Artificial photosynthetic reaction centers with porphyrins as primary electron acceptors

10566

J. Phys. Chem. B 2004, 108, 10566-10580

Artificial Photosynthetic Reaction Centers with Porphyrins as Primary Electron Acceptors^†

Stephanie L. Gould, Gerdenis Kodis, Rodrigo E. Palacios, Linda de la Garza, Alicia Brune,

Devens Gust,* Thomas A. Moore,* and Ana L. Moore*

The Center for the Study of Early EVents in Photosynthesis, Department of Chemistry and Biochemistry,

Arizona State UniVersity, Tempe, Arizona 85287-1604

ReceiVed: February 17, 2004; In Final Form: April 6, 2004

A triad consisting of a carotenoid (C), a dimesitylporphyrin (P), and a tris(heptafluoropropyl)porphyrin (P

C-P-P , has been synthesized and found to undergo rapid singlet-singlet energy transfer between the

porphyrin moieties so that their excited states are in equilibrium. Photoinduced electron transfer from the

F

),

F

•

+

•-

first excited singlet state of P, or hole transfer from the first excited singlet state of P

F

, yields C-P -P

with a quantum yield of

.73 and a lifetime of 500 ns in tetrahydrofuran solution at ambient temperature. The final charge-separated

F

.

•

+

•-

Electron transfer from C then yields the final charge-separated state C -P-P

F

0

3

F

state decays to form primarily a triplet excited state localized on the carotenoid, C-P-P , rather than the

ground state. A second triad in which P is metalated (PZn) has also been synthesized. In this system, the

excited singlet states of the porphyrins are no longer in equilibrium; fast electron transfer from excited PZn

•

+

•-

to P

F

to form C-PZn -P

F

and also fast energy transfer from PZn to P with subsequent hole transfer from

•

+

•-

•+

•-

P

F

to PZn converge to give the same C-PZn -P

yield of 0.14. This state decays to C-PZn-P

temperature. The charge recombination reaction follows a single-step mechanism for C -P-P

temperature to 77 K and for C -PZn-P

the recombination reaction of C -PZn-P

pathway involves an endergonic step to populate C-PZn -P

photophysical characteristics of these triads, in particular the recombination to the triplet state, are reminiscent

of those of artificial reaction centers with C60 as the primary electron acceptor.

F

species, which evolves to C -PZn-P

F

with a quantum

3

F

with a quantum yield of 0.06 in tetrahydrofuran at room

•+

•-

F

from room

below 250 K. Above 250 K, a two-step pathway is accessed for

•

+

•-

F

•

+

•-

F

in addition to the direct recombination mechanism. This new

•

+

•-

F

having an E

a

of ∼0.23 eV. Certain

Introduction

fullerene systems to occur at temperatures as low as 8 K.⁶^,¹⁸

The early steps of photosynthetic electron transfer can also occur

at very low temperatures.

Ensembles of covalently linked donors and acceptors have

been used successfully to model the stepwise electron transfer

process observed in the reaction centers of photosynthetic

organisms. In early reaction center models, quinones were used

extensively as primary electron acceptors.¹^-³More recently

fullerene derivatives emerged as appealing alternatives to

quinones as electron acceptors.⁴One of the attractive features

of C60 derivatives in photoinduced electron-transfer reactions

In our mimicry of photosynthesis, we have used energetic

charge-separated states generated with artificial reaction centers

inserted vectorially into model lipid bilayers to drive the

1

9-21

transport of ions across such membranes.

For our initial

investigations we employed artificial reaction centers (triads)

with quinones as the primary electron acceptors. Because of

their involved synthesis, inherent instability, and poor yield of

the final charge separation in low dielectric media, we designed

alternative reaction center mimics to use in model membranes.

-7

6,8-12

is their small total reorganization energy (λ).

to the Marcus-Hush treatment, λ is one of the key parameters

controlling the dynamics of electron-transfer processes.

According

13-16

Low

As mentioned above, C60-based constructs are in principle

reorganization energy results in faster forward rates at lower

driving force, which a priori shift charge recombination further

into the “inverted region” and thereby slow this energy-wasting

step.

In addition, the energy of the fullerene radical anion is less

affected by solvent dielectric properties and mobility than are

those of the anions of moieties such as quinones. Thus, the

thermodynamic driving force for electron-transfer reactions

involving fullerenes (-∆G°, another key parameter affecting

the rate of electron transfer) does not change as drastically as

appealing options. However, to date we and others²

2,23

have

found that the lipid aggregation/insolubility of C60 derivatives

limits their use in membranes.

Porphyrins are desirable alternatives to quinones and even

C60 as electron acceptors. Indeed, cyclic tetrapyrroles are the

initial electron acceptors in all types of reaction centers in natural

photosynthesis. Previously, we have reported the dependence

of electron-transfer rates in porphyrin dyads on structure,

2

4,25

thermodynamic driving force, and solvent.

Although there

are structural reasons to expect that the reorganization energies

of porphyrins could be larger than that of C60, they are also

expected to be considerably lower than those of quinones. Here,

we report the synthesis and spectroscopic study of triads 1 and

1

1,17

the solvent is made less polar or frozen into a glassy state.

This allows photoinduced electron transfer in some porphyrin-

†

Part of the special issue “Gerald Small Festschrift”.

1

′ in which porphyrins are the primary electron donors and

*

Corresponding authors. E-mail: gust@asu.edu (D.G.), tmoore@asu.edu

(T.A.M.), amoore@asu.edu (A.L.M).

acceptors; 1 and 1′ are models for amphipathic triads suitable

1

Published on Web 06/08/2004

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10567

for insertion into membranes. Triad 1 consists of a carotenoid

moiety (C) linked to a dimesityl free base porphyrin (P), which

is connected in series to a fluorinated free base porphyrin (PF);

a carboxylic acid attached to PF would render the triad

amphipathic. In 1′, P is replaced by a dimesityl zinc porphyrin

showed an oxidation at 1.59 V and a reduction at -0.63 V.

Model porphyrin 3′ exhibited an oxidation at 0.76 V and a

reduction at -1.34 V.

Triad 1 was found to exhibit a first oxidation potential of

0.52 V, which corresponds to the carotenoid oxidation,³⁰

a

(PZn). As described below, excitation of the free base porphyrin

second oxidation potential of 1.06 V, which corresponds to

oxidation of the free base dimesitylporphyrin moiety, a first

reduction potential of -0.62 V, which corresponds to the

fluorinated free base porphyrin, and a second reduction potential

of -1.18 V, which corresponds to the first reduction of the

dimesityl free base porphyrin. The potentials measured for the

triad are very similar to those measured for model porphyrins,

indicating that linking the moieties in the triad does little to

perturb the individual redox centers.

and/or the fluorinated porphyrin in C-P-PF in tetrahydrofuran

solution is followed by fast singlet-singlet energy exchange

between the porphyrins and slower electron and/or hole transfer,

•

+

•-

forming the initial charge-separated state C-P -PF . This

charge-separated state evolves into a long-lived charge-separated

•

+

•-

state, C -P-PF , which finally decays, forming mostly a

3

triplet excited-state localized on the carotenoid, C-P-PF, rather

than the ground state. Triad 1′ shows fast direct electron transfer

1

•+

•-

from PZn to PF to form C-PZn -PF and also fast energy

transfer from PZn to PF with subsequent hole transfer from PF

to PZn to converge to the same C-PZn -PF , which finally

evolves to C -PZn-PF . As discussed below, these photo-

chemical characteristics, in particular the recombination to the

triplet state, match quite closely those of triads with C60 as the

primary electron acceptor.⁶

•+

•-

•+

•-

•+

The energies of the C-P -PF , C -P -PF, C -P-

1

•-

•+

•-

•+

•-

•+

•-

PF , C-PZn -PF , C -PZn -PF, and C -PZn-PF charge-

separated states were estimated as 1.68, 1.70, 1.14, 1.38, 1.86,

and 1.14 eV, respectively, based on the electrochemically

determined oxidation and reduction potentials of porphyrins and

carotenoid in triad 1 and model porphyrin 3′.

•

+

•-

•

+

•-

,11,17,18,26

Absorption Spectra. The absorption spectrum of 1 in

tetrahydrofuran is shown in Figure 1a. Absorption maxima are

at 647, 593, 548, 509, 476, 421, 405(sh), 375(sh), and 300 nm.

For comparison, the spectra of model compounds 2, 3, and 4

are also shown. Absorption maxima for the dimesityl free base

porphyrin 3 and fluorinated free base porphyrin 4 are at 650,

Results

Synthesis. The required dimesityl porphyrin, 5,15-bis(4-carbo-

methoxyphenyl)-10,20-bis(2,4,6-trimethylphenyl)porphyrin (7),

2

7,28

and apocarotenoid were prepared by published procedures.

5

4

94, 550, 515, 419, 402(sh), 300 nm, and 644, 591, 546, 511,

10, and 375(sh) nm, respectively. The absorption spectrum of

Hydrolysis of one of the ester groups of 7 under acid conditions,

followed by a Curtius rearrangement and deprotection of the

remaining ester group, afforded the amino acid porphyrin (11),

which is the central building block of triad 1. The dipyrro-

methane required for the synthesis of the fluorinated porphyrin

was obtained by reaction of pyrrole and heptafluorobutaralde-

hyde in tetrahydrofuran with catalytic amounts of hydrochloric

model dyad 2 is essentially a linear combination of the

absorption spectra for 3 and 4, with maxima at 647, 593, 549,

514, 420, 405(sh), 375(sh), and 300 nm. The difference between

the absorption spectra of 1 and 2 corresponds to the absorption

due to the carotenoid in the triad.

The absorption spectrum of 1′ in tetrahydrofuran is shown

in Figure 1b. Absorption maxima are at 645, 598, 559, 509,

acid. Condensation of this dipyrromethane with heptafluoro-

_b_u_t_a_r_a_l_d_e_h_y_d_e_a_n_d₄_-_f_o_r_m_y_l_m_e_t_h_y_l_b_e_n_z_o_a_t_e₂9 resulted in three

4

76, 428, 405(sh), 375(sh), and 310 nm. For comparison, the

porphyrins, including 5-(4-carbomethyoxyphenyl)-10,15,20-tris-

spectra of model compounds 2′, 3′, and 4 are also shown.

Absorption maxima for the dimesityl zinc porphyrin 3′ are at

(heptafluoropropyl)porphyrin, 4. Because of the sensitivity of

the fluorinated porphyrin to bases, the hydrolysis of the methyl

ester of 4 had to be carried out under acidic conditions, with

trifluoroacetic acid/hydrochloric acid. Dyad 2 was obtained by

coupling the acid chloride of the fluorinated porphyrin with the

required dimesityl porphyrin. Similar coupling techniques were

employed in the syntheses of triad 1 and model dyad 5. Zinc

compounds 1′ and 2′ were synthesized by adding an excess of

zinc acetate in a methanol/dichloromethane solution to a

dichloromethane solution of the triad/dyad with careful monitor-

ing; when the dimetalated product started to form, the reaction

mixture was immediately chromatographed. Compounds 3′ and

6

00, 560, 522, 428, 408(sh), and 312 nm. A linear combination

of the absorption spectra for 3′ and 4 is virtually indistinguish-

able from the absorption spectrum of model dyad 2′ with max-

ima at 645, 598, 559, 516, 428, 405(sh), 375(sh), and 312 nm.

Fluorescence Emission. The fluorescence emission spectra

of 2, 3, and 4 with the same absorbance at the excitation

wavelength of 520 nm, in tetrahydrofuran, are shown in Figure

2

a. The emission spectra feature maxima for the dimesitylpor-

phyrin at 650 and 718 nm and for the fluorinated porphyrin at

6

7

46 and 713 nm. The fluorescence of 2 has maxima at 649 and

17 nm.

5

′ were prepared simply by treating the porphyrin or the

carotenoporphyrin with an excess of zinc acetate and allowing

the reaction to proceed until the metalation of the porphyrin

was completed.

The fluorescence emission spectra of 2′, 3′, and 4 dissolved

in tetrahydrofuran and having the same absorbance at the

excitation wavelength of 580 nm are shown in Figure 2b. The

emission spectra feature maxima for the dimesityl zinc porphyrin

at 604 and 655 nm. The emission of 2′, with maxima at 651

and 713 nm, is reduced compared to that of 3′ and 4, suggesting

that the excited singlet states of the porphyrin moieties in 2′

are quenched by some photochemical process(es).

Cyclic Voltammetry. To estimate the energies of the charge-

separated states, electrochemical measurements were undertaken

to investigate the redox properties of triad 1 and model

compounds 3, 3′, and 4. The measurements were performed in

benzonitrile containing 0.1 M tetra-n-butylammonium hexa-

fluorophosphate and ferrocene as an internal standard (oxidation

at 0.458 V vs SCE). The data below are reported vs SCE. The

oxidation and reduction processes of porphyrins were quasi-

reversible, whereas the carotenoid oxidation was irreversible

and the peak potential is given in this case. Free base

dimesitylporphyrin 3 featured oxidation waves at 1.03 and 1.32

V and reduction at -1.19 V. Free base fluorinated porphyrin 4

Time-Resolved Fluorescence. To learn more about the nature

of the fluorescence quenching, time-resolved fluorescence

experiments were undertaken, at room temperature, using the

single-photon-timing technique. A tetrahydrofuran solution of

model dyad 2 was excited at 625 nm, and the fluorescence decay

was measured at seven wavelengths in the 600-800 nm region.

