PROBING PICOSECOND PROCESSES WITH NANOSECOND LASERS - ELECTRONIC AND VIBRATIONAL-RELAXATION DYNAMICS OF HEME-PROTEINS

We discuss the technique of resonance Raman saturation spectroscopy and present experimental results that probe relaxation processes in heme proteins following electronic excitation in the Soret band. The observable relaxation time scales are limited by the laser excitation rate k(L) rather than by the laser pulse width (approximately 10 ns). Analysis of the data using a theory that separates the electronic and vibrational relaxation leads to electronic ground state recovery times tau = 6.4 +/- 2.0 ps for ferrocytochrome c, tau = 4.8 +/- 1.5 ps for deoxymyoglobin, and tau = 2.0 +/- 0.7 ps for deoxyhemoglobin. The Raman depolarization ratio is predicted to increase at high laser flux, due to the preparation of a partially oriented sample by photoselective excitation. Such effects are observed in heme systems and the relaxation times extracted from the depolarization analysis are in good agreement with those derived from measurements of Raman intensity saturation. Studies of the asymmetric broadening of the nu-4 mode of cytochrome c at high laser flux reveal that the line shapes in the Stokes and anti-Stokes region are inequivalent. Time-reversal symmetry dictates that this broadening is due to an underlying Raman band associated with an excited electronic state that is populated at high laser flux. Similar line broadening effects, observed in hemoglobin and myoglobin samples, are also shown to arise from Raman scattering of excited electronic states rather than Rabi broadening [Alden et al., J. Phys. Chem. 94, 85 (1990)] or anharmonic coupling to vibrationally hot low frequency modes [Petrich et al., Biochemistry 26, 7914 (1987) Quantitative analysis of Stokes and anti-Stokes Raman scattering determines the heme vibrational temperature as a function of laser flux and leads to a description of the Raman intensities that differs significantly from that of Lingle et al. [J. Phys. Chem. 95, 9320 (1991)], which ignores electronic saturation effects and is based on the scattering properties of a two-level system. For cytochrome c, we use a simple thermal transport model to extract a value for the product of the heme area and the coefficient of surface heat transfer between the heme and the surrounding protein. This leads to a approximately 4 ps time constant for the short-time exponential phase of heme cooling.