The Journal of
Physical Chemistry
Q Copyrighr, 1992, by the American Chemical Society
VOLUME 96, NUMBER 24 NOVEMBER 26, 1992
LETTERS Possibility of Self-organization in Photosynthetic Light Harvesting Antennae Karl K. Rebane Institute of Physics of the Estonian Academy of Sciences, Riia Street 142, Tartu, Estonia (Received: April 16, 1992; In Final Form: September 28, 1992)
Molecules in light-harvesting antenna form an inhomogeneous and to a certain extent random structure. Therefore, the energy-transfer routes may end in traps, Le., on molecules out of resonance with the potential acceptors. The electronic excitations stay on these molecules for a long time. This may result in photochemical changes and the molecules cease to be traps: their spectra orland the orientation of the dipole moments change in such a manner that the molecules become either “normal” acceptors-donors or “outsiders” of the energy-transfer process.
One of the problems of photosynthesis is what are the mechanisms and kinetics of energy transfer. Quite a number of models have been studied. As far as I know, all of them are models for lifeless world, like, e.g., models to describe luminescence of crystals. The destination of energy transfer in photosynthesis is to build and support living matter. It seems quite interesting to consider also some possibilities of self-organization, which is a very characteristic feature of life. It is not excluded that self-organization starts already at such an early stage of life as energy transfer in light harvesting antenna. Two extreme situations may be considered for the energy transfer. The first is random migration. The second is a strictly cotrelated structure and spatially directed energy transfer. The second one should be faster in energy transport and consequently should have higher yield. Further, in the second case the antenna may be larger, because a well-channeled and fast transfer takes the excitation energy over long distances to the photosynthetic unit before it is emitted as light or otherwise lost. But a system of the second kind may be both, in construction and in energy transfer, vulnerable to rather small defects and perturbations. A system of the first kind, the one with randomness in its construction, can be more stable against flaws in its structure and various happenings. But the first system may be slower in energy
transfer, which means more losses of energy via emission of light or nonradiative processes. On the other hand, if the randomness provides the ability for self-organization, the performance in the transfer process itself can be considerably improved. Nature starts with rather random systems, which in course of action “learn” to become more effective. Self-organization may provide also increase of stability. In this note I would like to call attention to a rather simple possibility of self-organization of the molecular system of the light-harvesting antenna. Such a mechanism of “self-education” could be photochemical transformations of the molecules in the course of energy transfer analogous to photoburning of spectral holes, but in addition to some spectral selectivity, high site selectivity should take place too. That mechanism can eliminate traps for excitation energy on its route of random migration to the reaction center. We suppose that the spectral bands of absorption and luminescence are bell-shaped, continuous, and without sharp-line structure. It is not difficult to take into account sharp lines, e.g., zero-phonon lines present usually in low (liquid helium) temperature spectra of impurity molecules.’ The mechanisms proposed below will also work and their influence should be even stronger in case the spectra are built of sharp lines.