2

Global analysis of these data (ø ) 1.15) yielded two significant

10568 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

CHART 1

decay components with lifetimes of 12 and 890 ps (the relative

amplitudes were wavelength dependent (60% and 35%) at the

fluorescence maximum, respectively). Additional components

with lifetimes of 0.12 and 11.5 ns (e5% of the decay) were

also required to fit the data; those were ascribed to minor

impurities. Similar results were obtained with triad 1, with two

2

major decay components of 12 and 785 ps (ø ) 1.19) and minor

components of 0.14 and 11.5 ns (<5% of the decay), most likely

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10569

a

TABLE 1: Excited Singlet State Lifetimes of Porphyrin

Moieties As Determined by Time-Resolved Fluorescence

Studies^b

compd

τ

1

(ns)

2

τ (ns)

1

2

3

4

5

0.012

11.50

2.06

0.785

0.890

c

3.27 (0.770)

2′

3′

5′

0.013 (0.049)^d

2.27

0.040 (2.52)^d

0.590 (0.370)^c

a

All lifetimes were determined in tetrahydrofuran unless otherwise

noted. ^bIn some cases, minor decay components were also observed

see text for details). ^cIn benzonitrile. ^dIn n-hexane.

(

ns (ø²) 1.19) and 770 ps (additional 2.2 and 12.2 ns

2

components <8% of the decay were required, ø ) 1.06),

respectively. The spectra of the major decay components are

consistent with the steady-state fluorescence spectra of the model

porphyrins. The significant shortening of the lifetime of the

singlet states of the porphyrins in dyad 2 relative to that of the

corresponding states in the model porphyrins is ascribed to

excited-state energy exchange and photoinduced electron transfer

Figure 1. (a) Absorption spectra in tetrahydrofuran of triad 1 (;;),

model dyad 2 (---), model porphyrin 3 (‚‚‚‚‚), and model porphyrin 4

(vide infra).

(

-‚-‚-). (b) Absorption spectra in tetrahydrofuran of triad 1′ (;;),

Tetrahydrofuran and n-hexane solutions of model dyad 2′

were excited at 590 nm, and the fluorescence decay was

model dyad 2′ (---), model porphyrin 3′ (‚‚‚‚‚), and model porphyrin 4

-‚-‚-). The insets are expansions of the Q-band absorption region.

(

measured at three wavelengths in the 600-800 nm region.

2

Global analysis of the data (ø ) 1.17) yielded two significant

decay components with lifetimes of 13 and 40 ps for the

tetrahydrofuran solution (the relative amplitudes were wave-

length dependent, at 600 nm 99% of 13 ps and at 720 nm 25%

of 13 ps and 75% of 40 ps) and 49 ps and 2.52 ns for the

n-hexane solution (the relative amplitudes were wavelength

dependent, at 600 nm 98% of 49 ps and at 720 nm 7% (negative

amplitude) of 49 ps and 90% of 2.52 ns). Additional components

with amplitude less than 3% were also required to fit the data;

those were ascribed to minor impurities. In both solvents, the

major decay component at 600 nm corresponds to emission from

the zinc porphyrin, while the major decay component at 720

nm corresponds to emission from the free base fluorinated

porphyrin.

Model porphyrin 3′ and model dyad 5′ in tetrahydrofuran

2

showed fluorescence lifetimes of 2.27 ns (ø ) 1.05) and 590

ps (additional 1.95 ns component <9% of the decay was

2

required, ø ) 1.00), respectively. The fluorescence lifetime of

dyad 5′ in benzonitrile is quenched to 370 ps (additional 0.9

and 2.1 ns components <10% of the decay were required, ø²

)

1.10). It is well-known that the attachment of carotenoid

moieties to porphyrins quenches the porphyrin singlet lifetime,

possibly due to singlet energy and/or electron transfer between

the chromophores (vide infra).

Transient fluorescence data for 1 and model compounds are

compiled in Table 1.

Figure 2. (a) Fluorescence emission spectra in tetrahydrofuran solution

of dyad 2 (;;), model porphyrin 3 (‚‚‚‚‚), and model porphyrin 4

Time-Resolved Absorption. Transient absorption experi-

ments were undertaken in order to characterize the various

excited states in these compounds and to allow detection of

nonemissive species such as charge-separated states.

Solutions of 1, 1′, 2, 2′, 5, and 5′ in tetrahydrofuran were

excited with ∼100 fs laser pulses, and the transient absorptions

were recorded using the pump-probe method. The spectra were

measured in the 930-1070 nm and 450-760 nm regions. A

total of 70 and 155 kinetic traces were recorded over those

wavelength regions, respectively, at times ranging from -50

to 4500 ps relative to the laser flash. The data were fitted

(

---) with excitation at 520 nm. The fluorescence of 4 has been

multiplied by 0.2. (b) Fluorescence emission spectra in tetrahydrofuran

solution of dyad 2′ (;;), model porphyrin 3′ (‚‚‚‚‚), and model

porphyrin 4 (---) with excitation at 580 nm. The fluorescence of 2′ has

been multiplied by 20.

due to impurities. The fluorescence lifetimes for model por-

phyrins 4 and 3, in tetrahydrofuran solution, were found to be

2

.06 ns (ø ) 1.09) and 11.50 ns (ø ) 1.13), respectively. The

fluorescence lifetimes of the dimesitylporphyrin in model dyad

, in tetrahydrofuran and benzonitrile, were found to be 3.27

3

1,32

5

globally using singular value decomposition.

10570 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

Figure 3. (a) Decay-associated spectra for 2 in tetrahydrofuran obtained

from a global analysis of the transient absorption data taken after

excitation at 550 nm with a ∼100 fs laser pulse. The lifetimes of the

components are 11.5 ps (4), 890 ps (9), and ∼4.2 ns (O). (b) Decay-

associated spectra for 2′ in tetrahydrofuran obtained from a global

analysis of the transient absorption data taken after excitation at 645

nm. The lifetimes of the components are 40 ps (0) and 70 ps (b).

Figure 4. (a) Transient absorption kinetics at 980 nm measured for

tetrahydrofuran solutions of triads 1 (b) and 1′ (0) with excitation by

a ∼100 fs laser pulse at 590 and 645 nm, respectively. A two-

exponential fit to the data of triad 1 with time constants of 820 ps and

a fixed, nondecaying component is shown as a smooth curve. A three-

exponential fit to the data of triad 1′ with time constants of 40, 60 ps

and a fixed, nondecaying component is shown as a smooth curve

(arrows indicate corresponding time axes). The inset shows the transient

absorption spectrum measured 4 ns after excitation of triad 1′ at 645

nm. (b) Transient absorption kinetics at 980 nm measured for a

benzonitrile solution of dyad 5 with excitation at 600 nm. A two-

exponential fit to the data of dyad 5 with time constants of 7 and 770

ps is shown as a smooth curve. The inset shows the transient absorption

spectrum measured 50 ps after excitation.

•+

•-

Verification of the formation of a P -PF charge-separated

state comes from the spectroscopic investigation of dyad 2 with

excitation at 550 nm. Figure 3a shows the decay-associated

spectra of dyad 2 obtained by a global analysis of the data. The

best fit was achieved with three components having lifetimes

33

of 11.5 ps, 890 ps, and ∼4.2 ns. The picosecond components

have characteristic porphyrin ground-state bleaching at ∼510,

5

50, 590, and 650 nm and stimulated emission bands at ∼650

and 720 nm. They are consistent with an equilibration process

mediated by singlet-singlet energy transfer between the two

•

+

•-

porphyrins, and subsequent formation of the P -PF charge

maximum at ∼980 nm, which is ascribed to the carotenoid

radical cation. When the spectrum is monitored at 980 nm, the

carotenoid radical cation absorption is found to appear with an

820 ps time constant. The carotenoid radical cation absorption

band does not decay over a period of 4.5 ns, which is the longest

time scale available on the spectrometer.

separated state. The nanosecond component shows spectral

34,35

characteristics of the porphyrin radical cation and anion

with

ground-state bleaching bands; stimulated emission was not

observed on the nanosecond time scale.

•

+

•-

Formation of the PZn -PF charge-separated state was

corroborated by spectroscopic investigation of dyad 2′ with

excitation at 645 nm. This light selectively excites PF of dyad

Figure 4a (squares) shows the results obtained from excitation

of triad 1′ in tetrahydrofuran solution with a 645 nm laser pulse.

This light, as mentioned before, is absorbed only by the

fluorinated porphyrin moiety. Over time, in this case also, a

new transient species grows in with an absorbance maximum

at ∼980 nm (inset Figure 4a), which is ascribed to the carotenoid

radical cation. When the spectrum is monitored at 980 nm, the

carotenoid radical cation absorption is found to grow in (negative

amplitude) with a 60 ps time constant. An additional time

constant of 40 ps (positive amplitude) due to decay of the

fluorinated porphyrin singlet excited state (vide infra) also was

needed to fit the kinetic results. The carotenoid radical cation

absorption band does not decay on the longest time scale

available to the spectrometer (4.5 ns).

2

′. Figure 3b shows the decay-associated spectra of dyad 2′

obtained by a global analysis of the data. The best fit was

obtained with two components, 40 and 70 ps. The spectra are

consistent with the decay of the initially formed PF singlet

•

+

•-

excited state in 40 ps and formation of the PZn -PF charge-

separated state, which consequently decays to the ground state

with a 70 ps lifetime. The 70 ps component clearly shows a

•

+

broad induced absorption band at ∼650 nm associated with PZn

and a ground-state bleaching band at 560 nm corresponding to

PZn.

Figure 4a (circles) presents the kinetics obtained at 980 nm

by excitation of triad 1 in solution with a 590 nm laser pulse.

This light is absorbed mainly by the porphyrin moieties. Over

time, a new transient species grows in with an absorbance

The formation of C -P -PF and C -PZn^•^--PF charge-

•

+

•-

•+

separated states or singlet energy transfer from the porphyrin

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10571

to the carotenoid moiety as quenching mechanisms for P or PZn

excited singlet states was investigated by studying model dyads

5

and 5′. No evidence of either process was observed in

tetrahydrofuran solution. Nevertheless, the formation of short-

•

+

•-

•+

•-

lived C -P or C -PZn charged-separated states cannot

be ruled out, because the formation of such states is energetically

favorable in both cases and the fluorescence lifetime of the

porphyrin is much more quenched in polar solvents, suggestive

36

of electron-transfer processes. It is possible that slow formation

•

+

•-

•+

•-

and a very short lifetime of the C -P or C -PZn charge-

separated states precludes the build up of sufficient concentration

to be observed by transient absorption spectroscopy. An

alternative explanation for the porphryrin fluorescence lifetime

quenching would be energy transfer to the carotenoid or an

increase of the rate constant for internal conversion to the ground

state and/or intersystem crossing, caused by the carotenoid

moiety.

To investigate electron transfer as a mechanism for the

fluorescence quenching of the porphyrin, dyads 5 and 5′ were

studied in the polar solvent benzonitrile. Figure 4b (diamonds)

presents the kinetics obtained at 980 nm by excitation of dyad

5

with a 650 nm laser pulse (a similar result was observed with

dyad 5′ and excitation at 605 nm). This light was absorbed

mainly by the porphyrin moiety. A new transient species grows

in with an absorbance maximum at ∼980 nm (inset in Figure

Figure 5. Normalized transient absorption kinetics measured for triads

1 (0) and 1′ (O) in deoxygenated tetrahydrofuran (room temperature)

at 980 nm (a) and 540 nm (b), following excitation with a 5 ns, 590

nm laser pulse. In part (a), a two-exponential fit to the data is shown

as a smooth curve with time constants of 500 ns (triad 1) and 210 ns

•

+

4

b), which is ascribed to the carotenoid radical cation of C -

•-

P . When the spectrum was monitored at 980 nm, the

carotenoid radical cation absorption was found to rise in with

a 7 ps time constant and decay over a period of 770 ps. This

observation is in agreement with time-resolved fluorescence

results for 5 provided that the rise of the charge-separated state

(triad 1′) (in each case a minor component, <5%, of tens of

microseconds was used). In part (b), a two-component fit to the data is

shown as a smooth curve with time constants for the rise of 500 ns

(triad 1) and 210 ns (triad 1′) and a decay time constant of 4.9 µs in

both cases. The inset in part (a) shows the transient absorption spectra

of 1 (1′ is similar) 45 ns after the laser flash in 2-methyltetrahydrofuran

at different temperatures (the carotenoid radical cation maximum is at

(770 ps) is slower than its decay (7 ps). In the case of the

metalated dyad 5′, the signal at 980 nm rises with a 20 ps time

constant and decays in 370 ps. This observation illustrates that

quenching of the porphyrin excited state in porphyrin-caro-

tenoid dyads 5 and 5′ can be explained by the formation of a

very short lived charge-separated state.

∼

980 nm): 300 K (9), 200 K (b), 105 K (2), and 77 K (1). The

inset in part (b) shows the transient absorption spectrum taken 900 ns

after the laser pulse (at room temperature), which is characteristic of

the carotenoid triplet excited state (maximum at ∼540 nm) of 1 (1′ is

similar).

Excitation of deoxygenated samples of 1 and 1′ in tetrahy-

drofuran with a ∼5 ns laser pulse at 590 nm permitted further

investigations of the fate of the long-lived charge-separated state.

