Classical optics describes light as waves which makes explaining the physical phenomenon of interference easy. In a quantum description light can be thought to consist of individual particles that are called photons. Obviously, this quantum mechanical picture based on photons must be equally valid. Each light particle carries a very small, but well-defined, amount of energy. Detecting a single photon is a formidable technological challenge but can be solved by modern single photon counting detectors that produce a 'click' whenever they detect a photon.

In experiments with single photons the behaviour of light regime is exactly predicted by the wave description of classical optics. Interference patterns (a typical wave phenemenon) emerge from the count statistics of the photons (as particles). In order to find effects that go beyond the realm of classical optics and are not easily explained by the wave picture one needs to investigate correlations between two or more photons.

In the quantum optics group at Leiden University we produce such special quantum states of light that contain two (or more) photons that are strongly correlated because they were created in a single process in the same crystal. For this purpose we use a nonlinear optical process called spontaneous parametric down-conversion that splits a single blue photon into two red photons, albeit with an overwhelmingly small probability. The process is called spontaneous as there is no obvious cause for the blue photon to decay into two lower energy photons.

The pairs produced by down-conversion have proven to be very useful in fundamental tests of quantum mechanics by violating Bell's inequalities. Using entangled photons these inequalities can be violated and no (classical) theories with local hidden variables can be constructed that explain the measured correlations between a property (e.g. polarization) of the photon. It is impossible to predict the outcome of a measurement on one particle, but if we measure a property of one particle we know with 100% certainty the outcome of the measurement on the other particle. To explain this we can use 'standard' quantum theory where a single wavefunction is used to described both particles. This wavefucntion contains a superposition of two (or more) states of the two photons. Each of these states represents a classical correlation between the two particles, but the superpostion state describes all possibilities simultaneously for the two photons together. Quantum mechanics postulates that a measurement on one particle collapses the wavefunction to one of the classically correlated states.

Similar reasoning has been used to show that the states of quantum particles cannot be copied and that the physical process of measurement changes the wavefunction that describes the particles. These insights make these photons an essential resource in quantum information and communication as well as in quantum computation.

NbN detectors

A surface plasmon, or plasmon-polariton is an electromagnetic wave that can exist on the interface between an insulating material and a good metal. The free electrons of the metal are accelerated by the field of the surface plasmon creating a coupled wave of electromagnetic field and charge density fluctuations on the metal surface.

These surface plasmons have received a lot of attention because they could become highly confined to the surface. Among others, this could be usefull in sensing application, non-linear optical studies or high resolution imaging. We have studied such surface plasmons on so-called metal hole arrays: a thin film of metal perforated by a regular array of subwavelength holes. The individual holes are not expected to transmit light because the holes are smaller than the wavelength of light. However, all the holes together do transmit much more light than expected. This phenemenon, first discovered by Thomas Ebessen and coworkers, is called extra-ordinary transmission and works because the holes on the regular grid couple the incoming light to the surface plasmon.

Photonic crystals are special optical materials that have a periodic variation in their refractive index on a length scale that is comparable to the wavelength of light. By far the easiest and most widespread example of such a material is a dielectric or Bragg mirror. These mirrors consist of alternating layers of high and low refractive index material. At each boundary between the two layers optical reflection occurs. If the layer thickness is tuned carefully so that the reflected light of each of the layers is exactly in phase constructive interference occurs and eventually all light will be reflected far a sufficient number of layers. This principle is commonly used in commercial dielectric mirrors that have high reflectivity that easily surpasses the reflectivity of metal mirrors. Such high reflectivity mirrors are an essential component in lasers.

However, the concept of periodic dielectrics is not limited to one-dimensional stacks of layers.