The Journal of
Physical Chemistry
0 Copyright, 1989, by the American Chemical Society
VOLUME 93, NUMBER 16 AUGUST 10, 1989
LETTERS Structure and Marker Mode of the Primary Electron Donor State Absorption of Photosynthetic Bacteria: Hole-Burned Spectra S. G . Johnson, D. Tang, R. Jankowiak, J. M. Hayes, G . J. Small,* Ames Laboratory-USDOE
and Department of Chemistry, Iowa State University, Ames. Iowa 5001 I
and D. M. Tiede Chemistry Division, Argonne National Laboratory, Argonne. Illinois 60439 (Received: April 19, 1989)
Structured photochemical hole-burned spectra are presented for P870 and P960 of the reaction centers (RC) of Rhodobacter sphaeroides and Rhodopseudomonas uiridis. A special pair marker mode Franck-Condon progression is identified. The zero-phonon holes yield P870*and P960*decay times in good agreement with the time domain values. Site excitation energy selection is used to establish correlation between a higher energy RC state of Rps. viridis and P960*.
The recently revealed structures of the reaction center (RC) of Rps. v i r i d i ~ l and - ~ Rb. sphaeroides4" have led to even greater activity7,* directed toward understanding the primary charge separation process which is triggered by excitation of the primary electron donor states P960*and P870*,respectively. It is generally accepted that the lowest energy component of the special pair Q, transition contributes significantly to the electronic structure of ( I ) Deisenhofer, J.; Epp, 0.;Miki, K.; Huber, R.; Michel, H. J . Mol. Biol. 1984, 180, 385. (2) Deisenhofer, J.; Epp, 0.; Miki, K.; Huber, R.; Michel, H. Nature (London) 1985, 318, 618. (3) Michel, H.; Epp, 0.; Deisenhofer, J. EMBO J . 1986, 5 , 2445. (4) Chang, C. H.; Tiede, D.; Tang, J.; Smith, U.; Norris, J.; Schiffer, M. FEBS Lett. 1986, 205, 82. (5) Allen, J. P.; Feher, G.; Yeates, T. 0.;Rees, D. C.; Deisenhofer, J.; Michel, H.;Huber, R. Proc. N a f l . Acad. Sci. U.S.A. 1986, 83, 8589. (6) Yeates, T. 0.; Komiya, H.; Chirino, A,; Rees, D. C.; Allen, J. P.; Feher, G. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 7993. (7) Kirmaier, C.; Holten, D. Photosynth. Res. 1987, 13, 225. (8) Budil, D. E.; Gast, P.; Chang, C. H.; Schiffer, M.; Norris, J. Annu. Reu. Phys. Chem. 1987, 38, 561.
0022-3654/89/2093-5953$01.50/0
P*. This component is often referred to as P- since the Q, transition dipoles of the monomers, which comprise the pair (PL,PM), would be antiparallel in the simplest model. It should be noted, however, that agreement on the structures of the higher energy Qr states of the six-pigment RC has not been arrived at on the basis of semiempirical calculation^.^^'^ In this Letter we present new photochemical hole-burned (PHB) spectra for P960 and P870 which reveal that their underlying structures are very similar, identify a low-frequency intermolecular marker mode for both special pairs, reveal the origin of the homogeneous broadening for the p960 and P870 absorption profiles, and determine the decay times of P960* and P870* from their zero-point levels at 4.2 K. The decay times are used to address the question of whether or not thermalization of vibrational modes occurs prior to charge separation. In addition, a site excitation energy correlation effect for Rps. viridis is observed which identifies an excited R C state that is correlated with P960*. An (9) Warshel, A.; Parson, W. W. J. Am. Chem. SOC.1987, 109, 6143. (10) Scherer, P. 0. J.; Fischer, S . F. J . Phys. Chem. 1989, 93, 1633.
