Bridge-dependent electron transfer in porphyrin-based donor-bridge-acceptor systems

Article abstract of DOI:10.1021/ja003820k

Photoinduced electron transfer in donor-bridge-acceptor systems with zinc porphyrin (or its pyridine complex) as the donor and gold(III) porphyrin as the acceptor has been studied. The porphyrin moieties were covalently linked with geometrically similar bridging chromophores which vary only in electronic structure. Three of the bridges are fully conjugated π-systems and in a fourth, the conjugation is broken. For systems with this bridge, the quenching rate of the singlet excited state of the donor was independent of solvent and corresponded to the rate of singlet energy transfer expected for a Foerster mechanism. In contrast, systems with a π-conjugated bridging chromophore show a solvent-dependent quenching rate that suggests electron transfer in the Marcus normal region. This is supported by picosecond transient absorption measurements, which showed formation of the zinc porphyrin radical cation only in systems with π-conjugated bridging chromophores. On the basis of the Marcus and Rehm-Weller equations, an electronic coupling of 5-20 cm^-1 between the donor and acceptor is estimated for these systems. The largest coupling is found for the systems with the smallest energy gap between the donor and bridge singlet excited states. This is in good agreement with the coupling calculated with quantum mechanical methods, as is the prediction of an almost zero coupling in the systems with a nonconjugated bridging chromophore.

Full text of DOI:10.1021/ja003820k

J. Am. Chem. Soc. 2001, 123, 3069-3080
3069
Bridge-Dependent Electron Transfer in Porphyrin-Based
Donor-Bridge-Acceptor Systems
Kristine Kilså,† Johan Kajanus,‡ Alisdair N. Macpherson,§ Jerker Mårtensson,‡ and
Bo Albinsson*,†
Contribution from the Department of Physical Chemistry and Department of Organic Chemistry,
Chalmers UniVersity of Technology, SE-412 96 Go¨teborg, Sweden, and Department of Biophysical
Chemistry, Umeå UniVersity, SE-901 87 Umeå, Sweden
ReceiVed October 30, 2000. ReVised Manuscript ReceiVed February 5, 2001
Abstract: Photoinduced electron transfer in donor-bridge-acceptor systems with zinc porphyrin (or its pyridine
complex) as the donor and gold(III) porphyrin as the acceptor has been studied. The porphyrin moieties were
covalently linked with geometrically similar bridging chromophores which vary only in electronic structure.
Three of the bridges are fully conjugated π-systems and in a fourth, the conjugation is broken. For systems
with this bridge, the quenching rate of the singlet excited state of the donor was independent of solvent and
corresponded to the rate of singlet energy transfer expected for a Fo¨rster mechanism. In contrast, systems with
a π-conjugated bridging chromophore show a solvent-dependent quenching rate that suggests electron transfer
in the Marcus normal region. This is supported by picosecond transient absorption measurements, which showed
formation of the zinc porphyrin radical cation only in systems with π-conjugated bridging chromophores. On
the basis of the Marcus and Rehm-Weller equations, an electronic coupling of 5-20 cm-1 between the donor
and acceptor is estimated for these systems. The largest coupling is found for the systems with the smallest
energy gap between the donor and bridge singlet excited states. This is in good agreement with the coupling
calculated with quantum mechanical methods, as is the prediction of an almost zero coupling in the systems
with a nonconjugated bridging chromophore.
Introduction
focused on geometric parameters (distance and orientation),13-17
the free energy of reaction,18-20 and temperature.21
Long-range electron transfer (ET) has attracted much attention
over the past few decades.1-4 Areas in which ET plays a major
role are, for example, in biological transport systems such as
photosynthesis5,6 and in the construction of artificial photosyn-
thetic complexes,7-9 solar cells,10,11 and optoelectronic devices.12
To understand and gain knowledge of the parameters that control
and affect ET, many supramolecular complexes have been
synthesized. Recent studies of porphyrin-based systems have
The work presented here is part of an ongoing project with
the purpose of investigating how the electronic structure of the
intervening medium (the bridge in our case) between a donor
and an acceptor affects photoinduced energy or electron transfer.
To focus solely on this factor, several series of donor-bridge-
acceptor (D-B-A) systems, in which the only parameter that
is varied within a series is the electronic structure of the bridge,
have been investigated. In other words, by keeping the geometric
parameters constant, we can determine how the bridging
chromophores affect the donor-acceptor electronic coupling.
All our series have porphyrin moieties as donor and acceptor,
and in previous studies we have focused on singlet-singlet22,23
and triplet-triplet24 energy transfer, as well as the enhancement
of intersystem crossing in the donor induced by a neighboring
* To whom correspondence should be addressed. E-mail: balb@
phc.chalmers.se. Phone: +46-31-772-3044. Fax: +46-31-772-3858.
† Department of Physical Chemistry, Chalmers University of Technology.
‡ Department of Organic Chemistry, Chalmers University of Technology.
§ Umeå University.
(1) Wasielewski, M. R. Chem. ReV. 1992, 92, 435-461.
(2) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996,
100, 13148-13168.
(3) De Cola, L.; Belser, P. Coord. Chem. ReV. 1998, 177, 301-346.
(4) Gust, D.; Moore, T. A. Intramolecular photoinduced electron-transfer
reactions of porphyrins. In The porphyrin handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: San Diego, 2000; Vol. 8, pp
153-190.
(13) Helms, A.; Heiler, D.; McLendon, G. J. Am. Chem. Soc. 1992, 114,
6227-6238.
(14) Osuka, A.; Kobayashi, F.; Maruyama, K.; Mataga, N.; Asahi, T.;
Okada, T.; Yamazaki, I.; Nishimura, Y. Chem. Phys. Lett. 1993, 201, 223-
228.
(5) Ort, D. R.; Yocum, C. F. Oxygenic Photosynthesis. The light
reactions; Ort, D. R., Yocum, C. F., Eds.; Kluwer Academic Publishers:
Dordrecht, The Netherlands, 1996.
(15) Hunter, C. A.; Hyde, R. K. Angew. Chem., Int. Ed. Engl. 1996, 35,
1936-1939.
(16) Tsuchiya, S. J. Am. Chem. Soc. 1999, 121, 48-53.
(17) Knyukshto, V.; Zenkevich, E.; Sagun, E.; Shulga, A.; Bachilo, S.
Chem. Phys. Lett. 1999, 304, 155-166.
(18) DeGraziano, J. M.; Liddell, P. A.; Leggett, L.; Moore, A. L.; Moore,
T. A.; Gust, D. J. Phys. Chem. 1994, 98, 1758-1761.
(19) DeGraziano, J. M.; Macpherson, A. N.; Liddell, P. A.; Noss, L.;
Sumida, J. P.; Seely, G. R.; Lewis, J. E.; Moore, A. L.; Moore, T. A.;
Gust, D. New J. Chem. 1996, 20, 839-851.
(6) Hoff, A. J.; Deisenhofer, J. Phys. Rep. 1997, 287, 1-257.
(7) Kurreck, H.; Huber, M. Angew. Chem., Int. Ed. Engl. 1995, 34, 849-
866.
(8) Harriman, A.; Sauvage, J.-P. Chem. Soc. ReV. 1996, 25, 41-48.
(9) Gust, D.; Moore, T. A.; Moore, A. L. Pure Appl. Chem. 1998, 70,
2189-2200.
(10) Kalyanasundaram, K.; Gra¨tzel, M. Coord. Chem. ReV. 1998, 177,
347-414.
(11) Bard, A. J.; Fox, M. A. Acc. Chem. Res. 1995, 28, 141-145.
(12) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley,
A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997,
97, 1515-1566.
(20) Osuka, A.; Noya, G.; Taniguchi, S.; Okada, T.; Nishimura, Y.;
Yamazaki, I.; Mataga, N. Chem. Eur. J. 2000, 6, 33-46.
(21) Harriman, A.; Heitz, V.; Sauvage, J.-P. J. Phys. Chem. 1993, 97,
5940-5946.
10.1021/ja003820k CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/13/2001

