Senge and co-workers addresses these shortcomings by
offering a simple one-step route to end-capped meso−meso
singly linked diporphyrinsthe immediate precursors to triply
fused diporphyrins.33,34 The general methodology involves the
reaction of a 5,15-disubstituted porphyrin with an alkyl or aryl
organolithium reagent, followed by oxidative dimerization to
afford a meso−meso singly linked diporphyrin.35 The singly
linked diporphyrin is readily converted to its triply fused
counterpart by oxidative cyclodehydrogenation.1,4,22,26,30
While there has been sustained interest in developing
synthetic methodologies toward triply fused diporphyrins,
there has been comparatively little work on their excited-state
dynamics. The research groups of Osuka, Kim, and Anderson
have measured the excited-state dynamics of zinc(II) triply
fused diporphyrins that differ only by their substitution at the
meso-like positions around the diporphyrin core.2,7,23,36
Interestingly, these structurally similar compounds display a
wide range of excited-state behaviors, suggesting that the
peripheral substituents have a pronounced influence on the
dominant pathways through which excited-state deactivation
occurs in these molecules. For instance, Osuka and co-workers
reported an extremely short S1 lifetime (4.5 ps) and no triplet
yield for a zinc(II) hexaaryl-substituted diporphyrin,2,36 whereas
Anderson and co-workers observed intersystem crossing and
relatively long triplet lifetimes for zinc(II) diporphyrins whose
terminal meso substituents had been replaced by iodo (τ = 52
ns), bromo (τ = 280 ns), and (triisopropylsilyl)alkynyl (τ = 177
μs) substituents.7 Furthermore, there is ongoing debate
regarding the dominant pathway of nonradiative S1 deactivation
in triply fused diporphyrins. Therefore, further work on the
excited-state dynamics of different triply fused diporphyrins is
crucial for understanding their photophysical behavior for the
aforementioned applications.
itself a variation of the Scholl reaction.37 We subsequently
measured the ultrafast excited-state dynamics of these
diporphyrins in order to observe the effect of both meso-alkyl
substitution and heavy metal chelation38 on the S1 decay
dynamics of these infrared chromophores. The excited-state
dynamics of these two compounds are compared to
diporphyrins reported in the recent literature and their
photophysical properties are discussed in terms of applicability
as sensitizers for PUC.
EXPERIMENTAL SECTION
■
Organic Synthesis. The synthesis details and character-
ization of the diporphyrins are described in the Supporting
Information.
Femtosecond Pump−Probe Laser Spectroscopy. The
dual-beam femtosecond time-resolved transient absorption
setup consisted of a self-mode-locked femtosecond Ti-sapphire
amplified fiber oscillator (Clark MXR, CPA 2210), two optical
parametric amplifier systems (Light Conversion, TOPAS), and
an optical detection system. The amplified output beam (780
nm) had ∼150 fs pulse width and a pulse energy of
approximately 1.5 mJ at 1 kHz repetition rate, which was
divided into two parts by a 1:1 beam splitter. Both beams were
color-tuned to the desired pump and probe wavelengths by
optical parametric amplification. The pump beam was directed
through a variable delay stage and an optical chopper operating
at 0.5 kHz and then focused to a ∼2 mm diameter spot on the
sample cuvette, and the laser fluence was adjusted to avoid
damage of the sample by using a variable neutral-density filter.
The time delay between pump and probe beams was controlled
by adjusting the length of the variable optical delay stage. Pump
and probe beams were monitored after and before transmission
through the sample by using high-speed photodiodes
(ThorLabs and SM1 series, respectively) with home-built
integration circuits. The difference between the signals with and
without the pump beam were calculated by in-house software
written in LabView.
In this contribution we have synthesized zinc(II) and
palladium(II) butyl-end-capped triply fused diporphyrins
using a simple two-step synthetic route that combines the
nucleophilic addition−dimerization methodology developed by
Senge and co-workers33 and the oxidative cyclodehydrogena-
tion protocol reported by Osuka and co-workers,5 which is
7899
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
RESULTS AND DISCUSSION
■
Diporphyrin Synthesis. Synthesis of the target compounds
is summarized in Scheme 1. Singly linked, butyl-end-capped
diporphyrin H-1DP was synthesized in one step from readily
available 5,15-bis(3,5-di-tert-butylphenyl)porphine using the
nucleophilic addition−dimerization methodology developed
by Senge and co-workers.33,35 The reaction is understood to
proceed through nucleophilic attack of n-butyllithium at a free
porphyrin meso-position to give an anionic intermediate, which
is then oxidized to its corresponding radical anion using DDQ
(2,3-dichloro-5,6-dicyano-1,4-benzoquinone).35 Homogenic
bond formation at the remaining free meso carbon atoms
and subsequent oxidation yields the singly linked product, H-
1DP, which was obtained in 89% yield, with 7% of a
monobutylated side product. As with other singly linked
diporphyrins, H-1DP displays four unique β-pyrrolic reso-
Figure 1. Electronic absorption spectra of H-3DP (red), Zn-3DP
(black), and Pd-3DP (blue) in chloroform. Bands I, II, and III are
consistent with reports by Osuka and co-workers.3 Ar = 3,5-di-tert-
butylphenyl.