0022-365419212096-9583$03.00/00 1992 American Chemical Society
9584 The Journal of Physical Chemistry, Vol. 96, No. 24, 1992
The energy of electronic excitation acquired by an antenna molecule via absorption of a photon starts migrating in a system of more or less similar molecules. The energy transfer from a molecule to the next one is supposed to be fast as compared to the migrating excited electronic state’s lifetime of the total antenna. However, on the other hand, the time spent at one particular molecule is sufficient for some vibrational relaxation to take place. We suppose that the transfer mechansim is a resonant one, e.g., the Forster-Galanin mechanism, but complete vibrational relaxation to the thermal equilibrium is not obligatory. The presence of coherent or hot transfer processes2 does not change the basic picture, provided the relaxation is not too small. Somewhere in the middle of the antenna is the reaction center, a certain structure of a pair of chlorophyll molecules with its specific surrounding, which, as soon as it has caught the excitation, performs the next step of photosynthesis, Le., utilizes the excitation energy to separate the elementary electrical charges. An antenna comprises a body of chemically identical molecules bound in complexes. The physical conditions are slightly different and the spectra of absorption and luminescence of the chromophores in the antenna are inhomogeneously broadened. The energy of the purely electronic and vibronic transitions varies from chromophore to chromophore. As long as the energy transfer is vibronic (electron-vibrational), the Franck-Condon principle governs the transfer probabilities and the overlapping of the luminescence spectrum (including the hot luminescence part) of donor with the absorption spectrum of the acceptor is the measure of the number of populated vibronic levels of donors, from which the energy can be resonantly given to an acceptor. This is also a measure of the efficiency of the transfer. Further, as long as some vibrational relaxation is present, the luminescence of donor is shifted to longer wavelengths in comparison to the acceptor’s absorption even if their electronic energies are equal. This shift increases with the amount of relaxed vibrational energy. If the relaxed energy is large in comparison with the sum of absorption and luminescence bandwidths, the overlap becomes small and the transfer is strongly restricted. The overlap increases when the purely electronic transition energy of the acceptor (e.g., because of inhomogeneity) is smaller than that of the donor and transfer goes faster. In the presence of notable inhomogeneous shift, the excitation is preferentially given to that particular acceptor in the donor’s surrounding whose absorption band has the largest shift to red. The probability of the opposite transition, from the excited molecule back to the initial donor, is in addition to the Stokes shift, suppressed by the inhomogeneous shift of the previous acceptor’s luminescence to red. The transfer obtains the character of a process directed along a line of molecules arranged in the order of the most rapidly decreasing electron energies. If the absorption spectrum of the reaction center lies sufficiently to the longer wavelengths and, therefore, has a good overlap with the luminescence of the antenna molecules, the last step of the transfer to the center is fast. Because of a large overlap it may even be so efficient that energy could be transferred over distances which cover many neighbor molecules. The whole antenna may be divided into an inner and an outer circle. As soon as the excitation appears in the first one, it will be directly and fast transferred to the center. In the outer circle, which enlarges the harvesting area, the energy has to migrate out of it first before entering the inner circle. The migration may be time and energy consuming. Self-organization, which improves the efficiency of the outer circle, is welcome. Energy migrating in the inhomogeneously broadened states of chromophores in the outer circle may be trapped: its path may end at a molecule, whose red shift is so strong that all the available nearby acceptors are out of resonance (no overlapping) and migration stops. Here is the point where self-organization may step in: as long as further transfer is strongly restricted, the excitation stays at this molecule very long compared to the normal transfer times. During this long waiting time the probability accumulates for some photochemical reaction in the molecule or its surroundings and it can take place bringing along changes of the spectra of the trap molecule. This may well detune