0 1989 American Chemical Society
5954
The Journal of Physical Chemistry, Vol. 93, No. 16, 1989
assignment of the former to a state which is significantly contributed to by P+ (upper special pair component) is suggested. The XB (burn-wavelength)-dependent PHB spectra of Boxer and his group”J2 first revealed that there is a significant homogeneous broadening contribution to the P870 and P960 absorption profiles. This result attracted considerable attenti~n’’.’~ since the broad (-400 cm-I) unstructured holes observed could be principally a manifestation of ultrafast electronic reIaxation1’J3J4or significant protein-pigment geometry change^'^^'^^'^ which accompany electronic excitation of P. We recently reported structured transient PHB spectra (4.2 K) for P960 for three glass-detergent host ~ y s t e m s . ’ ~The ~~~ structure included two relatively broad (- 100 cm-I) holes denoted as X and Y, with Y assigned as a -130-cm-I (=wsp in what follows) vibronic hole which builds to higher energy on X. The wSp mode was also identified from the spectrum of PBH-Q- (B BChl monomer, H BPheo, Q = quinone) produced by narrow line laser irradiation into P960 of PBHQ-.I8 The wsp feature in the (PBH-Q--PBHQ-) difference spectrum had been observed much earlier by Vermeglio and Paillotinlg under non-line-narrowing conditions; however, it was not assigned as a vibration. In addition, in ref 17 and 18 a weak but relatively sharp ( w 10 cm-’) zero-phonon hole (ZPH) coincident with AB was observed when AB is located in the vicinity of the low-energy shoulder (see below) of the P960 absorption profile for PBHQ. It was concluded that X correlates with this shoulder and, furthermore, that the 10-cm-’ hole is the ZPH of X. Hole X was assigned as the intramolecular zero-point level (utp)of P960. The ZPH width of N 10 cm-l yielded a P960* decay time of 1 ps at 4.2 K, which is in reasonable agreement with the 0.7 f 0.1 ps value measured in the time domain at 10 K.20*2’ The weak intensity of the ZPH relative to X (-1:lOO) was ascribed to moderately strong linear electron-phonon coupling (Huang-Rhys factor S 2) involving low-frequency protein phonon^.".'^ Arguments, based on the experimental gating employed and the AB dependence of the hole spectra, were presented17J8 for the PHB structure (including the ZPH) being an intrinsic property of the R C and not due to a chemical contaminant, e.g., stable PBH-Q- afforded by irradiation of PBHQ- in the presence of cytochrome. Observation of the same structure for P870 of Rb. sphaeroides would provide an even stronger argument against impurity since the cytochrome is absent and especially if the ZPH hole width were to yield a decay time for P870* in good agreement with the time domain value of 1.2 & 0.1 ps at I O K.20,2’ The results presented here provide definitive proof that the low-energy shoulder of P870 and P960 absorption profiles as well as the structure reported in ref 17 and 18 for the PHB spectra of P960 is intrinsic to the RC. The experimental systems employed for this study are the same as employed previously17J8except that the boxcar averager of the transient PHB system was replaced by a Stanford Research SR250 boxcar averager. The gate delay (triggered by the IO-ns laser pulse) for the boxcar was 2-3 ms, and the gate width was 150 MS. The AT (transmission) spectra were obtained by subtracting laser-on and laser-off spectra. AA (absorbance) spectra were
-
( 1 1) Boxer, S. G.; Lockhart, D. J.; Middendorf, T. R. Chem. Phys. Lett. 1986. 123. 476. (12) Boxer, S. G.; Middendorf, T. R.; Lockhart, D. J. FEBS L e f f .1986, 200. 237. ~. (13) Meech, S . R.; Hoff, A. J.; Wiersma, D. A . Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 9464. (14) Won, Y . ;Friesner, R. A. J . Phys. Chem. 1988, 92, 2214; 1989, 93, I.(11-17 __ . (15) Hayes, J. M.; Small, G. J. J . Phys. Chem. 1986, 90, 4928. (16) Hayes, J . M.; Gillie, J. K.; Tang, D.; Small, G. J. Biochim. Biophys. Acta 1988, 932, 287. (17) Tang, D.; Jankowiak, R.; Gillie, J. K.; Small, G. J.; Tiede, D.M. J . Phys. Chem. 1988, 92, 4012. (18) Tang, D.;Jankowiak, R.; Small, G. J.; Tiede, D.M. Chem. Phys. 1989, 131, 9-9. (19) Vermeglio, A,; Paillotin, G. Biochim. Biophys. Acta 1982, 681, 32. (20) Fleming, G. R.; Martin, J. L.; Breton, J. Nature (London) 1988,333, 190 (21) Breton, J.; Martin, J. L.; Fleming, G. R.; Lambry, J.-C. Biochemistry 1988, 27, 8276.