1
nances in its H NMR spectrum (Supporting Information,
Figure S1), characteristic of its C2 symmetric structure, and a
split Soret band in its electronic absorption spectrum due to
excitonic coupling along the long axis of the dimer.
according to the assignments by Osuka and co-workers: band I
in the visible region, whose transition dipole moment is
polarized along the short in-plane axis of the molecule, and
bands II (visible) and III (NIR), whose transition dipole
moments are both polarized along the long molecular axis.2,3
The purity of the triply fused diporphyrins was confirmed by
high-resolution LDI-FTICR mass spectrometry, UV−vis−NIR
spectroscopy, and 1H NMR analysis where possible (see
Supporting Information, Figure S3−S7). Both Zn-3DP and Pd-
The formation of a monobutylated byproduct is attributed to
protonation of the intermediate porphyrin anion, possibly as a
result of intermolecular proton transfer between two
intermediate porphyrin anions during the reaction. Senge and
co-workers have demonstrated that 5,10,15-trisubstituted
porphyrins are obtained as the major product of the reaction
if water is added as a proton source before DDQ oxidation.33
Accordingly, we found that extremely stringent control over the
use of pure, anhydrous reagents was crucial to the success of
this reaction. In particular, we found that using recrystallized
and dried DDQ gave yields in excess of 85%, whereas using
commercially purchased DDQthoroughly dried but without
further purificationresulted in heavily suppressed reaction
yields (23% of H-1DP) and an equal proportion of the
monobutylated side product (24%).
1
3DP displayed broad H NMR signals in deuterated chloro-
form, suggesting that the molecules aggregate in solution at the
concentrations required for obtaining readable NMR spectra.
Adding a small amount of pyridine-d5 to the sample in CDCl3
improved the quality of the 1H NMR spectrum of Zn-3DP but
did not noticeably ameliorate aggregation of Pd-3DP.
Recently, Senge et al. reported an attempt to combine the
addition−dimerization coupling methodology with the Sc-
(OTf)3−DDQ cyclodehydrogenation reaction employed by
Osuka and co-workers as a route to a hexasubstituted triply
fused porphyrin dimer.40 In this work, the authors reported that
the reaction of a zinc(II) hexyl-end-capped singly linked
diporphyrin with up to 7 equiv of Sc(OTf)3−DDQ failed to
afford any triply fused product. This result is in stark contrast to
our own experience with Zn-1DP, which was converted to its
triply fused counterpart under similar conditions. We did,
however, find that oxidative cyclodehydrogenation could not be
performed with H-1DP, which can be explained by the fact that
the first and second oxidation potentials of H-1DP are
significantly higher than those of Zn-1DP.3
Excited-State Dynamics. The excited-state dynamics of
Zn-3DP and Pd-3DP were measured by ultrafast transient
absorption spectroscopy using a femtosecond pump−probe
pulsed laser experimental setup. Both diporphryins were found
to have very short excited-state lifetimes, which is consistent
with previous work on other triply fused diporphyrins.2,7,23,36
Zinc(II) Triply Fused Diporphyrin (Zn-3DP). Ultrafast
excited-state absorption transients on Zn-3DP are shown in
Figure 2. In both cases, a deoxygenated chloroform solution of
the diporphyrin (5.7 × 10−5 M) was excited at a pump
wavelength of 1090 nm, exciting at the peak of band III to
prepare vibrationally excited molecules in their S1 electronic
state. The solutions were then probed at 580 and 700 nm to
monitor recovery of the ground state difference signal.
These results demonstrate that the Senge methodology is a
convenient and rapid route to meso−meso singly linked
diporphyrinsa precursor to triply fused diporphyrins. The
advantage of the Senge methodology over other routes to triply
fused diporphyrins is that the terminal meso positions of the
porphyrins are functionalized immediately prior to dimeriza-
tion, thus preventing the uncontrolled formation of porphyrin
oligomers that typically result from radical coupling methods,
such as the Ag(I)-mediated coupling procedure reported by
Aratani et al.39 This approach therefore circumvents the need
for 5,10,15-trisubstituted porphyrin starting materials. Further-
more, the end-capping groups of the diporphyrin can be
modified depending on the choice of organolithium reagent,
thus presenting the opportunity for installing further
functionality after dimerization.
Diporphyrin H-1DP was metalated quantitatively to the
zinc(II) chelate (Zn-1DP), which was planarized using
Sc(OTf)3−DDQ as an oxidant to give the corresponding triply
fused diporphyrin (Zn-3DP) in moderate yield (64%) (Scheme
1). Demetalation of Zn-3DP under acidic conditions afforded
the free-base triply fused diporphyrin (H-3DP), which was
smoothly converted to the palladium(II) chelate by treatment
with Pd(OAc)2 under basic conditions.
