Excited-state energy transfer and ground-state hole/electron hopping in p-phenylene-linked porphyrin dimers

Article abstract of DOI:10.1021/jp982729o

The ground- and excited-state properties of a series of p-phenylene-linked porphyrin dimers have been examined using a variety of static and time-resolved spectroscopic techniques. The dimers consist of a zinc porphyrin and a free base (Fb) porphyrin (ZnF

Full text of DOI:10.1021/jp982729o

9426
J. Phys. Chem. B 1998, 102, 9426-9436
Excited-State Energy Transfer and Ground-State Hole/Electron Hopping in
p-Phenylene-Linked Porphyrin Dimers
Sung Ik Yang,† Robin K. Lammi,† Jyoti Seth,‡ Jennifer A. Riggs,§ Toru Arai,§,| Dongho Kim,†,
David F. Bocian,*,‡ Dewey Holten,*,† and Jonathan S. Lindsey*,§
Department of Chemistry, North Carolina State UniVersity, Raleigh, North Carolina 27695-8204, Department
of Chemistry, Washington UniVersity, St. Louis, Missouri 63130-4899, and Department of Chemistry,
UniVersity of California, RiVerside, California 92521-0403
ReceiVed: June 23, 1998; In Final Form: September 23, 1998
The ground- and excited-state properties of a series of p-phenylene-linked porphyrin dimers have been examined
using a variety of static and time-resolved spectroscopic techniques. The dimers consist of a zinc porphyrin
and a free base (Fb) porphyrin (ZnFbΦ), two zinc porphyrins (Zn2Φ), or two Fb porphyrins (Fb2Φ). In each
array, the porphyrins are joined by the p-phenylene linker at one meso position, with the nonlinking meso
positions bearing mesityl groups. Three analogous dimers in which the mesityl groups are replaced with
pentafluorophenyl groups (F30ZnFbΦ, F30Zn2Φ, and F30Fb2Φ) were also synthesized and characterized. The
excited-state energy-transfer rate from the photoexcited Zn porphyrin to the Fb porphyrin is (3.5 ps)-1 for
ZnFbΦ and (10 ps)-1 for F30ZnFbΦ. The quantum yields of excited-state energy transfer are g99% for both
complexes. The energy-transfer rates in the p-phenylene-linked dimers are considerably faster than those
observed for the analogous dimers containing a diphenylethyne linker ((24 ps)-1, ZnFbU; (240 ps)-1,
F30ZnFbU). At these distances, both through bond and through space contributions to the electronic coupling
are important. The faster energy-transfer rates in the p-phenylene- versus diarylethyne-linked dimers are
attributed to enhanced electronic coupling between the porphyrins in the former dimers arising primarily
from the shorter inter-porphyrin separation. The electronic coupling in the p-phenylene-linked dimers is
sufficient to support ultrafast energy transfer in both ZnFbΦ and F30ZnFbΦ, but is not so large as to significantly
perturb the redox or inherent lowest excited-state photophysical properties of the porphyrin constituents.
Electronic perturbations resulting from fluorination have little effect on the energy-transfer rates in the
p-phenylene-linked dimers, but the rates of room-temperature ground-state hole/electron hopping processes
in the corresponding monocation radicals of the bis-Zn analogues of the p-phenylene-linked dimers (g(0.05
µs)-1, [Zn2Φ]+; e(2.5 µs)-1, [F30Zn2Φ]+) are significantly influenced by the fluorination-induced changes in
the electronic structure. Collectively, these characteristics make these constructs attractive candidates for
incorporation into extended multi-porphyrin arrays for a variety of molecular photonics applications.
1. Introduction
contribution from a through-space (TS) process. The rate of
energy transfer from zinc porphyrin to free base (Fb) porphyrin
in the ZnFb dimer, which contains a torsionally unhindered p,p′-
diphenylethyne linker (ZnFbU), is (24 ps)-1, giving an energy-
transfer yield of 99%.
Understanding the factors that control excited-state energy
transfer is essential for the rational design of a wide variety of
molecular photonic devices. We have synthesized light-
harvesting arrays,1 a molecular photonic wire,2 and optoelec-
tronic gates3 that all rely on excited-state energy transfer. Each
of these devices incorporates tetraarylporphyrins joined by the
semirigid4 p,p′-diarylethyne linker at the porphyrin meso
positions. Recently, we have synthesized several series of
porphyrin dimers in order to examine the factors that affect the
energy-transfer process. These dimers have probed the effects
of torsional constraints on the linker,5,6 the rotation of the
porphyrins about the ethyne linker,7 and the nature of the metal
in the metalloporphyrin.8 The major findings from these studies
are that energy transfer predominantly involves a through-bond
(TB) process mediated by the diarylethyne linker and a minor
More recently, we examined a series of dimers that revealed
the effects of the nature and electron-density distributions of
the frontier molecular orbitals (HOMOs and LUMOs) of the
porphyrin constituents.9 The HOMO of each porphyrin is either
a1u(π) or a2u(π), depending on factors such as the positions and
electron-donating/withdrawing character of the peripheral sub-
stituents. The energy-transfer rate in F30ZnFbU ((240 ps)-1),
which contains pentafluorophenyl groups at all nonlinking meso
carbons, is an order of magnitude slower than that in ZnFbU
((24 ps)-1), which contains mesityl groups at these positions.
The structures of these two dimers are shown in Figure 1. The
a1u HOMOs of the porphyrins in F30ZnFbU have nodes at the
meso positions, including those to which the linker is appended.
In contrast, the a2u HOMOs of the constituents of ZnFbU have
substantial electron density at these positions. This reversal of
orbital ordering, together with a redistribution of electron density
in the HOMOS and eg(π*) LUMOS accompanying incorpora-
† Washington University.
‡ University of California.
§ North Carolina State University.
| Permanent Address: Department of Applied Chemistry, Kyushu
Institute of Technology, Tobata, Kitakyushu 804, Japan.
Permanent Address: Spectroscopy Lab, Korea Research Institute of
Standards and Science, P.O. Box 102, Yusong, Taejon 305-600, Korea.
10.1021/jp982729o CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/30/1998

p-Phenylene-Linked Porphyrin Dimers
J. Phys. Chem. B, Vol. 102, No. 47, 1998 9427
exhibit an energy-transfer rate of (∼20 ps)-1 35
.
Another ZnFb
dimer bearing only meso substituents exhibits a rate estimated
to be faster than (10 ps)-1; however, the exact rate was not
determined.40 Many other architectures containing p-phenylene-
linked porphyrins have been prepared with the goal of inves-
tigating photoinduced electron transfer to an appended
acceptor.19,24,26,28-32,36,39-44 However, the electron-transfer
process in these arrays severely complicates the precise analysis
of underlying energy-transfer reactions.
In the present study, we have prepared and characterized two
types of p-phenylene-linked dimers substituted at the meso
positions, one with mesityl groups (ZnFbΦ) and the other with
pentafluorophenyl groups (F30ZnFbΦ). The structures of these
dimers are included in Figure 1. The goals of our study were
two-fold: (1) determination of how electronic communication
differs in p-phenylene- versus diarylethyne-linked dimers and
(2) the further examination of the effects of orbital characteristics
(ordering and electron density distributions) on electronic
communication in the arrays. Toward these objectives, we
compared the excited-state energy-transfer rates in ZnFbΦ
versus ZnFbU and F30ZnFbU versus F30ZnFbΦ, as well as the
ground-state hole/electron hopping rates in the monocation
radicals of the corresponding bis-Zn complexes of the respective
dimers.
2. Results
Figure 1. Structures of the diarylethyne- and p-phenylene-linked
dimers.
a. Synthesis. Prior syntheses of meso-(p-phenylene)-linked
multi-porphyrin architectures have employed various statistical
condensations,18-33 though one stepwise synthesis has recently
been developed.34 Most condensations require the synthesis of
porphyrin carboxaldehyde or dipyrromethane precursors,18-31
but the condensation of an aldehyde, terephthaldicarboxalde-
hyde, and pyrrole affords the p-phenylene-linked dimer in a one-
flask process using commercially available materials.32,33 We
have employed the latter approach in order to rapidly gain access
to p-phenylene-linked dimers bearing various nonlinking meso
substituents for spectroscopic analysis.
The p-phenylene-linked porphyrins were prepared using a
one-flask condensation of an aldehyde, terephthaldicarboxal-
dehyde, and pyrrole. This reaction is performed under the
standard conditions for the two-step one-flask synthesis of meso-
substituted porphyrins.49,50 A solution of pyrrole, mesitaldehyde,
and terephthaldicarboxaldehyde (molar ratio of 8:6:1; 10 mM
pyrrole) in chloroform (containing ethanol) at room temperature
was treated with boron trifluoride etherate (3.3 mM). The
reaction progress was monitored by the removal of aliquots and
their analysis by absorption spectroscopy. The yield of por-
phyrin was 10% after 30 min and 15% after 60 min. After 1 h,
the oxidant DDQ (7.5 mM) was added, and the mixture was
stirred at room temperature for an additional 1 h. The crude
reaction mixture consisted of small amounts of (uncharacterized)
non-porphyrinic materials in addition to the monomeric por-
phyrin, desired porphyrin dimer (Fb2Φ), and higher molecular
weight porphyrins as revealed by laser desorption mass spec-
trometry (LD-MS)51 and analytical size-exclusion column
chromatography (SEC).9 The purification scheme involved a
sequence of three consecutive chromatography procedures. Two
flash silica gel columns afforded a mixture of the porphyrin-
containing compounds, which were readily separated by pre-
parative SEC in toluene, affording the desired phenylene-linked
dimer Fb2Φ in 1.4% yield.52 This reaction at the 3 L scale
afforded 75 mg of Fb2Φ. This same approach was applied to
the reaction of pentafluorobenzaldehyde, terephthaldicarboxal-
dehyde, and pyrrole. The p-phenylene-linked dimer F30Fb2Φ
tion of the electron-withdrawing fluorines, substantially dimin-
ishes TB electronic coupling via the linker and reduces the rates
of both excited-state energy transfer (in the neutral F30ZnFbU
array) and ground-state hole/electron hopping (in the monocation
radical of the bis-Zn analogue, [F30Zn2U]+).