Figure 5 presents the kinetics and spectra (see insets) of the

carotenoid radical cation (a) and the carotenoid triplet excited

state with a maximum at ∼540 nm (b). An exponential fit to

the kinetics at 980 nm yielded time constants of 500 ns and

Low-Temperature Measurements. Excitation of deoxygen-

ated samples of 1 (Figure 6a) and 1′ (Figure 6b) in 2-methyl-

tetrahydrofuran at temperatures between 300 and 77 K with a

∼

5 ns laser pulse at 532 nm permitted investigation of the fate

of the long-lived charge-separated state at low temperatures.

For both triads, the decay kinetics are essentially single

2

10 ns for triads 1 and 1′, respectively. A minor (<5%) long

exponential above ∼120 K (gray circles in Figure 6). From room

•+

component of tens of microseconds was also required to fit the

data. This minor decay component was sensitive to the presence

of oxygen and therefore can be attributed to triplet states.

Exponential fits to the kinetics at 540 nm for triads 1 and 1′

gave the corresponding 500 and 210 ns rise times, respectively,

and a 4.9 µs decay component in both cases. Thus, the charge

temperature to ∼250 K, charge recombination in C -PZn-

•-

PF is an activated process with an Arrhenius activation energy

of ∼0.23 eV (see Figure 6b). From 250 to ∼120 K, the rates

for triads 1 and 1′ are similar and almost independent of

temperature. At around the glass transition temperature of

2-methyltetrahydrofuran (120-105 K) and below, the decay of

the transient absorption displays two major exponential rate

constants, ka and kb (gray and white circles in Figure 6) with

amplitudes a and a . The amplitude of the longer component,

•

+

•-

•+

•-

recombination process of C -P-PF and C -PZn-PF

result in population of the carotenoid triplet excited state (vide

infra).

a

b

ab, is one-half to one-third that of the shorter component over

the range of 120-105 K and becomes much smaller as the

temperature decreases to 77 K. This complex decay can be

ascribed to heterogeneity in the organic glass and/or aggregation

of the sample. Below ∼120 K, the transient absorption spectra

of the triads are perturbed in the carotenoid radical cation

absorption region (shift to longer wavelengths) due to an

apparent emission (maximum at ∼960 nm), which may arise

from molecular aggregates formed at these temperatures (see

inset in Figure 5a). The amplitudes of the spectra presented in

Quantum Yields. Using the comparative method and 5,10,

5,20-tetrakis(4-methylphenyl)porphyrin as a standard (ꢀT-G,

1

4

-1

37,38

50 nm ) 6.8 × 10 M cm , ΦT ) 0.67),

the quantum

yields of the final charge separated states in triads 1 and 1′ were

determined as 0.73 and 0.14, respectively, based on the

absorption of the carotenoid radical cation (ꢀC•+, 950 nm ) 1.6

10 M cm ). Using the same porphyrin as a standard, the

quantum yield of the carotenoid triplet (ꢀT-G, 540 nm ) 1 ×

0 M cm ) in triads 1 and 1′ was determined to be 0.85

and 0.06, respectively, at room temperature.

5

-1

×

5

-1

1

10572 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

lies at 1.92 eV, whereas the energy of C- P-PF is 1.91 eV

1

•+

•-

•+

above the ground state. The energies of the C-P -PF , C -

•

-

•+

•-

P -PF, and C -P-PF charge-separated states were esti-

mated from cyclic voltammetric data described above. The

energy of the triplet state C-P-PF of 1 is not known; but the

energy of the lowest triplet state of a closely related carotenoid

3

9

has been estimated to be 0.63 eV. The difference in energy

between the porphyrin first excited singlet states of triad 1 is

only 0.01 eV. Thus, the equilibrium constant (K) for singlet-

singlet energy transfer is 1.5 at ambient temperatures. If singlet-

singlet energy transfer (steps 2 and -2) occurs at rates

comparable to or greater than those of the other steps shown in

Figure 7, both excited states will be significantly populated

during the decay process. The general solution for coupled

reactions of this type reveals that two fluorescence lifetimes or

two components from transient absorption data are to be

expected for the decay of the two excited singlet states.⁴⁰As

mentioned above, two significant fluorescence lifetimes were

observed with triad 1 (12 and 785 ps) and dyad 2 (12 and 890

ps). The corresponding components from the transient absorption

measurements with dyad 2 were 11.5 and 890 ps (Figure 3a).

The reciprocals of these lifetimes (γ1 and γ2) do not represent

any of the rate constants in the scheme. Rather

Figure 6. Temperature dependence of the final charge separated state

decay in triads 1 (a) and 1′ (b) in 2-methyltetrahydrofuran after

excitation with 5 ns laser pulses at 532 nm. Two major rate constants,

shown as gray and white circles, were obtained from a three-exponential

fit to the data. For both triads at all temperatures a long-lived component

2

1/2

γ ) {(X + Y) + [(X - Y) + 4k k ] }/2

(1)

(2)

1

2 -2

2

1/2

(100 µs, <5%) was fixed as the third exponential to improve the fit.

γ ) {(X + Y) - [(X - Y) + 4k k ] }/2

2

2 -2

Above 120 K, the kinetics are essentially single exponential (minor

components, <5%, are not shown). The areas of the symbols indicate

the relative amplitudes at each temperature. The straight line in (b)

shows the Arrhenius fit (activation energy 0.23 eV) to the data for 1′

at higher temperatures (>250 K).

where

X ) k + k + k

(3)

(4)

9

1

5

2

Y ) k + k + k

3

4

-2

The quantities γ1 ) 8.3 × 10 s^-¹(12 ps) and γ2 ) 1.1 × 10

10

-1

s

(890 ps) are known from the fluorescence decay or transient

absorption results of 2. The rate constants k1 and k3 may be

estimated from the fluorescence lifetimes of model compounds

8

-1

7

-1

4

and 3 as k1 ) 4.9 × 10 s and k3 ) 8.7 × 10 s ,

respectively. The equilibrium constant for the singlet-energy

transfer between the porphyrins, K, gives k-2 as k2/1.5. Thus,

one is left with two equations and three unknowns (k2, k4, and

k5). Setting k5 ) Bk4 allows one to derive an analytical solution

in quadratic form

2

1/2

-

b ( (b - 4ac)

k_-₂

)

(5)

2

a

where

Figure 7. Transient states of triad 1 and relevant interconversion

pathways. The energies of the various states have been estimated as

discussed in the text, and no correction for Coulombic effects on the

energies of the charge-separated states has been made.

2

a ) 4[(K - B) + (B + 1) K]

(6)

b ) 4(K - B)[(B - 1)(γ + γ ) + 2(k - Bk )] (7)

1

2

1

3

Figure 5a illustrate the substantial decrease in quantum yield

of the final charge-separated state as the temperature is lowered.

2

c ) [(B - 1)(γ + γ ) + 2(k - Bk )] -

1

2

1

3

2

(

B + 1) (γ - γ ) (8)

1 2

Discussion

Analysis of the Kinetic Data. C-P-PF, Triad 1. The model

chosen for the interpretation of the kinetic and spectroscopic

data dealing with singlet energy and electron transfer in triad 1

is shown in Figure 7, which indicates the relevant high-energy

states and kinetic pathways. The energies of the various

spectroscopic states were determined from the wavenumber

average of the longest wavelength absorption and shortest

wavelength emission maxima of 1 and appropriate model

compounds. The fluorinated porphyrin first excited singlet state

The k2 and k-2 values are essentially independent of B, with k2

1

0

-1

10 -1

) 5.3 × 10

s

and k-2 ) 3.5 × 10 s . Molecular orbital

calculations (Hartree-Fock/STO-G3, Gaussian 98 and DFT,

Siesta) were used to compare the amplitudes of the HOMO-

HOMO (for hole transfer, step 4) and LUMO-LUMO (for

electron transfer, step 5) orbitals at the meso positions linking

P to PF. Qualitatively, the amplitudes are similar. Thus, it is

reasonable to assume that the electronic coupling is comparable

for both reactions. In addition, there is only a small difference

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10573

in thermodynamic driving force for reactions 4 and 5, suggesting

that the approximation B ) 1 and k4 ) k5 is reasonable. If this

assumption is made, the above equations yield k4 ) k5 ) 1 ×

9

-1

1

0 s . This is consistent with results from studies of a series

25

of porphyrin-porphyrin dyads related to 2.

On the basis of the transient absorption results for model

compound 2 (Figure 3a), the excited states of the porphyrins

•

+

•-

are quenched by electron/hole transfer to produce P -PF

,

which decays to the ground state with a ∼4.2 ns time constant,

8

-1

so k6 ≈ 1/4.2 ns ) 2.4 × 10 s . The charge shift rate k7 for

•

+

•-

the formation of C -P-PF in triad 1 (Figure 4 circles) is

not observed experimentally, because k7 . k4 or k5, so the

induced absorption rise at 980 nm is a single exponential with

•

+

•-

a lifetime associated with the formation of C-P -PF rather

•+

•-

•+

•-

than C -P-PF . The quantum yield of C -P-PF , obtained

using the comparative method (vide supra) is 0.73.

basis of this value, and assuming that C -P -PF decays only

to the ground state and does not form C -P-PF , the kinetic

3

7,38

On the

•+

•-

Figure 8. Transient states of triad 1′ and relevant interconversion

pathways. The energies of the various states have been estimated as

discussed in the text, and no corrections for Coulombic effects on the

energies of the charge-separated states have been made. The dotted

arrow shows a pathway of activated decay of the final charge-separated

state at temperatures >250 K.

•

+

•-

10

-1

scheme requires that k7 g 1 × 10 s . (Note that for k7 >

10

-1

1

0

s

the quantum yield is a weak function of the value of

k7. In a related triad, C-P-C60, having the same carotenoid

and porphyrin moieties and linkage between them, we measured

a rate constant for the charge shift equivalent to step 7 of 8.3

_t_r_a_n_s_f_e_r_a_n_d_e_l_e_c_t_r_o_n_t_r_a_n_s_f_e_r_t_o_p_r_o_d_u_c_e_t_h_e_P_Z_n•+_-_P_F•- state.

9

-1 41

×

10 s ). In principle, steps 11 and 13 in Figure 7 could

Assuming that energy transfer to PF dominates the relaxation

•

+

•-

also contribute to formation of C -P-PF . Fluorescence

quenching of the proximal porphyrin by the attached carotenoid

is indicated as electron-transfer process 11 in Figure 7. A value

1

of PZn-PF in n-hexane and that the energy transfer rate in

n-hexane is nearly equal to that in tetrahydrofuran, k2 ∼ 2.0 ×

10 -1

1

≈

0

s

[k2 ) (1/49 ps) - k3]. This allows the calculation of k4

8

-1

for the rate constant k11 of 2.2 × 10 s can be determined

from the fluorescence lifetime of the porphyrin moiety of 5 (3.27

ns). Kinetic simulations show that if steps 11 and 13 occur,

they could account for at most a 2% increase in the quantum

yield of C -P-PF . Experimentally it is not possible to

observe step 13 because, like in the case of step 7, it is much

faster than steps 4, 5, and 11.

10 -1

5.5 × 10

s

[k4 ) (1/13 ps) - k3 - k2].

The charge shift rate k7 for the formation of the final charge-

•+

•-

separated state C -PZn-PF in triad 1′ can be estimated as

9

-1

2

.4 × 10 s from the relationship k7 ) (1/60 ps) - k6. This

•

+

•-

relationship is based on the rise of the carotenoid radical cation

transient absorption signal at 980 nm, which displays 40 and

6

0 ps components (Figure 4). The 60 ps component is identified

•

+

•-

The final charge recombination process is dominated by

population of the carotenoid triplet excited state rather than direct

recovery of the singlet ground state. From the flash photolysis

with the decay of C-P_Z_n-P_Fbecause the 40 ps component

matches the lifetime of C-P - P (formation of C-P

P_F) as determined from fluorescence measurements. From the

nanosecond flash photolysis results we know that the lifetime

of the C -P -P state is 210 ns (assuming that k , k , k

1

•+

Zn

F

Zn

-

•-

•

+

•-

results we know that the lifetime of C -P-PF is 500 ns

6

•+

•-

(

Figure 5a). Assuming that k8 , k9 would yield k9 ) 2.0 × 10

Zn

F

8

9

-

1

6

-1

s

. The carotenoid triplet state decays to the ground state

+ k_-₇) 4.8 × 10 s , Figure 5a) at room temperature and

that the quantum yield of formation of the carotenoid triplet

state by charge recombination from the final charge separated

state is about Φ ∼ 0.06 (vide supra). The carotenoid triplet state

5

-1

with a rate constant k10 ) 2.0 × 10 s (from its lifetime of

4

.9 µs).

C-PZn-PF, Triad 1′. Figure 8 presents an energy level

5

decays to the ground state with rate constant k10 ) 2.0 × 10

diagram showing the pertinent high-energy states and kinetic

pathways used to interpret the kinetic and spectroscopic data

dealing with singlet energy and electron transfer in triad 1′. The

fluorinated porphyrin first excited singlet state lies at 1.92 eV,

-

1

s

(lifetime 4.9 µs).

As mentioned above, fluorescence quenching of the model

porphyrin by the attached carotenoid in dyad 5′ is likely due to

electron transfer from the carotenoid to the zinc dimesityl

porphyrin. The rate constant for this quenching is k11 ) 1.3 ×

1

whereas the energy of C- PZn-PF is 2.06 eV above the ground

state. The energies of the charge-separated states were deter-

mined from the electrochemical measurements given above. The

9

-1

10 s .

3

energy of the triplet state C-PZn-PF of 1′ is the same as that

The kinetic data described above are compiled in Table 2.

•

+

•-

of the corresponding state in 1.