Electronic absorption spectra of H-3DP, Zn-3DP, and Pd-
3DP are consistent with the spectra of hexaaryl-substitued
diporphyrins,1,3,24 displaying three distinct absorption bands
(Figure 1). For consistency, these bands are labeled I, II, and III
7900
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
3DP. In our experiment, molecules in the sample are excited
only to S1 using a pump wavelength of 1090 nm; as such,
deactivation processes from within the S1 manifold are
observable. Therefore, we assign the fast (3 ps) decay
component to nonradiative relaxation from an excited vibra-
tional level within the S1 state followed by rapid S1→S0 (and
some S1→T1) deactivation (14 ps), most likely dominated by
nonradiative processes based on the very short lifetime of the S1
state (Figure 3a). These decay constants are consistent with the
10−20 ps time scales observed for nonradiative decay in other
multiporphyrin arrays.41,42
Figure 3. Jablonski diagrams summarizing the ultrafast excited-state
dynamics of metallodiporphyrins (a) Zn-3DP and (b) Pd-3DP.
Figure 2. Ultrafast transient absorption dynamics of Zn-3DP in
chloroform, pumping at 1090 nm and probing at (a) 700 nm and (b)
580 nm. The experimental data are fitted to a biexponential decay
model (solid red line).
The reason for the different excited state dynamics we
observed for Zn-3DP, compared with the results of Cho et
al.,43 are not clear but appear to be strongly influenced by
substituent effects around the diporphyrin core. Indeed, reports
of other zinc(II) triply fused diporphyrins, differing only by
their peripheral substituents, display vastly different excited-
state behaviors.7,23,43 This point will be discussed in further
detail in the following sections.
Palladium(II) Triply Fused Diporphyrin (Pd-3DP). Transient
absorption dynamics at various probe wavelengths are displayed
in Figure S6 (Supporting Information). The resulting transient
absorption spectra from the S1 and T1 states are depicted in
Figure 4 with the decay lifetimes summarized in Figure 5.
At a probe wavelength of 580 nm, the transient absorption
consisted of ground state bleaching followed by complete
biexponential recovery to the prepump difference signal (ΔA =
0), with decay constants of 3.8
0.2 and 14.4
0.7 ps
contributing to 57 and 43% of the total decay profile,
respectively. Excited-state absorption was observed at the 700
nm probe wavelength, followed by near-complete biexponential
recovery to the prepump difference signal with decay constants
of 3.0 1.0 and 12.6 0.7 ps, contributing to 23 and 77% of
the observed decay profile, respectively. Close observation of
the decay profile at λpr = 700 nm reveals incomplete recovery of
the difference signal, thus suggesting that Zn-3DP might
possess a very small triplet yield.
Cho et al. have also reported biexponential decay processes
in an hexaaryl triply fused zinc(II) diporphyrin, whereby a short
decay component (4.5 ps) was attributed to extremely rapid
S1→S0 deactivation, and a much longer component (2.2 ns)
was tentatively assigned to the excited-state dynamics of a
monomeric porphyrin contaminant in the experimental
sample.2 The relative contributions of the fast and slow decay
components were 98 and 2%, respectively. These authors also
recorded fluorescence from the S2 state with a lifetime of 15 ps
and a very small fluorescence quantum yield of ∼10−5,
suggesting that either their diporphyrin disobeys Kasha’s rule
or that the putative contaminant in their sample gave rise to the
observed fluorescence. These findings differ substantially from
our own, not only in the decay components but, moreover, we
did not observe fluorescence in the 600−700 nm range upon
excitation of the Soret band that could be attributed to Zn-
Figure 4. Transient absorption spectra of the S1 (red circles) and T1
(blue triangles) states of Pd-3DP. Pump wavelength was fixed at 920
nm.
7901
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
heavy atom effect due to the larger atomic mass of palladium
compared to zinc.38
The T1 lifetime of Pd-3DP was measured using pump and
probe wavelengths of 560 and 780 nm, respectively, and
extending the pump−probe time delay to 900 ps (the
maximum delay on our instrumentation). The resulting ground
state bleach and biexponential recovery revealed a very short T1
lifetime of 1.7 0.3 ns, which we attribute to strong T1→S0
reverse ISC due both to spin−orbit coupling introduced by the
heavy palladium nucleus and the relatively small T1−S0 energy
gap expected for this molecule. This observation represents a
manifestation of the heavy atom effect that is highly undesirable
for applications such as photodynamic therapy and PUC, which
require molecules with long-lived triplet states. A similar result
was reported by Anderson and co-workers, who observed no
triplet yield on the nanosecond time scale for a lead(II) triply
fused diporphyrin.7 Interestingly, the authors did note a
significant triplet yield and lifetime on the order of 250 ns
for a meso-brominated zinc(II) triply fused diporphyrin.7 These
results underscore the care with which the heavy atom effect
should be employed when trying to synthesize compounds with
high triplet yields and lifetimes, especially since the heavy atom
effect is only one of many factors that govern intersystem
crossing efficiency and triplet lifetime in organic molecules.38
The S1 decay profile at each probe wavelength was fitted to a
monoexponential decay function, with the exception of the
experiments probing at 580 and 590 nm, which were better
described by a biexponential model (Supporting Information,
Figure S6). These experiments returned an average long decay
component with a time constant of 18.0 1.0 ps and a shorter
decay component with a time constant of 8.0 1.0 ps (Figure
5). The suitability of the biexponential fits at 580 and 590 nm
probe wavelengths suggests that the probe beam is illuminating
two concomitant deactivation processes. We ascribe the fast
decay process (8 ps) to vibrational cooling within the S1
manifold, while the slower decay process (18 ps) is assigned
to deactivation of S1 via both singlet and triplet channels
(Figure 3).