The diarylethyne linker used in our prototypical arrays and
dimers is quite attractive both from the standpoint of modular
construction10 as well as the fact that pairwise electronic
interactions are relatively weak and deleterious quenching
interactions are absent. However, it is desirable to construct
arrays with still faster energy-transfer rates. One approach to
increasing the energy-transfer rate is to shorten the distance
between the porphyrins. Multi-porphyrin arrays have been
constructed using several types of shorter linkers that are suitable
for preparing linear or extended architectures via meso position
attachment. These include porphyrins joined directly at the
meso positions (no linker)11 and those joined by ethene,12
ethyne,13butadiyne,13-16furan,14enyne,16hexatriene,17p-phenylene,18-
-
46
phenylethene,47 naphthalene,35,37 biphenyl,19,35-37 phenan-
threne,19 and ethynylphenylethyne16 groups. The attachment
of porphyrins via ethyne or butadiyne units results in strong
electronic coupling, as evidenced by the dramatic red-shifting
of the Q bands of the porphyrin. The direct meso-meso
connection results in weaker electronic coupling, as evidenced
by less perturbation of the optical spectrum, presumably due to
the orthogonality of the adjacent porphyrins. However, at
present, the construction of this connection lacks synthetic
control. Incorporation of a p-phenylene linker also results in
dimers that remain in the weak coupling limit as judged by their
absorption spectra.21,24,25,32,46 Equally important, the p-phe-
nylene group is amenable to a modular synthetic approach, a
key requirement for the systematic construction of molecular
devices.48
Relatively few of the previous p-phenylene-linked dimers
have been examined for their energy-transfer properties. One
ZnFb dimer bearing â-pyrrole substituents has been shown to

9428 J. Phys. Chem. B, Vol. 102, No. 47, 1998
Yang et al.
TABLE 1: Photophysical Data for the Arrays and
Reference Compoundsa
was obtained in 1.0% yield. Although these yields are quite
low, this synthetic procedure proved straightforward and
provided ready access to suitable quantities of p-phenylene-
linked dimers with the desired substituents at the nonlinking
meso positions.
Φf
τ (ns)b
compound
Zn
Fb
Zn
Fb
Dimers
0.10
The mono-Zn and bis-Zn complexes of the mesityl- and
pentafluorophenyl-substituted phenylene-linked porphyrin dimers
were prepared by treatment of the corresponding bis-Fb complex
with methanolic zinc acetate, resulting in a mixture of mono-
Zn and bis-Zn chelates that were readily separated by column
chromatography. In this way, the Fb2Φ and F30Fb2Φ complexes
were converted to the mono-Zn chelates ZnFbΦ and F30ZnFbΦ
in 37.8% and 44.0% yields, respectively. Treatment with excess
zinc acetate gave the desired bis-Zn porphyrin dimers Zn2Φ
and F30Zn2Φ in quantitative yield. The zinc insertion products
were readily characterized by absorption, fluorescence, and 1H
NMR spectroscopy. The 1H NMR spectra of the bis-Zn
porphyrin dimers (Zn2Φ and F30Zn2Φ) and the bis-Fb porphyrin
dimers (Fb2Φ and F30Fb2Φ) showed resonances that were
consonant with their symmetry, while in the mono-Zn dimers
(ZnFbΦ and F30ZnFbΦ) resonances were observed due to the
constituent Zn and Fb porphyrins.
ZnFbΦ
F30ZnFbΦ
Zn2Φ
F30Zn2Φ
Fb2Φ
F30Fb2Φ
c
c
0.0035
0.010
2.3
13.1
0.063d
0.038
0.027
1.6
0.10
13.0
11.8
0.060d
Monomers
ZnTMP
FbTMP
ZnTPP
0.039
0.033f
0.022e
0.019i
2.5
2.0
1.6e
1.4i
0.10
13.2
13.0
13.3e
11.2i
FbTPP
0.11g
ZnF15U′h
FbF15U′h
ZnF20TPP
FbF20TPP
0.060e
0.049i
a All measurements were made at room temperature in toluene.
b Lifetimes greater than 1 ns were determined by time-resolved emission
spectroscopy ((5%), and lifetimes less than 1 ns were determined by
time-resolved absorption spectroscopy ((10%). c Negligible emission
was observed from the Zn porphyrin. d The approximately 2-fold
reduced fluorescence yields of the Fb porphyrin in F30ZnFbΦ versus
ZnFbΦ and F30Fb2Φ versus Fb2Φ can be accounted for by a reduced
natural radiative lifetime of the lowest excited singlet state of the
fluorinated Fb porphyrin. The lower natural radiative rate for the latter
components corresponds to the reduced oscillator strength of corre-
sponding Qx(0,0) absorption band (when integrated along with the
Qx(1,0) band), observed in the corresponding ground-state absorption
spectra. A similar reduction in Fb emission yield has been found
previously for the diarylethyne-linked dimers, namely F30ZnFbU versus
ZnFbU.9 e From ref 53. f From ref 54a. This value is used as the
reference for the Zn porphyrin emission in all of the compounds studied.
g From ref 54b. This value is used as the reference for the Fb porphyrin
emission in all of the compounds studied. h These complexes contain
three meso-pentafluorophenyl groups and one meso-[4-(trimethylsilyl-
ethynyl)phenyl] group.9 i From ref 9.
b. Physical Properties of the Neutral Complexes. 1.
Absorption Spectra. The absorption spectra of the various
dimers were recorded in toluene at room temperature. The
absorption maxima are given in the Experimental Section. The
general spectral features of the nonfluorinated p-phenylene-
linked dimers (ZnFbΦ, Zn2Φ, and Fb2Φ) and the fluorinated
analogues (F30ZnFbΦ, F30Zn2Φ, and F30Fb2Φ) are as follows.
(1) The visible (Q) bands of the ZnFb dimers are essentially
the superposition of the spectra of the corresponding monomeric
Zn and Fb reference compounds (e.g., ZnTMP + FbTMP). (2)
The visible (Q) bands of the bis-Zn and bis-Fb dimers match
the spectra of the respective monomeric Zn or Fb reference
compounds. (3) Each of the dimers exhibits a splitting of the
Soret band that can be ascribed to exciton coupling of the two
porphyrin constituents, with the degree of splitting and relative
intensities of the bands depending on the particular complex.
excitation of the Zn versus Fb porphyrin component. However,
data were also acquired for ZnFbΦ with 580 nm excitation,
which increases the fraction of the Fb porphyrin that is excited
relative to the situation for 550 nm excitation. Although this
results in a corresponding increase in the contribution of the
Fb porphyrin spectral signatures to the transient spectra at early
times after excitation, the same kinetic parameters were obtained
with 580 nm excitation as with 550 nm excitation.
2. Fluorescence Spectra and Fluorescence Quantum Yields.
The fluorescence spectra and yields of the dimers and mono-
meric reference compounds were recorded in toluene at room
temperature. The emission yields are collected in Table 1, and
the emission wavelengths are given in the Experimental Section.
The fluorescence spectra and quantum yields of the bis-Fb and
bis-Zn dimers closely resembled those of the corresponding Fb
and Zn monomers, respectively. The emission spectrum and
yield of the Fb component of ZnFbΦ obtained upon illumination
at 515 nm, where the Fb porphyrin absorbs predominantly, are
essentially the same as those for the Fb monomer. However,
the fluorescence spectrum of ZnFbΦ obtained upon illumination
at 550 nm, where the Zn porphyrin absorbs predominantly,
showed emission almost exclusively from the Fb porphyrin. This
observation is consistent with a high yield of excited-singlet-
state energy transfer from the Zn to the Fb porphyrin. Es-
sentially identical results as those described for ZnFbΦ were
obtained for F30ZnFbΦ.
3. Transient Absorption Spectra. The energy-transfer rate
from the photoexcited Zn porphyrin to the Fb porphyrin subunit
in each of the dimers (ZnFbΦ and F30ZnFbΦ) was assessed
using transient absorption spectroscopy.55 For most measure-
ments, the complexes in toluene at room temperature were
excited using a 120 fs pulse at 550 nm, which predominantly
pumps the Zn porphyrin component of each array. This
excitation wavelength gives the best discrimination between
Hence, for both ZnFbΦ and F30ZnFbΦ, the transient absorp-
tion difference spectra observed shortly after excitation are
dominated by bleaching of the Q(1,0) and Q(0,0) ground state
absorption bands of the Zn porphyrin near 550 and 590 nm,
respectively (1 ps spectra in Figures 2 and 3). However, the
early-time spectra also contain some bleaching of the absorption
bands of the Fb porphyrin, since the Qy(0,0) absorption band
of this component also contributes to the absorption at the 550
nm excitation wavelength. This point is evidenced most clearly
by the small dip found near 515 nm in the 1 ps spectrum for
each array (Figures 2 and 3). This feature represents bleaching
of the Qy(1,0) ground-state band of the Fb porphyrin in the
fraction of the arrays in each sample in which this component,
rather than the Zn porphyrin, was excited by the excitation pulse.
(By 1 ps, the energy has also begun to flow from the
photoexcited Zn porphyrin to the Fb porphyrin, as discussed
below.) The Zn and Fb bleachings in these early-time spectra
are imbedded on a broad featureless transient absorption that
can be assigned to the respective excited states, Zn* and Fb*.

p-Phenylene-Linked Porphyrin Dimers
J. Phys. Chem. B, Vol. 102, No. 47, 1998 9429
Figure 3. Time-resolved spectra of F30ZnFbΦ. Data were acquired to
500 ps but are not shown to facilitate comparisons with the data for
ZnFbΦ given in Figure 2. Data acquired at t < 0 are also not shown
(see the legend to Figure 2).