3070 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
Figure 1. Structure of the systems studied. The donor is either ZnP (no ligand L) or pyZnP (L ) pyridine), and the acceptor is AuP+. The
donor-bridge-acceptor (D-B-A) systems are denoted ZnP-RB-AuP+ or pyZnP-RB-AuP+, respectively. RB is one of the bridging
chromophores: 1,4-bis(phenylethynyl)bicyclo[2.2.2]octane (OB), 1,4-bis(phenylethynyl)benzene (BB), 1,4-bis(phenylethynyl)naphthalene (NB) or
9,10-bis(phenylethynyl)anthracene (AB). The donor reference compounds consist of the porphyrin moiety and the bridging chromophore, D-B.
paramagnetic acceptor.25 The latter series, with iron(III) por-
phyrin as acceptor, were synthesized with the expectation that
long-range electron transfer would be observed. With the
exception of one system in highly polar solvents where stepwise
electron transfer via the bridge was observed,26 it turned out
that other deactivation pathways were dominant. We therefore
decided to prepare new series designed specifically for electron-
transfer studies. These were based on zinc and gold porphyrins
as donor and acceptor, respectively, because electron transfer
in such systems has previously been observed.
In elegant studies by Harriman, Sauvage, and co-workers,
oblique zinc/gold bis-porphyrins of quite short donor-acceptor
distance were fully characterized and ET, as well as its
temperature dependence, was reported.21,27,28 In these systems
it was shown that electron transfer could occur from both the
singlet and triplet excited states of the donor, while hole transfer
occurred only from the triplet excited acceptor. Many rotaxane-
type compounds with different distances, geometry, and elec-
tronic coupling between zinc and gold porphyrins have also been
studied.29-32 Finally, the energy gap dependence of electron
transfer has been investigated in noncovalently connected
sandwich-like heteroaggregates.33 In the study presented here,
electron transfer from the singlet excited state of the donor has
been investigated in a series of porphyrin-based D-B-A
systems where only the electronic structure of the bridging
chromophores has been varied. Thus, it has been possible to
focus on the role of the bridging chromophore on electron
transfer and donor-acceptor electronic coupling.
The D-B-A systems we have studied consist of 5,15-
diphenyl-2,8,12,18-tetraethyl-3,7,13,17-tetramethyl zinc por-
phyrin (ZnP) or its pyridine complex (pyZnP) as the donor, and
the corresponding gold(III) tetrafluoroborate porphyrin (AuP+)
as the acceptor. The porphyrin moieties are covalently connected
by geometrically well-defined bridging chromophores (RB).
Three of the four bridging chromophores are fully conjugated
systems: a bis(phenylethynyl)arylene where the central arylene
is either 1,4-phenylene (BB), 1,4-naphthylene (NB), or 9,10-
anthrylene (AB). In the fourth bridging chromophore the
conjugation is broken by replacing the arylene with 1,4-bicyclo-
[2.2.2]octylene (OB). The general structure of the D-B-A
systems is shown in Figure 1. Since these systems are similar
in structure to those previously used to investigate the mediating
role of the bridge in singlet energy transfer (the donor was ZnP
or pyZnP and the acceptor was the corresponding free base
porphyrin, H2P),22,23 the following characteristics are assumed:
(i) The donor-acceptor distance is constant throughout the series
(center-to-center 25 Å, edge-to-edge 19 Å). (ii) The relative
orientation of the donor and the acceptor is independent of the
bridging chromophore. (iii) The spectroscopic identities of the
donor, bridge, and acceptor chromophores are preserved in the
systems by minimizing the π-conjugation through the D-B-A
systems. This is ensured by the placement of methyl groups in
the porphyrin â-positions, forcing the porphyrin and adjacent
phenyl planes to be almost perpendicular (90° ( 18° at room-
temperature according to a Boltzmann distribution on a PM3
calculated surface).
(22) Jensen, K. K.; van Berlekom, S. B.; Kajanus, J.; Mårtensson, J.;
Albinsson, B. J. Phys. Chem. A 1997, 101, 2218-2220.
(23) Kilså, K.; Kajanus, J.; Mårtensson, J.; Albinsson, B. J. Phys. Chem.
B 1999, 103, 7329-7339.
(24) Andre´asson, J.; Kajanus, J.; Mårtensson, J.; Albinsson, B. J. Am.
Chem. Soc. 2000, 122, 9844-9845.
(25) Kilså, K.; Kajanus, J.; Larsson, S.; Macpherson, A. N.; Mårtensson,
J.; Albinsson, B. Chem. Eur. J. In press.
(26) Kilså, K.; Macpherson, A. N.; Gillbro, T.; Mårtensson, J.; Albinsson,
B. Spectrochim. Acta A. In press.
(27) Harriman, A.; Heitz, V.; Ebersole, M.; van Willigen, H. J. Phys.
Chem. 1994, 98, 4982-4989.
(28) Brun, A. M.; Harriman, A.; Heitz, V.; Sauvage, J.-P. J. Am. Chem.
Soc. 1991, 113, 8657-8663.
(29) Brun, A. M.; Atherton, S. J.; Harriman, A.; Heitz, V.; Sauvage,
J.-P. J. Am. Chem. Soc. 1992, 114, 4632-4639.
(30) Chambron, J.-C.; Harriman, A.; Heitz, V.; Sauvage, J.-P. J. Am.
Chem. Soc. 1993, 115, 7419-7425.
(31) Flamigni, L.; Armaroli, N.; Barigelletti, F.; Chambron, J.-C.;
Sauvage, J.-P.; Solladie, N. New J. Chem. 1999, 23, 1151-1158.
(32) Andersson, M.; Linke, M.; Chambron, J.-C.; Davidsson, J.; Heitz,
V.; Sauvage, J.-P.; Hammarstro¨m, L. J. Am. Chem. Soc. 2000, 122, 3526-
3527.
(33) Segawa, H.; Takehara, C.; Honda, K.; Shimidzu, T.; Asahi, T.;
Mataga, N. J. Phys. Chem. 1992, 96, 503-506.

Porphyrin-Based Donor-Bridge-Acceptor Systems
J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3071
Scheme 1
to promote the reaction and protect the acid labile zinc porphyrin
from demetalation, no gold insertion occurred. Approximately
50% of the starting material could be recovered by chromatog-
raphy (Al2O3/CH2Cl2), whereas the remaining part was de-
graded.37
The building block strategy of the second method was
feasible, although low-yielding (13-30%). Three of the four
ZnP-RB-AuP+ systems (R ) O, B, and N) were synthesized
with this method. However, since the gold porphyrins have very
low solubility in toluene, the reaction conditions developed by
Wagner et al.35 were slightly varied. A yield of 26% was
obtained for the phenyl-linked ZnP-BB-AuP+ from 1 and 5
with chloroform as solvent.
In an attempt to increase the yields, we investigated the
possibility of using a metal as a cocatalyst. Copper is tradition-
ally used,38 but we wanted to avoid the risk of transmetalation
of the zinc porphyrins,39 while the gold porphyrins are probably
stable toward metal exchange. Copper porphyrins would be
troublesome byproducts as they are nonfluorescent, making
small amounts of copper porphyrin contaminants hard to detect.
Zinc, on the other hand, would be the ideal cocatalyst since
exchange of zinc would have no net effect.
Crisp et al. have recently developed a method using ZnCl2
as a cocatalyst together with Pd(PPh3)4 and NaI in piperidine.40
This method is especially attractive since piperidine would
already be required in the synthesis of ZnP-OB-AuP+ from
alkylacetylene 6.23 Although we could reproduce their results
for the palladium-catalyzed coupling of iodobenzene with
phenylacetylene, giving a high yield of diphenylacetylene under
mild conditions, this procedure was not as efficient when applied
to porphyrins. Coupling of gold porphyrin 2 with zinc porphyrin
4 gave ZnP-NB-AuP+ in a 30% yield (40 °C, overnight).
ZnP-OB-AuP+ was afforded in 13% yield from gold por-
phyrin 1 and zinc porphyrin 6 (80 °C, 6 days). The synthesis
of ZnP-AB-AuP+ with the same method failed. The coupling
reaction between gold porphyrin 3 and zinc porphyrin 4 (40
°C, 38 h) gave only halide exchange, producing the chloro
analogue of 3, as shown by FAB-MS. Although the coupling
reaction can probably be optimized for gold porphyrins, we did
not consider it to be within the scope of this investigation.
Instead, we sought an alternative way to synthesize ZnP-AB-
AuP+.
The third method, starting from a bis(free base porphyrin)
dimer (Scheme 3), turned out to be the most convenient.
Although a greater number of reaction steps are used, compared
to the previous method, the synthetic procedure is easier. This
was the method used by Chambron, Sauvage, and co-workers
when they applied their gold insertion method to porphyrin
stoppered rotaxanes.41
ZnP-AB-AuP+ was synthesized from ZnP-AB-H2P,
which was demetalated to H2P-AB-H2P. Gold insertion gave
a mixture of non-, mono-, and dimetalated dimers, from which
H2P-AB-AuP+ was isolated by chromatography. The separa-
tion of the mono- and dimetalated species was surprisingly easy.
Results
In this study, the main goal has been to investigate the
(indirect) influence of the bridging chromophore on electron
transfer. We have concentrated on processes that originate from
the singlet excited state of the donor zinc porphyrin, and
examined the different possible relaxation pathways, such as
intersystem crossing S1 f T1, singlet energy transfer to the
acceptor, and electron transfer to the acceptor (Scheme 1). The
following sections systematically describe the synthesis, the
spectroscopic and electrochemical characterization of the com-
pounds, and their photophysical properties.
Synthesis. We have previously synthesized the zinc/free base
analogues (ZnP-RB-H2P) of ZnP-RB-AuP+ by a building
block synthesis employing the Sonogashira reaction.23,34 The
compounds linked by aromatic bridge chromophores (R ) N,
B, A) were prepared from the alkyne-substituted zinc porphyrins
4 and 5 (Scheme 2), and free base iodoporphyrins.34 The
building blocks were connected using the optimized reaction
conditions developed by Wagner et al.,35 i.e. treatment with
Pd2dba3‚CHCl3 (dba is dibenzylideneacetone) and AsPh3 in a
mixture of toluene and triethylamine. However, the synthesis
of ZnP-OB-H2P, which went via zinc porphyrin 6, required
the use of the stronger base piperidine because of the lower
acidity of alkylacetylenes compared to arylacetylenes.23
To prepare the ZnP-RB-AuP+ systems, three different
methods were evaluated, all utilizing the gold insertion method
recently developed by Chambron et al. for 5,15-diarylocta-
alkylporphyrins.36 In this procedure, [Au(tht)2]BF4 (tht is
tetrahydrothiophene) is used as the metalating agent. The gold
insertion takes place by a disproportionation of the Au(I) salt
to Au(0) and a Au(III) porphyrin and at least 3 equiv of
[Au(tht)2]BF4 and 2 equiv of base (2,6-lutidine) are required.
In the first method, gold insertion into the free base porphyrin
moiety of ZnP-RB-H2P was attempted, analogous to the
method used to produce ZnP-RB-FeP from ZnP-RB-H2P.25
In the second method, the Sonogashira reaction was used to
assemble the ZnP-RB-AuP+ systems from the gold porphyrin
(1-3) and zinc porphyrin (4-6) building blocks (analogous to
the synthesis of ZnP-RB-H2P).23,34 Finally, a sequence of gold
insertion followed by zinc insertion into a bis(free base
porphyrin) dimer was attempted (Scheme 3).
(37) A second fraction came off the chromatographic column as a brown
smear when MeOH was added to the eluent. This fraction contained a
mixture of compounds as judged by TLC. None of the compounds had the
chromatographic mobility or appearance of a gold porphyrin.
(38) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975,
4467-4470.
The first method was not successful. When the AB-linked
compound, ZnP-AB-H2P, was subjected to gold insertion with
[Au(tht)2]BF4 (3 equiv) and a 100-fold excess of 2,6-lutidine,
(34) Kajanus, J.; van Berlekom, S. B.; Albinsson, B.; Mårtensson, J.
Synthesis 1999, 1155-1162.
(39) Anderson, H. L.; Sanders, J. K. M. J. Chem. Soc., Chem. Commun.
1989, 1714-1715.
(35) Wagner, R. W.; Johnson, T. E.; Li, F.; Lindsey, J. S. J. Org. Chem.
1995, 60, 5266-5273.
(40) Crisp, G. T.; Turner, P. D.; Stephens, K. A. J. Organomet. Chem.
1998, 570, 219-224.
(36) Chambron, J.-C.; Heitz, V.; Sauvage, J.-P. New J. Chem. 1997, 21,
237-240.
(41) Solladie´, N.; Chambron, J.-C.; Sauvage, J.-P. J. Am. Chem. Soc.
1999, 121, 3684-3692.