Quantum Yields. The quantum yields of C-P -P and

F

•

+

•-

Rate constants k1 and k3 may be estimated from the

fluorescence lifetimes of model compounds 4 and 3′; k1 ) 4.9

C -P-PF based on the excitation of the porphyrins may be

determined by simulation of the kinetic pathways in Figures 7

and 8 using the rate constants given above. The simulations

were done using Chemical Kinetics Simulator created by IBM,

version 1.01 (see Experimental Section). Quantum yields of 0.73

for C-P -PF and 0.71 for C -P-PF were calculated.

Similarly, quantum yields of 0.98 for C-PZn -PF and 0.14

for C -PZn-PF were calculated. The results of these simula-

tions are in close agreement with the measured values of

quantum yields by the comparative method (vide supra).

Because of the slow formation of the initial charge-separated

8

-1

8 -1

×

10 s and k3 ) 4.4 × 10 s . On the basis of the transient

absorption results for model compound 2′, the fluorinated

porphyrin singlet excited state is quenched to 40 ps by hole

•

+

•-

•+

•-

•+

•-

transfer to produce PZn -PF , which decays to the ground state

•

+

•-

with a 70 ps time constant (Figure 3b). Therefore, the rate

1

0

-1

•+

•-

constants k5 and k6 can be estimated as k5 ) 2.5 × 10

s

[

1

assuming k5 ) (1/40 ps) - k1] and k6 ) 1/70 ps ) 1.43 ×

10 -1

0 s . On the basis of the fluorescence lifetime measurements

for model dyad 2′, the zinc porphyrin excited singlet state is

quenched to 13 ps by a combination of singlet-singlet energy

•

+

•-

8

state C-P -PF in the case of triad 1 (k4 ) k5 ) 9.1 × 10

10574 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

TABLE 2: Kinetic Data and Quantum Yields for the Formation of Charge-Separated States at Room Temperature of C-P-P

F

(Triad 1) and C-PZn-P

F

(Triad 1′) in Tetrahydrofuran

rate constants (ns)^-

1

quantum yields

Triad 1

Triad 1′

k

1

k

53

2

k

35

-2

k

3

k

1

4

k

1

5

k

6

Φ

C-P•+-P_F•-

Φ

C•+-P-P_F•-

0

.49

0.087

∼0.24

k

13

.k

0

.73

0.71

k

7

k

,k

8

k

9

k

10

k

11

k

12

g10

9

0.002

0.0002

0.22

<143

5

k

1

k

2

k

0.44

3

k

4

k

25

5

k

6

k

2.4

7

Φ

C-P_Z_n•+-P_F•-

Φ

C•+-P_Z_n-P_F•-

0

.49

∼20

k

∼55

∼14.3

0

.98

0.14

k

-7

8

k

9

k

10

k

11

k

12

k

13

∼

0.0023

,k

9

0.002

0.0002

1.3

<50

.k

5

-

1

s ), a considerable amount of the observed carotenoid triplet

14%) is formed by triplet-triplet energy transfer from the triplet

charge-separated states of 1′ and 1 above ∼250 K is due to the

operation of a two-step mechanism in addition to the single step

recombination in the case of 1′, but not in 1. Indeed, the intercept

of the Arrhenius plot in Figure 6, for triad 1′, gives a value of

∼3 × 10 s , which is consistent with the modest electronic

coupling associated with the linkage between C and PZn.

Simulations using measured rate constants (k6, k7, and k10) and

(

excited states of both porphyrins formed by normal intersystem

crossing.⁴²This accounts for the discrepancy between the

quantum yield of formation of the final charge-separated state

73%) and the quantum yield of the carotenoid triplet produced

85%). In triad 1′, the formation of the initial charge separation

1

0

-1

(

•

+

•-

10 -1

6

-1

6

C-PZn -PF is much faster (k4 ) 5.5 × 10

s

and k5 )

.5 × 10 s ) than in triad 1 (k4 ) 1 × 10 s and k5 ) 1 ×

0 s ); thus, little porphyrin triplet is formed by normal

assuming that k9 is 2 × 10 s result in a value of ∼2.3 × 10

10

-1

9

-1

2

1

s

for k-7 (the rate constant for the endergonic step) and

quantum yields of 0.14 for C -PZn-PF and 0.06 for C -

PZn-PF, in good agreement with measured values. The lifetime

of C -PZn-PF of 258 ns at 298 K obtained by the simulation

is in reasonable agreement with the observed lifetime of C -

PZn-PF of 210 ns.

Turning now to triad 1, recall that the energy of C-P -

PF is about ∼0.54 eV above that of C -P-PF . Because

the electronic coupling term associated with the carotenoid to

porphyrin linkage should be similar in 1 and 1′, the same

preexponential factor found for 1′ can be used to calculate the

expected rate for the two-step process in 1. The minimum

activation energy of 0.54 eV combined with a preexponential

9

-1

•+

•-

3

intersystem crossing, so the carotenoid triplet is essentially all

formed from recombination of the triplet radical pair state.

However, above 250 K almost 50% of the final charge-separated

state returns to the initial charge separated state (vide infra),

which in turn rapidly decays mostly to the ground state. Thus

the quantum yield of carotenoid triplet is 0.06, even though the

•+

•-

•+

•

-

•+

•-

•+

•-

•

+

•-

yield of C -PZn-PF is 0.14. Time-resolved EPR measure-

ments are underway to further investigate these findings.

•

+

•-

•+

Charge Recombination of C -P-PF and C -PZn-

PF^•. In 2-methyltetrahydrofuran, charge recombination has been

measured in 1 and 1′ from room temperature to ∼77 K (Figure

-

1

0

-1

6

). At temperatures above ∼250 K, charge recombination in

factor of 3 × 10

s

limits the Arrhenius rate constant for

•+

•-

-1

C -PZn-PF is an activated process with an Arrhenius

activation energy of ∼0.23 eV. From 250 to 120 K, the rate is

almost independent of temperature and similar to that of triad

recombination by the two-step mechanism to 22 s (lifetime

45 ms) at 298 K, a value much smaller than observed, and

establishes that the two-step mechanism is not kinetically

allowed in triad 1. These calculations together with similar

values of recombination rates at low temperatures for both 1

and 1′ argue against the multistep process for triads 1 and 1′ at

low temperatures and support it for triad 1′ at elevated

temperatures. Thus, a single-step recombination reaction mech-

anism is assigned for triad 1 and triad 1′ at <250 K.

1

, in which there is no evidence for an activated process.

In principle, the charge recombination step could involve

either a one-step electron transfer, as indicated by step 9 in

Figures 7 and 8, or a two-step process, whereby an electron

would be transferred endergonically from the central porphyrin

•

+

•-

to the carotenoid cation radical, yielding C-P -PF (C-

•

+

•-

PZn -PF ) as an intermediate (see Figure 8, step -7). From

the intermediate state recombination to the ground state would

again compete with formation of the final charge-separated state.

Such an activated, two-step pathway to the ground state has

The recombination rate of both triads changes abruptly around

120-105 K (see Figure 6), which is the glass transition temper-

ature of 2-methyltetrahydrofuran (the solvent dielectric constant

4

6

•+

•- •+

changes from 18.8 to 2.6). The decay of C -P-P_F/C -

•

-

been observed in some carotenoporphyrin-quinone triads and

P -P

is no longer a single exponential, possibly due to

Zn

F

porphyrin-pyromellitimide-C60 triads.⁴

3-45

In an activated,

molecular heterogeneity and/or aggregation that occurs as the

solvent becomes glassy. Because of the spectral shift (Figure

5a) and quenched lifetime, it is tempting to assign the short-

lived (∼60-100 ns, high amplitude) decay component to the

transient absorption signal from aggregated molecules. The long-

lived component (∼1 µs, small amplitude) is consistent with

decay of the final charge-separated state in nonaggregated triads

responding to the change in dielectric constant at the glass

transition temperature, which would likely move the recombina-

tion process further into the Marcus-Hush inverted region,

thereby slowing the process. The quantum yield of the final

charge-separated state at low temperatures is very low; never-

theless, charge separation at 77 K has been confirmed by inde-

pendent time-resolved EPR measurements (unpublished results).

two-step recombination, the rate should be limited by the energy

difference between the final charge-separated state and the

•

+

•-

•+

•-

higher energy C-P -PF /C-PZn -PF state (see Figures

, 8) plus any additional activation energy associated with the

electron transfer. In the case of triad 1′, cyclic voltammetric

7

•+

•-

measurements locate C-PZn -PF ∼0.24 eV higher in energy

•

+

•-

than C -PZn-PF , which is consistent with the activation

Arrhenius energy of 0.23 eV measured for this molecule (see

linear fit in Figure 6).

•

+

•-

In the case of triad 1, C-P -PF is ∼0.54 eV higher in

•

+

•-

energy than C -P-PF and therefore not accessible at room

temperature (vide infra), precluding a two-step process. This

suggests that the difference in recombination rate of the final

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10575

Conclusions

temperature are reminiscent of the behavior observed in other

•

+

•-

triads. The intermediate C-PZn -PF recombines directly to

Relatively simple models of photosynthetic reaction centers

consisting of two porphyrins and a carotenoid covalently linked

in series exhibit rather complex relaxation pathways following

excitation of the porphyrin moieties. The rate constants of all

major processes have been measured or estimated using

photophysical measurements on simpler model compounds. To

a first approximation, the electronic couplings through the

linkages connecting the components are the same in 1 and 1′,

so differences in rate constants for analogous steps in 1 and 1′

are due to thermodynamic effects, including reorganization

energy differences, as noted in previous studies.²⁵A major

difference between the triads is in the energies of the intermedi-

ate charge-separated states. Metalating the central porphyrin

3

the ground state rather than to the C-PZn-PF state.

The quantum yield for the formation of the final charge-

separated state in both triads is reduced considerably at lower

temperatures, the decrease being abrupt at 120-105 K, which

is the glass transition range of 2-methyltetrahydrofuran. But even

at 77 K, the signal associated with the carotenoid radical cation

can be detected. At this temperature in most quinone-based

artificial reaction centers the photoinduced electron-transfer

4

7,48

reaction does not occur.

On the other hand, photoinduced

electron transfer in C60-based systems is generally less sensitive

to temperature effects than the systems described herein. This

is consistent with the unusually low reorganization energy

associated with C60, so even as solvent motion is frozen out to

give a low dielectric environment, the process is still in a

favorable Marcus region.

•

+

•-

lowers the energy of C-PZn -PF 0.2 eV below that of

•+

•-

C-P -PF . This energy difference controls the dynamics and

quantum yields in predictable ways.

The fact that these porphyrin- and carotenoid-based systems

demonstrate photoinduced electron transfer at low tem-

peratures and charge recombination to form triplet states

demonstrates that such phenomena are characteristic not only

of natural photosynthesis and porphyrin-fullerene molecules

but also of other molecular constructs in which large delocalized

π-electron systems lead to low reorganization energies and a

weak dependence of the energy of the radical ions on the

environment.

The added driving force for photoinduced electron transfer

in the C-PZn-PF triad is useful to drive electron transfer in

low dielectric environments such as the interior of artificial

biological membranes. However, this added driving force comes

with a penalty in quantum yield and lifetime of the final charge-

separated species. The analysis of the energy and electron/hole

transfer processes in 1 and 1′ illustrates the interplay between

factors controlling electron-transfer rates that must be considered

for optimal function of artificial reaction centers in various

environments.

Formation of the initial charge separated state C-PZn^•⁺-PF^•^-

in 1′ is more than an order of magnitude faster than in triad 1,

due to the higher driving force for a process in the Marcus

normal region. Because this step is relatively slow and rate

limiting in 1, the rise time of the carotenoid radical cation

•

+

•-

absorbance is associated with the formation of C-P -PF

,

whereas in 1′ the rise time of the carotenoid radical cation

absorbance is associated with the decay of the analogous initial

•

+

•-

charge-separated species, C-PZn -PF

.

Triad 1 generates a long-lived charge-separated state with a

relatively high quantum yield of 73% at room temperature; the

yield is much lower, 14%, in triad 1′. This decrease in quantum

yield arises because lowering the energy of the intermediate

•

+

•-

•+

•-

C-PZn -PF compared with C-P -PF reduces the driving

force and thereby slows the charge shift in 1′, which occurs in

the Marcus normal region and yields the long-lived charge-

•

+

•-

separated species C -PZn-PF . The rate of recombination to

the ground state from C-PZn -PF is increased because

recombination is less inverted in 1′ than in the higher energy

intermediate in 1, C-P -PF . In other words, in accordance

with Marcus theory, the 0.20 eV shift in C-PZn -PF

increases k6 and decreases k7, thus lowering the quantum yield

•

+

•-

•

+

•-

Experimental Section

•

+

•-

1

Instrumental Techniques. The H NMR spectra were re-

•

+

•-

of C -PZn-PF

.

corded on Varian Unity spectrometers at 300 or 500 MHz. Un-

less otherwise specified, samples were dissolved in deuterio-

chloroform with tetramethylsilane as an internal reference. Mass

spectra were obtained on a matrix-assisted laser desorption/

ionization time-of-flight spectrometer (MALDI-TOF). Ultra-

violet-visible ground-state absorption spectra were measured

on a Shimadzu UV2100U UV-vis spectrometer.