Figure 2. Time-resolved spectral data for ZnFbΦ. Note, for clarity of
presentation, spectral and kinetic data at negative pump-probe delay
times (t < 0) are not shown but uniformly have ∆A ) 0.
At increasing times after excitation for both ZnFbΦ and F30-
ZnFbΦ, the contribution of Zn* to the transient spectra
diminishes and the contribution of Fb* increases. In particular,
the 550 nm bleaching of the Zn porphyrin decays and the 515
nm bleaching of the Fb porphyrin grows (see ∼20 ps spectra
in Figures 2 and 3). These spectral changes reflect energy
transfer from Zn* to the ground-state Fb component. In fact,
the amplitude ratios of the bleachings in the 20 ps spectrum for
ZnFbΦ match the amplitude ratios of the respective ground-
state absorption bands in the spectra of Fb porphyrin reference
compounds such as FbTPP and FbTMP. This observation
provides evidence that by 20 ps energy transfer from Zn* to
the Fb constituent of ZnFbΦ is essentially complete. Qualita-
tively similar results are found for F30ZnFbΦ, but the Zn
porphyrin contributions to the spectra remain at ∼20 ps,
indicating that energy transfer is slower in this complex than
in the nonfluorinated array.
The Zn* lifetimes in ZnFbΦ and F30FbΦ are obtained from
the time evolution of the spectra over the 450-600 nm region
investigated. Representative kinetic traces in the 550 nm region,
dominated by the Zn porphyrin, and in the 515 nm region,
dominated by the Fb porphyrin, are shown in Figure 2 for
ZnFbΦ and Figure 3 for F30ZnFbΦ. (Data for the F30ZnFbΦ
complex were acquired to ∼500 ps, but only the first 25 ps are
shown in Figure 3 to facilitate comparisons with the data for
ZnFbΦ shown in Figure 2.) Both the decay of the Zn porphyrin
bleaching (550 nm) and the growth of the Fb porphyrin
bleaching (515 nm) for ZnFbΦ occur with a time constant of
3.5 ( 0.5 ps. Similar results were also found using 580 nm
excitation. The Zn* lifetime is approximately 3-fold longer for
F30ZnFbΦ. In particular, the Zn bleaching decay at 550 nm
and the Fb porphyrin bleaching growth at 515 nm both occur
with time constants of 10 ( 1 ps. Measurements were also
carried out on ZnFbΦ in acetone/CH2Cl2 ) 4:1, using 580 nm
excitation flashes (the complex is not sufficiently soluble in a
pure polar solvent to carry out transient absorption measure-
ments). The Zn* lifetime obtained from these measurements
(4 ( 1 ps) is similar to that obtained for this complex in toluene.
c. Energy-Transfer Rates. The transient absorption data
and static fluorescence data all lead to the assessment that energy
transfer from Zn* to the Fb porphyrin in ZnFbΦ and F30ZnFbΦ
is a rapid and highly efficient process. This fact can be seen
from the following observations. (1) The kinetics of the decay
of the Zn porphyrin bleaching matches the growth of the Fb
porphyrin bleaching for both arrays. (2) The Zn* lifetime for
each array is over 150-fold shorter than the lifetime of ∼2 ns
for the relevant Zn porphyrin monomer reference complexes
(ZnTPP, ZnTMP, and ZnF20TPP). (3) The emission yield for
the Zn component of each ZnFb array is similarly reduced from
the values for the corresponding Zn porphyrin reference
compounds.
The rate and yield of energy transfer, Zn*Fb f ZnFb*, for
both ZnFbΦ and F30ZnFbΦ can be obtained from either the
transient absorption or static emission data. However, the
former measurements give a better assessment, given the
predominance of the energy-transfer channel for the decay of
Zn* in these complexes. The energy-transfer parameters are
obtained from the following expressions:
1/τD ) krad + kisc + kic
(1)
(2)
1/τDA ) krad + kisc + kic + ktrans

9430 J. Phys. Chem. B, Vol. 102, No. 47, 1998
ktrans ) 1/τDA - 1/τD
Yang et al.
TABLE 2: Energy-Transfer and Hole/Electron Hopping
(3)
(4)
Parameters
ktrans-1 (ps)a
Φtrans
khop-1 (µs)b
a
Φtrans ) ktransτDA ) 1 - τDA/τD
ZnFbΦ
Zn2Φ
F30ZnFbΦ
F30Zn2Φ
ZnFbU
Zn2U
F30ZnFbU
F30Zn2U
3.5
>0.99
c
where τDA is the excited-state lifetime of the donor in the
presence of the acceptor (i.e., the Zn* in the ZnFb dimer) and
τD is the excited-state lifetime of the zinc porphyrin monomer.
These equations assume that, besides energy transfer, there
are no pathways for depopulating Zn* in the arrays other than
the intrinsic processes (radiative decay (rad), intersystem
crossing (isc), and internal conversion (ic)) also present in the
benchmark Zn porphyrin monomer. Again, the transient
absorption and static emission data support this assumption.
Within experimental uncertainty, we cannot exclude the pos-
sibility of a small amount of electron transfer from the Zn*
state. However, the similarity of the Zn* lifetime of ZnFbΦ
in toluene and in acetone/CH2Cl2 (4:1) indicates any such
electron transfer must be e10%. The calculated energy-transfer
rates and yields for ZnFbΦ and F30ZnFbΦ are shown in Table
2. For comparison, this table also contains the results of similar
measurements we carried out previously on the diarylethyne-
linked analogues ZnFbU and F30ZnFbU.
d. Physical Properties of the Oxidized Complexes. The
oxidized arrays were investigated to gain insight into how the
p-phenylene linker modulates the rates of ground-state hole/
electron hopping.56 These rates are of interest because no
independent assessment of the rates of ground-state charge
transfer is available for the neutral arrays. Although the hole/
electron hopping rates in the ground electronic states of the
cations are not expected to be equal to the charge-transfer rates
in the excited states of the neutrals, they at least provide some
measure of the factors that control this type of process.
1. Electrochemistry. The E1/2 values for oxidation of the
phenylene-linked dimeric arrays are summarized in Table 3.
The two E1/2 values listed in the table correspond to the first
and second oxidations of the porphyrin ring.57 In general, the
values of the redox potentials and the redox characteristics of
the both the fluorine-containing and non-fluorine-containing
phenylene-linked dimers closely resemble those of the analogous
diarylethyne-linked dimers we have previously investigated.5,9
The redox behavior of these latter arrays is essentially identical
to that of monomeric Zn and Fb porphyrins, indicating that the
ground-state interaction between the porphyrin constituents of
the dimers are relatively weak. [The ground-state interaction
between the π systems of the diarylethyne linker and the
porphyrin rings is also quite weak.56] Accordingly, the replace-
ment of the diarylethyne linker with the shorter p-phenylene
linker does not result in any clearly detectable increase in the
magnitude of the ground-state interaction between the porphyrins
in the dimer.
e0.05
c
10
24d
0.99
g2.5
c
0.99d
0.87d
e0.05e
c
240d
g2.5d
a Obtained using the transient absorption data and eqs 1-4. b Esti-
mated from the 295 K EPR data (see text). c No hole/electron hopping
occurs in the ZnFb dimers because of the large disparity in oxidation
potential of the Zn and Fb constituents (Table 3). d From ref 9. e From
ref 56b.
TABLE 3: Half-Wave Potentialsa for Oxidation of the
Porphyrins of the Various Arrays
Zn porphyrin
Fb porphyrin
E
1/2(1)
E1/2(2)
E1/2(1)
E1/2(2)
·nFbΦ
·n2Φc
F30ZnFbΦ
0.58
0.59
0.99
0.99
0.88b
0.90
1.23b
1.20
0.72b
1.19
1.18b
1.41
c
F30Zn2Φ
a Obtained in CH2Cl2 containing 0.1 M TBAH. E1/2 vs Ag/Ag+; E1/2
of FeCp2/FeCp2+ ) 0.22 V; scan rate ) 0.1 V/s. Values are (0.01 V.
b Values are approximated due to overlap of the Zn and Fb porphyrin
waves. c The redox waves of the two Zn porphyrins are not resolved
by cyclic voltammetry.
2. Absorption Spectra. The UV-vis absorption character-
istics of the mono- and dications of the arrays (not shown) are
typical of those of other porphyrin π-cation radicals, namely
weaker, blue-shifted B bands (relative to the neutral complexes)
and very weak, broad bands in the visible and near-infrared
regions.57a The absorption spectra of the oxidized complexes
appear to be a superposition of the spectra of the neutral and
cationic species of the different porphyrin units. This observa-
tion, like the results of the electrochemical studies,58-61 is
consistent with weak interactions between the constituent
porphyrins.
Figure 4. EPR spectra of [ZnFbΦ]+ and [Zn2Φ]+ obtained at 295
and 100 K. Note that the magnetic field scales are different in the left
and right panels.
the same conditions are shown in Figure 5. The EPR spectra
of all of the cations were examined as a function of sample
concentration. No changes in hyperfine splittings or line shape
were observed for the concentrations used for the EPR studies
(e0.05 mM). Consequently, the differences observed in the
3. EPR Spectra. The EPR spectra of the one-electron
oxidation products of ·nFbΦ and ·n2Φ in CH2Cl2 at 295 and
100 K are shown in Figure 4. The spectra of the one-electron
oxidation products of F30ZnFbΦ and F30Zn2Φ obtained under

p-Phenylene-Linked Porphyrin Dimers
J. Phys. Chem. B, Vol. 102, No. 47, 1998 9431
analogous spectra of [F30ZnFbU]+ and [F30Zn2U]+.9 In par-
ticular, the liquid-solution EPR signal of [F30ZnFbΦ]+ is
relatively sharp and exhibits no resolved hyperfine splittings.