Letters WAVELENGTH (nm)
950
1000
1050
I
0 4 ,
0 3
LLI
z
p
s:
02
U
1
,/
“‘1
/
/ \
00 10400
lOSO0
10200
10000
860
900
,
05
9600
9800
940 I
B
ll800
116W
11400
11200
llOW
10800
10600
WAVENUMBERS (cm- 1 )
Figure 1. Primary donor state absorption at 4.2 K: (A) P960; (B) P870. Resolution = 4 cm-I. Insets are second-derivative spectra.
obtained by subtracting the logarithms of laser-off transmission spectra and the laser-on transmission spectra. It is important that readers comparing the AT and AA spectra (for the same AB) recognize that the AT spectrum would need to be “weighted” by the preburn transmission spectrum ( TPB)before it would mimic the AA spectrum. The approximate relationship AA = A T / T ~ B can be useful for this comparison. Only AT PHB spectra were reported for P960 in ref 17 and 18. By varying the gate delay, it was determined that the lifetimes of the charge-separated bottleneck state P’BHQ- for Rps. viridis and Rb. sphaeroides are 8 f 1 and 34 f 3 ms, respectively, at 4.2 K. These values are in good agreement with published values.12*22 Burn laser radiation (line width 0.2 cm-I) was provided by the Raman-shifted (H2 gas) output of an excimer-pumped dye laser. The excimer laser was operated at a pulse repetition rate of 16 and 20 H z for Rb. sphaeroides and Rps. viridis, respectively. Raman-shifted pulse energies utilized were 50.4 mJ (focused to either a 0.3cm-diameter spot or a 0.2 cm X 0.8 cm spot) and produced P870 and P960 transmission changes of less than 20%. By utilizing higher pulse energies, it was found that -80% bleaches could be attained for the samples utilized. All PHB spectra reported here are for AT changes less than 20% and forfresh samplesprepared by redissolving RC crystals in suitably buffered hosts. The spectra reported for Rps. viridis (glycero1:water glass, 0.1% LDAO detergent, 10 mM Tris, pH = 8.0) are very similar in structure to those obtained with poly(viny1 alcohol) (PVOH) host films (spectra not shown). The spectra reported for Rb. sphaeroides are for the g1ycerol:water glass (0.8% n-octyl-P-D-glucopyranosidedetergent, 10 mM Tris, 1 mM EDTA, pH = 8.0). The OD (optical density) of the samples utilized was less than 0.5 at the peak of the primary donor state absorption. Polarized PHB spectra have been obtained for both bacterial RC and will be reported on elsewhere.23 (22) Vermeglio, A,; Breton, J.; Paillotin, G.; Cogdell, R. Biochim. Biophys. Acta 1978, 501, 514. (23) Johnson, S . G.; Tang, D.; Jankowiak, R.; Hayes, J. M.; Small, G . J.; Tiede, D.M. To be published.
The Journal of Physical Chemistry, Vol. 93, No. 16, 1989 5955
Letters
I
,
9770
,
," ,
I
9790
9810
, , , I 9830 9850
1ww
850
830
870
890
910
lWB0
n
930
WAVELENGTH (nm)
llm
WAMNUNBEB (cm-1)
WAMNUUBW (em-1)
I
I
Figure 2. Hole-burned spectra for P870 at 4.2 K. Solid arrows locate A B = 9 18, 9 12, 9 I O , 9 IO, and 905 nm for spectra 1-4. All spectra are A T except spectrum 3b which is the AA spectrum corresponding to 3a. (See experimental details when comparing A T and AA.) Resolution = 4 cm-l. Dashed arrows in 2 and 3a locate wip satellite hole (see text). Two dashed arrows in 3b indicate approximate locations of w:p and wip satellites. Optical density at P870 maximum = 0.4 for all spectra. (See experimental details when comparing AT and AA.)
