Photons are natural carriers of quantum information due to their ease of distribution and long lifetime. This thesis concerns various related aspects of quantum information processing with single photons. Firstly, we demonstrate N-photon entanglement generation through a generalised N X N symmetric beam splitter known as the Bell multiport. A wide variety of 4-photon entangled states as well as the N-photon W-state can be generated with an unexpected non-monotonic decreasing probability of success with N. We also show how the same setup can be used to generate multiatom entanglement. A further study of multiports also leads us to a multiparticle generalisation of the Hong-Ou-Mandel dip which holds for all Bell multiports of even number of input ports. Next, we demonstrate a generalised linear optics based photon filter that has a constant success probability regardless of the number of photons involved. This filter has the highest reported success probability and is interferometrically robust. Finally, we demonstrate how repeat-until-success quantum computing can be performed with two distant nodes with unit success probability using only linear optics resource. We further show that using non-identical photon sources, robustness can still be achieved, an illustration of the nature and advantages of measurement-based quantum computation. A direct application to the same setup leads naturally to arbitrary multiphoton state generation on demand. Finally, we demonstrate how polarisation entanglement of photons can be detected from the emission of two atoms in a Young's double-slit type experiment without linear optics, resulting in both atoms being also maximally entangled.
Photons are natural carriers of quantum information due to their ease of distribution and long lifetime. This thesis concerns various related aspects of quantum information processing with single photons. Firstly, we demonstrate N-photon entanglement generation through a generalised N X N symmetric beam splitter known as the Bell multiport. A wide variety of 4-photon entangled states as well as the N-photon W-state can be generated with an unexpected non-monotonic decreasing probability of success with N. We also show how the same setup can be used to generate multiatom entanglement. A further study of multiports also leads us to a multiparticle generalisation of the Hong-Ou-Mandel dip which holds for all Bell multiports of even number of input ports. Next, we demonstrate a generalised linear optics based photon filter that has a constant success probability regardless of the number of photons involved. This filter has the highest reported success probability and is interferometrically robust. Finally, we demonstrate how repeat-until-success quantum computing can be performed with two distant nodes with unit success probability using only linear optics resource. We further show that using non-identical photon sources, robustness can still be achieved, an illustration of the nature and advantages of measurement-based quantum computation. A direct application to the same setup leads naturally to arbitrary multiphoton state generation on demand. Finally, we demonstrate how polarisation entanglement of photons can be detected from the emission of two atoms in a Young's double-slit type experiment without linear optics, resulting in both atoms being also maximally entangled.
Multiphoton entanglement is an important resource for linear optics quantum computing. Here we show that a wide range of highly entangled multiphoton states, including W-states, can be prepared by interfering single photons inside a Bell multiport beam splitter and using postselection. A successful state preparation is indicated by the collection of one photon per output port. An advantage of the Bell multiport beam splitter is that it redirects the photons without changing their inner degrees of freedom. The described setup can therefore be used to generate polarization, time-bin, and frequency multiphoton entanglement, even when using only a single photon source.
One of the main problems that optical quantum computing has to overcome is the efficient construction of two-photon gates. Theoretically these gates can be realized using Kerr-nonlinearities, but the techniques involved are experimentally very difficult. We therefore employ linear optics with projective measurements to generate these non-linearities. The downside is that the measurement-induced nonlinearities achieved with linear optics are less versatile and the success rate can be quite low. This project is mainly the result of a literature study but also a theoretical work on the physics behind quantum optical multiports which is essential for realizing two-photon gates. By applying different postcorrection techniques we increase the probability of success in a modifed non-linear sign shift gate which is foundational for the two photon controlled-NOT gate. We prove that it's not possible to correct the states by only using a single beam splitter. We show that it might be possible to increase the probability of success using a more complex setup with at least two error-correcting beam splitters.