The 14N hyperfine splittings are absent for [F30ZnFbΦ]+ (and
[F30ZnFbU]+) because the strongly electron-withdrawing pen-
tafluorophenyl groups at the nonlinking meso positions energeti-
cally stabilize the a2u molecular orbital,64,65 resulting in a ground
2
state that is solely, or mostly Α1u-like in character.9 The a1u
molecular orbital that contains the unpaired electron in these
latter type of radicals has appreciable electron density on the
â-pyrrole carbon atoms but nodal planes through the pyrrole
nitrogen atoms (and meso carbon atoms).64 Unlike the case
for [Zn2Φ]+ versus [ZnFbΦ]+ (Figure 4, left panel), the liquid-
solution signal of [F30Zn2Φ]+ is essentially superimposable on
the signal observed for [F30ZnFbΦ]+ (Figure 5, left panel). The
frozen-solution EPR signals for [F30Zn2Φ]+ and [F30ZnFbΦ]+
are also essentially superimposable, although broader than the
liquid solution signals (Figure 5, right panel). These spectral
characteristics indicate that the hole/electron in [F30Zn2Φ]+ is
localized on the EPR time scale in both liquid and frozen
solution. Similar conclusions were reached for [F30Zn2U]+.9
2
Given that the largest hyperfine couplings in the A1u-like
porphyrin π-cation radicals are ∼2 MHz,53,62 the value of khop
must be e(2.5 µs)-1. Accordingly, any acceleration in the hole/
electron hopping rate resulting from the replacement of the
diarylethyne linker with the shorter p-phenylene linker is not
sufficient to overcome the slower rate resulting from the effects
of the fluorine atoms on the orbital characteristics of the
porphyrins. For comparison, the khop values for [Zn2Φ]+ and
[F30Zn2Φ]+ are included in Table 2.
Figure 5. EPR spectra of [F30ZnFbΦ]+ and [F30Zn2Φ]+ obtained at
295 and 100 K.
spectral features of the various complexes are intrinsic and
cannot be ascribed to intermolecular interactions.
Examination of the EPR signatures of the cations of the
various phenylene-linked dimeric arrays reveals the following
features.
3. Discussion
In the present study, we have extended our work on the
synthesis and characterization of the building blocks for
porphyrin-based molecular photonic devices by exploring the
rates of excited-state energy transfer and ground-state hole/
electron hopping in p-phenylene-linked porphyrin dimers. These
studies complement our previous work on multiporphyrin
architectures constructed with diarylethyne linkers.5-9,56 Two
specific goals of the present study were as follows. (1) We
wished to determine whether the incorporation of a p-phenylene
linker would result in a significant increase in electronic
communication across a porphyrin dimer without significant
perturbation of the electronic properties of the constituent
porphyrins or deleterious quenching reactions. The desire to
minimize perturbation of the inherent photophysical and redox
properties of the lowest excited states of the porphyrin con-
stituents would thus permit incorporation of these constructs
into extended arrays without compromising the properties of
the individual elements. (2) We wished to further probe the
effects of the characteristics of the porphyrin frontier molecular
orbitals (energy ordering, electron density distributions) on
electronic communication in multiporphyrin arrays. In particu-
lar, we wanted to determine whether fluorination of the
nonlinking phenyl groups in a p-phenylene-linked porphyrin
dimer (which causes specific effects on porphyrin orbital
characteristic) would give rise to a decrease in the energy-
transfer rate, as we have previously observed for the analogous
chemical modification of diarylethyne-linked dimers. The
results obtained indicate that the study has been successful in
all respects, as is discussed in more detail below.
(1) The liquid- and frozen-solution EPR spectra of [ZnFbΦ]+
and [Zn2Φ]+ are very similar to those of the analogous spectra
of [ZnFbU]+ and [Zn2U]+.56 In particular, the liquid solution
signal of [ZnFbΦ]+ exhibits a poorly resolved nine-line hyper-
fine pattern due to interaction of the unpaired electron with the
four pyrrole 14N nuclei of the Zn constituent of the dimer (Figure
4, left panel). This hyperfine pattern is characteristic of a 2A2u
porphyrin π-cation radical.62 The unpaired electron is localized
on the Zn porphyrin due to the redox characteristics of the Zn
versus Fb porphyrins.57 In contrast, the liquid-solution signal
of [Zn2Φ]+ exhibits a much narrower line with no resolved
hyperfine structure. Simulations of the EPR spectrum of
[Zn2Φ]+ indicate that the line shape is well accounted for by
halving the hyperfine coupling and doubling the number of
interacting nuclei (while holding the line width constant). This
behavior indicates that the hole/electron of [Zn2Φ]+ is com-
pletely delocalized on the EPR time scale,63 as is also the case
for [Zn2U]+.56 This time scale is determined by the magnitude
of the 14N hyperfine coupling, which is ∼4.2 MHz.62 Thus,
the hole/electron hopping rate, khop, is greater than or equal to
(0.05 µs)-1. When the solution is frozen, the EPR signal of
[Zn2Φ]+ broadens and is essentially identical to the signal
observed for [ZnFbΦ]+ in frozen solution (Figure 4, right panel).
This behavior indicates that the hole/electron becomes localized
(on the EPR time scale) on one of the Zn constituents of
[Zn2Φ]+ (khop e (1 µs)-1). This behavior again parallels that
observed for [·n2U]+, in which the hole/electron hops rapidly
in liquid solution but becomes localized in frozen solution.5
(2) The liquid- and frozen-solution EPR spectra of [F30-
ZnFbΦ]+ and [F30Zn2Φ]+ are very similar to those of the
a. Effects of Linker Architecture on Electronic Com-
munication. The rate of excited-state energy transfer in the
p-phenylene-linked array ZnFbΦ ((3.5 ps)-1) is about 7-fold

9432 J. Phys. Chem. B, Vol. 102, No. 47, 1998
Yang et al.
faster than that for the diphenylethyne-linked analogue ZnFbU
((24 ps)-1). An even more dramatic 24-fold increase is found
for F30ZnFbΦ ((10 ps)-1) versus F30ZnFbU ((240 ps)-1).
Likewise, the incorporation of the p-phenylene linker facilitates
rapid ground-state hole/electron hopping (in liquid solution) in
the monocation radical of the bis-Zn complex ([Zn2Φ]+). This
latter rate may well be faster than that in the diarylethyne-linked
complex [Zn2U]+; however, the exact rate for either dimer
cannot be determined from the existing data because the hole/
electron hopping process is in the fast-exchange limit for both.
These effects, most notably the enhanced excited-state energy-
transfer rate, are not accompanied by significant perturbation
of the inherent properties of the photophysically relevant
electronic excited states (e.g., Q band absorption and emission
characteristics) or the ground-state redox potentials of the
porphyrin components. These findings parallel those previously
identified for dimers and larger arrays incorporating the longer
diarylethyne linker. Collectively, these observations indicate
that the enhanced excited-state energy-transfer rates found for
the p-phenylene-linked dimers do not arise from factors such
as enhanced spectral overlap/energy matching. Rather, the
enhanced rates derive principally (if not exclusively) from
enhanced electronic coupling between the porphyrin constituents
across the p-phenylene versus diarylethyne linker. This en-
hanced electronic coupling is attributed principally to a shorter
inter-porphyrin separation in the case of the former linker.
However, it should be noted that the different chemical
composition/architecture (and associated electronic character-
istics) of the p-phenylene versus diphenylethyne linker may also
modulate the magnitude of electronic coupling.66
the electron-withdrawing fluorine groups, which shift electron
density in both the HOMOs and LUMOs away from the linking
position in each porphyrin, must augment the HOMO ordering
effect on the reduction in electronic coupling. The reader is
referred to our previous work for a more detailed discussion of
these issues.6,9,53
The observation that the fluorination-induced reduction in the
excited-state energy-transfer rate in the p-phenylene-linked
dimers (ZnFbΦ, (3.5 ps)-1; F30ZnFbΦ, (10 ps)-1) is not as
substantial as that for the arylethyne-linked analogues (ZnFbU,
(24 ps)-1; F30ZnFbU, (240 ps)-1) warrants further comment.
From an empirical point of view, this result is advantageous
for the use of the p-phenylene-linked dimer motif in extended
architectures. In particular, the less substantial effect of
fluorination leads to an excited-state energy-transfer rate ((10
ps)-1) that remains 2-fold faster (and slightly more efficient,
see Table 2) than in the nonfluorinated diphenylethyne dimer
ZnFbU. Hence, F30ZnFbΦ has the advantage of an ultrafast
photoinduced energy-transfer rate coupled with the porphyrin
redox properties that disfavor competing charge-transfer reac-
tions that are deleterious when energy transfer is the desired
function.6,9,53 In general, the ability to use halogenation (or other
types of functionalization) to tune both the characteristic
properties of the chromophores and the energy-transfer and hole/
electron hopping rates is highly desirable. The combination of
favorable properties associated with F30ZnFbΦ is an unexpected
result of the present study and should find utility in future
applications.
c. Mechanism(s) of Electronic Communication. The final
issue concerns the mechanism(s) of electronic communication,
particularly the mechanism of energy transfer, in the p-
phenylene-linked versus diarylethyne-linked dimers. The ob-
servation that fluorination has a lesser impact on the energy-
transfer rate in the former class of dimers (a factor of 3 versus
10) has direct bearing on this issue and suggests that the energy-
transfer mechanism is different for the p-phenylene-linked versus
diarylethyne-linked dimers. This view is explained in more
detail below.
b. Effects of Fluorination on Electronic Communication.