Absorption profiles of P870 and P960 are shown in Figure 1. The low-energy shoulder of P960 is more evident than for P870. (The second derivative of the P870 profile clearly reveals the shoulder.) Again, we correlate the shoulder with hole X of P960.17318Further evidence for this and for it being the origin band of P960 is presented later. Transient AT spectra (1, 2, 3a, 4) for P870 are shown for different AB values (solid arrows) in Figure 2. The ZPH coincident with AB becomes more pronounced as AB is decreased to an optimum value located near the maximum of the P870 shoulder (910 nm from the second derivative of the absorption profile) and less pronounced as AB is decreased from the optimum value. Such behavior (confirmed with several other AB values) was reported for P96O"J8 and is consistent with the theory of hole burning in the presence of moderately strong linear electron-phonon coupling and inhomogeneous line broadening.I5J6 An average of several scans (2-cm-l read resolution) for AB in the range 907-912 nm yielded a ZPH width of 8.5 f 2.0 cm-' (corrected for read resolution); see Figure 3B for a typical profile. This width yields a P870* decay time of 1.3 f 0.3 ps, which is in good agreement with the time domain value of 1.2 f 0.1 ps at 10 K.20,21At this point it is appropriate to mention that the same procedure yielded a P960* decay time of 0.8 f 0.1 ps. The ZPH for P960 is more difficult to detect since it is broader ( 1 3.0 f 1.5 cm-I); see Figure 3A for a typical profile. The close agreement between the PHB and time domain decay times is interesting since the ZPH width measures decay from zero point while the time domain experiments initially prepare P* vibrationally excited. If charge separation occurred to a significant extent prior to thermalization (at 10 K), one would not expect a g r e e m e ~ ~ t The . ~ ~ observed ~~~ agreement suggests that thermalization occurs on a subpicosecond time scale. This suggestion is consistent with the photon echo results of Meech et a1.26 Returning to Figure 2, one observes in spectra 2 and 3a a broader hole (indicated by the dashed arrow) which is displaced 1 IO cm-I to higher energy of the ZPH (AB). It is hole Y for P870. Spectrum 3b of Figure 2 is the AA
-
(24) Jortner, J. Biochim. Eiophys. Acra 1980, 594, 193. (25) Bixon, M.; Jortner, J. Faraday Discuss. Chem. SOC.1982, 74, 17. (26) Meech, S. R.;Hoff, A. J.; Wiersma, D. A. Chem. Phys. Lett. 1985, 121, 287.
/
0
800
850
900
950
10'00
WAVELENGTH (nm)
Figure 3. Hole-burned spectra at 4.2 K: (A) ZPH for P960, AB = 1019 nm, resolution = 2 cm-', OD = 0.4 at P960 maximum; (B) ZPH for P870, AB = 91 1 nm, resolution = 2 cm-l, OD = 0.4 at P870 maximum; (C) Q, region for Rps. uiridis, AB = 980 nm (solid single arrow), resolution = 8 cm-l, OD = 0.4 at P960 maximum; (D) Q, region for Rps. uiridis, A B = 1020 nm (solid single arrow), resolution = 8 cm-l. Double-headed arrow locates hole corresponding to 850-nm shoulder in ab-
sorption spectrum. spectrum that corresponds to 3a. As expected (for a sufficiently high P870 OD, see caption), definition of structure diminishes in conversion to AA. The two dashed arrows indicate the locations of the unresolved ut, and wf (one- and two-quantum) satellite holes. The vibronic hole structure associated with wspis more apparent for P960, frame A of Figure 4, principally because wsp(P960)is significantly higher than for P870 ( 15%) and the P960 absorption profile is narrower than the P870 profile (420 vs 475 cm-I). Three quanta of wsp 140 cm-l are apparent (see arrows). Figure 5 compares the observed absorption and PHB spectra with spectra calculated by using the theory of Hayes and SmallI5 suitably modified to include ws along with the lower frequency phonons (w,) considered earlier.P6J8The parameter values utilized are given in the caption. The reader is referred to ref 16 for details of the theory. Allowance is made for subpicosecond decay of the d (j 2 1 ) levels (vide supra) with the decay time proportional t7j-I (Fermi Golden Rule prediction with cubic intermolecular anharmonicity) and a decay time of 350 fs for w : ~ .In the absence of subpicosecond decay, the calculations predict that vibronic satellite ZPHZ7 associated with the wsp progression should be observed. Repeated attempts to observe such features under optimum line-narrowing conditions met with no success for both
-
-
(27) Hayes, J. M.; Fearey, B. L.; Carter, T. P.; Small, G.J. Int. Reu. Phys. Chem. 1986, 5 , 175.
