One-dimensional quantum emitters and photons for quantum technologies: 1D QED

Quantum technologies exploit the intrinsic quantum nature of particles such as photons and electrons. It has been known for some time that the ability to control the exact state of these particles, and to precisely control how they interact, will lead to unprecedented breakthroughs in a variety of technological applications.

One of the immediate goals of quantum technologies is to exploit the fact that quantum particles can never be copied whilst retaining all of their information. Because the particle cannot be cloned, one may encode a cryptographic key in this way, as an eavesdropper would reveal their presence by changing the photon state as it is measured. Practical cryptographic "quantum key distribution", however, has limited information transfer rate by the fact that one needs to ensure that only one photon is transmitted per bit. To make sure that exactly one photon is generated, a "single photon source" from a quantum emitter such as an atom is required, or in our case, an "artificial atom", a quantum dot. We will fabricate single photon sources that output photons with very high efficiency into a fibre at a useful telecommunications wavelength (1300nm).

The long-term goal of quantum technologies is to create a "universal quantum computer". This would use quantum particles as "quantum bits" that show the property of "superposition" (the ability to prepare a particle two states at once) and "entanglement" (sharing the superposition between several particles). Manipulating quantum bit interactions leads to a way of performing calculations with a complexity that speeds up exponentially with number of quantum bits. Preparing states for a quantum computer that will perform any calculation (a "universal" computer), however, is very challenging.

Nevertheless, if one has a particular complex problem to solve, one may turn to quantum simulation instead. In this case, a calculation may be pre-programmed. It is known that by using photonic circuits (essentially a photon circuit consisting of the equivalent of mirrors and beamsplitters) one may perform a quantum simulation. A network of many channels is set up, and single photons input into chosen channels. However, an important requirement is that, again, controlled single photons must be available. The requirements are more stringent than for quantum communication. A second requirement is that each photon must be absolutely identical in bandwidth, wavelength and polarization - this is known as "indistinguishability". Indistinguishable photons input onto a beamsplitter undergo quantum interference that acts as a logic gate.
Truly indistinguishable single photons are extremely difficult to create. Nevertheless, a great deal of progress has been made in precisely controlling single photons using single atoms trapped in an optical cavity. However, atoms emit photons slowly, and collecting all photons is difficult. The rate at which single photons can be generated is presently still too low and the experimental setup involved very large, and unsuitable for anywhere except a laboratory.

However, quantum dots have very similar properties to atoms. These emit light far faster than atoms (at a rate of 1 billion photons per second) and may also be incorporated into semiconductor "cavities". In this proposal, I will show that one may collect the light extremely efficiently using similar optical fibre technology to that used in telecommunication networks. By doing this, I will provide single photon sources to quantum communication networks and quantum simulation devices. This will lead to absolutely secure communications, and the ability to calculate properties of novel materials or complex molecules to help design new drugs, and factorize large prime numbers used in cryptography.

In this proposal I will create (i) high quality indistinguishable (>99%) single photon sources (SPS) on demand (efficiency >90%) (ii) microsecond quantum memories and (iii) photonic cluster states, all with outputs to a single mode fibre. These are based on self-assembled quantum dots and will be easily interfaced with other quantum technology devices, such as quantum communication networks and quantum optical integrated circuits. Impact will be mainly on immediate and future quantum technologies, but will also impact fundamental physical science in the mid to long term.

In quantum key distribution and other quantum communication applications, both single photon sources and quantum memories are important. At present, a QKD system uses a strongly attenuated laser with decoy states that give approximately 0.3 photons per pulse, and significant post-processing is required. A single photon source of efficiency >70% produces significant increase in efficiency, and therefore effectively increases the feasible distance between repeater stations by ~50km. A quantum memory relay would allow concatenation of links of ~100km, which if loss resistant would allow quantum states to be transported over distances of >1000km. Thus this single photon source could be the route to enabling practical commercial QKD systems to become widespread.

Non-universal quantum computing is a class of computational techniques that address specific classically intractable tasks with qubits. Examples include quantum simulation, a technique which simulates classically intractable systems such as in quantum chemistry (where even the full quantum mechanical modelling of even 10 atom molecules is intractable classically) or solid-state physics, or for specific prime number factorisation tasks. Boson sampling involves inputting non-interacting bosons (in this case photons) into a linear-optic interferometer circuit (relatively easy to fabricate and at room temperature with multiple fibre input and output) that might, for example require input of n photons into an m channel device. Subsequent interference in the device performs a specific calculation. All these devices require indistinguishable on-demand single photons. With a 20 photon input (feasible using QDs and fibre output with a simple lossless 100MHz switch (eg EOM)) classically intractable problems are achievable. While development of protocols to map specific problems is still underway, the overwhelming barrier to the realisation of these devices are efficient and indistinguishable single photon sources.

In optics and photonics technologies the ability to achieve >70dB polarization extinction whilst retaining low losses would be a useful resource for many quantum and classical technologies, for example on-chip four-wave-mixing.

In the longer term, single photon sources may have extensive academic impact, for example in quantum simulations of molecules used in biochemistry and medicine, or in simulating complex phenomena etc. The extension of this technique to different wavelengths (eg to visible using colloidal QDs) may allow a direct delivery of well-defined packets of energy to single absorbers, allowing nanoscientists and biochemists to measure photochemical changes at the single molecule level.

Finally, efficient QD photon collection would allow a step-change in the capabilities of present QD-based basic Physics research. For instance, the ability to perform single shot measurements would allow sensitive quantum non-demolition techniques to be used to probe the electron spin dynamics. In certain circumstances, this would make, for instance, the probing of the nuclear spin state of the QD highly sensitive. In the long term, this may lead to an ability to use the millisecond coherence time nuclear spins as a quantum memory.