A 3-fold reduced rate of excited-state energy transfer is found
for F30ZnFbΦ ((10 ps)-1) versus ZnFbΦ ((3.5 ps)-1). This
factor is not as substantial as the 10-fold reduction found upon
fluorination of the diphenylethyne-linked analogue ((240 ps)-1
for F30ZnFbU versus (24 ps)-1 for ZnFbU). Nevertheless, the
general effect of fluorination on the energy-transfer rate in the
p-phenylene-linked dimers is consistent with the fluorine-
induced changes in the porphyrin orbital characteristics that we
have previously discussed in detail for the diarylethyne-linked
dimers.9 The same orbital effects underlie the observation that
hole/electron hopping in [F30Zn2Φ]+ slows considerably relative
to [Zn2Φ]+, similar to the situation for [F30Zn2U]+ versus
[Zn2U]+. The attenuation of both the excited-state energy-
transfer and ground-state hole/electron hopping rates occurs in
large part because fluorination of all the nonlinking phenyl
groups preferentially stabilizes the a2u orbital to the extent that
it drops below the a1u orbital in the case of the Zn porphyrins
and to an energy comparable to the a1u orbital in the case of
the Fb porphyrins.9
In both types of dimers, the observed energy-transfer rate is
assumed to be due to the additive effects of TB and TS
processes.
ktrans ) kTB + kTS
(5)
All of our previous work on the diarylethyne-linked porphyrins
indicates that excited-state energy transfer occurs predominantly
via an exchange-mediated TB mechanism with a much lesser
contribution from the TS mechanism.5-9 In particular, the TS
energy-transfer rate for ZnFbU is calculated to be on the order
of (1 ns)-1, which is approximately 2 orders of magnitude slower
than the observed rate of (24 ps)-1 9
[These rates were
.
The key point to be derived from the effects of fluorination
is that the a2u orbital has considerable electron density at the
porphyrin meso carbon to which the linker is attached whereas
the a1u orbital has nodes at this position. Hence, fluorination
increases contribution of the a1u orbital to the excited-state wave
function of the neutral dimers and to the ground-state wave
function of the monocation radicals, which results in reduced
electronic coupling between the two porphyrin constituents. This
in turn gives rise to a slowing of both photoinduced energy
transfer and ground-state hole/electron hopping. It should be
emphasized, however, that the reduced electronic coupling in
the fluorinated versus nonfluorinated dimers most likely does
not derive exclusively from the change in the relative energies
of the a1u and a2u orbitals. In particular, the general effect of
calculated using the Fo¨rster theory of resonance energy transfer
assuming a dipole-dipole approximation, which is reasonable
because of the large center-to-center distance (∼20 Å) between
the Zn and Fb constituents).] The TS energy-transfer rate
calculated for ZnFbU thus supports the empirical evidence5-9,56,57
that the TB mechanism dominates the excited-state energy
process for the diarylethyne-linked arrays. On the other hand,
Fo¨rster calculations (dipole approximation) for the TS energy-
transfer rate for the p-phenylene-linked dimers give a value on
the order of (50-100 ps)-1 (see Supporting Information). These
values are relatively close to the observed rates (ZnFbΦ, (3.5
ps)-1; F30ZnFbΦ, (10 ps)-1). Furthermore, previous studies
have shown that the 13 Å center-to-center distance between the
Zn and Fb porphyrins in these p-phenylene-linked dimers is

p-Phenylene-Linked Porphyrin Dimers
J. Phys. Chem. B, Vol. 102, No. 47, 1998 9433
sufficiently short that multipole corrections to the dipole-dipole
approximation could increase the energy-transfer rate by a factor
of 5 or more.67 These considerations indicate that the TS
contribution to excited-state energy transfer in the p-phenylene-
linked dimers is much more substantial than that in the
diarylethyne-linked dimers. In fact, the TS mechanism may
make a dominant contribution to the energy-transfer process in
the former arrays. This view is further consistent with the less
substantial effect of fluorination on the excited-state energy-
transfer rate in the p-phenylene-linked versus diarylethyne-linked
dimers. This effect may derive from two sources as the TS
mechanism becomes increasingly important. (1) The effect of
fluorination on the orbital characteristics and thus electronic
coupling for the TB transfer becomes less significant. (2) The
spectral overlap integral in the Fo¨rster TS mechanism is
approximately 2-fold less for the fluorinated versus nonfluori-
nated dimers (Supporting Information).9 The difference in
spectral overlap is inconsequential when the TB transfer process
is dominant but becomes increasingly significant as the TS
process becomes a larger contributor to the overall energy-
transfer rate.68
Finally, it should be noted that the characteristics of the
ground-state hole/electron hopping in the fluorinated versus
nonfluorinated dimers provides indirect evidence that the TS
mechanism plays an important role in the excited-state energy-
transfer process in the p-phenylene-linked dimers. In particular,
the hole/electron hopping rate in the p-phenylene-linked dimers
is diminished at least 50-fold by fluorination ([Zn2Φ]+, g(0.05
µs)-1; [F30Zn2Φ]+, g(2.5 ps)-1). The magnitude of this
attenuation is qualitatively in the same range as that which
occurs upon fluorination of the diarylethyne-linked dimers
([Zn2U]+, g(0.05 µs)-1, [F30Zn2U]+, g(2.5 ps)-1). This
behavior is quite different from the effect of fluorination on
the energy-transfer rate on the two classes of dimers (vide supra).
The ground-state hole/electron hopping process is most certainly
governed by TB electronic coupling over the large separation
distances in the diarylethyne-linked dimers. Although the
p-phenylene linker is shorter, the porphyrin-porphyrin separa-
tion is still sufficiently large that the TB process should
dominate. Thus, the fluorination-induced changes in HOMO
characteristics have a significant impact on the hole/electron
hopping rates in the p-phenylene-linked dimers. Accordingly,
the significantly smaller effect of fluorination on the excited-
state energy-transfer rate points to the increasing importance
of a new channel (i.e., the TS mechanism) for this process.
constituents are maintained in both ZnFbΦ and F30ZnFbΦ, both
of which exhibit ultrafast photoinduced energy-transfer rates
(∼(10 ps)-1 or faster). The fluorinated complex should be
particularly useful for applications where fast energy transfer
is required, charge transfer must be minimized, and chemical
stability must be enhanced. Overall, the results of this study
provide new architectures for the design of porphyrin-based
arrays for a variety of molecular photonics applications.
5. Experimental Section
a. Synthesis. 1. General. 1H NMR spectra were recorded
on either a Varian Gemini 300 MHz or General Electric GN-
300 MHz (IBM FT-300) spectrometers. Chemical shifts for
1
the H NMR spectra are reported in ppm using the deuterated
solvent resonances as the internal reference (CDCl3, δ ) 7.26
ppm; toluene-d8, δ ) 2.09 ppm; THF-d8, δ ) 3.58 ppm).
Porphyrins were analyzed by fast atom bombardment (FAB-
MS) on a JEOL, HX 110 HF or by laser desorption mass
spectrometry (LD-MS) using a Bruker Proflex II equipped with
a linear (1.2 m) flight tube. Absorption and emission spectra
were routinely obtained (as described below) to monitor purity.
The spectral characteristics of the purified products are sum-
marized in Table 1. Unless otherwise indicated, all reagents
were obtained from Aldrich Chemical Co., Milwaukee, WI, and
all solvents were obtained from Fisher Scientific.
2. SolVents. All solvents were dried by standard methods
prior to use. Toluene (Fisher, certified ACS or HPLC),
triethylamine (Fluka, puriss), and pyridine (Acros) were distilled
from CaH2. CHCl3 (Fisher, certified ACS) and CH2Cl2 (Fisher,
reagent grade) were subjected to simple distillation from K2CO3.
Pyrrole (Acros) was distilled under reduced pressure from CaH2.
Other solvents were used as received.
3. Chromatography. Adsorption column chromatography
was performed using flash silica gel (Baker, 60-200 mesh) or
alumina (Fisher A540-3, 80-200 mesh). Preparative scale
size-exclusion chromatography (SEC) was performed using
BioRad Bio-Beads SX-1. A preparative scale glass column (4.8
× 60 cm) was packed using BioRad Bio-Beads SX-1 in toluene
and eluted with gravity flow (∼1.2 mL/min). Following
purification, the SEC column was washed with 2 volume equiv
of toluene. Analytical scale SEC was performed to monitor
the dimer-forming reactions and assess the purity of the dimers.
Analytical SEC columns (styrene-divinylbenzene copolymer)
were purchased from Hewlett-Packard and Phenomonex. Ana-
lytical SEC was performed with a Hewlett-Packard 1090 HPLC
using 500 Å (300 × 7.8 mm), 500 Å (300 × 7.5 mm), and 100
Å (300 × 7.5 mm) columns (5 µm) in series eluting with THF
(flow rate ) 0.8 mL/min; void volume ≈15.0 min).
4. Conclusions
In previous studies, we have found that the rates and yields
of excited-state energy transfer and ground-state hole/electron
hopping in diarylethyne-linked porphyrin dimers can be modu-
lated in predictable ways through manipulation of the porphyrin
constituents (types of metal ion and substituents on the
nonlinking aryl rings) and the aryl rings of the linker. The
manipulation of these structural features affords control of the
electronic and steric factors that mediate the dynamic processes
(as well as other important features, such as solubility, chemical
stability, and redox properties) and permits tuning for specific
applications. Here, we have found that the desirable goal of
even faster rates of photoinduced energy transfer can be achieved
through the use of a simple p-phenylene linker. Furthermore,
the enhanced dynamic properties are obtained without perturbing
the desirable and well characterized electronic properties of the
porphyrin subunits. In particular, the inherent excited-state
photophysical and ground-state redox properties of the porphyrin
4. Compounds. 1,4-Bis(10,15,20-trimesitylporphin-5-yl)-
benzene (Fb2Φ). A 5 L, three-necked, round-bottom flask was
charged with 3 L of distilled CHCl3, pyrrole (2.08 mL, 30.0
mmol), mesitaldehyde (3.32 mL, 22.5 mmol), and terephthal-
dicarboxaldehyde (503 mg, 3.75 mmol). The resulting solution
was magnetically stirred at room temperature, and BF3‚OEt2
(1.25 mL, 9.90 mmol) was added dropwise via syringe. The
reaction vessel was shielded from ambient lighting. The reaction
was monitored by treating 25 µL aliquots with excess 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in toluene (150
µL) followed by absorption spectroscopy (418 nm, assuming ꢀ
) 5 × 105 M-1 cm-1 for all porphyrin species).49 The yield of
porphyrin was found to level off after 1 h, at which point DDQ
(5.11 g, 22.5 mmol, 3 equiv per porphyrinogen, 3/4 equiv per
pyrrole) was added. After the mixture was stirred at room

9434 J. Phys. Chem. B, Vol. 102, No. 47, 1998
Yang et al.