Electrochemical Measurements. Reduction and oxidation

potentials were measured by cyclic voltammetry in a two-

compartment glass cell, with a glassy carbon electrode as the

working electrode, a Pt wire as counter electrode, and Ag/

AgNO3 as a reference electrode. Electrode potentials are ex-

pressed vs SCE using ferrocene as internal standard (ferrocene/

ferrocenium potential vs SCE: 0.458 V). The electrolyte solution

was 0.1 M tetra-N-butylammonium hexafluorophosphate in

benzonitrile. Solutions were purged with nitrogen or argon prior

to measurements, and an inert atmosphere was maintained

thereafter. Cyclic voltammograms, at scan rates from 10 to 500

mV/s, were obtained with a Pine AFRDE4 potentiostat and an

X-Y recorder, or a PAR 173 potentiostat, a PAR 175 universal

programmer, and a computer with LabView software.

•+

The electron-hole recombination process whereby C -P-

PF^•and C -PZn-PF

-

•+

•-

relax features population of the

carotenoid triplet excited-state rather than recovery to the singlet

ground state. This is a rather unusual observation in artificial

photosynthetic systems but is observed in natural reaction

•

+

•-

•+

centers. Even though the energies of C -P-PF and C -

•-

PZn-PF are essentially the same, the rates of the recombination

reactions for 1 and 1′ at room temperature are different, 500

and 210 ns in tetrahydrofuran, respectively. Also, the temper-

•

+

•-

•+

ature dependence of the lifetimes of C -P-PF and C -

•

-

PZn-PF is significantly different, implying different mecha-

nisms in the two triads. Indeed, the charge recombination

•+

•-

reaction follows a single-step mechanism for C -P-PF over

the entire temperature range (300-77 K), while in the case of

•

+

•-

C -PZn-PF recombination occurs solely by a single-step

process only at temperatures below ∼250 K, where it has

•

+

•-

approximately the same rate observed for C -P-PF . The

different mechanisms arise from the 0.2 eV energy difference

between the intermediate charge-separated states in 1 and 1′.

•

+

•-

At temperatures above ∼250 K, the low-lying C-PZn -PF

•

+

•-

intermediate available to C -PZn-PF controls the dynamics

by providing an endergonic, activated step that operates in

addition to the direct process. An activated step at higher

temperature and a temperature independent process at low

Steady-state fluorescence emission spectra were measured

using a Photon Technology International MP-1 spectrophotom-

eter and corrected. Excitation was produced by a 75 W xenon

lamp and a single grating monochromator. Fluorescence was

10576 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

detected at 90° to the excitation beam via a single grating

monochromator and an R928 photomultiplier tube having S-20

spectral response and operating in the single-photon-counting

mode.

uses a stochastic method, based on reaction probabilities, to

calculate the time history of a chemical system using the reaction

mechanism and specified initial conditions. For more informa-

tion, see www.almaden.ibm.com/st/computational_science/ck/

msim/index.shtml.

Fluorescence decay measurements were performed on ∼1 ×

-5

10

M solutions by the time-correlated single photon counting

Synthesis. All chemicals employed in the preparation of the

following compounds were purchased from Aldrich, Acros, or

Lancaster. Solvents were bought from EM Science. Tetrahy-

drofuran was distilled from sodium and benzophenone in a

nitrogen atmosphere immediately prior to use. Toluene was

distilled from CaH2 and dichloromethane was distilled from

calcium carbonate. All solvents were stored over the appropriate

molecular sieves. Pyrrole was distilled prior to use. For thin-

layer chromatography (TLC), silica gel coated glass plates from

Analtech were used. Column chromatography was carried out

using EM Science silica gel 60 with 230-400 mesh. Unless

otherwise specified, all reactions were carried out under a nitro-

gen atmosphere. 5-Mesityldipyrromethane (6), 5,15-bis(4-carbo-

methoxyphenyl)-10,20-bis(2,4,6-trimethylphenyl)porphyrin (7),

method. The excitation source was a cavity-dumped Coherent

7

00 dye laser pumped by a frequency-doubled Coherent Antares

6s Nd:YAG laser. Fluorescence emission was detected at a

magic angle using a single grating monochromator and micro-

channel plate photomultiplier (Hamamatsu R2809U-11). The

instrument response time was ca. 60-90 ps, as verified by

scattering from Ludox AS-40. The spectrometer was controlled

by software based on a LabView program from National

Instruments.

Nanosecond transient absorption measurements were made

with excitation from an Opotek optical parametric oscillator

pumped by the third harmonic of a Continuum Surelight Nd:

YAG laser. The pulse width was ∼5 ns, and the repetition rate

was 5 Hz. The detection portion of the spectrometer has been

described elsewhere.⁴⁹

7

′-apo-7′-(4-acetamidophenyl)-â-carotene (16), and 7′-apo-7′-

(

4-aminophenyl)-â-carotene (17) were prepared by published

Low-temperature measurements were performed using an

27,28

procedures.

DN

Oxford Optistat liquid nitrogen-cooled optical cryostat equipped

5-(4-Carboxyphenyl)-15-(4-carbomethoxyphenyl)-10,20-bis-

with an Oxford ITC 601 temperature controller (accuracy (0.1

K).

(2,4,6-trimethylphenyl)porphyrin (8). Porphyrin 7 (200 mg,

0

.246 mmol) was dissolved in 40 mL of trifluoroacetic acid

and 80 mL of concentrated hydrochloric acid and heated to 60

C. When TLC indicated some formation of the diacid, the

The femtosecond transient absorption apparatus consists of

a kilohertz pulsed laser source and a pump-probe optical setup.

The laser pulse train was provided by a Ti:sapphire regenerative

amplifier (Clark-MXR, Model CPA-1000) pumped by a diode-

pumped CW solid-state laser (Spectra Physics, Model Millennia

V). The typical laser pulse was 100 fs at 790 nm, with a pulse

energy of 0.9 mJ at a repetition rate of 1 kHz. Most of the laser

energy (85%) was used to pump an optical parametric amplifier

°

reaction was diluted with dichloromethane and neutralized with

an aqueous solution of NaHCO3. The organic layer was washed

with water (× 3), dried over Na2SO4, and concentrated.

Porphyrins 7 (145 mg) and 8 (50 mg, 26% yield) were isolated

by column chromatography (silica gel, 1:9, methanol:dichloro-

methane). On the basis of recovered 7, a yield of 90% was

obtained. The recovered porphyrin (7) was recycled through

(IR-OPA, Clark-MXR). The excitation pulse was sent through

a computer-controlled optical delay line. The remaining laser

output (15%) was focused into a 1 cm flowing water cell to

generate a white light continuum. The continuum beam was

further split into two identical parts and used as the probe and

reference beams, respectively. The probe and reference signals

were focused onto two separate optical fiber bundles coupled

to a spectrograph (Acton Research, Model SP275). The spectra

were acquired on a dual diode array detector (Princeton

Instruments, Model DPDA-1024).⁵⁰

1

the same process: H NMR (300 MHz, CDCl3) δ -2.62 (2H,

s, pyrrole -NH), 1.84 (12H, s, 10,20-Ar o-CH3), 2.63 (6H, s,

1

0,20-Ar p-CH3), 4.11 (3H, s, 15-Ar -CO2CH3), 7.28 (4H, s,

0,20-Ar -H), 8.32 (2H, d, J ) 8 Hz, 5-Ar -H), 8.38 (2H, d,

J ) 8 Hz, 15-Ar -H), 8.44 (2H, d, J ) 8 Hz, 5-Ar -H), 8.57

(

2H, d, J ) 8 Hz, 15-Ar -H) 8.71-8.79 (8H, m, pyrrole H-2,

H-3, H-7, H-8, H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z

+

8

5

00.3 (C53H44N4O4 , calcd 800.3); UV-vis (CH2Cl2) 419, 515,

49, 590, 645 nm.

To determine the number of significant components in the

transient absorption data, singular value decomposition analy-

5

-(4-Carbomethoxyphenyl)-15-(N-tert-butoxy-4-amino-

31,32

phenyl)-10,20-bis(2,4,6-trimethylphenyl)porphyrin (9) was pre-

pared by dissolving 100 mg (0.125 mmol) of porphyrin 8 in 15

mL of tert-butyl alcohol and 25 µL (0.18 mmol) triethylamine.

After the addition of 33 µL (0.16 mmol) of diphenylphosphoryl

azide (DPPA) the solution was heated to 80 °C and stirred

overnight. After 30 h, the reaction was diluted with dichloro-

methane, washed successively with diluted aqueous citric acid,

saturated NaHCO3(aq), and water, and then dried with Na2SO4

and concentrated. The product was isolated by column chro-

matography (silica gel, 1:19 ethyl acetate:toluene) in 69% yield

sis

was carried out using locally written software based on

the MatLab 5.0 program (MathWorks, Inc.) Decay-associated

spectra were then obtained by fitting the transient absorption

change curves over a selected wavelength region simultaneously

as described by eq 9

n

∆

A(λ,t) ) A (λ) exp(-t/τ )

(9)

∑

i

i)1

1

where ∆A(λ,t) is the observed absorption change at a given

wavelength at time delay t and n is the number of kinetic

components used in the fitting. A plot of Ai(λ) versus wavelength

is called a decay-associated spectrum and represents the

amplitude spectrum of the ith kinetic component, which has a

lifetime of τi. Random errors associated with the reported

lifetimes obtained from transient absorption measurements were

typically e5%.

(75 mg): H NMR (300 MHz, CDCl3) δ -2.61 (2H, s, pyrrole

-NH), 1.65 (9H, s, 15-Ar -CO2CH(CH3)3), 1.85 (12H, s,

10,20-Ar o-CH3), 2.64 (6H, s, 10,20-Ar p-CH3), 4.11 (3H, s,

5-Ar -CO2CH3), 6.85(1H, s, 15-Ar -NH), 7.29 (4H, s, 10,20-

Ar -H), 7.77 (2H, d, J ) 8 Hz, 15-Ar -H), 8.15 (2H, d, J )

8 Hz, 15-Ar -H), 8.32 (2H, d, J ) 8 Hz, 5-Ar -H), 8.44 (2H,

d, J ) 8 Hz, 5-Ar -H), 8.70-8.75 (6H, m, pyrrole H-2, H-3,

H-7, H-8, H-12, H-18), 8.85 (2H, d, J ) 5 Hz, pyrrole H-13,

+

Simulations of the kinetic pathways were performed with the

Chemical Kinetics Simulator, Version 1.01, created by IBM. It

H-17); MALDI-TOF-MS m/z 871.4 (C57H53N5O4 , calcd 871.4);

UV-vis (CH2Cl2) 420, 516, 552, 591, 641 nm.

Artificial Photosynthetic Reaction Centers

-(4-Carbomethoxyphenyl)-15-(4-aminophenyl)-10,20-bis-

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10577

5

saturated NaHCO3(aq) (×2) and water, dried with Na2SO4, and

concentrated. Compound 12 (3.36 g, 42% yield) was isolated

by column chromatography (silica gel, 1:1 hexanes:dichloro-

(2,4,6-trimethylphenyl)porphyrin (10). Porphyrin 9 (50 mg,

0

1

.057 mmol) was dissolved in 10 mL of dichloromethane, and

0 mL of trifluoroacetic acid was added. The solution was stirred

1

methane): H NMR (300 MHz, CDCl ) δ 4.94 (1H, t, J ) 17

3

at room temperature. TLC indicated that no starting material

was present after 1 h. The solution was diluted with dichloro-

methane and neutralized with NaHCO3(aq). The organic layer

was washed with water (×3), dried with Na2SO4, and concen-

trated. Pure porphyrin 10 was obtained (44 mg, 99% yield):

Hz, meso H-5), 6.18-6.24 (4H, m, pyrrole H-3, H-4), 6.77-

6.79 (2H, m, pyrrole H-2), 6.12 (2H, bs, pyrrole -NH).

5

,10,15,20-Tetrakis(heptafluoropropyl)porphyrin (13), 5-(4-

carbomethoxyphenyl)-10,15,20-tris(heptafluoropropyl)por-

phyrin (4), and 5,15-bis(4-carbomethoxyphenyl)-10,20-bis-

1

H NMR (300 MHz, CDCl3) δ -2.59 (2H, s, pyrrole -NH),

(

heptafluoropropyl)porphyrin (14). In a 1000 mL round-

1

4

.85 (12H, s, 10,20-Ar o-CH3), 2.64 (6H, s, 10,20-Ar p-CH3),

.01 (2H, bs, 15-Ar -NH2), 4.11 (3H, s, 5-Ar -CO2CH3), 7.06

bottomed flask, 1.9 g (6.0 mmol) of 5-(heptafluoropropyl)-

dipyrromethane (12), and 500 mL of dichloromethane were

degassed by bubbling with N2 for 15 min. To this solution, 0.594

g (3.00 mmol) of heptafluorobutyraldehyde, 0.492 g (3.00

mmol) of methyl 4-formylbenzoate, and 0.03 mL (0.30 mmol)

of boron trifluoride diethyl etherate were added. After the

mixture turned a deep purple color (approximately 24 h), 0.681

g (3.00 mmol) of 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ)

was added and the stirring was continued for 24 h. The DDQ

was removed by chromatography through a pad of silica until

the eluent was tan-colored. The solvent was removed and the

product dissolved in 500 mL of toluene. The treatment with

DDQ was repeated with 0.681 g (3.00 mmol) and the solution

was refluxed for 48 h. Again, the DDQ was removed by

chromatography through a pad of silica. The residue was purified

by column chromatography (silica, 1:1 hexanes:dichloromethane).

