Ultracold atomic systems have been of great theoretical and experimental interest in the past few decades. They provide ideal paradigms to study quantum phenomena, simulate condensed matter systems, and could potentially become a platform for quantum computations. This thesis focuses on the theoretical investigation of universal and non-universal properties of ultracold few-atom systems. The goal of this thesis is to answer or provide insights into several interesting open questions. First, what is the role of the non-universal three-body parameter in weakly-interacting harmonically trapped Bose gases and at which order does it kick in? Second, how large is the ground state energy of small two-component Fermi gases with infinitely strong interactions compared to their non-interacting counterparts? Lastly, what is the interplay between the spin-orbit coupling and the
s-wave two-body interaction and how does it affect the energy spectrum of the two-atom system?In order to answer these questions, we develop and employ different analytical and numerical techniques. A perturbative effective field theory treatment to study the weakly-interacting trapped Bose gas is employed. The treatment is extended to up to fourth-order perturbation theory and techniques to regularize the diverging sums are developed. A logarithmic divergence in the three-body sector is identified and the need for a non-universal three-body parameter is shown.The explicitly correlated Gaussian method is used to calculate the ground state energy of small trapped two-component Fermi gases with up to ten particles. A code is developed from scratch and parallelized. An improved optimization scheme is developed. A novel technique to remove the finite-range dependence is proposed and tested. Our calculations serve as benchmarks for other numerical techniques. The explicitly correlated Gaussian method is extended to treat systems with periodic boundary conditions for the first time.Standard perturbation theory is used to treat the spin-orbit coupling term in trapped two-atom systems. Our treatment is based on the exact two-body zero-range solution in the trap. The interplay between the spin-orbit coupling and the two-body
s-wave interaction is investigated and shown to have a great impact on the energy spectrum for certain parameter combinations.
From celestial mechanics to quantum theory of atoms and molecules, perturbation theory has played a central role in natural sciences. Particularly in quantum mechanics, the amount of information needed for specifying the state of a many-body system commonly scales exponentially as the system size. This poses a fundamental difficulty in using perturbation theory at arbitrary order. As one computes the terms in the perturbation series at increasingly higher orders, it is often important to determine whether the series converges and if so, what is an accurate estimation of the total error that comes from the next order of perturbation up to infinity. Here we present a set of efficient algorithms that compute tight upper bounds to perturbation terms at arbitrary order. We argue that these tight bounds often take the form of symmetric polynomials on the parameter of the quantum system. We then use cellular automata as our basic model of computation to compute the symmetric polynomials that account for all of the virtual transitions at any given order. At any fixed order, the computational cost of our algorithm scales polynomially as a function of the system size. We present a non-trivial example which shows that our error estimation is nearly tight with respect to exact calculation.
A brief review of the history of many-body perturbation theory (MBPT) and its applications in nuclear physics is given. Problems regarding its application to nuclear-structure calculations are discussed and analysed. It is concluded that the usefulness of nuclear MBPT in terms of an expansion in the nuclear reaction matrix G for the calculation of effective interactions in shell-model investigations is severely challenged and restricted by the problems and uncertainties connected with this approach. New methods based on unitary transformation approaches have proven to be more accurate and reliable, particularly for light nuclei.