Figure 5. Summary of the excited state lifetimes of Pd-3DP at various
probe wavelengths between 560 and 660 nm. All traces were fitted to a
monoexponential decay function, except for λprobe at 580 and 590 nm,
which exhibited biexponential decay dynamics. Slow decay compo-
nents are shown in red; fast decay components are shown in blue.
Error bars indicate one standard deviation in the curve-fitting
procedure.
A chloroform solution of Pd-3DP (2.4 × 10−5 M) was
excited at a pump wavelength of 920 nm, exciting at the peak of
band III, to prepare molecules in their S1 electronic states. The
sample was then probed at 10 nm intervals from 560 to 660
nm, and the time-resolved transient absorption profiles were
recorded (Supporting Information, Figure S6). Ground-state
bleaching was observed for probe wavelengths at and below 590
nm, and excited-state absorption was observed at wavelengths
above 600 nm. In each case, the difference signal did not
recover to the prepump level but instead displayed a persistent
nonzero difference signal that decayed with a time constant of
1.7 ns (Figure 6). Since the pump beam was selected to excite
the absorbing molecules to S1, the most likely state that could
give rise to a persistent nonzero difference signal is T1. As such,
we propose that Pd-3DP crosses to the T1 state following
photoexcitation to S1, followed by rapid nonradiative decay
back to the ground state. This observation is consistent with the
The dynamics can be modeled in terms of three states with
differing absorption properties. Molecules are initially prepared
in S1*, which relaxes at rate k1 to S1. This vibrationally relaxed,
electronically excited level then decays back to S0 (and T1) at
rate k2. Thus
[S1*]t = [S1*]0 exp(−k1t)
(1)
[S1*]0k
k1 − k2
[S1]t =
1 (exp(−k2t) − exp(−k1t))
(2)
and
[S ] + [T] = [S *]
⎛
k2 exp(−k1t)
k1 − k2
⎜
⎝
0 t
1 t
1
0
⎞
k1 exp(−k2t)
k1 − k2
−
+ 1
⎟
⎠
(3)
The transient absorption signal may be modeled as the sum
of the contributions from the three species
Figure 6. Ultrafast transient absorption dynamics of Pd-3DP in
chloroform using an extended time delay, pumping at 560 nm, and
probing at 780 nm. Experimental data are fitted to a biexponential
decay model (red line).
3
ΔA (λ) = l
Δε(λ)[i]
∑
t
i
t
(4)
i
7902
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
where l is the path length, and Δεi(λ) is the difference of the
extinction coefficients of species i and the unexcited
chromophore at wavelength λ. The quantity ΔAt(λ) may thus
be fit to a constant and two exponential terms.
Intuitively, both the fast and the slow decay processes should
be occurring at each probe wavelength. However, they are only
observable at probe wavelengths of 580 and 590 nm. This can
be explained by considering the transient absorption spectra for
a vibrationally excited sublevel and the vibrational ground state
of S1, as depicted schematically in Figure 7. Upon initial
induced absorption between these two probe wavelengths
(Figure 4). Therefore, after blue-shifting due to vibrational
relaxation within S1, the molecule absorbs more strongly at 580
and 590 nm. As a result, the vibrational cooling processes
within S1on the order of 8 psare observable at these probe
wavelengths, manifesting in an apparent increase in the rate of
recovery of the ground-state difference signal. The reason as to
why this behavior was not observed at other wavelengths is that
the change in signal intensity upon vibrational cooling is only
significant at the band edge, particularly as the diporphyrin
excited-state absorption bands are quite sharp in the region
570−590 nm (cf. Figure 4). This behavior is reminiscent of the
ultrafast transient absorption dynamics of sterically crowded
nickel(II) porphyrins reported by Holten and co-workers, who
observed wavelength-dependent vibrational relaxation on the
order of 7 ps within the excited state manifolds of these
porphyrins.45,46
The longer S1 lifetime of Pd-3DP compared with Zn-3DP
can be explained by the larger S1−S0 energy gap in the former
compound. This is evident from the blue-shifting of band III in
the electronic absorption spectrum of Pd-3DP (Figure 1), with
respect to Zn-3DP, which suggests that palladium(II) has a
similar effect on stabilizing the HOMO (b2g)44 of the
diporphyrin as it does with the a2u orbital in monomeric
porphyrins.