temperature for 1 h, triethylamine (1.38 mL, 9.90 mmol) was
added to neutralize the acid and the solvent was removed under
reduced pressure. The resultant black solid from the reaction
mixture was dissolved in 100 mL of CH2Cl2, diluted with an
equal volume of hexanes, and loaded onto a silica gel column
(10 cm diameter × 28 cm) with CH2Cl2/hexanes (1:1) as the
eluent. The porphyrins as well as other pigments (uncharac-
terized) were eluted as a single fast-moving dark-red band. The
fractions were collected, combined, and concentrated to afford
a crude purple solid containing three major components:
monomeric porphyrin, desired dimeric porphyrin, and putative
trimeric porphyrin. The crude porphyrin mixture was further
purified by column chromatography (10 cm diameter × 20 cm)
over flash silica gel with CH2Cl2/hexanes (1:1) to remove small
amounts of slow moving pigments (uncharacterized). The three
porphyrin products were slightly visible on the column as the
separation proceeded. The porphyrin-containing fractions were
combined and concentrated to dryness. LD-MS and analytical
SEC data of this solid were consistent with the presence of
monomeric, dimeric, and trimeric porphyrins. The mixture of
porphyrins was then dissolved in 60 mL of toluene and placed
on top of a preparative SEC column. Gravity elution afforded
four major bands with the desired dimeric porphyrins as the
third band. The dimer-containing fractions were collected and
concentrated to dryness, affording 75.1 mg (1.42%) of a purple
solid. 1H NMR (CDCl3) δ -2.42 (s, 4H), 1.89 (s, 12H), 1.92
(s, 24H), 2.64 (s, 6H), 2.66 (s, 12H), 7.29 (s, 4H), 7.33 (s, 8H),
8.59 (s, 4H), 8.67 (s, 8H), 8.86 (d, 4H, J ) 5.1 Hz), 9.22 (d,
4H, J ) 5.1 Hz). C100H90N8: calcd mass 1402.7288, obsd m/z
1402.7196 (FAB-MS); calcd av mass 1402.9, obsd m/z 1402.6
(LDI-MS). λabs (toluene) 419, 428, 482, 516, 550, 592, 650
nm; λem (toluene, λex ) 514 m) 650, 717 nm. Anal. Calcd for
Fb2Φ‚2H2O (C100H94N8O2): C, 83.42; H, 6.58; N, 7.78.
Found: C, 83.61; H, 6.36; N, 7.73.
fractions containing the desired mono-zinc dimer ZnFbΦ were
combined and concentrated to give 7.92 mg (37.8%) of a red
solid. 1H NMR (toluene-d8) δ -1.70 (s, 2H), 1.89 (s, 6H), 2.01
(s, 6H), 2.02 (s, 12H), 2.10 (s, 12H), 2.48 (s, 3H), 2.52 (s, 3H),
2.53 (s, 6H), 2.56 (s, 6H), 7.16 (s, 2H), 7.22 (s, 2H), 7.24 (s,
4H), 7.28 (s, 4H), 8.38 (d, 2H, J ) 8.1 Hz), 8.48 (d, 2H, J )
8.1 Hz), 8.80 (dd, 4H, J ) 4.8 and 11.3 Hz), 8.93 (dd, 4H, J )
5.5 and 7.0 Hz), 8.94 (d, 2H, J ) 3.9 Hz), 9.05 (d, 2H, J ) 4.4
Hz), 9.31 (d, 2H, J ) 5.1 Hz), 9.39 (d, 2H, J ) 4.4 Hz).
C100H88N8Zn: calcd mass 1464.6423, obsd m/z 1464.6416
(FAB-MS); calcd av mass 1467.2, obsd m/z 1470.1, 1468.1 (LD-
MS). λabs (toluene) 420, 430, 480, 514, 550, 592, 650 nm; λem
(toluene, λex ) 551 m) 651, 720 nm.
1,4-Bis(zinc 10,15,20-Trimesitylporphin-5-yl)benzene (Zn2-
Φ). A sample of Fb2Φ (30.0 mg, 21.4 µmol) was dissolved in
12 mL of CHCl3, and then a methanolic solution of Zn(OAc)2‚
2H2O (14.1 mg, 64.2 µmol, 1.3 mL of methanol) was added.
The reaction mixture was stirred at room temperature for 20 h
then heated at reflux for 2 h. The progress of the reaction was
monitored by fluorescence excitation spectroscopy. When
complete, the reaction mixture was allowed to cool to room
temperature and washed with 10% NaHCO3, and the solvent
was removed under reduced pressure. Column chromatography
on silica gel (toluene, 1.5 cm diameter × 12 cm) gave 32.8 mg
(100%) of a red solid. 1H NMR (toluene-d8) δ 2.01 (s, 12H),
2.10 (s, 24H), 2.52 (s, 6H), 2.56 (s, 12H), 7.22 (s, 4H), 7.28 (s,
8H), 8.56 (s, 4H), 8.93 (dd, 8H, J ) 4.4 and 6.6 Hz), 9.06 (d,
4H, J ) 5.1 Hz), 9.44 (d, 4H, J ) 4.4 Hz). C100H86N8Zn2:
calcd mass 1526.5558, obsd 1526.5508 (FAB-MS); calcd av
mass 1530.6, obsd m/z 1530.3, 1531.5 (LD-MS). λabs (toluene)
420, 431, 513, 550, 590 nm; λem (toluene, λex ) 550 m) 597,
645 nm.
1-(Zinc 10,15,20-Tris(pentafluorophenyl)porphin-5-yl)-4-(10,-
15,20-tris(pentafluorophenyl)porphin-5-yl)benzene (F30ZnFbΦ).
A sample of F30Fb2Φ (33.5 mg, 19.8 µmol) in CHCl3 (35 mL)
was treated with a methanolic solution of Zn(OAc)2‚2H2O (13.0
mg, 59.4 µmol, 3.9 mL of methanol). The reaction mixture
was heated at reflux for 18 h, and workup proceeded as
previously described. Purification by column chromatography
on alumina (CHCl3/hexanes (1:1), 2 cm diameter × 20 cm)
afforded 15.3 mg (44.0%) of a red solid. 1H NMR (THF-d8) δ
-2.63 (s, 2H), 8.78 (s, 4H), 9.11 (dd, 2H, J ) 4.4 and 4.4 Hz),
9.18 (d, 2H, J ) 4.4 Hz), 9.19 (dd, 2H, J ) 4.4 and 4.4 Hz),
9.24 (d, 2H, J ) 4.4 Hz), 9.50 (d, 2H, J ) 4.4 Hz), 9.54 (d,
2H, J ) 4.4 Hz). C82F30H22N8Zn: calcd mass 1752.0780, obsd
1752.0031 (FAB-MS); calcd avg mass 1754.5, obsd m/z 1754.3
(LD-MS). λabs (toluene) 413, 426, 508, 552, 584, 643 nm; λem
(toluene, λex ) 508 m) 646, 711 nm.
1,4-Bis(zinc 10,15,20-Tris(pentafluorophenyl)porphin-5-yl)-
benzene (F30Zn2Φ). A sample of F30Fb2Φ (23.2 mg, 13.7 µmol)
in CHCl3 (54 mL) was treated with a methanolic solution of
Zn(OAc)2‚2H2O (15.1 mg, 68.6 µmol, 6.0 mL of methanol).
The reaction mixture was heated at reflux for 18 h, and workup
proceeded as previously described. Purification by column
chromatography on silica (CH2Cl2/hexanes (1:1), 2 cm diameter
× 20 cm) afforded 24.8 mg (100%) of a red solid. 1H NMR
(THF-d8) δ 8.75 (s, 4H), 9.11 (d, 8H, J ) 4.4 Hz), 9.18 (d, 4H,
J ) 4.4 Hz), 9.51 (d, 4H, J ) 4.4 Hz). C82F30H20N8Zn2: calcd
mass 1813.9915, obsd 1813.9969 (FAB-MS); calcd avg mass
1817.8, obsd m/z 1817.4 (LDI-MS). λabs (toluene) 421, 430,
508, 549, 581 nm; λem (toluene, λex ) 550 m) 587, 643 nm.
1,4-Bis[10,15,20-tris(pentafluorophenyl)porphin-5-yl]ben-
zene (F30Fb2Φ). In a 5 L, three-necked, round-bottom flask
containing 3 L of distilled CH2Cl2, pyrrole (2.08 mL, 30.0
mmol), pentafluorobenzaldehyde (2.77 mL, 22.5 mmol), and
terephthaldicarboxaldehyde (503 mg, 3.75 mmol) were con-
densed following the procedure outlined above. Workup
included chromatography by flash silica gel (CH2Cl2/hexanes
(1:1), 10 cm diameter × 28 cm), flash silica gel (CH2Cl2/hexanes
(1:1), 2 cm diameter × 20 cm), and preparative SEC (toluene),
affording a purple solid, 64.7 mg (1.0%). 1H NMR (THF-d8)
δ -2.64 (s, 4H), 8.80 (s, 4H), 9.18 (dd, 8H, J ) 4.4 and 4.4
Hz), 9.23 (d, 4H, J ) 4.4 Hz), 9.53 (d, 4H, J ) 4.4 Hz). 19F
NMR (CD2Cl2, CFCl3) δ -137.46 (m, 12F), -152.66 (t, 6F, J
) 22.3 Hz), -162.31 (m, 12F). C82F30H24N8: calcd mass
1690.1645, obsd m/z 1690.0620 (FAB-MS); calcd av mass
1691.1, obsd m/z 1691.3 (LD-MS). λabs (toluene) 413, 426, 512,
542, 587, 643 nm; λem (toluene, λex ) 511 m) 646, 713 nm.
