Yang et al. Reply:We are glad to respond to the interesting Comment by Kash [1] on our Letter [2]. Our model, described in detail in Ref. [3], states that while the exciton absorption profile is obtained from the spatial inhomogeneous broadening (distribution) of the optical band gap of a quantum well material, the luminescence profile mirrors the inhomogeneous broadening of classically localized excitons, on the assumption that there are no transitions between localized states.In order to discuss Kash's comment on the dynamics of exciton relaxation, it is important to distinguish two different processes: (i) hot excitons relax by losing kinetic energy (see, for example, [4]) to occupy local minima within a few tens of picoseconds or less, whereas (ii) classically localized excitons relax further by phonon-assisted hopping or tunneling between minima on a much longer time scale. Our Letter discussed the energy distribution of localized exciton states and assumed that the latter processes could be ignored. The decay time for exciton luminescence is an order of magnitude larger than that of the first process [4], so that relaxation into local minima is essentially complete before significant luminescence occurs. Kash points out that spectral diffusion is observed in certain samples during time-resolved luminescence, i.e., the Stokes shift between absorption and emission peaks is itself a function of time. He deduces that the near universal behavior explained by our model must provide only an approximate description, because the exciton relaxation is not completed before emission occurs.It is certainly the case that, as stated by Kash, samples where slow spectral diffusion is dominant will not conform to our model. The successful fit of our model to data from a wide range of samples suggests that in many cases the effect of these slow processes is in fact small, and that the shape of the emission spectrum therefore is not usually dominated by dynamical effects. Because of the lack of relevant time-resolved spectroscopic information on large numbers of samples, it is difficult to predict which samples will show a substantial slow-relaxation effect and hence not conform to our predicted S/W relation. It is, however, possible to make a comment on this point, based on existing spectroscopic data. Reviewing Fig. 2 in our Letter [2], samples of small linewidth Wshow Stokes shifts S which are uniformly slightly higher than are predicted by our model, while data for samples with larger linewidth agree well with the prediction. We offer a tentative interpretation of this as follows. A small "excess" Stokes shift is a marker for slow-relaxation processes, which can presumably occur more easily in systems with smaller inhomogeneous broadening, because the potential wells trapping the excitons are shallower.We would also like to comment on the relevance of the resonant Rayleigh scattering technique, which has been used by Hegarty and Sturge [5] as a probe of exciton dynamics. Resonant Rayleigh scattering from an exciton is pro...

No abstract

We have measured the differential cross sections for coherent Compton photon scattering in the reaction 4 He(/,y) 4 He at laboratory angles of 24°, 30°, 45°, and 60° with an average laboratory photon energy of 320 MeV, at 22° with 358 MeV, and at 30° with 260 MeV. These measurements are the first unambiguous test of the A-hole formalism for this reaction near the peak of the cross section for the A resonance. The results are compared to theoretical calculations in the isobar-hole model. Agreement is good for data at the energy corresponding to the A peak.PACS numbers: 25.20.Dc, 25.10,+s An important topic in intermediate-energy physics is the creation and propagation of the A(1232) resonance inside the nucleus. Compton scattering creates the A deep inside the nucleus, in contrast to pion scattering and photopion production which are dominated by interactions at the nuclear surface. Additionally, there are no initial-or final-state interactions to complicate the interpretation of the Compton-scattering data.tion. Koch, Moniz, and Ohtsuka 2 use a "spreading potential," whose parameters are adjusted to give a good fit to pion elastic-scattering data, to account for the more complicated decay channels. The model then gives predictions for /r° photoproduction and elastic photon scattering with no further adjustable parameters. Good elastic photon-scattering data over the whole angular range will help our understanding of the A resonance in li blocking of A decay or coupling to more complicated channels through absorption in the NN channels. The propagator can then be used to describe a variety of reactions. This has been applied with considerable success to pion-nucleus scattering and coherent /r° photoproduc-and Pb targets, but the data may have included a large contribution from incoherent channels due to insufficient photon energy resolution. The amount of incoherent scattering is consistent with the estimates calculated by Arenhovel. 6 We have previously published results for TOP VIEW 1922 FIG. 1. Schematic diagram of the experimental setup.