Trends in Excited-State Dynamics of Diporphyrins.
Comparing the excited-state lifetimes of Zn-3DP and Pd-
3DP with similar triply fused diporphyrins reported in the
literature provides further insight into the photophysical trends
within this class of molecule. The groups of Osuka, Kim, and
Anderson have measured the excited-state dynamics of other
triply fused diporphyrins, which exhibit vastly different behavior
despite very similar molecular structures (cf. Table 1).2,7,23,36
On the basis of the data compiled in Table 1, the substituents
Figure 7. Schematic depicting the positions of the transient absorption
spectra in a vibrationally excited state (red) and at the vibrationless
level (blue) of S1, which would give rise to the excited state dynamics
observed for Pd-3DP. Nonradiative decay of the hot-band spectrum
would be most clearly noticed in the shaded region.
excitation, the transient absorption is dominated by hot band
transitions from S1 (red spectrum). As nonradiative decay to
the vibrationless level in S1 occurs, the transient absorption
spectrum shifts to shorter wavelengths (blue spectrum). The
edge of the hot band spectrum is between 590 and 600 nm,
which can be inferred from the change from bleaching to
a
Table 1. Comparison of the Excited State Dynamics of Triply Fused Metallodiporphyrins
n
R1
R2
M
τ(S1)
τ(T1)
Φ(T1)
solvent
toluene
reference
1
1
0
0
0
0
0
0
0
Ar1
Ar1
Bu
Br/I
Bu
I
Ar2
Ar2
Ar1
Ar1
Ar1
Ar1
Ar1
Ar1
Zn(II)
Zn(II)
Zn(II)
Pb(II)
Pd(II)
Zn(II)
Zn(II)
Zn(II)
2.3 0.2 ps
−
−
−
−
−
2
2
4.5 0.5 ps
−
toluene
2
a
Zn-3DP
13.5 0.7 ps
CHCl3
herein
3
<5 ns
−
−
0.2
CHCl3/pyridine
CHCl3
7
Pd-3DP
18.0 1.0 ps
1.7 0.3 ns
herein
4
5
6
−
−
−
52 ns
CHCl3/pyridine
CHCl3/pyridine
CHCl3/pyridine
7
Br
280 ns
0.12
b
7
CC−Si(iPr)3
177 25 μs
23
a
b
For compounds 3−6, the solvent is 1% (v/v) pyridine in CHCl3. Specific excited-state values were not reported for dashed entries. Triplet yields
for these molecules were extremely small.
7903
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
around the periphery of the diporphyrin core appear to
influence the excited-state dynamics of the molecule strongly.
For instance, excited-state deactivation is extremely rapid (<5
ps) in porphyrin tapes with only aryl substituents (1 and 2),
and the decay constant of deactivation is the same regardless of
the pump wavelength.2 Changing the substituents at the
terminal meso-like positions to (triisopropylsilyl)alkynyl (4),
bromo (5), or iodo (6) groups results in diporphyrins with
measurable triplet yields and triplet lifetimes ranging from tens
of nanoseconds to hundreds of microseconds.7,23
The dominant pathway of nonradiative deactivation in triply
fused diporphyrins remains under debate. Osuka and colleagues
have attributed the ultrafast S1 decay of compounds 1, 2, and
related porphyrin tapes (n > 1), which have S1 lifetimes on the
order or 2−5 ps, to internal conversion as described by the
energy gap law.2,36 Anderson and co-workers have recently
reported, however, that extensive deuteration of the triply fused
diporphyrin framework has no effect on the fluorescence
quantum yield of the molecule.23 Since C−H vibrations
strongly influence the Franck−Condon factor of internal
conversion, this finding indicates that nonradiative deactivation
of the diporphyrin S1 state is not governed by S1−S0 internal
conversion. Instead, Anderson and co-workers propose an
alternative deactivation pathway involving a conical intersection
of the S1 and S0 potential energy surfaces.23
Comparing the excited state dynamics of Zn-3DP with
diporphyrin 2 provides an interesting insight to this ongoing
debate. Both compounds have identical ground-state electronic
absorption spectra, signifying that the positions of their
electronic energy levels are very similar. However, the S1
lifetime of Zn-3DP is more than three times that of 2 and
nearly seven times shorter than the ∼100 ps S1 lifetime that
Frampton et al. calculated for nonradiative deactivation
according to the energy gap law (based on an S1−S0 energy
gap of 109 kJ mol−1 or 1100 nm).23 Therefore, it follows that
the energy gap law alone is not sufficient to explain the rate of
S1→S0 deactivation in these triply fused diporphyrins. On the
basis of this apparent substituent effect, it seems likely that the
meso substituents strongly affect the access to the conical
intersection as put forward by Anderson and co-workers.23
Hence, while it is undoubtedly true that the S1−S0 energy gap
generally influences the rate of nonradiative decay in organic
molecules, the excited-state dynamics of Zn-3DP and 2 suggest
that there is not a simple relationship between the S1−S0
energy gap and nonradiative deactivation rates in triply fused
diporphyrins.
