Oct 8, 1981 - centers of photosynthetic bacteria. (energy transduction/biomimetic charge separation). I. FUJITAt, T. L. NETZELO§, C. K. CHANGI, AND C. -B.
Proc. Natl Acad. Sci. USA Vol. 79, pp. 413-417, January 1982 Biophysics
Picosecond photochemistry of a cofacial diporphyrin containing iron(III) and zinc(II): Mimicking electron transfer between cytochrome c and the primary electron donor in reaction centers of photosynthetic bacteria (energy transduction/biomimetic charge separation)
I. FUJITAt, T. L.
NETZELO§, C. K. CHANGI, AND C. -B. WANGO
Departments of tEnergy and Environment and tChemistry, Brookhaven National Laboratory, Upton, New York 11973; and ¶Department of Chemistry, Michigan State University, East Lansing, Michigan 48824
Communicated by Gerhart Friedlander, October 8, 1981
ABSTRACT Comparison of picosecond Idnetic and spectroscopic data for Zn octaethylporphine and Fe(III)CI octaethylporphine with that for Zn-Fe(EII)CI, a cofacial diporphyrin composed of a Zn porphyrin covalently bound to an Fe(lI)CI porphyrin with two chains offive atoms each, supports the assignment of a light-driven electron transfer (k > 1011 s )withinZn-Fe(III)CI to form [Znt-Fe(II)]CI. The kinetics (k 1010 so') and thermodynamics of the reverse electron transfer are compared to those of a similar electron transfer in bacterial photosynthesis, the reduction of an oxidized bacteriochlorophyll dimer, (BChl)2t, by Fe(II) cytochrome c. Recent picosecond spectroscopic experiments on cofacial diporphyrins (1, 2) provide substantial evidence to support the conclusion that electron transfer (ET) reactions occur from singlet excited states (S) in these molecules in 101" sol) ET from the SI state of porphyrin-type molecules. Indeed, a Mg-H2 diporphyrin provides a reasonable model for the primary electron donor-acceptor couple .(5, 6) in the reaction center (RC) of photosystem II (PS II) of green plants, In this paper we describe the photochemical properties of a Zn-Fe(III)Cl diporphyrin that has the same carbon skeleton as Mg-H2. However, this time one subunit has a central Fe(III) metal atom with.a Cl- ligand and the other a Zn(II) metal atom. The SI state of Mg-H2 is a ir-ir* type, whereas the Zn-Fe(III)Cl state of comparable energy has a mixed ir-wr* and ligand-to-metal charge transfer (LMCT) character. Thus, comparing the ET reactions in these diporphyrins may reveal differences due to the different types ofexcited states in the two cases. Also, the reverse ET in Zn-Fe(III)Cl to reform the. initial ground state, [Zn+ Fe(11)]CI-- Zr-Fe(I)C1, [1] is analogous to the reduction of (BChl2)t by cytochrome c (cytc) in bacterial photosynthesis (7, 8), [2] (BChl)2t Fe(II)cyt-c -- (BChl)2Fe(III)cyt-c. The oxidation of Fe(II)cyt-c is one. of the earliest-studied light-driven ETs associated with the "primary" events of bacterial photosynthesis (9-11). This investigation of the light-
driven ETs in Zn-Fe(III)Cl is intended to provide additional data for modeling the cytochrome c oxidation.
EXPERIMENTAL PROCEDURE The synthesis of Zn-Fe(III)Cl (structure I) will be reported elsewhere.
Each porphyrin in I is a 2,7,12, 17-tetramethyl porphine with RI = n-octyl groups at carbons 3 and 13 and R2 linking bridges at carbons 8 and 18. The two covalent bridges, R2, have the following.structure CH2-CH2-CH2-CH3
-CH2-CHI-N- C-CHj-
1
0
with the amine ends joined to the Fe porphyrin and the acid ends joined to the Zn porphyrin. The center-to-center separation of the porphyrin subunits in diporphyrin I is assumed to be the same. as that in a Cu-Cu diporphyrin with the same carbon skeleton. In that case, EPR measurements of copper-copper interactions showed the separation to be r4 Abbreviations: ZnEt8Por, zinc octaethylporphine; Fe(III)ClEt8Por, chloro(octaethylporphinato)iron(III); Ha-H2, Cu-Cu, Fe(II-Fe(II), Mg-H2, and Zn-Fe(III)Cl are cofacial diporphyrins composed of either free base (H2) or metallated porphyrin subunits joined with two covalent bridges five atoms in length (each has the same structure except for variation of the central metal or two H atoms); (BChl)g, a dimer of bacteriochlorophyll; Fe(II)cyt-c, iron(II) cytochrome c; RC, reaction center; PS II, photosystem II; ET, electron transfer; CT, charge transfer; LMCT, ligand-to-metal charge transfer state; H4furan, tetrahydrofuran.
