![]() |
This book is the first to develop a unified gauge theory of condensed matter systems dominated by vortices or defects and their long-range interactions. Gauge fields provide the only means of describing these interactions in terms of local fields, rendering them accessible to standard field theoretic techniques. Two particularly important examples, superfluid systems and crystals, are treated in great detail. The theory is developed in close contact with physical phenomena and evolves naturally from conventional descriptions of the systems. In addition to gauge fields, the book introduces the important new concept of disorder fields for ensembles of line-like defects. The combined field theory allows for a new understanding of the important phase transitions superfluid ‘normal and solid’ liquid. Apart from the above, the book presents the general differential geometry of defects in spaces with curvature and torsion and establishes contact with the modern theory of gravity with torsion. This book is written for condensed matter physicists and field theorists. It can be used as a textbook for a second-year graduate course or as supplementary reading for courses in the areas of condensed matter and solid state physics, statistical mechanics, and field theory.
https://doi.org/10.1142/9789814415606_fmatter01
The following sections are included:
https://doi.org/10.1142/9789814415606_0001
Most physical systems follow complicated nonlinear equations. For a small number of degrees of freedom, numerical methods may lead to a satisfactory theoretical understanding. In macroscopic many-body systems, however, this number is immensely large and such an approach is hopeless. The best we can achieve is an approximate understanding of the statistical behavior, averaged either with respect to thermal or to quantum mechanical fluctuations…
https://doi.org/10.1142/9789814415606_0002
The phenomena that will be studied take place in real space. The most economical way of describing such phenomena is based on attributing to each space point x one or more variables, collectively denoted by ϕα(x). The set of these variables is called a field. The field ϕα(x) undergoes thermal as well as quantum fluctuations and the development of this part of our text will be devoted to reviewing the general description of the statistical properties of such a fluctuating field. Since several good textbooks are already available on this subject we can afford to be somewhat sketchy. Wherever statements require elaborate proofs these will be omitted and we shall be content with simple illustrations and examples. Hopefully, our short treatment will be sufficient to serve as an introduction for those unfamiliar with the relevant basic field theoretical techniques. At the very least we hope to clarify the notation, conventions, and language to be used in the further development of the theory.
https://doi.org/10.1142/9789814415606_0003
The following sections are included:
https://doi.org/10.1142/9789814415606_0004
The following sections are included:
https://doi.org/10.1142/9789814415606_0005
The following sections are included:
https://doi.org/10.1142/9789814415606_0006
The following sections are included:
https://doi.org/10.1142/9789814415606_0007
The following sections are included:
https://doi.org/10.1142/9789814415606_0008
The simplest physical system for which a gauge theory of stresses and defects can be developed is superfluid 4He. It possesses precisely the properties described in the Introduction: there exists an ordered ground state at zero temperature. Small disturbances introduce soft long-range excitations which are observable as sound waves. Stronger disturbances, for instance, the rotation of the container of the superfluid, lead to the formation of vortex lines which may be viewed as line-like defects. Moreover there exists a critical temperature at which a phase transition takes place leading to the disappearance of the superfluid state…
https://doi.org/10.1142/9789814415606_0009
The following sections are included:
https://doi.org/10.1142/9789814415606_0010
The following sections are included:
https://doi.org/10.1142/9789814415606_0011
The following sections are included:
https://doi.org/10.1142/9789814415606_0012
While the high temperature expansion is quite accurate in describing almost the entire high temperature phase (except in the immediate vicinity of the critical point), it fails completely in the low temperature phase. Other methods are therefore needed in order to complete our understanding of the model. The most prominent among them is based on the mean field theory plus fluctuation corrections.
https://doi.org/10.1142/9789814415606_0013
In the general discussion in Chapter 2 we argued that superfluid 4He should be describable as an ensemble of vortex lines together with their long-range hydrodynamic interactions. Let us now verify this expectation within the XY model.
https://doi.org/10.1142/9789814415606_0014
The following sections are included:
https://doi.org/10.1142/9789814415606_0015
Up to now, we have dealt only with one type of gauge field, namely, the gauge field which describes superflow via a curl relation,
This gauge field was introduced in order to ensure that the bi, field lines are closed. They are the rings of superflow arising in the disordered phase due to fluctuations which carry the liquid toward the ordered state. In the vortex-line representation of the XY model, we encountered another set of closed lines, namely, those of vortex lines. Vortex lines give rise to a second important gauge structure which we are now going to discuss.