Three major products were formed. 5,10,15,20-Tetrakis(hepta-

(

2H, d, J ) 8 Hz, 15-Ar -H), 7.29 (4H, s, 10,20-Ar -H), 8.00

2H, d, J ) 8 Hz, 15-Ar -H), 8.32 (2H, d, J ) 8 Hz, 5-Ar

-

H), 8.43 (2H, d, J ) 8 Hz, 5-Ar -H), 8.69-8.74 (6H, m,

pyrrole H-2, H-3, H-7, H-8, H-12, H-18), 8.90 (2H, d, J ) 5

Hz, pyrrole H-13, H-17); MALDI-TOF-MS 772.6 m/z

+

(

5

C52H45N5O2 , calcd 771.9); UV-vis (CH2Cl2) 422, 517, 553,

94, 649 nm.

5-(4-Carboxyphenyl)-15-(4-aminophenyl)-10,20-bis(2,4,6-tri-

methylphenyl)porphyrin (11). In a 50 mL round-bottomed flask

fitted with a condenser, 30 mg (0.039 mmol) of porphyrin 10

in 12 mL of trifluoroacetic acid and 24 mL of concentrated

hydrochloric acid was heated to 60 °C for 4 days under inert

atmosphere. After TLC (1:9 methanol:dichloromethane) indi-

cated that no starting material remained, the solution was diluted

with dichloromethane and neutralized with NaHCO3(aq). The

organic layer was washed with water (×3) and dried with

Na2SO4. Distillation of the solvent at reduced pressure resulted

in 49 mg (99% yield) of porphyrin 11: ¹H NMR (300 MHz,

CDCl3) δ -2.57 (2H, s, pyrrole -NH), 1.86 (12H, s, 10,20-Ar

o-CH3), 2.65 (6H, s, 10,20-Ar p-CH3), 4.01 (2H, bs, 15-Ar-

NH2) 7.07 (2H, d, J ) 8 Hz, 15-Ar -H), 7.30 (4H, s, 10,20-Ar

1

fluoropropyl)porphyrin (13): 100 mg (3% yield); H NMR (300

MHz, CDCl3) δ -2.28 (2H, s, pyrrole -NH), 9.92 (8H, s,

pyrrole H-2, H-3, H-7, H-8, H-12, H-13, H-17, H-18); MALDI-

+

TOF-MS m/z 983.2 (C32H10F28N4 calcd 982.4); UV-vis

(

CH2Cl2) 409, 510, 541, 589, 649 nm. 5-(4-Carbomethoxy-

phenyl)-10,15,20-tris(heptafluoropropyl)porphyrin (4): 167 mg

-

H), 8.02 (2H, d, J ) 8 Hz, 15-Ar -H), 8.39 (2H, d, J ) 8

Hz, 5-Ar -H), 8.55 (2H, d, J ) 8 Hz, 5-Ar -H), 8.70-8.78

6H, m, pyrrole H-2, H-3, H-7, H-8, H-12, H-18), 8.92 (2H, d,

1

(

-

5

6% yield); H NMR (300 MHz, CDCl3) δ -2.40 (2H, s, pyrrole

NH), 4.13 (3H, s, 5-Ar -CO2CH3), 8.46 (2H, d, J ) 8 Hz,

-Ar -H), 8.81 (2H, d, J ) 8 Hz, 5-Ar -H), 9.38 (2H, bs,

(

J ) 5 Hz, pyrrole H-13, H-17); MALDI-TOF-MS m/z 758

+

pyrrole H-3, H-7), 9.57 (6H, bs, pyrrole H-2, H-8, H-12, H-13,

H-17, H-18); MALDI-TOF-MS m/z 947.4 (C37H17F21N4O2 ,

(

5

C51H43N5O2 , calcd 758); UV-vis (CH2Cl2) 421, 517, 554,

+

90, 647 nm.

-(4-Carbomethoxyphenyl)-15-(4-acetamidophenyl)-10,20-

bis(2,4,6-trimethylphenyl)porphyrin (3). Porphyrin 10 (14 mg,

.018 mmol) was stirred at 80 °C in 14 mL of acetic anhydride

calcd 848); UV-vis (CH2Cl2) 410, 512, 545, 589, 643 nm. 5,15-

5

Bis(4-carbomethoxyphenyl)-10,20-bis(heptafluoropropyl)por-

1

phyrin (14): 558 mg (20% yield); H NMR (300 MHz, CDCl3)

0

δ -2.55 (2H, s, pyrrole -NH), 4.15 (6H, s, 5, 15-Ar

and 0.5 mL of pyridine for 2 h. The reaction was quenched

with water while the solution was still warm. The organic layer

was washed with water (×3), dried with Na2SO4, and concen-

trated. Porphyrin 3 (13 mg, 88% yield) was obtained after

purification by column chromatography (silica gel, 1:4 ethyl

-CO2CH3), 8.26 (4H, d, J ) 8 Hz, 5, 15-Ar -H), 8.47 (4H, d,

J ) 8 Hz, 5, 15-Ar -H), 8.86 (4H, d, J ) 6 Hz, pyrrole H-3,

H-7, H-13, H-17), 9.48 (4H, bs, pyrrole H-2, H-8, H-12, H-18);

+

MALDI-TOF-MS m/z 915.2 (C42H24F14N4O4 , calcd 915);

1

UV-vis (CH2Cl2) 411, 512, 547, 591, 644 nm.

acetate:toluene): H NMR (300 MHz, CDCl3) δ -2.63 (2H, s,

pyrrole -NH), 1.84 (12H, s, 10,20-Ar o-CH3), 2.36 (3H, s, 15-

Ar -COCH3), 2.63 (6H, s, 10,20-Ar p-CH3), 4.11 (3H, s, 5-Ar

5-(4-Carboxyphenyl)-10,15,20-tris(heptafluoropropyl)por-

phyrin (15) was prepared by dissolving 160 mg (0.170 mmol)

of porphyrin 4 in 40 mL of trifluoroacetic acid and 80 mL of

concentrated hydrochloric acid. The solution was stirred at 60

°C overnight. After that time, the reaction mixture was diluted

with dichloromethane and neutralized with aqueous NaHCO3.

The organic layer was washed with water (× 3), dried with

Na SO , and concentrated. Porphyrin 15 (114 mg, 72% yield)

-

CO2CH3), 7.28 (4H, s, 10,20-Ar -H), 7.53 (1H, s, 15-Ar

NH), 7.89 (2H, d, J ) 8 Hz, 15-Ar -H), 8.18 (2H, d, J ) 8

-

Hz, 15-Ar -H), 8.31 (2H, d, J ) 8 Hz, 5-Ar -H), 8.43 (2H, d,

J ) 8 Hz, 5-Ar -H), 8.69-8.75 (6H, m, pyrrole H-2, H-3,

H-7, H-8, H-12, H-18), 8.82 (2H, d, J ) 5 Hz, pyrrole H-13,

+

H-18); MALDI-TOF-MS m/z 814.1 (C54H47N5O3 , calcd 814);

2

4

UV-vis (CH2Cl2) 420, 516, 551, 590, 647 nm.

was isolated by column chromatography (silica gel, 1:9 metha-

5

-(Heptafluoropropyl)dipyrromethane (12). Pyrrole (3.5 mL,

nol:dichloromethane). The starting material, porphyrin 4, could

also be recovered (30 mg): H NMR (300 MHz, CDCl3) δ

1

5

0 mmol), heptafluorobutyraldehyde (5.0 g, 25 mmol), and 100

mL of freshly distilled tetrahydrofuran were degassed for 15

min by bubbling with N2. Concentrated hydrochloric acid (2

mL) was added and the mixture was refluxed for 2 h under N2.

The solution was neutralized with NaHCO3(aq) and then diluted

with dichloromethane. The organic layer was washed with

-2.38 (2H, s, pyrrole -NH), 8.57 (2H, d, J ) 8 Hz, 5-Ar -H),

8.84 (2H, d, J ) 8 Hz, 5-Ar -H), 9.40 (2H, bs, pyrrole H-3,

H-7), 9.58 (6H, bs, pyrrole H-2, H-8, H-12, H-13, H-17, H-18);

MALDI-TOF-MS m/z 934.8 (C36H15F21N4O2 , calcd 934.5);

UV-vis (CH2Cl2) 410, 512, 545, 590, 642 nm.

+

1

0578 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

PF-P-COOH, Dyad 18. In a dry 100 mL round-bottomed

porphyrin 10 in 15 mL of toluene and 0.5 mL of pyridine. After

1 h the reaction was quenched with water and washed with water

(×3). The organic layer was dried with Na2SO4 and concen-

trated. Column chromatography (silica gel, dichloromethane)

flask under an argon atmosphere, 35 mg (0.037 mmol) of

fluorinated porphyrin (15) was dissolved in 25 mL of dry toluene

and 1 mL of pyridine. To this solution, 4 µL (0.05 mmol) of

thionyl chloride was added, and the stirring was continued. The

acid chloride formation was monitored by quenching a small

aliquot in 1:9 methanol:dichloromethane and a TLC was run

every 15 min. After the acid chloride formed, about 1 h, excess

thionyl chloride and solvent was removed by evaporation at

reduced pressure. The acid chloride in 5 mL of toluene and 0.5

mL of pyridine was added dropwise to a solution of 14 mg

1

gave 15 mg (57% yield) of compound 2. H NMR (500 MHz,

CDCl3) δ -2.57 (2H, s, PM pyrrole -NH), -2.34 (2H, s, PF

pyrrole -NH), 1.87 (12H, s, PM 10,20-Ar o-CH3), 2.65 (6H, s,

PM 10,20-Ar p-CH3), 4.11 (3H, s, PM 5-Ar -CO2CH3), 7.31

(4H, s, PM 10,20-Ar -H), 8.23 (2H, d, J ) 8 Hz, PM 15-Ar

-H), 8.33 (2H, d, J ) 8 Hz, PM 5-Ar -H), 8.35 (2H, d, J ) 8

Hz, PM 15-Ar -H), 8.44 (2H, d, J ) 8 Hz, PM 5-Ar -H), 8.47-

8.51 (4H, m, PF 5-Ar -H), 8.73-8.77 (6H, m, PM pyrrole H-2,

H-3, H-7, H-8, H-12, H-18), 8.91-8.94 (4H, m, PM pyrrole

H-13, H-17, PF pyrrole H-3, H-7), 9.45 (2H, bs, PF pyrrole H-2,

H-8), 9.60 (4H, bs, PF pyrrole H-12, H-13, H-17, H-18);

(

0.018 mmol) of porphyrin 11 in 15 mL of toluene and 0.5 mL

of pyridine. After 1 h the reaction was quenched with water

and washed with water (×3). The organic layer was dried with

Na2SO4 and concentrated. Column chromatography (silica gel,

+

1

:9 methanol:dichloromethane) resulted in 15 mg (50% yield)

MALDI-TOF-MS m/z 1687.4 (C88H58F21N9O3 , calcd 1687.4);

1

of 18: H NMR (300 MHz, CDCl3) δ -2.57 (2H, s, PM pyrrole

NH), -2.34 (2H, s, PF pyrrole -NH), 1.87 (12H, s, PM 10,20-

Ar o-CH3), 2.66 (6H, s, PM 10,20-Ar p-CH3), 7.31 (4H, s, PM

0,20-Ar -H), 8.22-8.57 (12H, m, PM 5, 15-Ar -H, PF 5-Ar

UV-vis (CH2Cl2) 420, 514, 548, 590, 644 nm.

-

5-(4-Carboxyphenyl)-15-(4-trifluoroacetamidophenyl)-10,20-

bis(2,4,6-trimethylphenyl)porphyrin (19). At 0 °C 1 mL of

trifluoroacetic anhydride was added to a solution of 50 mg

(0.066 mmol) of porphyrin 11 in 25 mL of dichloromethane.

The dark green solution was quenched after 1 h by pouring the

solution into 50 mL of dichloromethane and washing with water

(× 6). The organic layer was dried over Na SO and concen-

1

-

H), 8.74-8.80 (6H, m, PM pyrrole H-2, H-3, H-7, H-8, H-12,

H-18), 8.91-8.94 (4H, m, PM pyrrole H-13, H-17, PF pyrrole

H-2, H-7), 9.45 (2H, bs, PF pyrrole H-3, H-8), 9.60 (4H, bs, PF

pyrrole H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z 1673.8

2

4

+

(

5

C87H56F21N9O3 , calcd 1674.4); UV-vis (CH2Cl2) 421, 515,

trated under reduced pressure. Porphyrin 19 was purified by

49, 591, 645 nm.

column chromatography (silica gel, 1:9 methanol:dichloro-

1

C-P-PF, Triad 1. In a dry 100 mL round-bottomed flask

methane) to yield 54 mg (96% yield): H NMR (300 MHz,

under argon atmosphere, 15 mg (0.01 mmol) of dyad 18 was

dissolved in 20 mL of dry toluene and 1 mL of pyridine. A

portion of thionyl chloride (2 µL, 0.03 mmol) was added and

the mixture was kept well-stirred. TLC was used to check the

progress of the reaction, in the same fashion as for dyad 18.