3072 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
Scheme 2
Scheme 3
Zinc insertion gave ZnP-AB-AuP+ in 34% overall yield from
ZnP-AB-H2P (Scheme 3). The intermediate H2P-AB-H2P
could, in principle, be synthesized in a single step by using the
palladium-catalyzed reaction between 9,10-diiodoanthracene and
an alkyne-substituted free base porphyrin. We did not use this
procedure because of the difficulties of removing butadiyne-
linked byproducts that may be formed in the palladium-catalyzed
reaction.35 As was pointed out by Lindsey et al.,42 even if a
bis(zinc porphyrin) dimer or a bis(free base porphyrin) dimer
is the target molecule, it is advantageous to prepare a mixed
zinc/free base porphyrin dimer first. The mixed dimer is easier
to purify, and can then be metalated or demetalated. Therefore,
we started the synthesis of ZnP-AB-AuP+ from ZnP-AB-
H2P, which was synthesized in a stepwise manner and was free
of butadiyne-linked byproducts.34
Ground-State Absorption. In the Q-band region of the
porphyrins (λ > 500 nm), the absorption spectra of the D-B-A
series with zinc porphyrin as a donor are very similar to a linear
combination of D-B and A (Figure 2). The absorption maxima
in chloroform are seen at 518 and 553 nm for the gold porphyrin
acceptor and at 538 and 574 nm (551 and 586 nm upon addition
of pyridine) for ZnP and all the bridge-substituted zinc por-
phyrins. This supports the assumption that D, B, and A can be
regarded as isolated chromophores in the D-B-A systems.
Although the sum of the components does not match the
D-B-A spectra quite as well in the Soret-band region (λ ≈
410 nm), the agreement is still satisfactory. The overlap of the
donor and acceptor Q-bands and the presence of a AuP+ tail
extending into the red means that there is no wavelength where
(42) Lindsey, J. S.; Prathapan, S.; Johnson, T. E.; Wagner, R. W.
Tetrahedron 1994, 50, 8941-8968.

Porphyrin-Based Donor-Bridge-Acceptor Systems
J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3073
emission (Figure 3, Table 1). The largest quenching is seen for
the AB-bridged systems, followed by the systems with NB and
BB as the bridge. In the ZnP-RB-AuP+ series, the donor
quenching was measured in solvents of different polarity
(although the solubility of the systems in toluene and 2-MeTHF
is poor). ZnP-OB-AuP+ showed almost the same rate constant
(kDBA ≈ 2 × 108 s-1) for all solvents. For the other systems,
the quenching was dependent upon solvent, with the largest rate
constants (kDBA ≈ 2-9 × 109 s-1) found for the most polar
solvents and for the AB-bridged system (Table 2). In the
nonpolar solvent toluene, the quenching of the donor emission
for the AB-, NB-, and BB-bridged systems was substantially
smaller and approached that observed for ZnP-OB-AuP+.
Donor Intersystem Crossing. In a study of a similar
D-B-A series with zinc porphyrin as the donor and the
paramagnetic high-spin iron(III) chloride porphyrin, FeP, as the
acceptor, we found that the presence of this acceptor enhanced
the S1 f T1 intersystem crossing rate of the donor contributing
significantly to the observed fluorescence quenching.25 Although
AuP+ is diamagnetic and we did not expect the same effect for
the systems investigated here, the intersystem crossing could
be affected by the heavy-atom effect of gold. This was
investigated by using nanosecond transient absorption to detect
the amount of triplet formed in D-B relative to D-B-A
immediately after the excitation pulse. This was done by
comparing the differential absorbances, ∆ADBA and ∆ADB
,
immediately after the excitation pulse.
In the series with iron porphyrin as the acceptor, a minimum
value43 of enhanced intersystem crossing was determined in the
OB-bridged system because triplet energy transfer, which would
have complicated the analysis, does not occur in this system.25
For a similar reason, the OB-bridged system was also chosen
to establish if the presence of AuP+ affects the yield of ZnP
triplet formation. Assuming that the intersystem crossing
quantum yield for ZnP in polar solvents is similar to that
Figure 2. Absorption spectra of the ZnP-RB-AuP+ systems (s),
resolved into ZnP-RB (‚‚‚) and AuP+ (- - -) in CHCl3 at 20 °C. From
top to bottom RB is (a) OB, (b) BB, (c) NB, and (d) AB.
ZnP can be selectively excited. In all the fluorescence or
transient absorption measurements, the compounds were there-
fore excited at a wavelength in the Q-band region where the
donor absorption dominates. However, since the acceptor has
short-lived excited states and is nonfluorescent, excitation of
AuP+ will not influence the analysis of the emission properties.
Emission Properties. The steady-state fluorescence spectra
of the D-B-A systems show that the donor fluorescence is
quenched compared to the D-B reference compounds. The
steady-state quenching corresponds to the decrease in the donor
lifetime (Table 1). The decrease in both the fluorescence
intensity and the lifetime of the singlet excited donor was used
to calculate the efficiency (E) and the rate constant (kDBA) for
donor quenching.
reported for zinc tetraphenylporphyrin (φ0 ) 0.90),44 the
isc
intrinsic rate constant for intersystem crossing is k0
)
isc
φ0isc/τ0 ) 6 × 108 s-1 in butyronitrile. As a result of quench-
f
ing of the singlet excited donor state in the D-B-A
systems, the quantum yield for intersystem crossing, φisc
)
(∆ADBA/∆ADB)φ0isc, is reduced to 0.78 in ZnP-OB-AuP+. With
use of the fluorescence lifetime of ZnP-OB-AuP+, τf ) 1.12
ns, the rate constant for intersystem crossing, kisc ) 7 ×108
s-1, is obtained. Since this rate is essentially the same as that
for ZnP, within the limits of experimental uncertainty, we do
not believe that the donor quenching can be a result of enhanced
intersystem crossing in any of the gold porphyrin containing
systems.
Energy Transfer. The lowest singlet excited state of AuP+,
as determined from the maximum of the lowest energy
absorption band (Figure 2), is approximately 700 or 1000 cm-1
higher in energy than that in ZnP or pyZnP, respectively.
However, since AuP+ is nonfluorescent and the singlet lifetime
short, there is no possibility of singlet energy transfer from AuP+
to ZnP or pyZnP. Although energy transfer in the reverse
direction, from ZnP or pyZnP to AuP+, does not at first seem
likely, the AuP+ absorption has a red tail that overlaps the
emission from ZnP and pyZnP (Figure 4). Using Fo¨rster theory,
τf
F
E ) 1 - ) 1 -
F0
E
kDBA
)
(1)
τ0
(1 -E)τ0
f
f
F and F0 are the total donor fluorescence intensities in the
D-B-A systems and in the D-B references, respectively, and
τf and τ0 are the corresponding fluorescence lifetimes. The
f
fluorescence lifetimes of the reference compounds were, within
experimental error, independent of the bridging chromophore.
The efficiencies calculated from the steady-state and time-
resolved measurements were, within 20%, the same and the
(43) An increase in the rate of intersystem crossing from (6 ( 1) × 108
s-1 in ZnP-OB to (11 ( 2) × 108 s-1 in ZnP-OB-FeP was determined
in 2-methyl-THF. In the systems with the other bridging chromophores the
enhancement could be larger, although the results suggested by not more
than a factor of 2-3.
average value was used to calculate kDBA
.
In both series, ZnP-RB-AuP+ and pyZnP-RB-AuP+, it
is clear that the systems with OB as the bridging chromophore
show only a minor quenching, while introduction of aromatic
bridging chromophores has a dramatic effect on the donor
(44) Harriman, A.; Porter, G.; Wilowska, A. J. Chem. Soc., Faraday
Trans. 2 1983, 79, 807-816.