1-(Zinc 10,15,20-Trimesitylporphin-5-yl)-4-(10,15,20-trimesi-
tylporphin-5-yl)benzene (ZnFbΦ). A sample of dimeric por-
phyrin Fb2Φ (20.0 mg, 14.2 µmol) was dissolved in 8 mL of
CHCl3, and then a methanolic solution of Zn(OAc)2‚2H2O (2.5
mg, 11.4 µmol, 0.9 mL of methanol) was added. The reaction
mixture was heated at reflux and was monitored by fluorescence
excitation spectroscopy. After 90 min, the reaction mixture was
cooled to room temperature, washed with 10% NaHCO3, dried
(Na2SO4), and filtered and the solvent was removed under
reduced pressure. Column chromatography on alumina (CHCl3/
hexanes (1:1), 7 cm diameter × 12 cm) gave Fb2Φ as the first
band and ZnFbΦ as the second porphyrin band. Final elution
with CHCl3 afforded the bis-zinc porphyrin (Zn2Φ). The
b. Physical Methods. The static and time-resolved absorp-
tion and fluorescence studies were performed on samples

p-Phenylene-Linked Porphyrin Dimers
J. Phys. Chem. B, Vol. 102, No. 47, 1998 9435
prepared in toluene at room temperature. Toluene (EMI,
Omnisolve) was distilled from sodium. The samples for time-
resolved fluorescence measurements were degassed by several
freeze-pump-thaw cycles on a high-vacuum line. The samples
for static fluorescence and time-resolved absorption measure-
ments were not degassed. The electrochemical and EPR studies
were performed on samples prepared in CH2Cl2. CH2Cl2
(Aldrich, HPLC Grade) was purified by vacuum distillation from
P2O5, followed by another distillation from CaH2. For the
electrochemical studies, tetrabutylammonium hexafluorophos-
phate (TBAH; Aldrich, recrystallized 3 times from methanol
and dried under vacuum at 110 °C) was used as the supporting
electrolyte. The solvents were degassed thoroughly by several
freeze-pump-thaw cycles prior to use.
a sabbatical leave. D.K. was supported by the Creative Research
Initiatives of the Ministry of Science and Technology of Korea.
Supporting Information Available: A table showing the
results of the Fo¨rster energy-transfer calculations for several
dimers is provided (1 page). Ordering information is given on
any current masthead page.
References and Notes
(1) Prathapan, S.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc.
1993, 115, 7519-7520.
(2) Wagner, R. W.; Lindsey, J. S. J. Am. Chem. Soc. 1994, 116, 9759-
9760.
(3) Wagner, R. W.; Lindsey, J. S.; Seth, J.; Palaniappan, V.; Bocian,
D. F. J. Am. Chem. Soc. 1996, 118, 3996-3997.
(4) Bothner-By, A. A.; Dadok, J.; Johnson, T. E.; Lindsey, J. S. J.
Phys. Chem. 1996, 100, 17551-17557.
1. Static Absorption and Fluorescence Spectroscopy. Static
absorption (HP8451A, Cary 3, Perkin-Elmer Lamba 3B) and
fluorescence (Spex Fluoromax or Fluorolog II) measurements
were performed as described previously.5-9
(5) Hsiao, J.-S.; Krueger, B. P.; Wagner, R. W.; Delaney, J. K.;
Mauzerall, D. C.; Fleming, G. R.; Lindsey, J. S.; Bocian, D. F.; Donohoe,
R. J. J. Am. Chem. Soc. 1996, 118, 11181-11193.
(6) Strachan, J.-P.; Genteman, S.; Seth, J.; Kalsbeck, W. A.; Lindsey,
J. S.; Holten, D.; Bocian, D. F. Inorg. Chem. 1998, 37, 1191-1201.
(7) Wagner, R. W.; Seth, J.; Yang, S. I.; Kim, D.; Bocian, D. F.; Holten,
D.; Lindsey, J. S. J. Org. Chem. 1998, 63, 5042-5049.
(8) Li, F.; Gentemann, S.; Kalsbeck, W. A.; Seth, J.; Lindsey, J. S.;
Holten, D.; Bocian, D. F. J. Mater. Chem. 1997, 7, 1245-1262.
(9) Strachan, J. P.; Gentemann, S.; Seth, J.; Kalsbeck, W. A.; Lindsey,
J. S.; Holten, D.; Bocian, D. F. J. Am. Chem. Soc. 1997, 119, 11191-
11201.
2. Time-ResolVed Absorption Spectroscopy. Transient ab-
sorption data were acquired as described elsewhere.8,9,55 Samples
(∼0.2 mM in toluene) in 2 mm path length cuvettes at room
temperature were excited at 10 Hz with a 120 fs, 550 nm or
580 nm, 2-4 µJ flash from an optical parametric amplifier
(OPA) pumped by an amplified Ti:sapphire laser system
(Spectra Physics). Absorption changes were monitored using
a white-light probe flash generated in water with the residual
800 nm light (∼0.4 mJ per pulse) from the Ti:sapphire/OPA
system. On each laser flash, spectral data across the 450-600
nm region were obtained, and data from 300 flashes were
averaged to achieve a ∆A resolution of (0.005. Spectra at
different pump-probe delay times were acquired by changing
the path length over which the probe light traveled with respect
to the pump pulse. The kinetic data shown in Figures 2 and 3
were generated by averaging the ∆A values in the range 520-
525 nm (for the Fb porphyrin bleaching) or in the range 550-
555 nm (for the Zn porphyrin bleaching) at each delay time
(-10 to 200 ps) and then plotting these values as a function of
time. The kinetic traces were then fit to a function consisting
of a single exponential plus a constant using a nonlinear least-
squares algorithm.
(10) Wagner, R. W.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc.
1996, 118, 11166-11180.
(11) (a) Osuka, A.; Shimidzu, H. Angew. Chem., Int. Ed. Engl. 1997,
36, 135-137. (b) Khoury, R. G.; Jaquinod, L.; Smith, K. M. Chem.
Commun. 1997, 1057-1058. (c) Ogawa, T.; Nishimoto, Y.; Yoshida, N.;
Ono, N.; Osuka, A. Chem. Commun. 1998, 337-338.
(12) (a) Ponomarev, G. V.; Borovkov, V.; Sugiura, K.-I.; Sakata, Y.;
Shul’ga, A. Tetrahedron Lett. 1993, 34, 2153-2156. (b) Senge, M.;
Gerzevske, K.; Vicente, M.; Forsyth, T.; Smith, K. Angew. Chem., Int. Ed.
Engl. 1993, 32, 750-753. (c) Ponomarev, G.; Borovkov, V.; Shul’ga, A.;
Sakata, Y. J. Chem. Soc., Chem. Commun. 1994, 1927-1928. (d) Senge,
M.; Vicente, M.; Gerzevske, K.; Forsyth, T.; Smith, K. Inorg. Chem. 1994,
33, 5625-5638. (e) Higuchi, H.; Takeuchi, M.; Ojima, J. Chem. Lett. 1996,
593-594.
(13) (a) Lin, V. S.-Y.; DiMagno, S. G.; Therien, M. J. Science 1994,
264, 1105-1111. (b) Lin, V. S.-Y.; Therien, M. J. Chem.sEur. J. 1995, 1,
645-651.
(14) Arnold, D.; Nitschinsk, L. Tetrahedron 1992, 48, 8781-8792.
(15) Anderson, H. L. Inorg. Chem. 1994, 33, 972-981.
(16) Arnold, D.; Nitschinsk, L. Tetrahedron Lett. 1993, 34, 693-696.
(17) Vicente, M. G. H.; Smith, K. J. Org. Chem. 1991, 56, 4407-4418.
(18) Osuka, A.; Liu, B.-L.; Maruyama, K. Chem. Lett. 1993, 949-952.
(19) Heiler, D.; McLendon, G.; Rogalskyj, P. J. Am. Chem. Soc. 1987,
109, 604-606.
3. Electrochemistry. The oxidized complexes were prepared
and manipulated in a glovebox as described previously.9 The
integrity of the samples was checked by cyclic voltammetry
after each successive oxidation. In all cases, the cyclic
voltammograms were reproducible upon repeated scans and
exhibited no scan rate dependence in the 20-100 mV/s range.
Studies were performed immediately after oxidation and transfer
of the samples to an optical cuvette (absorption) or quartz
capillary (EPR).
(20) Wennerstro¨m, O.; Ericsson, H.; Raston, I.; Svensson, S.; Pimlott,
W. Tetrahedron Lett. 1989, 30, 1129-1132.
(21) (a) Sessler, J.; Johnson, M. R.; Lin, T.-Y. Tetrahedron 1989, 45,
4767-4784. (b) Sessler, J.; Johnson, M.; Creager, S.; Fettinger, J.; Ibers,
J. J. Am. Chem. Soc. 1990, 112, 9310-9329.
(22) Nagata, T.; Osuka, A.; Maruyama, K. J. Am. Chem. Soc. 1990,
112, 3054-3059.
(23) Osuka, A.; Tanabe, N.; Zhang, R.-P.; Maruyama, K. Chem. Lett.
1993, 1505-1508.
4. EPR Spectroscopy. The EPR spectra were recorded as
described previously.9 The sample concentration for all the
experiments was typically 0.05 mM. The microwave power
and magnetic field modulation amplitude were typically 5.7 mW
and 0.32 G, respectively.
(24) Sessler, J.; Capuano, V.; Harriman, A. J. Am. Chem. Soc. 1993,
115, 4618-4628.
(25) Osuka, A.; Tanabe, N.; Nakajima, S.; Maruyama, K. J. Chem. Soc.,
Perkin Trans. 2 1996, 199-203.
(26) Osuka, A.; Nagata, T.; Maruyama, K. Chem. Lett. 1991, 481-
484.
(27) Osuka, A.; Nagata, T.; Maruyama, K. Chem. Lett. 1991, 1687-
1690.
(28) Osuka, A.; Zhang, R.-P.; Maruyama, K.; Ohno, T.; Nozaki, K.
