Excitation Wavelength Dependence of Bacterial Reaction Center Photochemistry. 2. Low-Temperature Measurements and Spectroscopy of Charge Separation

Abstract

Excitation with spectrally narrow (5 nm), temporally short duration (200 fs) laser pulses at a variety of wavelengths between 848 and 903 nm results in substantial excitation wavelength dependent differences in the evolution of the bacterial reaction center absorbance spectrum both before and after charge separation occurs. The transient holes in the initial electron donor band showed a more resolved vibrational band structure at 20 K, when compared to those of earlier room temperature transient hole-burning experiments (Peloquin, J. M.; Lin, S.; Taguchi, A. K. W.; Woodbury, N. W. J. Phys. Chem. 1995, 99, 1349). The dominant vibrational band observed is at 120 cm-1, in agreement with dynamic measurements of coherent oscillations on this time scale (Vos, M. H.; Rappaport, F.; Lambry, J.-C.; Breton, J.; Martin, J.-L. Nature 1993, 363, 320). At both room temperature and low temperature, there is a distribution of P to P* transition energies due to a distribution of the protein conformations in the ground state. At 20 K, one can also see a distribution of P* to P stimulated emission energies. As might be expected, the barriers to conformational interconversion are more easily crossed at room temperature, resulting in a smaller difference between the mean transition energies of the photoselectable subpopulations on the picosecond time scale relative to those at low temperature. At 20 K, much of this conformational interconversion is apparently lost when exciting near the 0−0 transition wavelength of P. More conformational interconversion appears to take place at low temperature when higher energy excitation is used, implying that P* is vibrationally hot for at least hundreds of femtoseconds following excitation on the blue side of its QY band. Of particular interest is the insensitivity of the overall charge separation kinetics to the selection of ground state conformational subpopulations by different excitation wavelengths. There appears to be very little coupling between the nuclear motion that defines the ground state transition distribution and the charge separation reaction itself. Given the apparently slow rate of vibrational relaxation of some of the excited state modes most strongly coupled to the P to P* transition, the lack of excitation wavelength dependence of the charge separation rate also suggests that these modes are not strongly coupled to the charge separation reaction. What the ground state (and possibly excited state) conformational distributions and interconversions do affect is the kinetic complexity of the absorbance changes in the 800 nm region. Specifically, the extent of involvement of fast, multiexponential decay components in this region depends strongly on the wavelength of excitation.