Letters the resonance that brought excitation to this particular trap molecule. For example, this is the case when the absorption band of the phototransformed trap molecule shifts considerably to shorter wavelengths and the overlap with the luminescence of potential donors drops well below limits of effective competition with other potential acceptors in the vicinity of the donors. If the first trapped excitation does not perform the photochemistry, the following ones will. A trap molecule, which was a normal acceptor and could not work as donor, is converted into a very bad acceptor and becomes an outsider of the migration. The probability for a photochemical transformation is approximately proportional to the time that the excitation spends at a particular molecule. It means that a trap molecule on a busy road of migration will be burned out sooner than a trap in a quiet street. We can see here a self-organizing mechanism, clearing away the traps from the pathway of the electronic excitations through the inhomogeneous body of chromophore molecules in the outer circle to the inner circle and to the reaction center. What happens under strong illumination? Should not all the antenna molecules then be turned into outsiders? Indeed, the reaction center cannot accept the next excitation before transformation of the previous one is completed. A number of excitations may be rejected and they have to continue migrating and initiating photochemistry outside the reaction center. Fast degradation may be avoided if the rejected excitations remain migrating in the inner circle. The electronic energies of the molecules in the vicinity of the reaction center, Le., in the inner circle, may have quite strong red shifts caused by nearby presence of the center. Excitations will be captured in the inner circle. Further, we assume that molecules close to the center are more stable to photochemistry or that the photochemistry works and causes many changes, but in the inner circle it does not matter. The second assumption seems reasonable. The red shifts determined by position (not far from the reaction center) are large enough to preserve overlapping and the ability to accept energy from the outer antenna area even after the photochemically induced shifts have taken place. As soon as the reaction center becomes active again, the excitation waiting in a migration trap will be accepted by it. Energy levels of the reaction center are still more red-shifted and one-step fast transfer is possible. What should be observed, if this mechanism really works? Firstly, the efficiency of a newly born antenna should increase during some period of time after it is exposed to illumination. After a longer working time, efficiency may gradually fade away, because the integrated time of being excited becomes long even for the normally transferring molecules. Secondly, the antenna molecules must show photochemical changes. Experiments confirm the presence of sensibility to persistent spectral hole burning in systems of chlorophyll molecules in low-temperature solids3 and in photosynthetic units.4 Thirdly, the product of trap removal-phototransferred antenna molecules-should be present, their number increasing initially with the extent of “learning” afterwards with the amount of molecules worn out in long service. Up to now we agreed that the photochemical changes are irreversible. Self-organization will work also if reverse or almost reverse reactions are present. Of course, these reactions should not be too fast. A proper amount of reversibility makes the model even more flexible-the antenna would last longer. An almost reverse reaction may change an outsider into a normal acceptor and normal donor. Further, if a considerable part of the molecules is converted to be outsiders, they may form a new hierarchy of energy-transferring molecules and establish new effective energy routes. Reversible mechanisms may work creating (and reversing again) photoproducts, which are quite difficult to distinguish from the educts. If in place of the energy differences or in addition to them an inhomogeneous distribution of the transition dipole moments is present, the self-organization procedure would work essentially in the same way. The main question seems to be whether the Stokes and inhomogeneous shifts are large enough to really influence the transfer rates at photosynthetic temperatures via the overlapping of spectra. Even when they are too small to guarantee complete trapping,
J. Phys. Chem. 1992, 96,9585-9587 the effect of a longer stay at bad donors remains and those will be preferably “burned out”. What is presented above is a rather speculative idea. Nevertheless it might not be useless if it stimulates one to look for features of self-organization in photosynthesis. New possibilities for speculations on kinetic equations and, in particular, for computer simulations, are also opened. Let us note in conclusion that not only the photosynthetic units but also other energy-transferring systems with proper donor, acceptor, trap, and luminescence quenching centers properties may have the abovedescribed ability to self-organize faster energy-transfer pathways to the quenching centers.
9585
Acknowledgment. I am grateful to Professor Josef Friedrich and Dr. Alexander Rebane for discussion and valuable remarks.
References and Notes (1) Rebane, K.K.Impurity Spectra of Solids; Plenum: New York, 1970. ( 2 ) Hizhnyakov, V. V. Phys. Starus Solidi 1976, B 76, K69. (3) Avarmaa, R.; Mauring, K.;Suisalu, A. Chem. Phys. Lett. 1981, 77, 88. Avarmaa, R.; Renge, I.; Mauring, K. FEBS Lett. 1984, 167, 186. Avarmaa, R. A.; Rebane, K. K.Sov. Phys. Usp. 1988, 154, 433.
(4)Gillie, J. K.;Lyle, P.A.; Small, G.J.; Golbeck, J. H. Photosynth. Res.
1989, 22. 233. Jankowiak, R.;Tang, D.; Small, G.J.; Seibert, M. J . Phys. Chem. 1989, 93, 1649. Tang, D.;Jankowiak, R.; Seibert, M.; Small, G. J. Photosynth. Res. 1991, 27, 19.