Acknowledgment. This research was supported by a grant
from the Division of Chemical Sciences, Office of Basic Energy
Sciences, Office of Energy Research, U.S. Department of Energy
(J.S.L.) and grants from the NSF (CHE-9707995) (D.F.B., D.H.,
and J.S.L.) and the NIH (GM 34685) (D.H.). Partial funding
for the Mass Spectrometry Laboratory for Biotechnology at
North Carolina State University was obtained from the North
Carolina Biotechnology Center and the NSF. T.A. thanks
Kyushu Institute of Technology and Prof. Norikazu Nishino for
Chem. Lett. 1993, 1727-1730.
(29) Osuka, A.; Nakajima, S.; Maruyama, K.; Mataga, N.; Asahi, T.;
Yamazaki, I.; Nishimura, Y.; Ohno, T.; Nozaki, K. J. Am. Chem. Soc. 1993,
115, 4577-4589.
(30) Osuka, A.; Zhang, R.-P.; Maruyama, K.; Ohno, T.; Nozaki, K. Bull.
Chem. Soc. Jpn. 1993, 66, 3773-3782.
(31) Osuka, A.; Nakajima, S.; Okada, T.; Taniguchi, S.; Nozaki, K.;
Ohno, T.; Yamazaki, I.; Nishimura, Y.; Mataga, N. Angew. Chem., Int. Ed.
Engl. 1996, 35, 92-95.

9436 J. Phys. Chem. B, Vol. 102, No. 47, 1998
Yang et al.
(32) Rempel, U.; von Maltzan, B.; von Borczyskowski, C. Chem. Phys.
Lett. 1995, 245, 253-261.
(33) Chernook, A.; Shulga, A.; Zenkevich, E.; Rempel, U.; von
Borczyskowski, C. J. Phys. Chem. 1996, 100, 1918-1926.
(34) Zhou, X.; Chan, K. S. J. Org. Chem. 1998, 63, 99-104.
(35) Osuka, A.; Maruyama, K.; Yamazaki, I.; Tamai, N. Chem. Phys.
Lett. 1990, 165, 392-396.
(54) (a) Gouterman, M. In The Porphyrins; Dolphin, D., Ed.; Academic
Press: New York, 1978; Vol. III, pp 1-165. (b) Seybold, P. G.; Gouterman,
M. J. Mol. Spectrosc. 1969, 31, 1-13.
(55) (a) Kirmaier, C.; Holten, D. Biochemistry 1991, 30, 609-613. (b)
Drain, C. M.; Kirmaier, C.; Medforth, C. J.; Nurco, D. J.; Smith, K. M.;
Holten, D. J. Phys. Chem. 1996, 100, 11984-11993.
(56) (a) Seth, J.; Palaniappan, V.; Johnson, T. E.; Prathapan, S.; Lindsey,
J. S.; Bocian, D. F. J. Am. Chem. Soc. 1994, 116, 10578-10592. (b) Seth,
J.; Palaniappan, V.; Wagner, R. W.; Johnson, T. E.; Lindsey, J. S.; Bocian,
D. F. J. Am. Chem. Soc. 1996, 118, 11194-11207.
(57) (a) Felton, R. H. In The Porphyrins; Dolphin, D., Ed.; Academic
Press: New York, 1978; Vol. V, pp 53-126. (b) Davis, D. G. In The
Porphyrins; Dolphin, D., Ed.; Academic Press: New York, 1978; Vol. V,
pp 127-152.
(58) Heath, G. A.; Yellowlees, L. J.; Braterman, P. S. J. Chem. Soc.,
Chem. Commun. 1981, 287-289.
(59) Elliot, C. M.; Hershenhart, E. J. Am. Chem. Soc. 1982, 104, 7519-
7526.
(60) Edwards, W. D.; Zerner, M. C. Can. J. Chem. 1985, 63, 1763-
1772.
(61) (a) Angel, S. M.; DeArmond, M. K.; Donohoe, R. J.; Wertz, D.
W. J. Phys. Chem. 1985, 89, 282-285. (b) Donohoe, R. J.; Tait, C. D.;
DeArmond, M. K.; Wertz, D. W. Spectrochim. Acta 1986, 42A, 233-240.
(c) Tait, C. D.; MacQueen, D. B.; Donohoe, R. J.; DeArmond, M. K.; Hanck,
K. W.; Wertz, D. W. J. Phys. Chem. 1986, 90, 1766-1771. (d) Donohoe,
R. R.; Tait, C. D.; DeArmond, M. K.; Wertz, D. W. J. Phys. Chem. 1986,
90, 3923-3926. (e) Donohoe, R. J.; Tait, C. D.; DeArmond, M. K.; Wertz,
D. W. J. Phys. Chem. 1986, 90, 3927-3930.
(62) Fajer, J.; Davis, M. S. In The Porphyrins; Dolphin, D., Ed.;
Academic Press: New York, 1979; Vol. IV, pp 197-256.
(63) Huber, M.; Kurreck, H.; von Maltzan, B.; Plato, M.; Mo¨bius, K. J.
Chem. Soc., Faraday Trans. 1990, 86, 1087-1094.
(64) Spellane, J. P.; Gouterman, M.; Antipas, A.; Kim, S.; Liu, Y. C.
Inorg. Chem. 1980, 19, 386-391.
(36) (a) Osuka, A.; Nakajima, S.; Maruyama, K.; Mataga, N.; Asahi,
T. Chem. Lett. 1991, 1003-1006. (b) Osuka, A.; Maruyama, K.; Mataga,
N.; Asahi, T.; Yamazaki, I.; Tamai, N.; Nishimura, Y. Chem. Phys. Lett.
1991, 181, 413-418. (c) Osuka, A.; Nagata, T.; Maruyama, K.; Mataga,
N.; Asahi, T.; Yamazaki, I.; Nishimura, Y. Chem. Phys. Lett. 1991, 185,
88-94. (d) Osuka, A.; Zhang, R.-P.; Maruyama, K.; Mataga, N.; Tanaka,
Y.; Okada, T. Chem. Phys. Lett. 1993, 215, 179-184.
(37) Osuka, A.; Maruyama, K.; Mataga, N.; Asahi, T.; Yamazaki, I.;
Tamai, N. J. Am. Chem. Soc. 1990, 112, 4958-4959.
(38) Greiner, S.; Winzenburg, J.; von Maltzan, B.; Winscom, C.; Mo¨bius,
K. Chem. Phys. Lett. 1989, 155, 93-98.
(39) Rempel, U.; von Maltzan, B.; von Borczyskowski, C. Chem. Phys.
Lett. 1990, 169, 347-354.
(40) Rempel, U.; von Maltzan, B.; von Borczyskowski, C. J. Lumin.
1991, 48/49, 415-418.
(41) Rempel, U.; von Maltzan, B.; von Borczyskowski, C. J. Lumin.
1992, 53, 175-178.
(42) (a) Jaegermann, P.; Plato, M.; von Maltzan, B.; Mo¨bius, K. Mol.
Phys. 1993, 78, 1057-1074. (b) Rempel, U.; Meyer, S.; von Maltzan, B.;
von Borczyskowski, C. J. Lumin. 1998, 78, 97-110.
(43) Sessler, J.; Johnson, M.; Lin, T.-Y.; Creager, S. J. Am. Chem. Soc.
1988, 110, 3659-3661.
(44) Rodriguez, J.; Kirmaier, C.; Johnson, M.; Friesner, R.; Holten, D.;
Sessler, J. J. Am. Chem. Soc. 1991, 113, 1652-1659.
(45) Sessler, J.; Capuano, V. Tetrahedron Lett. 1993, 34, 2287-2290.
(46) Eriksson, S.; Ka¨llebring, B.; Larsson, S.; Mårtensson, J.; Wenner-
stro¨m, O. Chem. Phys. 1990, 146, 165-177.
(47) Burrell, A.; Officer, D.; Reid, D. Angew. Chem., Int. Ed. Engl.
1995, 34, 900-902.
(65) Ghosh, A. J. Am. Chem. Soc. 1995, 117, 4691-4699.
(66) Davis, W. B.; Svec, W. A.; Ratner, M. A.; Wasielewski, M. R.
Nature, in press.
(48) Lindsey, J. S. In Modular Chemistry; Michl, J., Ed.; NATO ASI
Series C, Mathematical and Physical Sciences, Vol. 499; Kluwer Academic
Publishers: Dordrecht, 1997; pp 517-528.
(67) Chang, J. J. Chem. Phys. 1977, 76, 3901-3909.
(68) The assumptions inherent in the Forster calculation for the
phenylene-linked dimers preclude a precise analysis of the TB and TS rates.
However, the observed rate for F30ZnFbΦ is likely dominated by a TS
process, as seen in the following comparison. Assume that the TS energy
transfer is the sole contributor to the observed (10 ps)-1 rate in F30ZnFbΦ.
The TS rate in ZnFbΦ would then be about (5 ps)-1 based on the larger
spectral overlap for ZnFbΦ than for F30ZnFbΦ (Table 1 of Supporting
Information).9 Accordingly, the measured (3.5 ps)-1 rate in ZnFbΦ would
suggest an ∼(12 ps)-1 TB rate. Hence, the TB rate in F30ZnFbΦ would be
about 120 ps (based on previous finding of a 10-fold greater measured rate
for F30ZnFbU versus ZnFbU, both of which are dominated by a TB
mechanism).
(49) Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827-836.
(50) Lindsey, J. S.; Wagner, R. W. J. Org. Chem. 1989, 54, 828-836.
(51) Fenyo, D.; Chait, B. T.; Johnson, T. E.; Lindsey, J. S. J. Porphyrins
Phthalocyanines 1997, 1, 93-99.
(52) When the same reaction was performed at 10× the concentration
(100 mM pyrrole), the isolated yield of Fb2Φ varied significantly with the
BF3‚OEt2 concentration: 0.15% (3.3 mM), 0.72% (10 mM), 0.84% (33
mM), and 0.01% (100 mM).
(53) Yang, S. I.; Seth, J.; Strachan, J. P.; Genteman, S.; Kim, D.; Holten,
D.; Lindsey, J. S.; Bocian, D. F. J. Porphyrins Phthalocyanines, in press.