3074 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
Table 1. Fluorescence Intensity (F) and Lifetime (τ) of the Donor Porphyrin Moiety, Together with the Calculated Efficiency (E) and Total
Quenching Rate Constant (kDBA) in Chloroform and, in Parentheses, Butyronitrilea at 20 °C
compd
F (au)
τD (ns)
Eb
kDBA (s-1
)
kFo¨rster (s-1
)
ZnP-OB-AuP+
ZnP-OB
70 (78)
100 (100)
18 (20)
100 (100)
13 (11)
100 (100)
6
0.99 (1.12)
1.29 (1.46)
0.46 (0.34)
1.28 (1.42)
0.37 (0.29)
1.28 (1.47)
0.21
0.27 (0.23)
0.3 × 109 (0.2 × 109)
2.1 × 109 (2.5 × 109)
3.1 × 109 (3.9 × 109)
5.8 × 109
0.2 × 109 (0.2 × 109)
6.7 × 109 (3.3 × 109)
5.6 × 109 (5.5 × 109)
13.2 × 109
0.1 × 109 (0.1 × 109)
0.1 × 109 (0.1 × 109)
0.1 × 109 (0.1 × 109)
0.1 × 109
0.1 × 109 (0.1 × 109)
0.1 × 109 (0.1 × 109)
0.1 × 109 (0.1 × 109)
0.1 × 109
ZnP-BB-AuP+
ZnP-BB
0.73 (0.78)
0.80 (0.85)
0.88
ZnP-NB-AuP+
ZnP-NB
ZnP-AB-AuP+
ZnP-AB
100
1.26
pyZnP-OB-AuP+
pyZnP-OB
83 (86)
100 (100)
5 (20)
100 (100)
5 (14)
100 (100)
3
100
0.93 (1.03)
1.21 (1.25)
0.20 (0.23)
1.20 (1.30)
0.26 (0.13)
1.20 (1.12)
0.13
0.20 (0.16)
0.89 (0.81)
0.87 (0.86)
0.94
pyZnP-BB-AuP+
pyZnP-BB
pyZnP-NB-AuP+
pyZnP-NB
pyZnP-AB-AuP+
pyZnP-AB
1.19
a Fluorescence measurements were not made in butyronitrile for systems containing AB because of electron transfer from ZnP or pyZnP to AB
(see ref 26). b Based on an average of steady-state and lifetime measurements.
Table 2. The Refractive Index (n), Dielectric Constant (ꢀ), Calculated Driving Force (∆G°) and Reorganization Energy (λ), Observed
Quenching Rate Constants (kDBA), and Calculated Fo¨rster Rate Constants (kFo¨rster) for ZnP-RB-AuP+ in Different Solvents at 20 °C
d
∆G° b λc
E00 ZnP-OB-AuP+ ZnP-BB-AuP+ ZnP-NB-AuP+ ZnP-AB-AuP+
ZnP/AuP+
solvent
na
ꢀa
(eV) (eV) (eV)
kDBAe (s-1
)
kDBAe (s-1
)
kDBAe (s-1
)
kDBAe (s-1
)
kFo¨rsterf (s-1
)
toluene
CHCl3
1.496 2.379
0.21 0.26 2.15 (3.4 ( 0.7) × 108 (0.5 ( 0.1) × 109 (0.8 ( 0.2) × 109 (1.3 ( 0.3) × 109 (2.0 ( 0.4) × 108
1.446 4.807 -0.43 0.86 2.15 (2.9 ( 0.6) × 108 (2.1 ( 0.4) × 109 (3.1 ( 0.6) × 109 (5.8 ( 1.2) × 109 (1.1 ( 0.2) × 108
2-MeTHF 1.405 6.97 -0.61 1.08 2.13 (2.0 ( 0.4) × 108 (1.0 ( 0.2) × 109 (1.6 ( 0.3) × 109 (6.2 ( 1.3) × 109 (1.7 ( 0.4) × 108
CH2Cl2
1.424 8.93 -0.72 1.13 2.15 (2.3 ( 0.5) × 108 (2.2 ( 0.4) × 109 (4.3 ( 0.9) × 109 (9.1 ( 1.8) × 109 (1.5 ( 0.3) × 108
C3H7CN 1.384 24.83 -0.92 1.37 2.13 (2.2 ( 0.4) × 108 (2.5 ( 0.5) × 109 (3.9 ( 0.8) × 109
g
g
(0.8 ( 0.2) × 108
DMF
1.431 38.25 -0.95 1.32 2.12 (1.3 ( 0.3) × 108 (2.3 ( 0.5) × 109 (4.9 ( 1.0) × 109
(0.6 ( 0.1) × 108
a From ref 47. b Calculated from eq 5. c Calculated from eq 6. d Determined from absorption and fluorescence measurements. e Uncertainties
based on the differences in steady-state and lifetime measurements. f Calculated by using eqs 2 and 3. Uncertainties based on uncertainties in
lifetime, quantum yield, and molar absorptivity. g Not determined due to electron transfer from ZnP to AB in polar and coordinating solvents (see
ref 26).
Figure 3. Fluorescence emission spectra of the (a) ZnP-RB-AuP+ (λex ) 538 nm) and (b) pyZnP-RB-AuP+ (λex ) 550 nm) systems compared
to the fluorescence spectrum of the donor alone (scaled to equal donor optical density) in CHCl3 at 20 °C. Donor (- ‚‚ -), RB ) OB (‚‚‚), RB ) BB
(- - -), RB ) NB (s) and RB ) AB (- ‚ -).
the expected rate of singlet energy transfer (kFo¨rster) in the
absence of any mediating influence of the bridge can be
calculated.45,46
orientation factor (κ2), which depends on the orientation of the
donor and acceptor transition dipoles, the refractive index of
the solvents used (n),47 and the fluorescence quantum yield (φ0 )
f
and lifetime (τ0 ) of the donor in the absence of the acceptor. If
f
φ0 κ2J
we assume that the transition dipoles of the metal porphyrin
f
kFo¨rster ) 8.79 × 10-25
6s-1
(2)
moieties are degenerate in the porphyrin plane23,48 and the planes
τ0 n4RDA
f
(45) Fo¨rster, T. Naturwissenschaften 1946, 33, 166-175.
(46) Fo¨rster, T. Ann. Phys. (Paris) 1948, 2, 55-75.
(47) Handbook of Chemistry and Physics, 75th ed.; CRC Press: Boca
Raton, FL, 1994.
The D-A center-to-center distance (RDA) in our bridged systems
is 25.3 Å. The other parameters in this equation are the

Porphyrin-Based Donor-Bridge-Acceptor Systems
J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3075
from the structure (λi) and solvent (λs). If V is small compared
to λ and the energy of the promoting vibration, the rate constant
49-51
is given by:
-(∆G° + λ)2
π
2
kET
)
|V| exp
(4)
(
)
p2λk T
4λkBT
x
B
where T is the temperature and kB is the Boltzmann constant.
As a result of the quadratic form of (∆G° + λ), this expression
leads to a normal region where kET increases with more negative
∆G° and, after a maximum in kET is reached at ∆G° ) -λ, an
inverted region where kET decreases with more negative ∆G°.
The calculation of ∆G° and λ is based on the following
parameters: the donor and acceptor redox potentials (Eox and
E
red), the 0-0 excitation energy of the donor (E00), the donor-
Figure 4. Spectral overlap of AuP+ Q-band absorbance (‚‚‚) and
normalized ZnP (s) or pyZnP (- - -) emission in CHCl3 at 20 °C.
acceptor center-to-center distance (RDA), and the average radii
of the donor and acceptor (r). The dielectric constant (ꢀs) and
refractive index of the solvent (n) used in the ET measurements,
together with the dielectric constant of the solvent used in the
electrochemical measurements (ꢀsref), are also required:52-56
are almost freely rotating with respect to each other, the average
value of κ2 ) 2/3 for random orientations can be used. The
final term is the spectral overlap integral (J), which is calculated
from the molar absorptivity of the acceptor (ꢀ(λ)) and the
normalized donor emission (F(λ)) in the wavelength (λ) region
show in Figure 4.
e2
1
1
ꢀsref
1
∆G° ) e(Eox - Ered) - E00
+
-
(5)
(6)
4πꢀ ꢀ
r
(
)( )
0
s
e2
1
1
1
1
J ) ∞ꢀ(λ) F(λ) λ4 dλ
(3)
∫
λ ) λi + λs ) λi +
-
-
0
2
(
)(
)
4πꢀ0 r RDA
ꢀs
n
The calculated rate constants for Fo¨rster energy transfer are
8 × 107 and 11 × 107 s-1 for ZnP-RB-AuP+ in butyronitrile
and chloroform, respectively, and 6 × 107 s-1 for pyZnP-RB-
AuP+ in both of these solvents. In the other solvents used to
study the fluorescence emission from the ZnP-RB-AuP+
systems, kFo¨rster varies between 0.5 and 2 × 108 s-1 (Table 2).
These calculations indicate that the quenching observed in the
OB-linked systems is the result of singlet energy transfer, while
in the other systems, an additional pathway must be operating.
Electrochemistry. Cyclic voltammetric measurements were
made so that the energy of the charge separated states could be
estimated from the redox potentials of the chromophores. The
cyclic voltammogram of AuP+ in CH2Cl2 revealed two (revers-
ible) reduction potentials of -1.05 and -1.76 V vs Ag/Ag+
electrode (E ) 0.45 vs SHE) at a sweep rate of 100 mV/s. This
is in agreement with previously reported data.36 The oxidation
potentials of ZnP and pyZnP are 0.38 and 0.29 V, respectively,26
indicating a slightly larger driving force for electron transfer in
the pyZnP-RB-AuP+ systems (vide infra). Previously, the AB
bridging chromophore has been shown to exhibit reversible
potentials: Eox ) 0.86 V and Ered ) -1.68 V, which is a
sufficiently small reduction potential for AB to be a possible
electron acceptor (at least on thermodynamic grounds) for the
singlet excited states of ZnP and pyZnP.26 However, for the
AB-linked reference compounds, this stepwise ET was only
observed in highly polar solvents. BB showed potentials of Eox
) 1.38 V and Ered ) -2.35 V, while NB gave Eox ) 1.20 V.
The reduction potential for NB was not clearly observed, but if
it follows the same trend as for the oxidation potentials, it will
lie between BB and AB at approximately -2.1 V.
In our calculations we have used an average radius of 4.8 Å, as
found for other porphyrin-based donor-acceptor systems,19 and
an internal reorganization energy (λi) of 0.2 eV, which was used
in other systems having zinc and gold porphyrins.21 Since ET
in our systems is a charge transfer, as opposed to a charge
separation, we have omitted the Coulombic stabilization term
usually present in eq 5. However, inclusion of this term would
in the polar solvents only change ∆G° by 0.1 eV, or less,
because the donor-acceptor distance is relatively large. The
calculated ∆G° and λ values for all the solvents studied are
listed in Table 2.
Transient Absorption. Transient absorption measurements
on the picosecond time scale were made to determine if electron
transfer occurs in the systems with π-conjugated bridging
chromophores. DMF was chosen for these experiments because
its high polarity favors electron transfer and the solubility and
stability of the D-B-A systems is high in this solvent. The
kinetics recorded at 645 nm and the spectra recorded at 800 ps
for the D-B-A systems in DMF are shown in Figure 5a,b.
The formation of a clear peak at ∼680 nm, indicative of
ZnP•+,13,28 on the time scale of several hundreds of picoseconds
for ZnP-BB-AuP+ and ZnP-NB-AuP+ is consistent with a
dominant electron-transfer quenching pathway for these systems
in polar solvents. In contrast, no obvious signal from the radical
cation could be detected for ZnP-OB-AuP+ at 800 ps (Figure
(49) Marcus, R. A. J. Chem. Phys. 1956, 24, 966-978.
(50) Levich, V. G. Present state of the theory of oxidation-reduction in
solution (bulk and electrode reactions). In AdVances in Electrochemisty and
Electrochemical Engineering; Delahay, P., Ed.; Interscience Publishers:
New York, 1966; Vol. 4, pp 249-371.
(51) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265-
322.
(52) Marcus, R. A. Can. J. Chem. 1959, 37, 155-163.
(53) Marcus, R. A. J. Chem. Phys. 1965, 43, 679-701.
(54) Rehm, D.; Weller, A. Ber. Bunsen-Ges. Phys. Chem. 1969, 73, 834-
839.
Electron Transfer. In Marcus theory for diabatic electron
transfer, the rate constant for electron transfer (kET) is related
to the electronic coupling (V), the driving force (∆G°) for the
reaction, and the reorganization energy (λ), with contributions
(48) Gouterman, M. Optical spectra and electronic structure of porphyrins
and related rings. In The porphyrins; Dolphin, D., Ed.; Academic Press:
New York, 1978; Vol. III, Chapter 1.
(55) Weller, A. Z. Phys. Chem. NF 1982, 133, 93-98.
(56) Schmidt, J. A.; Liu, J. Y.; Bolton, J. R.; Archer, M. D.; Gadzekpo,
V. P. Y. J. Chem. Soc., Faraday Trans. 1 1989, 85, 1027-1041.

