Attosecond shot noise

Complete text of publication follows. The spatio-temporal characteristics of electromagnetic radiation can be monitored on the basis of first order interference or higher-order correlation techniques, by measuring the moments of the fields strength at different space-time points. In most cases the information is mediated by photoelectrons whose current - owing to the granular nature of electricity - is subject to particle noise. An attosecond pulse train (or an isolated pulse) is usually viewed as an extreme signal stemming from a nonlinear dipole moment associated to bound-continuum-bound electron transitions, containing very high-order harmonics of the exciting radiation. As long as this source is approximated by an expectation value of the otherwise quantal dipole moment, the resulting attosecond radiation is automatically set to be a many-mode coherent radiation in the strict sense of quantum electrodynamics. Then, according to Glauber's quantum coherence theory, all the moments characterizing the correlations in such signals would factorize, giving no possibility to overcome the usual shot-noise limit. The question naturally arises, whether to what extent can attosecond pulses be considered as simply broad-band classical signals? If they are not classical , then by what means could one sample them to study their spatio-temporal structure? Instead of first-order interference, we propose to carry out intensity-intensity correlation measurements on split nano-pixels, which are not sensitive against phase distortions. By changing the tilt angle of the impinging attosecond pulses with respect to the detector array, the modal structure of the pulses could be monitored by measuring the current-current correlations, and this would yield a delayed coincidence curve. We have modelled the attosecond pulses by many-mode quantum-mechanical phase eigenstates, and a contrast ratio 4/3 relative to the Poissonian shot noise level of the coincidence curve has been derived. Acknowledgements. This work has been supported by the Hungarian National Scientific Research Foundation, OTKA, Grant No. K73728.