New Pattern of Volume Phase Transition in Polymer Gels Cross-Linked by Bifunctional Monomer with an Ionizable Group Seiji Katayama,* Fumiko Yamazaki, and Yukio Akahori University of Shizuoka, School of Pharmaceutical Sciences, 52-1, Yada Shizuoka, 422 Japan (Received: July 16, 1992; In Final Form: October I , 1992) Volume behaviors of 2-(acry1oyloxy)ethylacid phosphate (AEAP)/acrylamide (Am) and 2,2’-bis(acry1amido)aceticacid (BAA)/Am copolymer gels immersed in acetonewater mixtures were examined, where AEAP and BAA were bifunctional monomers with an ionizable group. It was found that the equilibrium volume of the gels decreased with an increase in AEAP (or BAA) and that the observed discontinuous volume change at the transition point decreased with an increase in AEAP (or BAA), after which the transition point shifted toward a higher level of acetone composition. This pattern of volume phase transition differed from those observed for conventional cross-linked gels ionized along linear polymer chains. The new pattern could be considered consistent with a structural specificity of the gels which were produced by synchronous introduction of an ionizable group and a cross-link group.
Introduction It is well-known that polymer gels with an ionizable group’ undergo a reversible, discontinuous volume change upon changes in solvent composition, temperature, pH, and salt cor~centration.~-~~ This behavior is a volume phase transition between a swollen gel and a shrunken gel and has been theoretically considered as one of the fmt-order phase transition^.^,^ In recent years, volume phase transition, which is regarded as a mechanochemical property of polymer gels, has become of interest because of its scientific and technological importan~e.~ According to a number of experimental studies of volume phase transition,’-’O most changing patterns are characterized by a swollenshrunken behavior with a discontinuous volume change, except isopropylacrylamide gels which show a changing pattern with a reentrant curve profile.8 While investigating various patterns of volume phase transition, a new pattern of volume phase transition was incidentally found for a 2-(acryloy1oxy)ethyl acid phosphate (AEAP)/acrylamide (Am) copolymer gel, where AEAP is a specific bifunctional monomer having an ionizable group. The
purpose of the present experiment is to thus demonstrate clearly the new pattern of volume phase transition observed exclusively for such structurally specific gels and to consider the theoretical factors involved in determining the new pattern of volume phase transition. For this purpose, the following gel samples were prepared: AEAP/Am copolymer gel, 2,2’-bis(acrylamido)acetic acid (BAA)/Am copolymer gel, and Am/AEAP copolymer gel as a control.
Experimental Section
Acrylamide (A), N,N’-methylenebisacrylamide (Bis), and 2,2’-bis(acrylamido)acetic acid (BAA) were commercially obtained and were used without further purification because of adequate purities (>99.0%). 2-(Acryloy1oxy)ethyl acid phosphate (AEAP) was obtained from Daihachi Co. Ltd. The sodium salt of BAA was prepared by reacting BAA with sodium carbonate. AEAP/Am copolymer gel samples were prepared as follows. Variable amounts of AEAP (3200, 1500,600,210, 120, and 60 mg) and a given amount of acrylamide (1 g) were dissolved in distilled water to a final volume of 20 mL. AEAP/Am copolymer gel samples (A-F) were then prepared by radical copolymerization of the solutions at 50 OC for 1 h, after adding ammonium perCH2=CH-C-O-CH2-CH2-O-~-O-CH2-CH2-O-C-CH=CH~ sulfate (initiator, 50 mg). BAA/Am copolymer gel samples were OH also prepared by the same procedure as mentioned above. Variable amounts of BAA (880,480, 280, 120,40, 12, and 4 mg) and a given amount of acrylamide (1 g) were dissolved in distilled water 2-acryloyloxyethyl acid phosphate (AEAP) to a final volume of 20 mL. BAA/Am copolymer gel samples (A-G) were then prepared by solution copolymerization. Similarly, control Am/AEAP gel samples (A-E) were prepared by copolymerization of solutions containing a given amount of AEAP (0.25 M) and variable amounts of acrylamide (1000, 500, 250, 100, and 50 mM). The prepared gel samples were all washed in CH,=CH-C-NH-YH-NH-C-CH=CHz distilled water for a few days and immersed in acetone-water COOH mixtures of desired compositions. After achieving equilibrium, the gel diameters were measured and the volumes then estimated 2,2’-bis(acrylamido)acetic acid (BAA) by cubing the diameters.
8
8
51
8
9
0022-365419212096-9585$03.00/0 0 1992 American Chemical Society