https://doi.org/10.1142/9789814415606_0016
The following sections are included:
https://doi.org/10.1142/9789814415606_0017
In the last chapters we showed that the calculation of correlation functions of the order parameter eiγ(x) amounts to calculating the behaviour of a fixed set of magnetic monopoles within an ensemble of vortex lines. It is worthwhile pointing out that the model can be enriched in a very simple manner to contain, from the outset, a whole ensemble of such monopoles. All we have to do is to discretize the angle γ(x) so that it no longer covers the interval [0, 2π) continuously but only the fractional angles…
https://doi.org/10.1142/9789814415606_0018
The following sections are included:
https://doi.org/10.1142/9789814415606_0019
The following sections are included:
https://doi.org/10.1142/9789814415606_0020
The following sections are included:
https://doi.org/10.1142/9789814415606_bmatter01
The following sections are included:
https://doi.org/10.1142/9789814415606_fmatter02
The following sections are included:
https://doi.org/10.1142/9789814415606_others01
The description of superfluid 4He in terms of a disorder field theory developed in Part II can serve as a prototype for the treatment of many other physical systems. For this to be true, these systems have to possess the following fundamental properties…
https://doi.org/10.1142/9789814415606_0021
The following sections are included:
https://doi.org/10.1142/9789814415606_0022
The following sections are included:
https://doi.org/10.1142/9789814415606_0023
The following sections are included:
https://doi.org/10.1142/9789814415606_0024
The following sections are included:
https://doi.org/10.1142/9789814415606_0025
The following sections are included:
https://doi.org/10.1142/9789814415606_0026
So far we have studied only static defect lines. For a complete understanding of the plastic properties of materials caused by defect lines it is necessary to know also their kinematic properties. In this text we shall be mainly interested in equilibrium thermodynamics of defects and stresses, where these properties are, to a large extent, irrelevant. For completeness, however, it will be useful to recapitulate the way defects move in a crystal.
https://doi.org/10.1142/9789814415606_0027
So far we have studied specific defect configurations with their elastic long range stress interactions. These interactions were described by a gauge field coupled locally to the defect densities. In the previously investigated case of superfluid 4He we saw that given a system of random lines with such a coupling, it is relatively easy to develop a disorder field theory for a grand canonical ensemble of such lines. This field theory permitted the study of phase transitions in which vortex lines condensed. In 4He this transition carried the superfluid into the normal state…
https://doi.org/10.1142/9789814415606_0028
The Lindemann criterion gave us some important information on the nature of the melting transition: the transition occurs at a temperature slightly below which the atoms still perform, to a very good approximation, harmonic fluctuations about their mean positions. The nonlinearities of the interatomic potential are not very important. They merely provide a shift in the mean position and a softening of the elastic constants. If these two effects are taken into account, the crystal, even right below the melting temperature, can be treated practically as an ideal crystal. Defects are quite rare, due to their high energy, and we can describe the interactions of defect lines by the methods developed in the previous chapters…
https://doi.org/10.1142/9789814415606_0029
The following sections are included:
https://doi.org/10.1142/9789814415606_0030
In the last chapter we focused attention mostly upon the gauge structure of stresses, since it leads to a simple defect representation of the melting model. Initially, however, when constructing the model for linear elasticity plus jump numbers nij, it was the defect gauge fields which played a primary role. The partition function (9.14) was invariant under the defect gauge transformations (9.15)–(9.18) and required a gaugefixing functional Φ which removed the gauge degeneracy. We stated that it was always possible to choose a gauge in which nij is quasi-symmetric and in which the symmetrized jump numbers have three components
,
,
vanishing identically with the non-zero components satisfying the boundary conditions (9.20), (9.21). With Φ
denoting the Kronecker δ enforcing these conditions, the partition function reads [note the differences with (9.14)]…
https://doi.org/10.1142/9789814415606_0031
In the following two chapters we are going to analyze the thermal properties of the lattice model of defect melting in a way similar to that employed in the Villain model of the superfluid phase transition. For this we shall first expand the partition function Z of (9.40) into a high temperature series.
https://doi.org/10.1142/9789814415606_0032
Let us now study the properties of the defect model and see whether it properly describes the phase transition of melting. We shall first consider the leading approximations at high and low temperatures, thereby obtaining crude estimates for the location and type of the transition. Then we shall proceed and calculate the corrections due to stress and defect graphs (which, in three dimensions, will be found to produce amazingly small effects).
