No abstract

No abstract

We demonstrate that solid-state material when microstructured on a nanometer scale leads not only to discrete electronic energy levels but to selective coupling of the electronic states via local operators due to acoustic-phonon or dipole interaction. As a result the system can be designed to allow for multistable dissipative quantum dynamics realized by only a few electronic degrees of freedom. Our model sheds some new light on the problem of fundamental constraints on information-processing systems.

Obermayer, Mahler, and Haken Reply: In our Letter we have discussed a microscopic model of an optically controlled bistable element. Control here means that we can induce an electron transfer process from the given initial state to the desired final state. The question raised by Saito refers to the conditions under which the-unwanted-back reaction can be avoided.In principle, optical selectivity can be realized in real space and/or in frequency space. Clearly, for a submicroscopic element only the latter possibility remains, which has thus been the basis of our investigations (though possibly not made clear enough in the Letter).The optically induced single-electron transfer process is a stochastic process 1 : Its nature is most suitably characterized in terms of an error probability per preparation step. It would seem that this probability could be made as small as one likes in the limit of infinitely long laser pulses, 2 but then, besides the finite lifetime of the metastable states, the competing back transfer comes into play, and possibly also off-resonance transitions within other bistable elements not supposed to switch (i.e., more and more processes on longer time scales become relevant).In the present context it can be safely assumed that the laser linewidth is small compared with all transition-frequency separations, which, by the structural constraints as given, should be in the range of some millielectronvolts. The width of the optical transitions is (in our model) due to the spontaneous decay of the transient state (i.e., in the order of 10 ~6 eV) and is therefore orders of magnitude smaller than the transition frequency separation. Of course, other mechanisms not considered here might contribute. In particular, manufacturing errors could severely impair the selectivity; for an ensemble of switches this would lead to inhomogeneous line broadening. Future developments will show whether such limitations can be avoided.

No abstract