3076 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
attributed to differences in the electronic structure of the bridging
chromophore, and thus in the electronic coupling between the
donor and acceptor.
Two series were investigated, with either the donor ZnP or
its pyridine complex pyZnP, and these differed only slightly in
their spectroscopic and electrochemical properties. In both series,
the lifetime of the singlet excited state of the zinc porphyrin is
reduced when the gold porphyrin is covalently linked and the
quenching is substantially larger for the fully conjugated
bridging chromophores (BB, NB, and AB) compared to the OB-
linked systems. Singlet-singlet energy transfer is one of the
quenching pathways, although the spectral overlap between the
AuP+ absorption and ZnP or pyZnP emission is small. Fo¨rster
theory was used to calculate a rate constant kFo¨rster ≈ 1 × 108
s-1 for singlet energy transfer in butyronitrile and chloroform,
independent of the bridging chromophore. The red shift in the
emission spectrum of pyZnP compared to ZnP results in only
a small reduction in the Fo¨rster energy transfer rate. Experi-
mentally, the reduction in the measured rate constant when
changing the donor was only observed for the system with OB
as bridging chromophore (Table 1).
In the D-B-A systems previously investigated (D ) ZnP
or pyZnP, A ) free base porphyrin, H2P), energy transfer rates
larger than predicted by the Fo¨rster theory (kFo¨rster ) 3.3 × 108
or 0.9 × 108 s-1 for ZnP or pyZnP, respectively) were observed
and the enhancement was clearly dependent upon the bridging
chromophore.23 The total rate constant for excitation energy
transfer (kEET) could be expressed as kEET ) kFo¨rster + kMed, where
Figure 5. (a) Transient absorption kinetics of ZnP-OB-AuP+, ZnP-
k
Med is the mediating contribution from the bridge. For systems
BB-AuP+, and ZnP-NB-AuP+ in DMF (λpump ) 585 nm, λprobe
)
645 nm). (b) Transient absorption spectra of ZnP-OB-AuP+ (‚‚‚),
ZnP-BB-AuP+ (s), ZnP-NB-AuP+ (- - -), and ZnP-AB-AuP+
(- ‚ -) at 800 ps in DMF (λpump ) 585 nm).
with OB, BB, NB, and AB as the bridging chromophore, the
average kMed was determined to be 0.5 × 108, 1.7 × 108, 2.2 ×
108, and 8.7 × 108 s-1, respectively. Assuming the mediating
contribution from the bridge to be the same when AuP+ is the
acceptor, the donor quenching rates (kDBA) of Tables 1 and 2
are in reasonable agreement with those expected for singlet
energy transfer (kEET) when OB is the bridging chromophore.
The contribution from any other quenching pathways, such as
electron transfer or enhanced intersystem crossing, must be
rather small for ZnP-OB-AuP+. The absence of any significant
enhancement in the donor intersystem crossing by the dia-
magnetic gold porphyrin is not surprising, since this effect has
previously only been found for metalloporphyrins with unpaired
electrons, such as iron(III) porphyrin25,59 or copper(II) por-
phyrin.60,61
5b). For ZnP-AB-AuP+ in DMF, the major route of ZnP•+
formation is by ET to the bridge, giving the charge-separated
state ZnP•+-AB•--AuP+, followed by a second ET step
resulting in ZnP•+-AB-AuP•. However, in less polar solvents
such as CH2Cl2 where stepwise ET does not occur,26 formation
of ZnP•+ was observed, indicating direct ET in ZnP-AB-AuP+
as is the case for the systems containing BB and NB as bridging
chromophores.
Discussion
The donor-bridge-acceptor systems that we have studied
were designed to isolate the effect of the bridging chromophore
from the other parameters that could influence the rate of
electron transfer. To this end, the structure of the donors and
acceptor and the overall geometry of the D-B-A systems were
kept essentially constant. Previous studies on related systems
indicate that both the donor-acceptor distance and relative
orientation are independent of the bridging chromophore
throughout the series.22,23 Furthermore, by placement of methyl
groups on the porphyrin rings adjacent to the phenyl substituents,
direct conjugation through the D-B-A systems is minimized.
Although different porphyrin substitution patterns can affect the
overall donor/acceptor coupling, as has been found for energy
transfer studies of other porphyrin-based systems,57,58 this does
not prevent comparisons of the role of the bridging chro-
mophores being made within our series. Therefore, any change
in the quenching rate constant within a series can be directly
In contrast to the OB-linked systems, the quenching rate for
systems with π-conjugated bridging chromophores is about an
order of magnitude larger than the predicted kEET in polar
solvents. In the case of toluene, however, the quenching rate is
substantially smaller and is of the same order of magnitude as
for ZnP-OB-AuP+. The rates in this solvent follow the trend
ZnP-AB-AuP+ > ZnP-NB-AuP+ > ZnP-BB-AuP+
>
ZnP-OB-AuP+ and correspond, within the experimental
uncertainty, to an energy transfer rate constant calculated as
the sum of a Fo¨rster contribution and bridge mediation. In the
more polar solvents, energy transfer must make only a minor
contribution to the donor quenching, implying that the electron-
transfer pathway is dominant. To directly establish electron
transfer as a quenching pathway transient absorption measure-
ments on a picosecond time scale were made in DMF. The zinc
porphyrin radical cation, ZnP•+, has a characteristic absorption
(57) 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.
(58) Yang, S. I.; Seth, J.; Balasubramanian, T.; Kim, D.; Lindsey, J. S.;
Holten, D.; Bocian, D. F. J. Am. Chem. Soc. 1999, 121, 4008-4018.
(59) Brookfield, R. L.; Ellul, H.; Harriman, A. J. Chem. Soc., Faraday
Trans. 2 1985, 81, 1837-1848.
(60) Asano-Someda, M.; Kaizu, Y. Inorg. Chem. 1999, 38, 2303-2311.
(61) Toyama, N.; Asano-Someda, M.; Ichino, T.; Kaizu, Y. J. Phys.
Chem. A 2000, 104, 4857-4865.

Porphyrin-Based Donor-Bridge-Acceptor Systems
J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3077
band around 680 nm,13,28 and this spectral feature was only
observed for the systems with fully conjugated bridging
chromophores. The transient absorption spectrum of ZnP-OB-
AuP+ in the same wavelength region was very similar to that
of ZnP at 800 ps. The spectra show a stimulated emission feature
at ∼645 nm superimposed on a broad and featureless excited-
state absorption, typical of the excited singlet state. The direct
charge-transfer pathway from ZnP f AuP+ could not be studied
in the ZnP-AB-AuP+ system in highly polar solvents, because
formation of the ZnP•+-AB•--AuP+ charge-separated state is
rapid.26 However, formation of ZnP•+ was also observed in
CH2Cl2, where ET to the bridge does not take place. In view of
this and the relatively large quenching rates measured for this
system in other solvents, we believe that direct electron transfer
to AuP+ does occur in ZnP-AB-AuP+, and in all the other
systems (including the systems with pyZnP as the donor) with
π-conjugated bridging chromophores, in all investigated solvents
except toluene.
For ZnP-BB-AuP+, ZnP-NB-AuP+, and ZnP-AB-
AuP+ a marked solvent dependence of kDBA is found between
toluene and polar solvents. An increase in the rate constant with
polarity is indicative of electron transfer in the Marcus normal
region. In the nonpolar solvent toluene, the calculated ∆G° for
ET is positive and thus energy transfer is likely to be the only
quenching pathway in this solvent, regardless of the bridging
chromophore. In butyronitrile and chloroform, where both series
(D ) ZnP or pyZnP) have been investigated, the rate constant
for the systems with fully conjugated bridging chromophores
is larger in the systems with pyZnP as the donor. This is
consistent with the larger driving force for electron transfer and
stabilization of the zinc porphyrin radical cation,62 and implies
that energy transfer, which is expected to decrease upon pyridine
coordination, makes a negligible contribution to the total
quenching rate constant. The difference between the two series
is largest in chloroform, since butyronitrile can also coordinate
to ZnP in the same manner as pyridine.
The driving force, ∆G°, depends on the reduction potential
of the acceptor, the oxidation potential of the donor, and the
energy of the donor’s lowest singlet excited state. These are
the first terms in eq 5. The oxidation and reduction potentials
of the separate donor and acceptor species were measured in
dichloromethane and are specific to this one solvent. The driving
force for ET in the other solvents in which the photophysics
were measured can be estimated by compensating for the change
in electrochemical potentials with solvent. A correction for the
different solvation energies of the ions can be made by using
the solvent dielectric constant and the Born sphere radii of the
chromophores involved. This is the last term of eq 5. The
validity of using the Born sphere correction has been discussed,
in terms of both the spherical shape and the size of the
chromophore radii.56 However, in changing from almost non-
polar to highly polar solvents, we find it appropriate to include
this kind of correction. We have used a radius of 4.8 Å, since
systems with similar porphyrins showed good agreement
between the experimentally determined and calculated driving
forces when using this value.19 In cases where a charge-separated
state is formed from a neutral system, a work term is also
included in the expression for ∆G°. This term allows for the
Coulombic stabilization of the ion pair compared to the separate
ions produced in the electrochemical measurements and depends
inversely on the distance of D-A charge separation. For our
systems, this term is of the order of 0.1 eV and will result in a
Figure 6. Solvent dependence of kDBA in the ZnP-RB-AuP+
systems: RB ) OB ([), RB ) BB (2), RB ) NB (b), and RB ) AB
(9). The lines are linear fits with a fixed slope of 0 or -1.
more negative ∆G°. However, for systems containing gold(III)
porphyin, it has been argued that the Coulombic term should
not be included because the transfer process is a charge shift
ZnP-AuP+ f ZnP•+-AuP•.21,28,30 The reorganization energy
of the solvent, λs, is also based on the Born sphere radius and
distance, and is generally agreed to be of the form found in eq
6.
The solvent dependence of kDBA in the polar solvents can be
analyzed in several ways, but one simple approach is to
rearrange the Marcus equation (eq 4) to a logarithmic form.56
(∆G° + λ)2
π
2
ln(kETλ1/2) ) ln
V
-
(7)
2
| |
(
)
4λkBT
p k T
x
B
In this way, a plot of Y ) ln(kETλ1/2) vs X ) (∆G° +λ)2/(4λkBT)
should give a straight line with slope -1. Such a plot of the
total deactivation rate constant, kDBA, for ZnP-RB-AuP+ is
shown in Figure 6 with the data for toluene omitted. This clearly
shows the solvent independence of kDBA in ZnP-OB-AuP+
(slope 0), compared to the solvent dependence of kDBA in ZnP-
BB-AuP+, ZnP-NB-AuP+, and ZnP-AB-AuP+ (slope -1).
The lines with fixed slopes fit the data satisfactorily. This
therefore supports our assertion that the mechanism for donor
quenching in the systems with BB, NB, and AB as bridging
chromophores is photoinduced electron transfer from ZnP to
AuP+. The solvent dependence can also yield information about
the electronic coupling, V. The couplings calculated from the
intercepts of Figure 6 are 6.3, 8.4, and 13.1 cm-1 for ZnP-
BB-AuP+, ZnP-NB-AuP+, and ZnP-AB-AuP+, respec-
tively. However, there are many parameters that can influence
the intercepts determined from this plot, and changing λi from
0.2 to 0.1 eV, for example, decreases the coupling by 2-4 cm-1
Including the Coulombic stabilization term in the estimation of
.
∆G° also affects the scale of the plot, but not the linearity, and
the resulting coupling again decreases 2-4 cm-1
.
The data in Figure 6 are based on an average of steady-state
and lifetime quenching measurements of fluorescence emission.
In general, the steady-state quenching results gave slightly higher
rate constants, with the largest difference found for ZnP-AB-
AuP+. This is probably caused by a limitation in the time
resolution of our phase/modulation spectrofluorimeter, as the
donor quenching efficiency becomes larger than 90%. Calculat-
ing the electronic coupling V from the solvent dependence of
either the steady-state or lifetime measurements gives the
following ranges: ZnP-BB-AuP+, V ) 5.5-7.5 cm-1; ZnP-
NB-AuP+, V ) 7-10 cm-1; and ZnP-AB-AuP+, V ) 11-
17 cm-1. For an oblique system based on zinc and gold(III)
(62) Gust, D.; Moore, T. A.; Moore, A. L.; Kang, H. K.; DeGraziano, J.
M.; Liddell, P. A.; Seely, G. R. J. Phys. Chem. 1993, 97, 13637-13642.