https://doi.org/10.1142/9789814415606_0033
For comparison we shall now investigate the melting transition in the cosine form of our model as defined in Eq. (9.147). For simplicity, let us omit the β dependent prefactors and consider the slightly modified partition function,…
https://doi.org/10.1142/9789814415606_0034
Let us compare the properties of our defect models of melting with previous predictions on the behavior of two-dimensional defect melting as advanced by Kosterlitz, Thouless, Halperin, Nelson and Young (KTHNY).
https://doi.org/10.1142/9789814415606_0035
Now that we possess a lattice model which respects all stack-up properties of the defects in a crystal it is possible to derive a consistent disorder field theory of crystalline defects. This field theory turns out to be rather different from the tentative construction presented in Chapter 8. In contrast to that theory for which a good deal of effort was spent in finding possible mechanisms for making the transition first-order, the proper disorder field theory to be proposed in this chapter will have a natural way of undergoing a first-order transition right at the mean field level. We shall see that as was the case with the field theory involving order fields [see Section 13.1] the disorder field theory will contain a temperature dependent quartic term which naturally causes a first-order transition (recall Fig. 13.1). The disorder field theory contains D complex disorder fields, one for every lattice direction. It describes all possible configurations of the defect tensor ηij(x) and consequently a grand canonical ensemble of dislocations as well as disclinations.
https://doi.org/10.1142/9789814415606_0036
In Chapter 15 we constructed a model of defect melting which, after a duality transformation, yielded a partition function with a sum over symmetric discrete defect tensors satisfying the conservation law
. The diagonal parts of
were integer, the off-diagonal parts half-integer. Such a sum contained only three independent sets of integer numbers which were not able to distinguish the full variety of all possible defect lines. So the question arises as to how we have to modify the model so that all defect lines appear explicitly in the partition function. In order to work towards an answer to this question let us first try and reformulate the continuous decomposition of the defect tensor according to dislocations and disclinations in such a way that it can be used on a lattice.
https://doi.org/10.1142/9789814415606_0037
The following sections are included:
https://doi.org/10.1142/9789814415606_0038
Our goal is to investigate how our model of defect melting must be modified if we are to account for the higher gradient terms in linear elasticity. For this it is first necessary to understand the elastic properties of a continuous material which has undergone a given set of plastic deformations.
https://doi.org/10.1142/9789814415606_0039
The following sections are included:
https://doi.org/10.1142/9789814415606_0040
The defects in different physical systems have the common property that, in the continuum limit, certain closed contour integrals over field variables do not vanish due to singularities. If the field variables are spatial distortions, as in the case of crystals, the defects in the continuum may be efficiently described by means of differential geometry. A crystal filled with dislocations and disclinations turns out to have the same geometric properties as an affine space with torsion and curvature, respectively. Now, according to Einstein and later researchers, such a space forms the basis for a coordinate independent description of gravity. Mass points generate curvature, spinning matter gives rise to curvature and torsion. We may therefore expect many features of gravitational matter to coincide with those of crystalline defects…
https://doi.org/10.1142/9789814415606_0041
The following sections are included:
https://doi.org/10.1142/9789814415606_0042
The following sections are included:
https://doi.org/10.1142/9789814415606_0043
The following sections are included:
https://doi.org/10.1142/9789814415606_0044
According Noether’s theorem, the invariance of the action under general coordinate transformations and local Lorentz transformations must be associated with certains conservation laws. For the following considerations, it will be convenient to consider and
as independent variables and rename
as
. Then, from the derivation in (4.71) and (4.72), it follows that varying the action in
at fixed
gives the canonical energy-momentum tensor…
https://doi.org/10.1142/9789814415606_0045
The following sections are included:
https://doi.org/10.1142/9789814415606_0046
After the preparations in the previous chapters it is now easy to construct a geometric theory of stresses and defects. With the defects described by the geometry of an affine space, we have to find an appropriate way of incorporating the correct long-range elastic interactions between the defects into the theory.
https://doi.org/10.1142/9789814415606_bmatter02
The following sections are included:
“It is written in clear and interesting way and stresses the importance of field theoretical ideas and methods in the areas of condensed matter and solid state physics.”