Excess thionyl chloride and solvent were removed under

vacuum. To a solution of the acid chloride in 15 mL of dry

toluene and 0.5 mL pyridine was added 10 mg (0.02 mmol) of

aminocarotenoid 17 in 5 mL of dry toluene and 0.5 mL of

pyridine. After 1 h of stirring, the reaction mixture was quenched

with water and then washed with water (× 3). The organic layer

was dried with Na2SO4 and concentrated. Triad 1 (17 mg, 89%

yield) was isolated by column chromatography (silica gel, 3:7

CDCl3) δ -2.63 (2H, s, pyrrole -H), 1.83 (12H, s, 10,20-Ar

o-CH3), 2.62 (6H, s, 10,20-Ar p-CH3), 7.28 (4H, s, 10,20-Ar

-H), 7.98 (2H, d, J ) 9 Hz, 15-Ar -H), 8.21 (1H, s, 15-Ar

-NH) 8.27 (2H, d, J ) 9 Hz, 15-Ar -H), 8.78 (2H, d, J ) 9

Hz, 5-Ar -H), 8.56 (2H, d, J ) 9 Hz, 5-Ar -H), 8.71-8.78

(8H, m, pyrrole H-2, H-3, H-7, H-8, H-12, H-13, H-17, H-18);

+

MALDI-TOF-MS m/z 853.3 (C53H42F3N5O3 , calcd 853.3);

UV-vis (CH2Cl2) 420, 516, 551, 592, 648 nm.

C-P-NHCOCF , Dyad 20, was prepared by dissolving 35

3

mg (0.041 mmol) of porphyrin 19 in 10 mL of dry toluene and

0.2 mL of pyridine under argon atmosphere. A portion of 10

µL (0.12 mmol) of thionyl chloride was added and the stirring

continued for 1 h. After the acid chloride was formed, the excess

thionyl chloride and solvent were removed. The acid chloride

was dissolved in 10 mL of dry dichloromethane and 0.2 mL of

pyridine. Finally, 41 mg (0.082 mmol) of carotenoid 17 was

added and the stirring was continued for 1 h. After 1 h, the

reaction was quenched with water. The organic layer was

washed with water (×3), dried over Na SO , and concentrated

1

hexanes:dichloromethane): H NMR (500 MHz, CDCl3) δ

-

2.56 (2H, s, PM pyrrole -NH), -2.34 (2H, s, PF pyrrole

NH), 1.04 (6H, s, car H-16, H-17), 1.44-1.49 (2H, m, car

-

H-2), 1.58-1.64 (2H, m, car H-3), 1.72 (3H, s, car H-18), 1.88

12H, s, PM 10,20-Ar o-CH3), 1.98-2.01 (14H, m, car H-4,

H-19, H-20, H-20′, H-19′), 2.65 (6H, s, PM 10,20-Ar p-CH3),

(

2

4

6

-

8

.12-6.86 (14H, m, car vinyl -CH), 7.32 (4H, s, PM 10,20-Ar

under reduced pressure. After column chromatography (silica

gel, 1:0.1:9 ethyl acetate:triethylamine:toluene), 17 mg (32%

H), 7.55 (2H, d, J ) 8 Hz, car H-1′, H-5′), 7.79 (2H, d, J )

1

Hz, car H-2′, H-4′), 8.14-8.49 (12H, m, PM 5, 15-Ar -H, PF

yield) of dyad 20 was obtained: H NMR (300 MHz, CDCl₃)

5

-Ar -H), 8.74-8.79 (6H, m, PM pyrrole H-2, H-3, H-7, H-8,

δ 2.63 (2H, s, pyrrole -NH), 1.04 (6H, s, car H-16, H-17),

1.47-1.49 (2H, m, car H-2), 1.60-1.62 (2H, m, car H-3), 1.73

H-12, H-18), 8.91-8.94 (4H, m, PM pyrrole H-13, H-17, PF

pyrrole H-3, H-7), 9.45 (2H, bs, PF pyrrole H-2, H-8), 9.60 (4H,

(3H, s, car H-18), 1.85 (12H, s, P_M10,20-Ar o-CH ), 1.98-

3

bs, PF pyrrole H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z

2.01 (14H, m, car H-4, H-19, H-20, H-19′, H-20), 2.63 (6H, s,

+

2

4

162.7 (C124H101F21N10O2 , calcd 2162); UV-vis (CH2Cl2) 421,

PM 10,20-Ar p-CH3), 6.12-6.96 (14H, m, car vinyl -CH), 7.29

(4H, s, PM 10,20-Ar -H), 7.54 (2H, d, J ) 8 Hz, car H-1′,

H-5′), 7.78 (2H, d, J ) 8 Hz, car H-2′, H-4′), 7.99 (2H, d, J )

77, 511, 591, 644 nm.

PF-P-COOCH3, Dyad 2. In a dry 100 mL round-bottomed

8

Hz, PM 15-Ar -H), 8.17 (1H, s, PM 15-Ar -NH), 8.25-8.28

flask under argon atmosphere, 34 mg (0.037 mmol) of fluori-

nated porphyrin (15) was dissolved in 25 mL of dry toluene

and 1 mL of pyridine. To this solution, 10 µL (0.06 mmol) of

thionyl chloride was added, and the stirring was continued for

(

5

4H, m, PM 5-Ar -H, 15-Ar -H), 8.37 (2H, d, J ) 8 Hz, PM

-Ar -H), 8.72-8.79 (8H, m, PM pyrrole H-2, H-3, H-7, H-8,

H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z 1340.7

+

(

4

C90H87F3N6O2 , calcd 1340.7); UV-vis (CH2Cl2) 300, 374,

1

h. After the acid chloride formed, excess thionyl chloride and

22, 480, 512, 592, 650 nm.

solvent were removed by evaporation under reduced pressure.

The acid chloride in 5 mL of toluene and 0.5 mL of pyridine

was added dropwise to a solution of 14 mg (0.02 mmol) of

C-P-NH2, Dyad 21. Dyad 20 (17 mg, 0.013 mmol) was

deprotected by stirring in 6 mL of tetrahydrofuran and 20 mL

Artificial Photosynthetic Reaction Centers

J. Phys. Chem. B, Vol. 108, No. 29, 2004 10579

+

of freshly prepared methanolic potassium hydroxide for 36 h

at 40 °C. The reaction was quenched by pouring the solution

into ethyl ether and then washing the organic phase with water

(C124H99F21N10O2Zn , calcd 2222.7); UV-vis (CH2Cl2) 422,

478, 512, 545, 587, 641 nm.

PF-PZn-COOCH3, Dyad 2′. A portion of 2 mg (0.001 mmol)

of dyad 2 was dissolved in 0.5 mL of dichloromethane. To this

solution was added 0.5 mL of 0.0025 M zinc acetate dissolved

in 10% methanol/dichloromethane, and the stirring continued

at room temperature for 12 h. After 12 h, an additional 0.5 mL

aliquot of the zinc acetate solution was added and the stirring

was continued for another 6 h until a third compound was

apparent by TLC (the dimetalated triad). The reaction mixture

was immediately purified by column chromatography (silica gel,

dichloromethane) to yield 0.5 mg (24% yield): H NMR (500

MHz, CDCl3) δ -2.34 (2H, s, PF pyrrole -NH), 1.87 (12H, s,

PM 10,20-Ar o-CH3), 2.66 (6H, s, PM 10,20-Ar p-CH3), 4.12

3H, s, PM 5-Ar -CO2CH3), 7.23 (4H, s, PM 10,20-Ar -H),

.22 (2H, d, J ) 8 Hz, PM 15-Ar -H), 8.35 (2H, d, J ) 8 Hz,

PM 5-Ar -H), 8.37 (2H, d, J ) 8 Hz, PM 15-Ar -H), 8.45 (2H,

d, J ) 8 Hz, PM 5-Ar -H), 8.47-8.51 (4H, m, PF 5-Ar -H),

,82-8.86 (6H, m, PM pyrrole H-2, H-3, H-7, H-8, H-12, H-18),

.93 (2H, d, J ) 5 Hz, PM pyrrole H-13, H-17), 9.02 (2H, d, J

5 Hz, PF pyrrole H-3, H-7), 9.46 (2H, bs, PF pyrrole H-2,

H-8), 9.60 (4H, bs, PF pyrrole H-12, H-13, H-17, H-18);

MALDI-TOF-MS m/z 1749.4 (C88H56F21N9O3Zn , calcd 1749.3);

UV-vis (CH2Cl2) 422, 510, 549, 587, 642 nm.

Zinc 5-(4-Carbomethoxyphenyl)-15-(4-acetamidophenyl)-

(

×6). The organic solution was dried with Na2SO4 and

concentrated under reduced pressure. Purification by column

chromatography (silica gel, 1:0.1:99 ethyl acetate:triethylamine:

toluene) resulted in 14 mg (91% yield) of dyad 21: H NMR

1

(

300 MHz, CDCl3) δ -2.63 (2H, s, pyrrole -NH), 1.04 (6H,

s, car H-16, H-17), 1.44-1.49 (2H, m, car H-2), 1.61-1.66

2H, m, car H-3), 1.73 (3H, s, car H-18), 1.85 (12H, s, PM 10,20-

(

Ar o-CH3), 1.98-2.08 (14H, m, car H-4, H-19, H-20, H-19′,

H-20), 2.63 (6H, s, PM 10,20-Ar p-CH3), 4.03 (2H, s, PM 15-

Ar -NH2), 6.15-6.96 (14H, m, car vinyl -CH), 7.06 (2H, d,

J ) 8 Hz, PM 15-Ar -H), 7.29 (4H, s, PM 10,20-Ar -H), 7.54

1

(2H, d, J ) 8 Hz, car H-1′, H-5′), 7.78 (2H, d, J ) 8 Hz, car

(

8

H-2′, H-4′), 8.00 (2H, d, J ) 8 Hz, PM 15-Ar -H), 8.25 (2H,

d, J ) 8 Hz, PM 5-Ar -H), 8.36 (2H, d, J ) 8 Hz, PM 5-Ar

-

H), 8.68-8.75 (6H, m, PM pyrrole H-2, H-3, H-7, H-8, H-12,

H-18) 8.89-8.91 (2H, m PM pyrrole H-13, H-18); MALDI-

8

)

+

TOF-MS m/z 1244.7 (C88H88N6O , calcd 1244.7); UV-vis

(CH2Cl2) 300, 373, 422, 509, 550, 594, 649 nm.

C-P-NHCOCH3, Dyad 5. Acetic anhydride (1 µL, 0.01

mmol) was added to a solution of 7 mg (0.006 mmol) of dyad

+

2

1 in 5 mL of dichloromethane and 2 µL of pyridine. After 2

h of stirring, the solution was diluted with dichloromethane and

washed with water (×5). The organic layer was dried over

Na2SO4 and concentrated. After column chromatography (silica

gel, 1:4 ethyl acetate:toluene), 5 mg (70% yield) of dyad 5 was

1

0,20-bis(2,4,6-trimethylphenyl)porphyrin (3′). A portion of 7

mg (0.009 mmol) of porphyrin 3 was stirred overnight in 1 mL

of dichloromethane and an excess of 0.01 M zinc acetate

dissolved in 10% methanol/dichloromethane. The reaction was

worked up by diluting the solution with dichloromethane and

washing (×3) with H2O. The organic layer was dried with

Na2SO4 and concentrated. The resulting 3′ was obtained in

1

obtained: H NMR (300 MHz, CDCl3) δ -2.62 (2H, s, pyrrole

-NH), 1.04 (6H, s, car H-16, H-17), 1.42-1.49 (2H, m, car

H-2), 1.61-1.64 (2H, m, car H-3), 1.73 (3H, s, car H-18), 1.85

12H, s, PM 10,20-Ar o-CH3), 1.98-2.08 (14H, m, car H-4,

H-19, H-20, H-19′, H-20), 2.36 (3H, s, PM 15-Ar -COCH3),

.63 (6H, s, PM 10,20-Ar p-CH3), 6.14-6.96 (14H, m, car vinyl

CH), 7.30 (4H, s, PM 10,20-Ar -H), 7.51-7.55 (3H, m, car

(

1

quantitative yield (8 mg): H NMR (500 MHz, CDCl3) δ 1.82

2

-

(

12H, s, 10,20-Ar o-CH3), 2.36 (3H, s, 15-Ar -COCH3), 2.63

6H, s, 10,20-Ar p-CH3), 4.11 (3H, s, 5-Ar -CO2CH3), 7.28

4H, s, 10,20-Ar -H), 7.47 (1H, s, 15-Ar -NH), 7.86 (2H, d,

J ) 8 Hz, 15-Ar -H), 8.16 (2H, d, J ) 8 Hz, 15-Ar -H), 8.30

2H, d, J ) 8 Hz, 5-Ar -H), 8.39 (2H, d, J ) 8 Hz, 5-Ar -H),

.71-8.76 (6H, m, pyrrole H-2, H-3, H-7, H-8, H-12, H-18),

H-1′, H-5′, PM 15-Ar -NH), 7.78 (2H, d, J ) 8 Hz, car H-2′,

H-4′), 7.90 (2H, d, J ) 8 Hz, PM 15-Ar -H), 8.17 (2H, d, J )

8

Hz, PM 15-Ar -H), 8.26 (2H, d, J ) 9 Hz, PM 5-Ar -H),

.37 (2H, d, J ) 9 Hz, PM 5-Ar -H), 8.70-8.83(8H, m, PM

(

8

pyrrole H-2, H-3, H-7, H-8, H-12, H-13, H-17, H-18); MALDI-

+

8.83 (2H, d, J ) 5 Hz, pyrrole H-13, H-17); MALDI-TOF-MS

m/z 875.3 (C54H45N5O3Zn , calcd 875.3); UV-vis (CH2Cl2)

TOF-MS m/z 1286.7 (C90H90N6O2 , calcd 1286.7); UV-vis

+

(CH2Cl2) 300, 370, 417, 477, 510, 590, 648 nm.

4

19, 550, 591 nm.