5956
The Journal of Physical Chemistry, Vol. 93, No. 16, 1989
Letters
:
W 720
760
Bw
840
880
920
WAVELENGTH (nm)
Figure 4. Hole-burned spectra at 4.2 K. A: AA (1) and AT (2) spectra for Rps. uiridis, Ab = 1020 nm (first arrow), resolution = 8 cm-'. OD = 0.4 at P960 maximum. Second through fourth arrows locate 1-3 quantum w,, satellite holes. B: AA ( I ) and AT (2) spectra for R b . sphaeroides, Ab = 910 nm (arrow), resolution = 8 cm-I, OD = 0.2 at P870 maximum (low OD results in similar shape for AT and L LPHB ~ spectra).
P870 and P960. Subpicosecond decay is also suggested by the agreement between the decay time for P960* and P870* measured in the time domain and by hole burning (vide supra). In the calculations each member of the os,progression was assigned the same values for r i n h , w,, r, and S (see Figure 5 caption). The values of S and w, were constrained by the condition that 2Sw, 150 cm-l, the low-T Stokes shift for P960.28 The value determined for wSpis 150 cm-' with a Huang-Rhys factor (SSp) of 1. I. The Franck-Condon factors for the progression are given by exp(-Ssp)Sip/j!, j L 0. Considering that the theoretical model includes just two modes (wSp, a,), we consider the agreement between the calculated and observed spectra to be satisfactory especially since the disagreement on the low-energy tails of the two spectra is due mainly to the utilization of a Lorentzian profile in the theory for the site excitation energy distribution function (rinh). The use of a Lorentzian may also be responsible for both the hole and absorption calculated spectra being broader than the experimental spectra. Lorentzians were employed so that an analytic expression for the hole profile, which significantly reduced computation time, could be ~ b t a i n e d . l ~ The * ' ~ structure in the central region of the calculated PHB spectrum should not be strongly influenced by the employment of a Lorentzian. Comparable fits for the PHB and absorption spectra of P870 have been achieved with wSp = 125 cm-', S,, = 1.5, w, = 35 cm-l, S = 2.0, r = 50 cm-I, and r i n h = 130 ~ m - ' . ~ ~ The basic conclusion from our earlier work,16 which interpreted the unstructured hole spectra for P870 and P960 reported in ref 11 and 12, was that the spectra can be satisfactorily understood in terms of inhomogeneous broadening (resulting from RC to RC heterogeneity) and linear electron-phonon coupling without having to invoke ultrafast electronic relaxation of P*. The results reported here are consistent with this model since wSpis most reasonably assigned as an intermolecular mode (vide infra). In ref 16 a single
-
(28) Maslov, V . G.; Klevanik, A. V.; Ismailov, M. A,; Shuvalov, V. A. Dokl. Acad. Nauk. SSSR 1983, 269, 1217. (29) Tang, D.; Johnson, S . G.; Jankowiak, R.; Hayes, J. M.; Small, G. J.; Tiede, D. M. In Twenty-Second Jerusalem Symposium: Perspectives in Photosynthesis; Jortner, J., Pullman, B., Eds.; Kluwer Academic: Dordrecht, to be published.
