The quantum revolution of the 20th century culminated in the development of quantum optics and the study of nonclassical light. By the beginning of this century, scientific laboratories could produce and handle various quantum states of light: from single photons and photon pairs to squeezed states and bright twin beams. Gradually, this toolbox has become part of a range of emerging quantum technologies. The first real application of quantum optics was quantum key distribution,1 commercialized almost two decades ago. This is not surprising because photons are the ideal messengers and very convenient for communications as they can make use of existing optical fiber network infrastructure. A major milestone for quantum sensing is its application to new disciplines: this has excitingly been showcased by the next-generation of gravitational-wave detectors that employ squeezed light to enhance detector sensitivity;2,3 we anticipate other achievements will follow and span a huge range of disciplines, from fundamental physics to biomedicine and more. This special topic collection samples key research areas from the wider literature on quantum imaging, sensing, and spectroscopy. It also showcases developments in enabling technologies including novel nonlinear light sources as well as the growing use of enhanced CCD cameras for quantum imaging.
Quantum optical sensing relies on sources of correlated photon pairs or bright twin beams, generally produced by spontaneous parametric downconversion (SPDC). Indeed, second order nonlinear crystals are the well-established workhorses of the majority of quantum sensing protocols. For example, in this special topic collection, the Setzpfandt group report on the use of SPDC to perform ghost imaging using a pinhole to form an image,4 removing the need for lenses in imaging. This is important for eliminating chromatic aberration problems and lens technologies in hard to access wavelength ranges. Challenges have also emerged that have spurred on research into novel photon pair sources. In particular, non-degenerate photon pairs linking the visible and infrared through their correlations are currently sought to enable infrared imaging with visible detectors. The Kitaeva group explore extreme non-degenerate SPDC to produce photon pairs spanning visible and THz frequencies.5 The group achieve phase matching via temperature tuning of a lithium niobate crystal under cryogenic conditions. Although only the visible photons are evaluated directly, the team are able to infer correlations with THz photons through a Kirchhoff–Klyshko analysis. They find that quantum correlations prevail over thermal fluctuation below temperatures of 50 K. This is an exciting development that could enable imaging at THz frequencies, but by detection at visible frequencies.
In some applications, the SPDC process is not available; for example, few integrated quantum sensing technologies are constructed with materials exhibiting the necessary second-order nonlinearity. In these cases, non-degenerate correlated photon-pairs may be produced by the third-order process of spontaneous four wave mixing (SFWM). In this Special Topic, the Takeuchi group demonstrate a SFWM pair source at telecom wavelengths in doped glass ring resonators.6 Correlated photon-pairs were demonstrated across a 24 nm bandwidth, showing promise for quantum cryptography and quantum information applications. Meanwhile, the Clark group demonstrate a tunable correlated photon pair source by SFWM in a polarization maintaining fiber.7 The fiber's birefringence enables phase matching over a wide wavelength range spanning the visible and telecoms wavelengths, simply by varying the pump wavelength. The group has highlighted the utility of their tunable photon-pair source for quantum enhanced spectroscopy.
As long as quantum sensing relies on relatively weak streams of photons generated by nonlinear crystals, it equally leans on detector technologies that must be sensitive to isolated photons. Photon multipliers (PMs) and avalanche photodiodes (APDs) have thus been integral to this field, but this has meant slow imaging protocols that rely on scanning. It is thus noteworthy that this special topic highlights the emergence of quantum sensing techniques that utilize enhanced CCD cameras in three contributions.8–10 Electron Multiplying CCD (EMCCD) cameras, which boast near unity quantum efficiency, are able to recover photon correlations in space across the camera's array, while sacrificing temporal correlations, due to the limited time resolution of each pixel. The perspective article, by Padgett's group, poses the question, “How many photons does it take to form an image?” and so it is apt that this question should be addressed through their various studies exploiting EMCCD cameras.10 In past work, the Genovese group deployed EMCCD technology in sub-shot noise quantum imaging under wide field and real-time operation. In this special issue, the group explore the trade-off between quantum enhancement and resolution.9 While EMCCDs offer exceptional sensitivity, their response time is still limited by readout electronics and the sheer density of pixels and potential detection events. The Nomerotski group have tackled this issue, by exploiting fast acquisition camera technology from high energy particle detectors.8 Here, the readout electronics allow for a high data acquisition for a limited number of individual events across an intensified CCD. The team showcase their camera's capability by measuring continuous time and energy variables simultaneously, a capability not possible with EMCCDs and difficult to implement with scanned PMs/APDs.
The purpose of any quantum sensing scheme is to illicit an enhancement over a related classical approach. This special topic collection includes a variety of examples of how improvements can be won and optimized, from quantum imaging to quantum spectroscopy. Padgett's group explore the idea of quantum compressive sensing to speculate on the minimum number of photons, and by extension information, necessary to form an image.10 In doing so, they highlight an intriguing link between quantum sensing and machine learning. In a related fashion, the Genovese group's sub-shot noise quantum imaging work also highlights the relevance of computational post-processing in maximizing the resolution capabilities of quantum imaging.9 Here, the group considers how the variations of photon flux across an image, projected onto an EMCCD, influence the visibility of post-processed photon correlations. Regions of an object with high absorption, yielding lower imaging flux, typically require larger groups of pixels to be used in the correlation post-processing, which, in turn, influences the resolution. Through optimization, they show that the Rayleigh limit may still be achieved. The Matthews group have identified the ideal conditions required to maximize correlations in quantum absorption sensing using imperfect correlated intensity measurements.11 In general, the precision in absorption sensing depends on the variation in amplitude of the transmitted probe light signals and when quantum properties of light are used to enhance this, optical loss and detector efficiency are known to be important factors. The Matthews group shows in practice that for the presence of low uncorrelated experimental noise (such as noisy signals present in the detection), maximizing each detector's efficiency is best, even if this imbalances efficiency. While for higher levels of noise, balancing detection efficiencies is best, even if this means reducing efficiency of one of the detectors.