3078 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
tetraphenylporphyrins an electronic coupling of 85 cm-1 has
been found, which decreased to 30 cm-1 for the same porphyrin
system as part of a rotaxane.21,30 In this system both the through-
space and through-bond (8.5 Å edge-to-edge) distances are much
shorter than in the systems presented here. Also the coupling
to the bridge in these systems might be larger due to the reduced
steric hindrance afforded by the different porphyrin substitution
pattern. In view of these differences, the values of the electronic
coupling in the ZnP-RB-AuP+ systems seem reasonable.
The two mechanisms that are generally suggested for ET are
the through-space and the through-bond (or superexchange)
mechanisms. Because of the large donor-acceptor distance (25
Å center-to-center, 19 Å edge-to-edge) in the systems investi-
gated here, the through-space mechanism, requiring a direct
orbital overlap between the donor and acceptor,63 is not likely
to make a contribution. In contrast, the superexchange mech-
anism, in which virtual states of the bridging chromophore mix
with electronic states of the donor and/or the acceptor, can
promote long-range electron transfer.64,65
Figure 7. The correlation between the measured (b) or calculated (/)
electronic coupling for electron transfer and the inverse energy gap
between the first singlet excited states of the donor and bridge. The
lines are the linear fits with a fixed intercept of 0.
The electronic coupling within the superexchange mechanism
can be analyzed in terms of the couplings between the donor
and bridge, âDB, and between the bridge and acceptor, âBA, and
also the energy difference between the initial state and the bridge
state involved, δE.65
systems investigated in this study are 11600, 8600, and 3900
cm-1 for ZnP-BB-AuP+, ZnP-NB-AuP+, and ZnP-AB-
AuP+, respectively.23 Figure 7 shows a plot of V vs δE. For
the systems with BB and NB as bridges, the correlation
including the intercept at zero is very good (δEBB/δENB ≈ 1.3).
The deviation of the point for the AB bridge is larger, but so is
the experimental uncertainty as a result of the few solvent points
in the plot of Figure 6. Furthermore, as has already been
mentioned, the measured V could be too small because of the
high degree of fluorescence quenching (approaching 100%).
The electronic coupling predicted by quantum mechanical
calculations on structurally similar, but symmetric, ZnP-RB-
ZnP systems25 is also shown in Figure 7. We believe that ZnP-
RB-ZnP is a reasonable model for ZnP-RB-AuP+ since the
absorption spectrum of AuP+ is similar to that of ZnP (blue-
shifted ca. 700 cm-1) and the reduction is believed to be in the
porphyrin ring in both cases, rather than at the metal.66-68
In the calculations, the electronic coupling for the different
bridging chromophores was determined from the minimum
energy gap between the near-degenerate LUMO and LUMO#
of the entire symmetric D-B-A system. The coupling varies
with both the overall conformation of the systems (statistically
weighed for a Boltzmann distribution of angles between the
porphyrin and phenyl plane) and the relative orientation of the
central unit in the bridging chromophore.23 The (room temper-
ature) couplings obtained on averaging these distributions were
2.2, 3.5, and 8.1 cm-1 for RB ) BB, NB, and AB, respec-
tively.25 The electronic couplings estimated from the experi-
mental results are larger than the couplings calculated by a
quantum mechanical approach by a factor of 2 to 3, but the
relative trend seems to be correctly captured by the theoretical
model. The reasonable correlation between the calculated
couplings and the measured inverse energy gap δE also indicates
that the model based on the bridge excitation energies is
reasonable.
â
DBâBA
V
(8)
δE
The correlation between V and 1/δE gives, therefore, a straight
line with zero coupling at infinite energy gap. For our systems,
δE of the virtual state D+-B--A may be calculated as δE )
e(EoxD - EredB) - E00D, an approach that has successfully been
used to discriminate between electron and hole transfer in
another ZnP/AuP+ system.21 However, for our systems a similar
calculation of δE fails in two ways. The first is that δE for
ZnP-AB-AuP+ becomes negative for all solvents, even if a
solvent correction is included in the same manner as for ∆G°
in eq 5. In highly polar solvents, the negative δE is in agreement
with the observed stepwise electron-transfer ZnP f AB,26 but
in the less polar solvents the activation energy for this process
is most likely too high and only direct ZnP f AuP+ electron
transfer can occur. The second failure is that a significant (≈4
cm-1), and thus nonzero, coupling for an infinite energy gap is
indicated when comparing the ratio of electronic couplings with
the ratio of inverse energy gap for BB and NB (VNB/VBB ≈ 1.3
and δEBB/δENB ≈ 1.8).
We believe that both of these problems have their origin in
the fact that δE is calculated from the redox potentials, which
are values for the relaxed species (including solvent relaxation),
and not for the vertical Franck-Condon states, which are
involved in the electron tunneling process. Therefore, we will
assume that the energy of the virtual bridge state involved in a
superexchange mechanism for photoinduced electron transfer
is proportional to the LUMO energy of the bridge. For the
bridging chromophores investigated here, the first transition is
a pure HOMO f LUMO transition and the energy of the
LUMO can in our case be shown to be proportional to the
energy of the first excited state of the bridge. In our case, we
The calculations predict that there will be no electronic
coupling (0.05 cm-1) in the case of OB as the bridging
B
D
therefore suggest δE E00 - E00 to be a better measure of
the energy term in the first-order perturbation formulation of
the superexchange mechanism (eq 8). The δE values for the
(66) Antipas, A.; Dolphin, D.; Gouterman, M.; Johnson, E. C. J. Am.
Chem. Soc. 1978, 100, 7705-7709.
(67) Jamin, M. E.; Iwamoto, R. T. Inorg. Chim. Acta 1978, 27, 135-
143.
(68) Felton, R. H. Primary redox reactions of metalloporphyrins. In The
porphyrins; Dolphin, D., Ed.; Academic Press: New York, 1978; Vol. V,
Chapter 3.
(63) Clayton, A. H. A.; Scholes, G. D.; Ghiggino, K. P.; Paddon-Row,
M. N. J. Phys. Chem. 1996, 100, 10912-10918.
(64) McConnell, H. M. J. Chem. Phys. 1961, 35, 508-515.
(65) Larsson, S. J. Am. Chem. Soc. 1981, 103, 4034-4040.