Rapid nonradiative deactivation of the S1 state in
diporphyrins 1, 2, and Zn-3DP, and their negligible triplet
yields, severely limits their suitability for many photophysical
and photochemical applicationsin particular as sensitizer
compounds for PUC. A common approach for increasing the
triplet yield of an organic compound is to incorporate heavy
atoms into its molecular structure.38 The heavy atom
introduces electronic spin−orbit coupling, thus accelerating
the rate of ISC. The success of this concept can be seen by the
increased triplet yields of heavy atom-containing diporphyrins
Pd-3DP, 4, and 5 as compared to diporphyrins 1, 2, and Zn-
3DP (Table 1). The heavy atom effect also has the adverse
consequence of drastically decreasing the triplet lifetime due to
the increased rate of T1→S0 reverse ISC. Notably, the triplet
lifetimes decrease in the order of τ5 > τ4 > τ3 as the atomic
number of the corresponding heavy atom increases ZBr < ZI <
ZPb. The fact that Pd-3DP has a shorter triplet lifetime than 4,
despite palladium having a lower atomic number than iodine,
suggests that the position of the heavy atom within the
diporphyrin framework influences the degree to which spin−
orbit coupling is affected. A possible reason for this observation
could originate from the large spin density on the pyrrolic
nitrogen atoms predicted for the HOMO and LUMO orbitals
of the triply fused diporphyrin,44 thus resulting in a strong
interaction between the metal center and the π-electrons of the
porphyrin excited state. In terms of its applicability for PUC,
the short triplet lifetime of Pd-3DP precludes its use as a
sensitizer molecule. Indeed, an emitter concentration of ∼1 M
would be needed for sufficient triplet energy transfer.
Zinc(II) diporphyrin 6, with (triisopropylsilyl)alkynyl
terminal meso substituents, has a very long triplet lifetime
(177 μs) compared to the other diporphyrins in Table 1.
However, Anderson and co-workers found that the triplet yield
of this compound is extremely low,7 which is consistent with
the absence of a heavy nucleus from the molecular structure
and also with the low triplet yields typically observed in zinc(II)
porphyrins.38 Hence, the rate of ISC in 6 is slow, which results
in both a low triplet yield (slow S1→T1) and a long triplet
lifetime (slow T1→S0). Although the triplet state of 6 is
sufficiently long-lived for undergoing energy transfer, the very
low triplet yield limits its utility for PUC. On the basis of the
data in Table 1, incorporation of heavy atoms around the
periphery of the diporphyrin, as exemplified by meso-
brominated diporphyrin 5, represents the most promising
concept for achieving an acceptable balance between triplet
yield and lifetime.
CONCLUSIONS
■
We have synthesized butyl-end-capped triply fused diporphyr-
ins Zn-3DP and Pd-3DP in good yields using an efficient two-
step strategy that combines the work of the Senge and Osuka
research groups.5,33 The ultrafast excited-state dynamics of
these diporphyrins were subsequently measured and compared
with similar compounds in the recent literature, revealing that
the substituents in the terminal (15,15′) meso positions of the
diporphyrin strongly influence the excited-state dynamics of the
molecule. Different excited-state dynamics observed for Zn-
3DP and literature2 diporphyrin 2, despite similar chemical
structures and identical electronic absorption spectra, suggests
that the energy gap law alone does not explain the rapid
excited-state deactivation observed for these molecules. The
hypothesis of nonradiative deactivation via a crossing of the S1−
S0 potential energy surfaces, as proposed by Anderson and co-
workers,23 is consistent with our data.
Triply fused diporphyrins are potential candidates for
application as NIR sensitizer materials in PUC due to their
intense NIR absorption features. While the compounds
reported herein lack the triplet yield and lifetime for immediate
application, comparison with different diporphyrins reveals
trends for improvement. Notably, metalation of the diporphyrin
with palladium(II) has the consequence of increasing T1→S0
reverse ISC significantly, thus resulting in a very short triplet
lifetime. However, replacing meso-substituents with halogen
(Br, I) atoms produces a reasonable balance between triplet
yield and triplet lifetime, also indicating that the position of the
heavy atom within the diporphyrin framework influences how
the heavy atom effect manifests. Hence, modifications at the
periphery of the diporphyrin framework seem to offer the most
promising route to NIR-PUC sensitizers.
7904
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905

The Journal of Physical Chemistry A
Article
(16) Cheng, Y. Y.; Khoury, T.; Clady, R.; Tayebjee, M. J. Y.; Ekins-
Daukes, N. J.; Crossley, M. J.; Schmidt, T. W. Phys. Chem. Chem. Phys.
2010, 12, 66−71.
ASSOCIATED CONTENT
* Supporting Information
Full synthetic details and characterization of the diporphyrins,
selected 1H NMR and mass spectra and time-resolved transient
absorption dynamics traces for Pd-3DP. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
(17) Cheng, Y. Y.; Fuckel, B.; Khoury, T.; Clady, R. G. C. R.;
̈
Tayebjee, M. J. Y.; Ekins-Daukes, N. J.; Crossley, M. J.; Schmidt, T. W.
