Limitations on building single-photon-resolution detection devices

Abstract

Single-photon resolution (SPR) detectors can tell the difference between incoming wave packets of n and n + 1 photons. Such devices are especially important for linear optical quantum computing with projective measurements. However, in this paper I show that it is impossible to construct a photodetector with single-photon resolution when we are restricted to single-photon sources, linear optical elements and projective measurements with standard (non-photon-number discriminating) photodetectors. These devices include SPR detectors that sometimes fail to distinguish one-and two-photon inputs, but at the same time indicate this failure. Recently, several groups have demonstrated singlephoton sources in quantum-dot microcavities and magneto-optical traps [2,3,4]. The importance of these experiments for linear optical quantum computing can hardly be overstated, even though the road to highvisibility interference between independent sources is still long and arduous. Furthermore, Hockney, Dowling and I assessed the quality of single-photon sources by defining the suitability of a source with respect to a given application [5]. In this paper, I turn my attention to the other essential component of LOQC: a photodetector with single-photon resolution.Most detectors that are currently used in optical quantum communication and computation experiments cannot tell the difference between one or more photons. Single-photon resolution (SPR) detectors are devices that can distinguish between wave packets containing n and n + 1 photons [6,7]. They are important because the LOQC research program relies heavily on projective measurements, which in turn involve photon-number measurements [8,9]. We therefore need a way to efficiently distinguish between different photon number states. Very often, the output of an optical gate is significantly different when postselected on a single-photon or a two-photon detection outcome. Perhaps the most dramatic example of this is the teleportation experiment by Bouwmeester et al. [10], in which the lack of single-photon resolution reduces the non-postselected fidelity of the teleported out- * Electronic address: Pieter.Kok@jpl.nasa.gov put state to a value lower than the clasical limit [11,12]. As a consequence, postselection was needed. Recently, Pan et al. modified the pair-production rates of the two down-converters in this experiment, such that the nonpostselected output fidelity surpasses the classical limit [13].However, it may not always be possible to modify the quantum gate in such a way that non-SPR detectors can be used, and we really would like to have cheap, reliable, and efficient SPR detection devices. But the current experimental detectors with single-photon resolution are expensive and operate at low temperatures [14]. In this paper, I will therefore investigate whether we can build an SPR detector with single-photon sources, linear optics and ordinary photodetection without single-photon resolution. It turns out that such a device is impossible.Before I proceed with the proof o...

Photon-number resolution using time-multiplexed single-photon detectors

“…The net result was the creation of N = 16 pulses of approximately equal amplitude. Unlike earlier approaches involving detector arrays [11,12,13,14], this technique requires only two detectors, and it avoids the need for optical switches and a storage loop [15,16,17].…”

Towards photostatistics from photon-number discriminating detectors

“…Standard photodetectors can measure only the presence or absence of light (single-photon sensitivity), and generally do not have the capability of discriminating the number of incoming photons (single-photon number resolution). There have been suggestions of accomplishing single-photon resolution using many singlephoton-sensitive detectors arranged in a detector array or detector cascade [20,21,22,23]. For example, the VLPC (visible light photon counter) is based on a confined avalanche breakdown in a small portion of the total detection area and, hence, can be modeled as a detector cascade [24].…”