Porphyrin-Based Donor-Bridge-Acceptor Systems
J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3079
chromophore. For ZnP-OB-AuP+ the donor-bridge energy
gap is 18 000 cm-1 which, according to Figure 7, should give
a measurable electronic coupling for electron transfer of V ≈ 4
cm-1. However, for this to be true, it must be assumed that the
â values of eq 8 are comparable for the OB and π-conjugated
bridges. Since we have shown that ET does not occur in ZnP-
OB-AuP+, this suggests that â is dependent on the type of
bridge. It is perhaps not unexpected that the OB systems do
not behave in the same manner as the other systems, since the
conjugation within the bridge is broken, making the nature of
the first excited state of the bridge quite different.23 OB may
be regarded as a series of bridge segments, instead of one bridge,
and it will therefore belong to a class of bridging chromophores
that is different from the fully conjugated bridges, BB, NB, and
AB. It is interesting to note that this observation is in agreement
with an investigation of triplet energy transfer in similar systems
(ZnP-RB-H2P, RB ) OB, BB, NB). The systems with an OB
bridging chromophore showed a rate constant that was remark-
ably smaller than that of the systems with conjugated bridging
chromophores.24 This qualitative agreement in the mediating
effect of the bridge is in accordance with the proposed similarity
between triplet energy transfer and electron transfer.69,70
through the solution for 30 min. Palladium-catalyzed coupling reactions
were performed under argon and protected from light.
Gold Porphyrin 1. A solution of the free base analogue34 of 1 (50
mg, 58 µmol), [Au(tht)2]BF4 (152 mg, 330 µmol), and 2,6-lutidine (27
µL, 232 µmol) in CHCl3 (10 mL) was heated to reflux for 1 h. The
solvent was evaporated. The residue was dissolved in CH2Cl2 and added
to a chromatographic column (Al2O3, grade III). Residues of the free
base porphyrin were removed by elution with pure CH2Cl2. The gold
porphyrin was collected by elution with 2% MeOH/CH2Cl2. Recrys-
tallization of the gold porphyrin from CH2Cl2/hexane gave 44 mg (66%)
of 1. 1H NMR δ 1.52 (s, 18 H, t-Bu), 1.83 (t, J ) 7.6 Hz, 12 H,
-CH2CH3), 2.57 (s, 6 H, pyrrole-CH3), 2.65 (s, 6 H, pyrrole-CH3),
4.13 (q, J ) 7.6 Hz, 8 H, -CH2CH3), 7.82 (dm, J ) 8.3 Hz, 2 H,
phenyl), 7.85 (d, J ) 1.8 Hz, 2 H, phenyl), 7.94 (t, J ) 1.8 Hz, 1 H,
phenyl), 8.23 (dm, J ) 8.3 Hz, 2 H, phenyl), 10.64 (s, 2 H, meso);
HRMS calcd for C52H59AuN4I [(M - BF4)+] 1063.345, found 1063.343.
Gold Porphyrin 2. The free base analogue34 of 2 was subjected to
gold insertion by using the same procedure as described for 1. Yield
49%. 1H NMR δ 1.53 (s, 18 H, t-Bu), 1.85 (m, 12 H, -CH2CH3), 2.58
(s, 6 H, pyrrole-CH3), 2.72 (s, 6 H, pyrrole-CH3), 4.13 (m, 8 H,
-CH2CH3), 7.54-7.69 (m, 3 H, naphthyl), 7.87 (d, J ) 1.8 Hz, 2 H,
phenyl), 7.95 (t, J ) 1.8 Hz, 1 H, phenyl), 8.05 (m, 1 H, naphthyl),
8.10 (d, J ) 7.5 Hz, naphthyl), 8.16 (m, 4 H, phenyl), 8.51 (m, 1 H,
naphthyl), 10.63 (s, 2 H, meso).
ZnP-BB-AuP+. Pd2dba3‚CHCl3 (4 mg, 4 µmol) and AsPh3 (10
mg, 32 µmol) were added under argon flushing to a deaerated solution
of zinc porphyrin 534 (21 mg, 23 µmol) and gold porphyrin 1 (25 mg,
22 µmol) in 15 mL of CHCl3/Et3N (2:1). The reaction mixture was
stirred at 40 °C overnight. The solvents were evaporated. The crude
product was purified by chromatography on alumina (grade III) eluting
with CH2Cl2 f 2% MeOH/CH2Cl2. The fraction collected with 2%
MeOH/CH2Cl2 was evaporated and triturated with toluene, removing
toluene-soluble material. The residue was recrystallized from CH2Cl2/
Conclusion and Remarks
The D-B-A systems with fully conjugated bridging chro-
mophores that have been described in this work show fast
photoinduced electron transfer at a center-to-center distance of
25 Å. When the conjugation within the bridge is broken, the
quenching rate constant of the donor is dramatically decreased
and the residual quenching corresponds to that expected for
singlet-singlet energy transfer. However, energy transfer makes
a negligible contribution to the deactivation rate in the systems
where electron transfer occurs. In these systems, the electronic
coupling of the donor and acceptor is found to be dependent
on the bridging chromophore and an inverse relationship with
the energy gap between the excited states of the donor and
bridge was found. The coupling constants predicted by quantum
mechanical calculations showed the same behavior, but were
smaller than those determined from the experimental results.
These calculations also predict a decrease in coupling with
decreasing temperature, and measurements to investigate this
are in progress.
1
hexane giving 11 mg (26%) of ZnP-BB-AuP+. H NMR δ 1.53 (s,
18 H, t-Bu), 1.54 (s, 18 H, t-Bu), 1.77 (m, 12 H, -CH2CH3), 1.84 (m,
12 H, -CH2CH3), 2.46 (s, 6 H, pyrrole-CH3), 2.50 (s, 6 H, pyrrole-
CH3), 2.59 (s, 6 H, pyrrole-CH3), 2.65 (s, 6 H, pyrrole-CH3), 4.03 (m,
16 H, -CH2CH3), 7.81 (s, 4 H, phenyl), 7.83 (t, J ) 1.8 Hz, 1 H,
phenyl), 7.88 (d, J ) 1.8 Hz, 2 H, phenyl), 7.94-8.10 (m, 11 H,
phenyl), 10.13 (s, 2 H, meso), 10.56 (s, 2 H, meso); FAB-MS calcd
for C114H122AuN8Zn [(M - BF4)+] 1863.88, found 1863.88.
ZnP-NB-AuP+. Pd(PPh3)4 (3 mg, 3 µmol), ZnCl2 (4 mg, 15 µmol),
and NaI (1 mg, 7 µmol) were added under argon flushing to a deaerated
solution of zinc porphyrin 434 (9 mg, 11 µmol) and gold porphyrin 2
(7 mg, 5 µmol) in 6 mL of piperidine. The reaction mixture was stirred
at 40 °C overnight and then poured into CH2Cl2 (20 mL). The organic
phase was washed with saturated aqueous NH4Cl, and the aqueous phase
was extracted with CH2Cl2 (20 mL). The combined organic layers were
dried over Na2SO4 and evaporated. The crude product was purified by
chromatography on Al2O3 (grade III) eluting with CH2Cl2 f 2% MeOH/
CH2Cl2. The fraction collected with 2% MeOH/CH2Cl2 was evaporated
and triturated with toluene, removing toluene-soluble material. The
residue was recrystallized from CH2Cl2/hexane giving 3 mg (30%) of
ZnP-NB-AuP+. 1H NMR δ 1.53 (s, 18 H, t-Bu), 1.54 (s, 18 H, t-Bu),
1.79 (m, 12 H, -CH2CH3), 1.86 (m, 12 H, -CH2CH3), 2.46 (s, 6 H,
pyrrole-CH3), 2.58 (s, 6 H, pyrrole-CH3), 2.60 (s, 6 H, pyrrole-CH3),
2.72 (s, 6 H, pyrrole-CH3), 4.02 (m, 8 H, -CH2CH3), 4.13 (m, 8 H,
-CH2CH3), 7.82-7.87 (m, 3 H, 2 naphthyl + 1 phenyl), 7.89 (d, J )
1.8 Hz, 2 H, phenyl), 7.95 (m, 3 H, phenyl), 8.03 (s, 2 H, naphthyl),
8.11 (d, J ) 8 Hz, phenyl), 8.16 (m, 4 H, phenyl), 8.22 (d, J ) 8 Hz,
phenyl), 8.74 (m, 1 H, naphthyl), 8.78 (m, 1 H, naphthyl), 10.19 (s, 2
H, meso), 10.62 (s, 2 H, meso); FAB-MS calcd for C118H124AuN8Zn
[(M - BF4)+] 1913.89, found 1913.90.
Experimental Section
Synthesis. (a) Materials. Triethylamine was dried by distillation
from calcium hydride under nitrogen and used immediately after
distillation. Commercially available reagents were purchased from
Aldrich and used without further purification. The syntheses of the zinc
porphyrins 4-6 (Scheme 2), ZnP and ZnP-RB, have been described
elsewhere.23,34 AuP+ was synthesized as described by Chambron et al.36
(b) Methods. Column chromatography of the gold porphyrins and
ZnP-RB-AuP+ systems was performed with aluminum oxide (acti-
vated, neutral, approximate 150 mesh) deactivated by addition of water
to Brockmann grade III.71 Proton (400 MHz) NMR spectra were
recorded at room temperature in CDCl3, using a Varian UNITY-400
NMR spectrometer. Chemical shifts are reported relative to tetra-
methylsilane (δH 0 ppm). Mass spectra were recorded with a VG
ZabSpec instrument. The substances were analyzed by positive FAB-
MS (matrix 3-nitrobenzyl alcohol) and high-resolution FAB-MS
(HRMS) was performed with PEG 1000 as an internal standard.
Deaeration of reaction mixtures was achieved by bubbling argon
ZnP-OB-AuP+. ZnP-OB-AuP+ was prepared in 13% yield from
1 and 6 by using the procedure described for ZnP-NB-AuP+ but
employing a higher reaction temperature (80 °C) and a longer reaction
1
time (6 days). H NMR δ 1.52 (s, 18 H, t-Bu), 1.53 (s, 18 H, t-Bu),
(69) Closs, G. L.; Piotrowiak, P.; MacInnis, J. M.; Fleming, G. R. J.
Am. Chem. Soc. 1988, 110, 2652-2653.
1.78 (m, 12 H, -CH2CH3), 1.83 (m, 12 H, -CH2CH3), 2.20 (s, 12 H,
-CH2CH2-), 2.45 (s, 6 H, pyrrole-CH3), 2.50 (s, 6 H, pyrrole-CH3),
2.58 (s, 6 H, pyrrole-CH3), 2.64 (s, 6 H, pyrrole-CH3), 4.01 (m, 8 H,
-CH2CH3), 4.10 (m, 8 H, -CH2CH3), 7.79-8.02 (m, 14 H, phenyl),
(70) Closs, L. C.; Johnson, M. D.; Miller, J. R.; Piotrowiak, P. J. Am.
Chem. Soc. 1989, 111, 3751-3753.
(71) Brockmann, H.; Schodder, H. Chem. Ber. 1941, 74, 73-78.

