If quantum computers and networks are ever to be realized, they likely will be made of different types of parts that will need to share information with one another, just like the memory and logic circuits in today's computers do.
The goal to develop quantum computers—a long-awaited type of computer that could solve otherwise intractable problems, such as breaking complex encryption codes—has inspired scientists the world over to invent new devices that could become the brain and memory of these machines. Many of these tiny devices use particles of light, or photons, to carry the bits of information that a quantum computer will use.
But while each of these pieces of hardware can do some tasks well, none are likely to accomplish all of the functions necessary to build a quantum computer and, eventually, the global quantum network. This implies that several different types of quantum devices will need to work together for the computer or network to function. The trouble is that these tiny devices frequently create photons of such different character that they cannot transfer the quantum bits of information between one another. Transducing two vastly different photons into two similar ones would be a first step toward permitting quantum information components to communicate with one another over large distances. We study parametric processes to make photons from dissimilar quantum systems compatible.
We study the indistinguishability of photons produced by highly dissimilar sources: photons produced from a single quantum dot (QD) and from parametric down-conversion (PDC) in a nonlinear crystal. A QD source is a deterministic source of single photons. On the other hand, a PDC process generates photon pairs, and the detection of one of the photons is used to herald the presence of the other. After spectral and temporal control of the PDC photons to optimize their overlap with the QD photons, we measure a two-photon coalescence and show partial indistinguishability of these photons.
We study properties of parametric up- (and down-) conversion for frequency translation of various states of light: from faint laser beams to entangled states. The goal is to try to faithfully preserve the state. To make a frequency converter efficient requires a strong laser beam that contains approximately 1018 photons per second. Yet, the typical quantum input and output of this converter are single photon states! To date, we have demonstrated the feasibility of nearly-noiseless up-converting for this and other purposes. Particularly, we developed an up-converter that adds so few background photons, that we cannot detect their presence. To characterize noise, we have developed a new measurement method, suitable to distinguish between a few photons of visible light per hour and the complete darkness. This measurement uses a calibrated transition edge sensor detector and a special dark count reduction algorithm.
With our unique measurement system, we characterized our optical up-converter and demonstrated that its background noise is on the order of 100 photons per hour. This frequency converter enables a new generation of nonlinear devices that can be used with extra-faint quantum states of light. We are currently designing them.
Practically noiseless frequency conversion enables the use of nonlinear optics phenomena for quantum applications. In our research we found that direct noiseless frequency transduction between two arbitrary wavelengths is not always possible. We investigate three-wave mixing (TWM) processes in a 1D array of nonlinear waveguides evanescently coupled to one another. We demonstrate an analogy of this system to an atom interacting with an external optical field using both classical and quantum models of the optical fields. Using this framework, we adapt well-known coherent processes from atomic optics, such as electromagnetically induced transparency (EIT) and stimulated Raman adiabatic passage (STIRAP) to design and simulate novel nearly-noiseless photonic devices. This approach allows the implementation of devices that are very difficult or impossible to implement by conventional techniques.
The robust frequency converter. We propose using the all-optical STIRAP analog for noiseless frequency conversion between the states that are hard to noiselessly couple in a single-step TWM transduction. In particular, it turns out that a single-step TWM transduction of telecom photons to most material cubits and back cannot be noiseless. Still, noiseless transduction can be achieved, but in a two-step TWM transduction. Thanks to a STIRAP approach both transduction steps occur simultaneously and in one device. The generation of the field with the intermediate frequency is significantly suppressed, and the process is robust against the experimental and technological imperfections, such as pump intensity fluctuations.
Nonlinear optical integrated devices can perform multiple different manipulations with the photonic states. The goal may be not only to faithfully preserve the state, but in some cases to make a quantum state that is more useful, i.e. accomplish a few steps in one physical system. One such idea is an integrated photonic switch. We show how an all-optical analog of EIT in the system of two evanescently coupled waveguides can be used as an ultrafast broadband all-optical switch or on-chip quantum memory. A fast low-noise switch switches the input to one of the two waveguides based on presence (or absence) of the strong optical control pump at a different wavelength. Switching is based on an all-optical analogy of electromagnetically induced transparency. With no pump, weak light couples from input waveguide to the output waveguide evanescently. The strong pump turns the “transparency” on through nonlinear parametric interaction, i.e., nearly instantaneously, so that the input stays in its waveguide. The large wavelength separation makes it easy to nearly-noiselessly spectrally filter the faint signal from the strong pump. In addition, because wavelengths of pump and input signals can be far-detuned, noise is not generated at the signal wavelength.
One possible switch application is the optically controlled delay loop – based quantum memory. The nonlinear waveguide and the waveguide loop are evanescently coupled. When the control field is off, photons injected in the nonlinear waveguide will tunnel to the loop and vice versa. The control field locks the loop.
A demonstration of indistinguishability of photons from dissimilar sources, ref. Coalescence of Single Photons Emitted by Disparate Single-Photon Sources: The Example of InAs Quantum Dots and Parametric Down-Conversion Sources, Sergey V. Polyakov, Andreas Muller, Edward B. Flagg, Alex Ling, Natalia Borjemscaia, Edward Van Keuren, Alan Migdall, and Glenn S. Solomon, Phys. Rev. Lett. 107, 157402 – Published 5 October 2011
A demonstration of background-free upconversion in a nonlinear waveguide, refs. Statistically background-free, phase-preserving parametric up-conversion with faint light, Y.-H. Cheng, Tim Thomay, Glenn S. Solomon, Alan L. Migdall, and Sergey V. Polyakov, Optics Express Vol. 23, Issue 14, pp. 18671-18678 (2015).
Quantum frequency bridge: high-accuracy characterization of a nearly-noiseless parametric frequency converter, Ivan A. Burenkov, Thomas Gerrits, Adriana Lita, Sae Woo Nam, L. Krister Shalm, and Sergey V. Polyakov, Opt. Express 25, 907-917 (2017)