Quantum states of light can be used to enhance spectroscopy of biological and chemical systems. In this Special Topic collection, two groups employ intensity squeezing to enhance microscopy of organic molecules. The Agarwal group generates squeezing via seeded four-wave mixing in an atomic vapor of rubidium and uses this light to excite fluorescent biomarkers via two-photon absorption (TPA).12 The TPA efficiency is enhanced by the quantum correlations generated in the source, and they test this by changing the arrival time of correlated photons at the molecules. This work has applications in low intensity biosensing and bioimaging, especially for in vivo deep tissue studies. The Walmsley group also generate squeezing, but in this case using a solid-state optical parametric oscillator.13 They use the generated pulses of light to perform stimulated emission detection of triphenyl methane organic dye molecules in solution. This technique has the advantage of not requiring a fluorescent biomarker, while also enhancing the signal-to-noise ratio of their measurements. Finally, Ye and Mukamel show that higher photon number quantum states can be used to access new regimes of two-photon absorption spectroscopy.14 Their theoretical and numerical investigation shows that photon triplets generated through cascaded SPDC can be manipulated using three-port beam splitters and phase shifters in an interferometer to enhance or suppress various excitation pathways in the multi-level system being investigated.
Recent years have seen a growing body of work in the area of quantum sensing with nonlinear interferometry. The incorporation of a pair of nonlinear correlated light sources within an interferometer arrangement induces coherence between the sources at the output,15 even if the photon pairs are at distinct wavelengths. The result is that changes in transmission and phase in one arm of the interferometer can be seen in both photon paths at the output. In the past, this has been used for imaging an object by detecting photons that never even interacted with it16 finding applications in optical coherence tomography17 and biological imaging in the mid-infrared.18 In this special topic collection, the Krivitsky group uses nonlinear interferometry for semiconductor device testing. Here, metal structures buried in silicon were imaged by detecting photons at 810 nm, a wavelength at which silicon is opaque, and thus, the structures would be otherwise obscured using direct imaging techniques.19 This imaging is made possible as the idler photons in the nonlinear interferometer are at 1550 nm, a wavelength at which silicon is transparent. The detection at 810 nm is possible in standard CMOS camera technology, making this form of imaging cheap and efficient. The work also explores the use of compressive sensing to speed up the imaging process.
The Kaiser group reports an asymmetric nonlinear interferometer with two different SPDC processes and shows that they can see a sensitivity to phase in single-photon detection that would normally only be seen for two photons detected in coincidence.20 This and the inherent brightness of their SPDC sources result in their ability to use standard photodiodes for detection of phase variations, making this scheme more practical. Meanwhile, the Chekhova group investigates the use of a nonlinear interferometer in the high parametric gain regime for optical coherence tomography, which holds promise for use in, for example, ophthalmology and art restoration.21 The high gain operation results in signals that can be detected using standard cameras and detectors, again making this system more suitable for mass markets. The Dorfman group shows that four-wave mixing in various configurations, including a nonlinear interferometer, can be used to characterize the third-order nonlinear susceptibility of a medium.22 They perform a numerical study of this method to probe silicon-vacancy centers in diamond, paving the way for the investigation of more complex biological samples. Nonlinear interferometers can also find application in quantum information systems. The Zhang group combine a nonlinear interferometer with a Mach–Zehnder interferometer so that it can be used to measure both phase and amplitude modulation to enable quantum dense metrology.23
Sensing and imaging remain cornerstones of modern science, technology, healthcare, and manufacturing, so advances in this field continue to have a significant impact on society. This special topic samples the wealth of activity that is being undertaken around the world in advancing techniques in imaging and sensing using quantum phenomena and techniques born from quantum technologies, which have seen a large global investment in recent years.24 A broad range of imaging applications are emerging and are being explored with a wide scope of techniques from analysis with machine learning to engineering nonlinear optical processes. In many cases, development focuses on the constituent components of light sources and their properties, interaction with imaged and measured objects, as well as detection and data analysis; in others, partial and full system development is required to advance the techniques to realistic application. Regardless, collaboration across disciplines and with end-users is key, especially where requirements of the “killer-application” defines the underpinning hardware that needs to be developed.
R.F.O., A.S.C., J.C.F.M., and J.G.R. acknowledge support from the EPSRC UK Quantum Technology Hub in Quantum Enhanced Imaging (QuantIC) (No. EP/T00097X/1). J.C.F.M. acknowledges support from European Research Council starting Grant No. ERC-2018-STG 803665. A.S.C. acknowledges a Royal Society University Research Fellowship (UF160475).
