The heat engine, a machine that extracts useful work from thermal sources, is one of the basic theoretical constructs and fundamental applications of classical thermodynamics. The classical description of a heat engine does not include coherence in its microscopic degrees of freedom. By contrast, a quantum heat engine might possess coherence between its internal states. Although the Carnot efficiency cannot be surpassed 1-3 , and coherence can be performance degrading in certain conditions 4-9 , it was recently predicted that even when using only thermal resources, internal coherence can enable a quantum heat engine to produce more power than any classical heat engine using the same resources 10,11 . Such a power boost therefore constitutes a quantum thermodynamic signature. It has also been shown that the presence of coherence results in the thermodynamic equivalence of different quantum heat engine types 10,12 , an effect with no classical counterpart. Microscopic heat machines have been recently implemented with trapped ions 13,14 , and proposals for heat machines using superconducting circuits 15,16 and optomechanics 17,18 have been made. When operated with standard thermal baths, however, the machines implemented so far have not demonstrated any inherently quantum feature in their thermodynamic quantities. Here we implement two types of quantum heat engines by use of an ensemble of nitrogen-vacancy centres in diamond, and experimentally demonstrate both the coherence power boost and the equivalence of different heat-engine types. This constitutes the first observation of quantum thermodynamic signatures in heat machines. 2A quantum heat engine consists of a microscopic system, or an ensemble of such systems, whose internal state can be a coherent superposition of energy states. The engine cycle consists of a sequence of operations (strokes), which include the interaction of the system (or part thereof) either with a thermal bath (cold or hot), or with an external classical/semi-classical field responsible for work extraction. Interactions with the thermal baths act to change the populations of the energy states of the heat engine incoherently, in contrast to the field, which changes the populations coherently. Fig. 1 schematically presents three basic quantum heat-engine types: continuous, two-stroke and four-stroke, which differ by the ordering of the different strokes. Of these types, the four-stroke engine bears the strongest resemblance to macroscopic classical engines such as the Otto engine. It can be described (classically) by a two level system undergoing a four part cycle, illustrated in the top panel of Fig. 1a, consisting of alternating couplings to the hot and cold baths, interspersed with couplings to the work reservoir, whose effect is to change the spacing between the levels. It can be shown that this dynamics is equivalent to classical swap operations in a multilevel system 10 (multilevel embedding), as shown in the middle panel of Fig. 1a (taking U = swap). In a quantum heat engine, the oper...
Broadband quantum memories, used as temporal multiplexers, are a key component in photonic quantum information processing, as they make repeat-until-success strategies scalable. We demonstrate a prototype system, operating on-demand, by interfacing a warm vapour, high timebandwidth-product Raman memory with a travelling wave spontaneous parametric down-conversion source. We store single photons and observe a clear influence of the input photon statistics on the retrieved light, which we find currently to be limited by noise. We develop a theoretical model that identifies four-wave mixing as the sole important noise source and point towards practical solutions for noise-free operation.
Broadband quantum memories hold great promise as multiplexing elements in future photonic quantum information protocols. Alkali vapour Raman memories combine high-bandwidth storage, on-demand read-out, and operation at room temperature without collisional fluorescence noise. However, previous implementations have required large control pulse energies and suffered from fourwave mixing noise. Here we present a Raman memory where the storage interaction is enhanced by a low-finesse birefringent cavity tuned into simultaneous resonance with the signal and control fields, dramatically reducing the energy required to drive the memory. By engineering anti-resonance for the anti-Stokes field, we also suppress the four-wave mixing noise and report the lowest unconditional noise floor yet achieved in a Raman-type warm vapour memory, (15 ± 2) × 10 −3 photons per pulse, with a total efficiency of (9.5 ± 0.5)%. Quantum information technologies such as quantum key distribution and random number generators are beginning to transition into the commercial sphere, where key requirements are the ability to function at high speed, and in non-laboratory settings. The high carrier frequency of optical signals enables photonic quantum devices to operate noise-free at room temperature whilst offering GHz-THz operational bandwidths. However, direct photon-photon interactions are prohibitively weak, which has held back the development of photonic quantum processors. One solution to this problem has been to use probabilistic measurement-induced non-linearities [1], but the probability of success decreases exponentially with system size, limiting the state of the art to < 10 photons [2]. Further scaling photonic devices requires a multiplexing strategy. Quantum memories capable of storing photons and releasing them on-demand provide the ability to temporally multiplex a repeat-until-success architecture to achieve a freely scalable photonic quantum information platform operable at room temperature.There are many types of quantum memory for light under development: electromagnetically induced transparency [3], the full atomic-frequency comb protocol [4,5], gradient-echo memories [6,7], and the far-offresonant Raman memory [8,9]. Each of these protocols have advantages and challenges, see Ref.[10] for a recent review. A helpful metric for temporal multiplexing is the time-bandwidth product B = τ δ, with δ the acceptance bandwidth and τ the memory lifetime. B is the maximum number of time-bins over which a memory can synchronise an input signal. Time-bandwidth products of B > 1000 enable a dramatic enhancement in the multiphoton rate from parametric photon sources [11], as required to utilise multiphoton interference for computational speed-up (e.g. boson sampling [12]). Large timebandwidth products have been achieved with cold atom memories [13,14], cryogenic rare earth ion memories [15] and room-temperature Raman memories [8,16]. In this paper we present an implementation of an alkali-vapour Raman memory, operating at 75• C, and show the lo...
Optical quantum memories are devices that store and recall quantum light and are vital to the realisation of future photonic quantum networks. To date, much effort has been put into improving storage times and efficiencies of such devices to enable long-distance communications. However, less attention has been devoted to building quantum memories which add zero noise to the output. Even small additional noise can render the memory classical by destroying the fragile quantum signatures of the stored light. Therefore noise performance is a critical parameter for all quantum memories. Here we introduce an intrinsically noise-free quantum memory protocol based on two-photon off-resonant cascaded absorption (ORCA). We demonstrate successful storage of GHz-bandwidth heralded single photons in a warm atomic vapour with no added noise; confirmed by the unaltered photon number statistics upon recall. Our ORCA memory meets the stringent noise-requirements for quantum memories whilst combining high-speed and room-temperature operation with technical simplicity, and therefore is immediately applicable to low-latency quantum networks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.