Since the awareness of entanglement was raised by Einstein, Podolski, Rosen and Schrödinger in the beginning of the last century, it took almost 55 years until entanglement entered the laboratories as a new resource. Meanwhile, entangled states of various quantum systems have been investigated. Sofar, their biggest variety was observed in photonic qubit systems. Thereby, the setups of today's experiments on multi-photon entanglement can all be structured in the following way: They consist of a photon source, a linear optics network by which the photons are processed and the conditional detection of the photons at the output of the network. In this thesis, two new linear optics networks are introduced and their application for several quantum information tasks is presented. The workhorse of multi-photon quantum information, spontaneous parametric down conversion, is used in different configurations to provide the input states for the networks. The first network is a new design of a controlled phase gate which is particularly interesting for applications in multi-photon experiments as it constitutes an improvement of former realizations with respect to stability and reliability. This is explicitly demonstrated by employing the gate in four-photon experiments. In this context, a teleportation and entanglement swapping protocol is performed in which all four Bell states are distinguished by means of the phase gate. A similar type of measurement applied to the subsystem parts of two copies of a quantum state, allows further the direct estimation of the state's entanglement in terms of its concurrence. Finally, starting from two Bell states, the controlled phase gate is applied for the observation of a four photon cluster state. The analysis of the results focuses on measurement based quantum computation, the main usage of cluster states. The second network, fed with the second order emission of non-collinear type II spontaneous parametric down conversion, constitutes a tunable source of a whole family of states. Up to now the observation of one particular state required one individually tailored setup. With the network introduced here many different states can be obtained within the same arrangement by tuning a single, easily accessible experimental parameter. These states exhibit many useful properties and play a central role in several applications of quantum information. Here, they are used for the solution of a four-player quantum Minority game. It is shown that, by employing four-qubit entanglement, the quantum version of the game clearly outperforms its classical counterpart. Experimental data obtained with both networks are utilized to demonstrate a new method for the experimental discrimination of different multi-partite entangled states. Although theoretical classifications of four-qubit entangled states exist, sofar there was no experimental tool to easily assign an observed state to the one or the other class. The new tool presented here is based on operators which are formed by the correlations between local measurement settings that are typical for the respective quantum state.
Photon entanglement is an essential ingredient for linear optics quantum computing schemes, quantum cryptographic protocols and fundamental tests of quantum mechanics. Here we describe a setup that allows for the generation of polarisation-entangled N-photon states on demand. The photons are obtained by mapping the entangled state of N atoms, each of them trapped inside an optical cavity, onto the free radiation field. The required initial state can be prepared by performing postselective measurements on the collective emission from the cavities through a multiport beamsplitter.
Entanglement lies at the heart of quantum mechanics and challenged the intuition of physicists ever since it was discovered. At the same time, it is a powerful tool that serves as a key resource for quantum communication and quantum computation schemes. Many of these applications rely on multiparticle entanglement, whose description, generation and manipulation became therefore a very active field in theoretical and experimental quantum information science. The goals are here to classify and understand the different types of entanglement, to find new applications and to control and analyze the quantum states experimentally. In this thesis, the experimental observation and analysis of two different types of four-photon polarization entangled states is presented: The cluster state and the symmetric Dicke state with two excitations. For this purpose, experimental setups based on spontaneous parametric down conversion and linear optics with conditional detection were designed. They allowed to observe the cluster state with a fidelity of 74.1 % and the symmetric Dicke state with a fidelity of 84.4 %. The cluster state experiment included the development of a new instrument that is of interest for linear optics quantum logic in general: A probabilistic controlled phase gate that is, due to the simplification of a previous approach, highly stable and can actually be used in multiphoton experiments. The quality of the gate is evaluated by analyzing its entangling capability and by performing full process tomography. The achieved results demonstrate that this device is well suited for implementation in various multiphoton quantum information protocols. In order to study the observed quantum states, efficient analysis tools are introduced. It was possible to verify that essential properties of the ideal states are indeed reproduced in the experiment, among others, the presence of genuine four-partite entanglement. A particular focus is put on the behavior of the states under projective measurements and photon loss. Several new insights in their entanglement structure are revealed and verified experimentally. We further demonstrate properties that are characteristic for the entanglement classes of the states. These can be used to infer the applicability of the observed states for certain distributed quantum communication applications. The presented experiments are generic for the design of setups to observe cluster- and symmetric Dicke states with a higher number of photons. Furthermore, also the efficient non-tomographic methods for state analysis we employ can directly be generalized to experiments with higher qubit numbers, where the reduction of the experimental effort for state analysis is even more crucial.
The Hong–Ou–Mandel (HOM) dip has played an important role in recent linear optics experiments. It is crucial for quantum computing with photons and can be used to characterize the quality of single photon sources and linear optics setups. In this paper, we consider generalized HOM experiments with N bosons or fermions passing simultaneously, i.e. within their coherence time, through a symmetric Bell multiport beam splitter. It is shown that for an even number of bosons, the HOM dip occurs naturally in the coincidence detection in the output ports. In contrast, fermions always leave the setup separately, exhibiting perfect coincidence detection. Our results can be used to verify or employ the quantum statistics of particles experimentally.