3080 J. Am. Chem. Soc., Vol. 123, No. 13, 2001
Kilså et al.
10.17 (s, 2 H, meso), 10.58 (s, 2 H, meso). FAB-MS calcd for C116H130
-
solvent).26 The emission spectra of the reference compounds (D-B)
were numerically scaled to take into account the partitioning of the
incident excitation light between the donor and the acceptor in the
D-B-A systems. The quenching efficiency was calculated from the
decrease in the integrated intensity of the donor emission.
AuN8Zn [(M - BF4)+] 1897.94, found 1897.97.
H2P-AB-H2P. A solution of the zinc/free base analogue34 (18 mg)
was dissolved in CH2Cl2 (70 mL). The solution was extracted with 1
M HCl (50 mL), washed with 5% aqueous NaHCO3 (2 × 50 mL), and
dried (Na2SO4). Evaporation of the solvent gave the bis(free base
Fluorescence lifetimes were measured by the phase/modulation
technique on a SPEX Fluorolog τ2 spectrofluorimeter, with a diluted
silica sol scattering solution as the reference. On a logarithmic scale in
the range 10-300 MHz, 20 modulation frequencies were selected and
the emission was collected through a 550 nm cutoff filter. The 528.7
nm line from an argon laser (Spectra Physics 165-08) was used as
excitation source, except for samples containing AB or pyZnP, where
the 514.5 nm line was used. For all the compounds (D-B-A and
D-B), the observed demodulations and phase shifts could be satis-
factorily fitted to a single-exponential model since the acceptor is
nonfluorescent. However, even better fits for the D-B-A systems were
found if a small fraction (5-10%) was fitted to a much longer lifetime,
irrespective of compound. This is most likely due to impurities or
instrumental artifacts. Including this second lifetime did not change
the donor lifetime significantly. The goodness-of-fit was evaluated from
the value of ø2 and by visual inspection of the fit to the data points. In
frequency domain measurements it is not the absolute value of ø2, but
rather the relative change in ø2 with different models, that is used as
an indication of goodness-of-fit.72
Transient absorption kinetic traces on the nanosecond time scale were
measured as described previously on an Applied Photophysics flash
photolysis instrument.25 The pump pulse was the second harmonic of
a Nd:YAG laser (Spectron Laser Systems, SL803G, 532 nm) with a
pulse width of about 10 ns fwhm. The concentration of the samples
was approximately 5 µM. However, in the analysis, differences in donor
absorbance were corrected for. Butyronitrile was used as the solvent,
and all samples were degassed by 5 freeze-pump-thaw cycles to a
final pressure of 10-4 mbar.
Transient absorption spectra on the picosecond time scale were made
by using the pump-probe technique with the setup described previ-
ously.25 Excitation pulses at 585 nm and a 5 kHz repetition rate were
provided by frequency doubling the signal output of an optical
parametric amplifier (Spectra Physics, OPA-800) pumped by a Ti:
Sapphire regenerative amplifier (Positive Light, Spitfire). Transient
absorption spectra were recorded between 470 and 740 nm with a 2
nm step size. Kinetics were determined over an 800 ps time range (the
maximum available with the setup) and thus charge recombination times
could not be reliably determined. The sample was held in a static 1
mm path length cuvette and the concentration was approximately 160
µM in DMF.
1
porphyrin) dimer in a quantitative yield. H NMR δ -2.36 (bs, 4 H,
NH), 1.52 (s, 36 H, t-Bu), 1.82 (m, 24 H, -CH2CH3), 2.48 (s, 12 H,
pyrrole-CH3), 2.67 (s, 12 H, pyrrole-CH3), 4.06 (m, 16 H, -CH2CH3),
7.82 (t, J ) 1.8 Hz, 2 H, phenyl), 7.85 (m, 4 H, anthryl), 7.94 (d, J )
1.8 Hz, 4 H, phenyl), 9.02 (m, 4 H, anthryl), 10.28 (s, 4 H, meso).
FAB-MS calcd for C122H131N8 [(M + H)+] 1709.05, found 1708.88.
H2P-AB-AuP+. [Au(tht)2]BF4 (18 mg, 40 µmol) dissolved in
CHCl3 (5 mL) was added to a solution of H2P-AB-H2P (17 mg, 10
µmol) and 2,6-lutidine (3 µL, 3 µmol) in CHCl3 (5 mL). The reaction
mixture was heated to reflux for 1 h. The solvent was evaporated. The
residue was dissolved in CH2Cl2 and added to a chromatographic
column (Al2O3, grade III). Elution with CH2Cl2 gave unreacted H2P-
AB-H2P (5 mg). Elution with 1% MeOH/CH2Cl2 gave 9 mg (45%)
of H2P-AB-AuP+. The bis-gold analogue (5 mg) was eluted with
1
5% MeOH/CH2Cl2. H NMR δ -2.40 (bs, 2 H, NH), 1.52 (s, 18 H,
t-Bu), 1.54 (s, 18 H, t-Bu), 1.81 (m, 12 H, -CH2CH3), 1.87 (m, 12 H,
-CH2CH3), 2.49 (s, 6 H, pyrrole-CH3), 2.60 (s, 6 H, pyrrole-CH3),
2.66 (s, 6 H, pyrrole-CH3), 2.79 (s, 6 H, pyrrole-CH3), 4.05 (m, 8 H,
-CH2CH3), 4.16 (m, 8 H, -CH2CH3), 7.77 (m, 4 H, anthryl), 7.83 (t,
J ) 1.8 Hz, 1 H, phenyl), 7.90 (d, J ) 1.8 Hz, 2 H, phenyl), 7.94 (d,
J ) 1.8 Hz, 2 H, phenyl), 7.96 (t, J ) 1.8 Hz, 1 H, phenyl), 8.24 (m,
6 H, phenyl), 8.36 (dm, J ) 8 Hz, phenyl), 8.88 (m, 4 H, anthryl),
10.26 (s, 2 H, meso), 10.65 (s, 2 H, meso). FAB-MS calcd for C122H128
-
AuN8 [(M - BF4)+] 1903.00, found 1903.01.
ZnP-AB-AuP+. Zn(OAc)2‚H2O (10 mg, 46 µmol) dissolved in
MeOH (1 mL) was added to a solution of H2P-AB-AuP+ (9 mg, 4.5
µmol) in CHCl3 (5 mL). The reaction mixture was stirred at room
temperature for 2.5 h and diluted with CHCl3 (30 mL). The organic
layer was washed with 5% aqueous NaHCO3 (2 × 30 mL) and water
(30 mL), dried (Na2SO4), and evaporated. The crude product was
purified by chromatography (Al2O3, grade III) eluting with CH2Cl2 f
4% MeOH/CH2Cl2. The fraction collected after addition of MeOH to
the eluent was recrystallized from CH2Cl2/hexane giving 7 mg (76%)
1
of ZnP-AB-AuP+. H NMR δ 1.53 (s, 18 H, t-Bu), 1.55 (s, 18 H,
t-Bu), 1.80 (m, 12 H, -CH2CH3), 1.87 (m, 12 H, -CH2CH3), 2.46 (s,
6 H, pyrrole-CH3), 2.60 (s, 6 H, pyrrole-CH3), 2.63 (s, 6 H, pyrrole-
CH3), 2.78 (s, 6 H, pyrrole-CH3), 4.03 (m, 8 H, -CH2CH3), 4.14 (m,
8 H, -CH2CH3), 7.80 (m, 4 H, anthryl), 7.83 (t, J ) 1.8 Hz, 1 H,
phenyl), 7.90 (d, J ) 1.8 Hz, 2 H, phenyl), 7.96 (m, 3 H, phenyl), 8.22
(m, 6 H, phenyl), 8.35 (dm, J ) 8 Hz, phenyl), 8.92 (m, 4 H, anthryl),
Cyclic Voltammetry. The redox potentials of AuP+, BB, and NB
were measured in a similar way as described earlier for AB, ZnP, and
pyZnP by using a PAR 273A potentiostat with a platinum working
electrode, a platinum coil counter electrode, and a double junction Ag/
Ag+ reference electrode (E ) 0.45 V vs SHE).26 The chromophores
were dissolved in CH2Cl2 to a concentration of 1 mM (AuP+) or 2
mM (BB and NB), with 0.1 M (Bu)4NBF4 as supporting electrolyte.
The solutions were N2-purged before measurement and the sweep rates
varied between 0.1 and 1 V/s.
10.20 (s, 2 H, meso), 10.64 (s, 2 H, meso). FAB-MS calcd for C122H126
-
AuN8Zn [(M - BF4)+] 1965.91, found 1965.88.
Spectroscopic Measurements. All solvents (CH2Cl2, N,N-dimethyl-
formamide (DMF) (from LabScan), CHCl3, butyronitrile (from Merck),
and 2-methyltetrahydrofuran (2-MeTHF) (from Acros)) were used as
purchased. All measurements were made at 20 °C. Pyridine (from
LabScan) was distilled before use. Samples containing pyZnP were
prepared immediately before spectroscopic measurements by adding
pyridine to the corresponding ZnP solution. At a final pyridine
concentration of approximately 2 M full conversion of ZnP to pyZnP
was obtained, as judged from the changes in the absorption spectra.26
Pyridine does not coordinate with AuP+.
Absorption spectra were recorded on a Cary 4 Bio spectrophotometer.
Fully corrected steady-state emission spectra were recorded on a SPEX
Fluorolog τ2 spectrofluorimeter. Quantum yields were determined
relative to ZnP in chloroform (0.025).23,26 The concentration of the
chromophores was kept at approximately 5 µM to exclude the possibility
of intermolecular interactions and to ensure that inner filter effects were
negligible. To facilitate immediate comparison of the emission spectra
of the D-B-A systems, the optical densities of the samples were
matched at the excitation wavelength (537-550 nm, dependent upon
Acknowledgment. This work was supported by grants from
the Swedish Natural Science Research Council (NFR) and the
Swedish Research Council for Engineering Sciences (TFR).
K.K. is grateful to the Danish Research Agency in Århus for a
fellowship. We thank Dr. Gunnar Stenhagen for help concerning
mass spectrometry and Prof. Elisabeth Ahlberg for help with
the electrochemical measurements.
JA003820K
(72) Lakowicz, J. R. Principles of fluorescence spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999.