J. Phys. Chem. Lett. 2010, 1, 1795−1799.
(18) Baluschev, S.; Yakutkin, V.; Miteva, T.; Avlasevich, Y.; Chernov,
S.; Aleshchenkov, S.; Nelles, G.; Cheprakov, A.; Yasuda, A.; Mullen,
K.; Wegner, G. Angew. Chem., Int. Ed. 2007, 46, 7693−7696.
AUTHOR INFORMATION
Corresponding Author
*E-mail: maxwell.crossley@sydney.edu.au (M.J.C.), timothy.
schmidt@sydney.edu.au (T.W.S.).
Notes
■
(19) Fuckel, B.; Roberts, D. A.; Cheng, Y. Y.; Clady, R.; Piper, R. B.;
̈
Ekins-Daukes, N. J.; Crossley, M. J.; Schmidt, T. W. J. Phys. Chem. Lett.
2011, 2, 966−971.
(20) Singh-Rachford, T. N.; Nayak, A.; Muro-Small, M. L.; Goeb, S.;
Therien, M. J.; Castellano, F. N. J. Am. Chem. Soc. 2010, 132, 14203−
11.
The authors declare no competing financial interest.
(21) Englman, R.; Jortner, J. Mol. Phys. 1970, 18, 145−164.
(22) Brennan, B. J.; Kenney, M. J.; Liddell, P. A.; Cherry, B. R.; Li, J.;
Moore, A. L.; Moore, T. A.; Gust, D. Chem. Commun. 2011, 47,
10034−10036.
(23) Frampton, M. J.; Accorsi, G.; Armaroli, N.; Rogers, J. E.; Fleitz,
P. A.; McEwan, K. J.; Anderson, H. L. Org. Biomol. Chem. 2007, 5,
1056−1061.
(24) Kamo, M.; Tsuda, A.; Nakamura, Y.; Aratani, N.; Furukawa, K.;
Kato, T.; Osuka, A. Org. Lett. 2003, 5, 2079−2082.
(25) Nakamura, Y.; Aratani, N.; Tsuda, A.; Osuka, A.; Furukawa, K.;
Kato, T. J. Porphyrins Phthalocyanines 2003, 7, 264−269.
(26) Ouyang, Q.; Zhu, Y. Z.; Zhang, C. H.; Yan, K. Q.; Li, Y. C.;
Zheng, J. Y. Org. Lett. 2009, 11, 5266−5269.
(27) Blake, I. M.; Krivokapic, A.; Katterle, M.; Anderson, H. L. Chem.
Commun. 2002, 1662−1663.
(28) Bonifazi, D.; Scholl, M.; Song, F. Y.; Echegoyen, L.; Accorsi, G.;
Armaroli, N.; Diederich, F. Angew. Chem., Int. Ed. 2003, 42, 4966−
4970.
(29) Inokuma, Y.; Ono, N.; Uno, H.; Kim, D. Y.; Noh, S. B.; Kim, D.;
Osuka, A. Chem. Commun. 2005, 3782−3784.
ACKNOWLEDGMENTS
■
D.A.R. acknowledges the Australian government for an
Australian Postgraduate Award and the University of Sydney
for an Honours Scholarship and the Vice Chancellor’s Research
Scholarship. D.A.R. thanks Prof. Jeffrey Reimers for helpful
discussions. B.F. acknowledges the Alexander von Humboldt-
Foundation for a Feodor Lynen fellowship. R.G.C.R.C.
acknowledges the Australian Solar Institute for a postdoctoral
research fellowship. Y.C. acknowledges The University of
Sydney for a Gritton Fellowship. The authors thank Duncan
Wild for assistance with building the integrating photodiodes.
This research project is funded by the Australian Solar Institute,
with contributions from The New South Wales Government
and The University of Sydney. Aspects of this research were
supported under Australian Research Councils Discovery
Projects funding scheme (DP110103300 to T.W.S. and
DP1092560 to M.J.C.). Equipment was purchased with support
from the Australian Research Council (LE0668257).
(30) Sahoo, A. K.; Nakamura, Y.; Aratani, N.; Kim, K. S.; Noh, S. B.;
Shinokubo, H.; Kim, D.; Osuka, A. Org. Lett. 2006, 8, 4141−4144.
(31) Osuka, A.; Shimidzu, H. Angew. Chem., Int. Ed. 1997, 36, 135−
137.
REFERENCES
■
(1) Tsuda, A.; Furuta, H.; Osuka, A. Angew. Chem., Int. Ed. 2000, 39,
2549−2552.
(2) Cho, H. S.; Jeong, D. H.; Cho, S.; Kim, D.; Matsuzaki, Y.;
Tanaka, K.; Tsuda, A.; Osuka, A. J. Am. Chem. Soc. 2002, 124, 14642−
14654.
