Many-particle systems, computational scheme

The theoretical description of any many body system is usually approached in two distinct stages. First, the solution of some independent particle model yielding a set of quasi-particles, or dressed particles, which are then used to formulate a systematic scheme for describing the corrections to the model. Perturbation theory, when developed with respect to a suitable reference model, affords the most systematic approach to the correlation problem which today, because it is non-iterative and, therefore, computationally very efficient, forms the basis of the most widely used approaches in contemporary electronic structure calculations, particularly when developed with respect to a Moller-Plesset zero order Hamiltonian. 

The quantities lnd>(r (r)) are defined as the quotients (-Sp,lh), where is the so-called action for the problem under consideration and involves an integration of kinetic and potential contributions over the period 0dimensionless quantity - In (r" (t)), its relation to the product of the density matrix elements in Eqs. (14) and (16) being clear [28]. A few simple examples (e.g., free particle and harmonic oscillator) admit the exact application of the PI formahsm in the P t form [12, 13], but for general many-body quantum systems this is not possible. However, some analytic developments related to Eq. (15) have given rise to the so-called Feynman s semiclassical approaches, which will be considered in Section 111. To exploit the power of the PI formahsm computational schemes utilize finite-P discretizations. In this regard, given that approximations to calculate density matrix... 

Numerical calculations using Kapuy s partitioning scheme have shown that for covalent systems the role of one-particle localization corrections in many-body perturbation theory is extremely important. For good quality results several orders of one-particle perturbations have to be taken into account, although the additional computational power requirement is much less in these cases than for the two-electron perturbative corrections. Another alternative for increasing the precision of the calculations is to estimate of the asymptotic behavior of the double power series expansion (24) from the first few terms by applying Canterbury approximants [31], which is a two-variable generalization of the well-known Pade approximation method. It has also been found [6, 7] that in more metallic-like systems the relative importance of the localization corrections decreases, at least in PPP approximation. 

We have seen that many-body-based methods provide an ab-initio way to treat the Coulomb correlation in an N electron system without the expensive cost of QMC calculations. However, they are computationally more demanding than routine LDA-KS calculations and, hence, the feasibility of their application to complex systems is unclear, especially in the context of ab-initio molecular dynamics calculations, where many total-energy evaluations are required. As described in Sect. 5.3, the main problem when constructing approximations to E c [n] is related to its inherent non-analytical character which is due to the specific way in which the KS mapping between the real and the fictitious systems is done. However, this is not the only possible realisation of DFT and recently, new DFT methods have been proposed [112,113]. In these generalised Kohn-Sham schemes (GKS) the actual electron system is mapped onto a fictitious one in which particles move in an effective non-local potential. As a result of this, it is possible to describe structmal properties at the same (or better) level than LDA/GGA but improving on its description of quasiparticle properties. 

---