The publication costs -ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
§ To whom reprint requests should be addressed.
413
4144Proc. NatL Acad. Sci. USA 79 (1982) Biophysics: Fujita et aL
A (12, 13). The present samples were outgassed and studied in sealed cells with 2-mm optical path lengths. The range ofsample concentrations was 0.2-0.5 mM. One sample, Zn-Fe(III)Cl in pyridine, decomposed slowly. Therefore, whenever that sample's absorption spectrum changed by more than 5%, it was replaced with a fresh one of the same concentration. Note that chloro(octaethylporphinato)iron(III) [Fe(III)ClET8Por] in pyridine did not decompose during the course of these experiments. The electronic processes after light absorption by the diporphyrin and its monomeric subunits were monitored with a previously described picosecond absorption spectrometer (14). Samples were excited with 6-ps pulses of 527-nm laser light. It will become apparent in the next sections that the excited states of Fe(III)ClET8Por are extremely short lived (r < 20 ps). This makes the effects oftime dispersion in the white probe light important. It is well known that light ofshorter wavelength travels more slowly in dense media such as water and glass than does light of longer wavelength. In other words, if 775-nm light is coincident in the sample with the excitation pulse, 600-nm light will not get to the sample for several more picoseconds. In our apparatus there is about 1 ps of delay for each 35 nm of spectral increment in the 550- to 775-nm region. These delays are not significant for optical transients with lifetimes greater than 25 ps. However, this is not the case for these compounds. To deal with this problem, we adopted the procedure of specifying the time of arrival of the 775-nm light at the sample relative to the excitation pulse. Negative time means that the 775-nm light arrived before or during the excitation pulse. Positive time means that it arrived after the excitation pulse. To tell the arrival time ofany other wavelength, the delay at that wavelength must be added to the 775-nm arrival time. The delays as a function of wavelength are given at the top of each AA spectral plot. For example, a AA spectrum labelled -4 ps is actually probing the 600-nm region +1 ps after excitation. RESULTS AND DISCUSSION This study contrasts the picosecond kinetics and absorption spectra of a Zn-Fe(III)Cl diporphyrin to those of zinc octaethylporphine (ZnEt8Por) and Fe(III)ClEt8Por monomers to discover photochemical processes unique to the diporphyrin. Because the goal of this study is to learn about light-driven ET reactions, an estimate ofthe free energy change for the following reaction is appropriate: [Zn+ Fe(II)]CI, [3] in which * denotes an electronic excited state produced after light absorption. An estimate of the free-energy change, AE1/2, for forming [Znt-Fe(II)]Cl from Zn-Fe(III)Cl in the ground state can be obtained from separate one-electron oxidations and reductions of the diporphyrin. This approximation gave realistic predictions for a series of Mg-H2 diporphyrins that have the same carbon skeleton (1). However, several attempts to measure these redox potentials for Zn-Fe(III)Cl failed because addition of the supporting electrolyte caused the diporphyrin to precipitate. Thus redox values for the porphyrin monomers ZnEt8Por and Fe(III)ClEt8Por were used to estimate AE1/2 for this diporphyrin. Electrochemical measurements on Mg-H2 diporphyrins showed that the AE1/2 values estimated from diporphyrin data were 200 meV smaller than the AE1/2 values estimated from data on the corresponding monomers (refs. 1 and 15; unpublished data). Therefore this correction was applied to the AE1/2 estimates (see below) for the Zn-Fe(III)Cl
[Zn 7Fe(III)Cl]
diporphyrin.
The reduction potential data show that, while there is some difference in midpoint potentials for ZnEt8Port/ZnEt8Por on going from tetrahydrofuran (H4furan) (+0.73 V versus the saturated calomel electrode) to pyridine (+0.66 V), the difference for Fe(III)ClEt8Por/Fe(II)ClEt8Por is over 6 times greater (-0.50 V in H4furan compared to -0.05,V in pyridine). This may reflect a change in the spin state of the Fe(III) porphyrin. Observation of a g = 6 EPR signal at .77 K for both Fe(III)ClEt8Por and the diporphyrin in H4furan implies that both molecules are high spin. Therefore a reasonable estimate of the AEI/2 for forming [Zn!-Fe(II)]Cl in H4furan is 1.0 eV. Good EPR signals were not obtained for either molecule in pyridine. Therefore only a range of AE1/2 values for forming [Znt-Fe(II)]Cl in this solvent can be given: 0.5-1.0 eV. Because the first, strongly allowed excited state of ZnFe(III)Cl seen in absorption has 1.9 eV of energy, the free energy change for Reaction 3 from this state will be -0.9 eV in H4furan and -0.9 to -1.4 eV in pyridine. In Mg-H2 diporphyrins (1) ET reactions from the SI state were observed to occur in