(32) Aratani, N.; Osuka, A.; Kim, Y. H.; Jeong, D. H.; Kim, D. Angew.
Chem., Int. Ed. 2000, 39, 1458−1462.
(33) Senge, M. O.; Feng, X. D. J. Chem. Soc., Perkin Trans. 1 2000,
3615−3621.
(34) Senge, M. O.; Feng, X. D. Tetrahedron Lett. 1999, 40, 4165−
4168.
(3) Tsuda, A.; Furuta, H.; Osuka, A. J. Am. Chem. Soc. 2001, 123,
10304−10321.
(35) Senge, M. O. Acc. Chem. Res. 2005, 38, 733−743.
(36) Kim, P.; Ikeda, T.; Lim, J. M.; Park, J.; Lim, M.; Aratani, N.;
Osuka, A.; Kim, D. Chem. Commun. 2011, 47, 4433−4435.
(37) Scholl, R.; Mansfeld, J. Ber. Dtsch. Chem. Ges. 1910, 43, 1734−
1746.
(38) Solovyov, K. N.; Borisevich, E. A. Phys.−Usp. 2005, 48, 231−
253.
(39) Aratani, N.; Osuka, A. Chem. Rec. 2003, 3, 225−234.
(40) Senge, M. O.; Pintea, M.; Ryan, A. A. Z. Naturforsch. B 2011, 66,
553−558.
(41) Kumble, R.; Palese, S.; Lin, V. S. Y.; Therien, M. J.;
Hochstrasser, R. M. J. Am. Chem. Soc. 1998, 120, 11489−11498.
(42) Rodriguez, J.; Kirmaier, C.; Holten, D. J. Am. Chem. Soc. 1989,
111, 6500−6506.
(43) Cho, H. S.; Song, N. W.; Kim, Y. H.; Jeoung, S. C.; Hahn, S.;
Kim, D.; Kim, S. K.; Yoshida, N.; Osuka, A. J. Phys. Chem. A 2000, 104,
3287−3298.
(44) Cho, S.; Yoon, M.-C.; Kim, K. S.; Kim, P.; Kim, D. Phys. Chem.
Chem. Phys. 2011, 13, 16175−16181.
(45) Drain, C. M.; Gentemann, S.; Roberts, J. A.; Nelson, N. Y.;
Medforth, C. J.; Jia, S.; Simpson, M. C.; Smith, K. M.; Fajer, J.;
Shelnutt, J. A.; Holten, D. J. Am. Chem. Soc. 1998, 120, 3781−3791.
(46) Drain, C. M.; Kirmaier, C.; Medforth, C. J.; Nurco, D. J.; Smith,
K. M.; Holten, D. J. Phys. Chem. 1996, 100, 11984−11993.
(4) Tsuda, A.; Osuka, A. J. Inclusion Phenom. Macrocycl. Chem. 2001,
41, 77−81.
(5) Tsuda, A.; Osuka, A. Science 2001, 293, 79−82.
(6) Ikeda, T.; Aratani, N.; Osuka, A. Chem.−Asian J. 2009, 4, 1248−
1256.
(7) McEwan, K. J.; Fleitz, P. A.; Rogers, J. E.; Slagle, J. E.; McLean, D.
G.; Akdas, H.; Katterle, M.; Blake, I. M.; Anderson, H. L. Adv. Mater.
2004, 16, 1933−1935.
(8) Diev, V. V.; Hanson, K.; Zimmerman, J. D.; Forrest, S. R.;
Thompson, M. E. Angew. Chem., Int. Ed. 2010, 49, 5523−5526.
(9) Bonifazi, D.; Accorsi, G.; Armaroli, N.; Song, F. Y.; Palkar, A.;
Echegoyen, L.; Scholl, M.; Seiler, P.; Jaun, B.; Diederich, F. Helv. Chim.
Acta 2005, 88, 1839−1884.
(10) Dougherty, T. J.; Gomer, C. J.; Henderson, B. W.; Jori, G.;
Kessel, D.; Korbelik, M.; Moan, J.; Peng, Q. J. Natl. Cancer Inst. 1998,
90, 889−905.
(11) Singh-Rachford, T. N.; Castellano, F. N. Coord. Chem. Rev.
2010, 254, 2560−2573.
(12) Zhao, J. Z.; Ji, S. M.; Guo, H. M. RSC Adv. 2011, 1, 937−950.
(13) Yakutkin, V.; Aleshchenkov, S.; Chernov, S.; Miteva, T.; Nelles,
G.; Cheprakov, A.; Baluschev, S. Chem.−Eur. J. 2008, 14, 9846−9850.
(14) Crossley, M. J.; Sheehan, C. S.; Khoury, T.; Reimers, J. R.;
Sintic, P. J. New J. Chem. 2008, 32, 340−352.
(15) Khoury, T.; Crossley, M. J. Chem. Commun. 2007, 4851−4853.
7905
dx.doi.org/10.1021/jp3043895 | J. Phys. Chem. A 2012, 116, 7898−7905