Future applications of optical quantum technologies depend to a large extent on the availability of miniaturised and reliable optical components with specially tailored functions. In our group, we are working on the development of so-called integrated optical circuits for various quantum applications. The aim is - analogous to integrated electrical circuits - to integrate various optical functional elements in a common substrate and thus realise optical chips with complex functionality, high efficiency and compact design.

Sources for the generation of single photons or photon pairs are fundamental requirements for many quantum optical applications. In recent years, we have developed a variety of different integrated optical versions of such sources [2-4]. All of these sources utilise the chi(2) nonlinearity of _LiNbO3, which enables the generation of photon pairs by parametric down-conversion (PDC). In this process, a pump photon decays into two daughter photons, known as the signalling and idler photon. This PDC process is subject to the rules of energy conservation and phase matching (momentum conservation), whereby the latter can be realised by quasi-phase matching in a periodically polarised structure.

Work in recent years has focussed on the development of specific customised sources. These include, for example, user-friendly plug-and-play sources with fibre pigtails or spectrally narrow-band PDC sources that can be efficiently coupled to quantum memories, for example.

Future work will focus on the development of sources in TFLN. By utilising waveguide dispersion or resonant structures, for example, completely new design possibilities arise in TFLN with a wide range of applications in quantum computing, metrology and cryptography.

The main goal of "integrated quantum optics" is the monolithic integration of complex optical systems into "optical chips" with specific functionality. Such miniaturisation, accompanied by simplified handling, significantly increased stability and low-cost fabrication, is essential to make optical quantum technologies suitable for everyday use.

We have presented a "HOM chip" as an example of an integrated optical circuit for a typical quantum experiment [5]. On the chip, photon pairs are generated by a non-linear process. These are spatially separated and fed to different waveguides. Voltage signals applied to the electrodes of the chip cause a polarisation conversion and simultaneously enable a time delay between the photons. At the end of the chip, the two photons are then recombined in an (integrated optical) beam splitter. Quantum interference causes the so-called Hong-Ou-Mandel effect (HOM effect): If the photons reach the beam splitter at the same time and their properties are identical, they always leave the beam splitter as a pair. This is a fundamental quantum property of light and serves as the basis for many then much more complex quantum logic operations, e.g. in future quantum computers.

While the circuit described above was still realised with conventionally manufactured titanium monodiffusers, we are now increasingly working on circuits in TFLN. The significantly stronger waveguide should make it possible to realise even more compact circuits with higher integration densities.