C-PZn-PF, Triad 1′. A portion of 2 mg (0.001 mmol) of

C-PZn-NHCOCH3, Dyad 5′. A portion of 2 mg (0.002

mmol) of dyad 5 was stirred overnight in 1 mL of dichloro-

methane with an excess of a solution of 0.01 M zinc acetate in

0% methanol/dichloromethane. The reaction was worked up

triad 1 was dissolved in 0.5 mL of dichloromethane. To this

solution 0.5 mL of 0.001 M zinc acetate dissolved in 10%

methanol/dichloromethane was added and the stirring was

continued at room temperature for 12 h. After 12 h, an additional

1

by diluting the solution with dichloromethane and washing (×3)

0

.5 mL aliquot of the zinc acetate solution was added and the

with H2O. The organic layer was dried with Na2SO4 and

stirring was continued for 6 h until a third compound was

apparent by TLC (the dimetalated triad). The reaction mixture

was immediately purified by column chromatography (silica gel,

concentrated. The resulting product was obtained in 95% yield

2 mg): H NMR (500 MHz, CDCl3) δ 1.04 (6H, s, car H-17,

H-16), 1.42-1.49 (2H, m, car H-2), 1.61-1.64 (2H, m, car H-3),

.73 (3H, s, car H-18), 1.84 (12H, s, PM 10, 20 o-CH3), 1.99-

2.09 (14H, m, car H-4, H-19, H-20, H-19′, H-20′), 2.33 (3H, s,

PM 15-Ar -COCH3), 2.64 (6H, s, PM 10,20-Ar p-CH3), 6.16-

6.95 (14H, m, car vinyl -CH), 7.29 (4H, s, PM 10,20-Ar -H),

7.47 (1H, s, PM 15-Ar -NH), 7.54 (2H, d, J ) 9 Hz, car H-1′,

H-5′), 7.77 (2H, d, J ) 9 Hz, car H-2′, H-4′), 7.86 (2H, d, J )

8 Hz, PM 15-Ar -H), 8.18 (2H, d, J ) 8 Hz, PM 15-Ar -H),

8.24 (2H, d, J ) 8 Hz, PM 5-Ar -H), 8.38 (2H, d, J ) 8 Hz,

PM 5-Ar -H), 8.77-8.90 (8H, m, PM pyrrole H-2, H-3, H-7,

H-8, H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z 1348.7

1

(

1

dichloromethane) to yield 0.5 mg of 1′ (25% yield): H NMR

1

(

1

2

500 MHz, CDCl3) δ -2.34 (2H, s, PF pyrrole, -NH), 1.04

6H, s, car H-16, H-17), 1.44-1.49 (2H, m, car H-2), 1.59-

.64 (2H, m, H-3), 1.87 (12H, s, PM 10,20-Ar o-CH3) 1.98-

.09 (14H, m, car H-4, H-19, H-20, H-19′, H-20′), 2.66 (6H, s,

PM 10,20-Ar p-CH3), 6.12-6.95 (14H, m, car vinyl -CH), 7.32

4H, s, PM 10, 20 Ar -H), 7.54 (2H, d, J ) 9 Hz, car H-1′,

(

H-5′), 7.79 (2H, d, J ) 9 Hz, car H-2′, H-4′), 8.14-8.49 (12H,

m, PM 5, 15-Ar -H, PF 5-Ar -H), 8.83-8.87 (6H, m, PM

pyrrole H-2, H-3, H-7, H-8, H-12, H-18), 8.92 (2H, d, J ) 6

Hz, PM pyrrole H-13, H-17), 9.02 (2H, d, J ) 5 Hz, PF pyrrole

H-3, H-7), 9.45 (2H, bs, PF pyrrole H-2, H-8), 9.60 (4H, bs, PF

H-12, H-13, H-17, H-18); MALDI-TOF-MS m/z 2222.7

+

(C90H88N6O2Zn , calcd1348.6); UV-vis (CH2Cl2) 422, 477,

505, 546, 588 nm.

10580 J. Phys. Chem. B, Vol. 108, No. 29, 2004

Gould et al.

Acknowledgment. This work was supported by a grant from

(24) 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.

by the U.S. Department of Energy (DE-FG02-03ER15393). We

thank Roche for the generous gift of carotenoid samples. This

is publication 590 from the ASU Center for the Study of Early

Events in Photosynthesis.

(

25) DeGraziano, J. M.; Liddell, P. A.; Leggett, L.; Moore, A. L.; Moore,

T. A.; Gust, D. J. Phys. Chem. 1994, 98, 1758-1761.

26) Kuciauskas, D.; Liddell, P. A.; Moore, A. L.; Moore, T. A.; Gust,

(

D. J. Am. Chem. Soc. 1998, 120, 10880-10886.

(

27) Lee, C. H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427-11440.

References and Notes

28) Gust, D.; Moore, T. A.; Moore, A. L.; Liddell, P. A. Methods

Enzymol. 1992, 213, 87-100.

(

1) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 1993, 26,

(29) Wijesekera, T. P. Can. J. Chem. 1996, 74, 1868-1871.

1

98-205.

(30) Fungo, F.; Otero, L.; Durantini, E. N.; Silber, J. J.; Sereno, L.;

(

2) Wasielewski, M. R. Chem. ReV. 1992, 92, 435-461.

Marino-Ochoa, E.; Moore, T. A.; Moore, A. L.; Gust, D. J. Phys. Chem. B

(3) Gust, D.; Moore, T. A. In The Porphyrin Handbook; Kadish, K.

2

001, 105, 4783-4790.

M., Smith, K. M., Guilard, R., Eds.; Academic Press: New York, 1999;

(

31) Golub, G. H.; Reinsch, C. Numer. Math. 1970, 14, 403-420.

32) Henry, E. R.; Hofrichter, J. Methods Enzymol. 1992, 210, 129-

pp 153-190.

(

4) Guldi, D. M. Pure Appl. Chem. 2003, 75, 1069-1075.

192.

(

33) The time range used in this experiment (4.5 ns) is too short to

accurately determine this time constant.

34) Luo, C.; Guldi, D. M.; Imahori, H.; Tamaki, K.; Sakata, K. J. Am.

Chem. Soc. 2000, 122, 6535-6551.

35) Mosseri, S.; Mialocq, J. C.; Perly, B. Radiat. Phys. Chem. 1992,

9, 223-233.

36) Hermant, R. M.; Liddell, P. A.; Lin, S.; Alden, R. G.; Kang, H.

K.; Moore, A. L.; Moore, T. A.; Gust, D. J. Am. Chem. Soc. 1993, 115,

080-2081.

37) Bensasson, R. V.; Land, E. J.; Truscott, T. G. Flash Photolysis

Ito, O.; Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 2607-

617.

6) Liddell, P. A.; Kuciauskas, D.; Sumida, J. P.; Nash, B.; Nguyen,

D.; Moore, A. L.; Moore, T. A.; Gust, D. J. Am. Chem. Soc. 1997, 119,

400-1405.

7) Kuciauskas, D.; Liddell, P. A.; Lin, S.; Johnson, T. E.; Weghorn,

2

(

1

3

(

S. J.; Lindsey, J. S.; Moore, A. L.; Moore, T. A.; Gust, D. J. Am. Chem.

Soc. 1999, 121, 8604-8614.

2

(

8) Guldi, D. M. Chem. Soc. ReV. 2002, 31, 22-36.

(

(9) Ohkubo, K.; Imahori, H.; Shao, J. G.; Ou, Z. P.; Kadish, K. M.;

and Pulse Radiolysis. Contribution to the Chemistry of Biology and

Medicine; Pergamon Press: New York, 1983.

Chen, Y. H.; Zheng, G.; Pandey, R. K.; Fujitsuka, M.; Ito, O.; Fukuzumi,

S. J. Phys. Chem. A 2002, 106, 10991-10998.

(

38) Bonnett, R.; McGarvey, D. J.; Harriman, A.; Land, E. J.; Truscott,

T. G.; Winfield, U.-J. Photochem. Photobiol. 1988, 48, 271-276.

39) Lewis, J. E.; Moore, T. A.; Benin, D.; Gust, D.; Nicodem, D.;

Nonell, S. Photochem. Photobiol. 1994, 59, 35S.

40) Demas, J. N. Excited-state lifetimes; Academic Press: New York,

983.

41) Kodis, G.; Liddell, P. A.; Moore, A. L.; Moore, T. A.; Gust, D. J.

(

10) Imahori, H.; Yamada, H.; Guldi, D. M.; Endo, Y.; Shimomura,

A.; Kundu, S.; Yamada, K.; Okada, T.; Sakata, Y.; Fukuzumi, S. Angew.

Chem., Int. Ed. Engl. 2002, 41, 2344-2347.

(

11) Kuciauskas, D.; Liddell, P. A.; Lin, S.; Stone, S. G.; Moore, A. L.;

(

Moore, T. A.; Gust, D. J. Phys. Chem. B 2000, 104, 4307-4321.

1

(

12) Fukuzumi, S.; Nakanishi, I.; Suenobu, T.; Kadish, K. M. J. Am.

(

Chem. Soc. 1999, 121, 3468-3474.

Phys. Org. Chem. 2004, in press.

(

13) Hush, N. S. Trans. Faraday Soc. 1961, 57, 557-580.

14) Marcus, R. A. J. Chem. Phys. 1956, 24, 966-978.

15) Hush, N. S. J. Chem. Phys. 1958, 28, 962-972.

(42) Gust, D.; Moore, T. A.; Moore, A. L.; Krasnovsky, A. A.; Liddell,

P. A.; Nicodem, D.; Degraziano, J. M.; Kerrigan, P.; Makings, L. R.; Pessiki,

P. J. J. Am. Chem. Soc. 1993, 115, 5684-5691.

(43) Gust, D.; Moore, T. A.; Makings, L. R.; Liddell, P. A.; Nemeth,

G. A.; Moore, A. L. J. Am. Chem. Soc. 1986, 108, 8028-8031.

(44) Osuka, A.; Yamada, H.; Shinoda, S.; Nozaki, K.; Ohno, T. Chem.

Phys. Lett. 1995, 238, 37-41.

(45) Imahori, H.; Tamaki, K.; Araki, Y.; Sekiguchi, Y.; Ito, O.; Sakata,

Y.; Fukuzumi, S. J. Am. Chem. Soc. 2002, 124, 5165-5174.

(46) Furutsuka, T.; Imura, T.; Kojima, T.; Kawabe, K. Technol. Rep.

Osaka UniVersity 1974, 24, 367-369.

16) Marcus, R. A. Can. J. Chem. 1959, 37, 155-163.

17) Kuciauskas, D.; Liddell, P.; Moore, T. A.; Moore, A. L.; Gust, D.

In Recent AdVances in the Chemistry and Physics of Fullerenes and Related

Materials; Kadish, K. M., Ruoff, R. S., Eds.; The Electrochemical

Society: Pennington, NJ, 1998; pp 242-261.

(

18) Carbonera, D.; Di Valentin, M.; Corvaja, C.; Agostini, G.;

Giacometti, G.; Liddell, P. A.; Kuciauskas, D.; Moore, A. L.; Moore, T.

A.; Gust, D. J. Am. Chem. Soc. 1998, 120, 4398-4405.

(

19) SteinbergYfrach, G.; Liddell, P. A.; Hung, S. C.; Moore, A. L.;

Gust, D.; Moore, T. A. Nature 1997, 385, 239-241.

20) Steinberg-Yfrach, G.; Rigaud, J. L.; Durantini, E. N.; Moore, A.

L.; Gust, D.; Moore, T. A. Nature 1998, 392, 479-482.

21) Bennett, I. M.; Vanegas Farfano, H. M.; Bogani, F.; Primak, A.;

(47) Gaines, G. L.; O’Neil, M. P.; Svec, W. A.; Niemczyk, M. P.;

Wasielewski, M. R. J. Am. Chem. Soc. 1991, 113, 719-721.

(48) Wasielewski, M. R.; Gaines, G. L.; O’Neil, M. P.; Svec, W. A.;

Niemczyk, M. P.; Prodi, L.; Gosztola, D. J. In Dynamics and Mechanisms

of Photoinduced Transfer and Related Phenomena; Mataga, N., Okada, T.,

Masuhara, H., Eds.; Elsevier Science Publishers: New York, 1992; pp 87-

103.

(

Liddell, P. A.; Otero, L.; Sereno, L.; Silber, J. J.; Moore, A. L.; Moore, T.

A.; Gust, D. Nature 2002, 420, 401-403.

(22) Janot, J. M.; Bienvenue, E.; Seta, P.; Bensasson, R. V.; Tome, A.

C.; Enes, R. F.; Cavaleiro, J. A. S.; Leach, S.; Camps, X.; Hirsch, A. J.

Chem. Soc. Perkin Trans. 2 2000, 301-306.

(49) Davis, F. S.; Nemeth, G. A.; Anjo, D. M.; Makings, L. R.; Gust,

D.; Moore, T. A. ReV. Sci. Instrum. 1987, 58, 1629-1631.

(50) Freiberg, A.; Timpmann, K.; Lin, S.; Woodbury, N. W. J. Phys.

Chem. B 1998, 102, 10974-10982.

(

23) Tronel-Peyroz, E.; Miquel-Mercier, G.; Vanel, P.; Seta, P. Chem.

Phys. Lett. 1998, 285, 294-298.

Article Doi

Artificial photosynthetic reaction centers with porphyrins as primary electron acceptors

DOI: 10.1021/jp040143y

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jp040143y

Full text of DOI:10.1021/jp040143y

Products guided by the article

R&D Labs maybe for 736173-35-6

Relevant to this article

Hot Product