p+++y 9600
9800
10000
10200
.oo
WAVENUMBERS
Figure 5. Calculated and experimental absorption and hole spectra for Rps. uiridis: upper frame, hole-burned spectra for AB = 1020 nm; lower frame, absorption. Parameters used for both calculations were F (onephonon profile width) = 50 cm-I, w, (mean phonon frequency) = 40 cm-I, S (Huang-Rhys factor) = 1.5, rmh (inhomogeneous line broadening) = 120 cm-I, wSp= 150 cm-I, and s,,= I . 1. First four overtones of wsp were included in the calculations. mean phonon frequency approximation was utilized with w, = 80 cm-I for both P870 and P960. This value is close to the mean of the w, and w,, values reported here. The value of S determined in ref 16 was 4.5 which yields Sw, = 360 cm-I. The latter value is about 1.5 times larger than the values of Sw, + SspwSpreported in this paper for P870 and P960. Thus, the contribution from r l n h to the unstructured spectra of ref 1 1 and 12 was underestimated in ref 16. The total linear electron-phonon coupling for P870 and P960 may now be classified as moderately strong since s s,, = 3. On the basis of their theory and calculations, Won and Friesner14 have suggested that the absence of a ZPH in the hole spectra of ref 11 and 12 is due to ultrafast electronic relaxations (5100 fs) of P* into a charge-transfer state manifold. Our ZPH data show that this is not the case for the zero-point level of P*. It is conceivable that the subpicosecond decay we impose on the dP(j I 1) levels might be due to a wSpmode number dependent relaxation of P* into a C T vibronic manifold. We are inclined not t o favor such an interpretation because of the agreement between the time domain and hole-burning (ZPH) values for the decay times of P870* and P960* and because there are no time domain data that support ultrafast relaxation for P* prior to formation of P+BH-. We hasten to add, however, that the theory of Won and Friesner reduces to oursis in the limit where electronic relaxation of P* into the CT vibronic manifold becomes negligible. In this limit their theory should be capable of accounting for the data presented here. The assertion in ref 18 that a charge-transfer (CT) state, lying -300 cm-' higher in energy than the low-energy shoulder of P960 ) figures importantly in the PHB (Le., the origin ( u ! ~band), spectrum is brought into question by the present results. This assertion was mainly based on a AT PHB spectrum (Figure 3) for which the OD of P960 was sufficiently high (0.6) t o render
+
Letters the AT spectrum an unfaithful representation of the A4 spectrum at higher energies (confirmed by measurements on the same but diluted sample).30 The present work establishes that the principal features of the PHB spectra can be accounted for without invoking a C T state although the possibility that a weakly absorbing C T state does contribute to the high-energy side of the primary donor state absorption profile cannot be excluded. We consider next the question of the nature of the wsp mode for P870 and P960. We assign it as a special pair intermolecular Franck-Condon marker mode for the following reasons: recent hole-burned spectra for the antenna Chl a and b for photosystem I have shown that31 the intramolecular S factors for all modes are 10.04. With these data and the fluorescence excitation spectra of BChl a reported in ref 31, the S factors for BChl a modes possessing a frequency of 1200 cm-' can be estimated to be 10.02, in sharp contrast to the value of S,, = 1.1 for us,reported here. Furthermore, no intramolecular BChl a, Chl a, and Chl b monomer modes with frequencies close to the ws, values reported here have been observed in ref 3 1 and 32. Since the us,values reported are for P*, it would be worthwhile to attempt resonance Raman or coherent four-wave mixing experiments on P870 and P960 in order to determine the corresponding ground-state frequencies. More important for future work, however, are the questions of the dynamical nature of the oSp mode and a possible role for the geometry change associated with wsp, which accompanies electronic excitation of P, in the primary charge separation process. Recently, nonphotochemical hole-burning experiments on the antenna complex of the green algae Prosthecochloris aestuarii have proven that there is a high degree of site excitation energy correlation between different exciton components of a subunit characterized by strong excitonic interaction^.^^ It is with this in mind that we consider the PHB spectra for Rps. uiridis shown in the bottom frame of Figure 3. Notice that the centroid of the P960 hole profile for the higher AB value is lower in energy than the centroid for the lower AB value. This AB "tracking" behavior has been studied for many AB values, for the glass-detergent system considered here as well as glycerol-HzO/NGP, PVOH/LDAO, and PVOH/NGP. A full account of these data will be given elsewhere.29 As reported in ref 18, the tracking ceases when AB is decreased below a maximum value (Amax) which defines the excitation frequency at which line narrowing is effectively lost. For AB < A,, the P960 hole profile is invariant.ls The gradual loss of line narrowing and, concomitantly, tracking as AB approaches A, from below is due, in part, to the accompanying increase in the probability for multiphonon excitation relative to (30) Also, experiments performed on the undiluted sample yielded PHB spectra that agree with the ATspectra obtained earlier,"*'*yet result in AA spectra that are consistent with those presented here. The feature in Figure 3 (which should be labeled as a AT spectrum) of ref 18 assigned to a CT state is actually the satellite hole. (31) Gillie, J. K.; Small, G. J.; Golbeck, J. H. J . Phys. Chem. 1989, 93, 1620. (32) Renge, I.; Mauring, K.; Avarmaa, R. J . Lumin. 1987, 37, 207. (33) Johnson, S.G.; Small, G. J. Chem. Phys. Lett. 1989, 155, 371.
The Journal of Physical Chemistry, Vol. 93, No. 16, 1989 5957 zero-phonon excitation.16,'8In addition, the probability for exciting overlapping wsp progression members also increases. Because rinh for the PVOH hosts is greater than for the glass hosts (-470 vs 420 cm-I), the tracking is more pronounced for the former29 as expected.I6 Of particular interest is the observation that for the four host-detergent systems only one feature of the Qyregion of the transient spectra lying higher in energy than P960 exhibits an observable dependence on AB. It is the hole located by the double arrow near 850 nm in the lower frame of Figure 3. The centroid of this hole ( b l e a ~ h ~tracks ~ , ~ ~AB) in a manner similar to that described above. This is the reason why the 850-nm hole is best resolved for higher values of AB, as exemplified by the lower frame of Figure 3. The bleach (hole) at -850 nm corresponds to an absorption feature at -850 nm (which appears as a shoulder on the 830-nm BChl monomer band18) which has been assigned by Vermeglio et al.I9 to P+, the upper dimer component of the Qr transition of the special pair. Our results indicate that there is a significant degree of positive correlation between the site excitation energy distribution functions of the 850 and P960 bands. Because of the aforementioned studies on P. aestuarii and because we are not aware of any line-narrowing studies on molecular systems which establish any correlation for excited states of different electronic parentage, we suggest that our results provide support for the 850-nm band assignment of Vermeglio et al.I9 Furthermore, it must be emphasized that the features in spectra C and D of Figure 3 which lie to higher energy of 850 nm do not exhibit a measurable AB d e p e n d e n ~ e .These ~ ~ features are due to electrochromic shifts of the BChl and BPheo monomer transitions produced by formation of P'BHQ-. Further experiments are required before any inferences from the AB independence can be drawn. In conclusion, the results presented here reveal the origin of the homogeneous broadening associated with the primary donor state absorption profile of bacterial RC, identify a special pair intermolecular marker mode (the dynamical nature of which has yet to be determined), further the utility of PHB for determining electron-transfer dynamics from zero point, and establish the potential of site excitation energy selection for the determination of correlation between different R C states.
-
Acknowledgment. Ames Laboratory is operated for the U S . Department of Energy by Iowa State University under Contract No. W-7405-Eng-82. The research was supported by the Director for Energy Research, Office of Basic Energy Science. The research at Argonne Laboratory was supported by the Division of Chemical Sciences, Office of Basic Energy Sciences, U S . Department of Energy, under Contract No. W-31-109-Eng-38. (34) Breton, J. Biochim. Biophys. Acta 1985, 810, 235. (35) Breton, J.; Vermeglio, A. In Photosynthesis: Energy Conuersion by Plants and Bacteria; Govindjee, Ed.; Academic Press: New York, 1982; Vol. 1, p 153. (36) Jankowiak, R.; Tang, D.; Small, G. J.; Seibert, M. J . Phys. Chem. 1989, 93, 1649.
