Point-group symmetries play a very important role in nuclear mean-field theories. Two distinct aspects of this role can be singled out. Firstly, the point symmetries of a Hamiltonian provide good quantum numbers that can conveniently be used to label its eigenstates. They help to formulate selection rules for electromagnetic and/or other types of transitions, and allow for solving the stationary problems in subspaces rather than in the complete Hilbert space of the problem in question. In that respect the use of point symmetries in nuclear physics resembles their use in other branches of physics. Secondly, however, and this aspect is more specific to nuclear structure domain, the use of the self-consistent Hartree-Fock (HF) or Hartree-Fock-Bogolyubov (HFB) mean field approximations invariably leads to the problem of self-consistent symmetries and related spontaneous symmetry-breaking mechanisms[1].
In this article we aim at describing properties of nuclear one-body densities under the action of point symmetries. For the time-even densities we calculate the electric multipole moments which give information about nuclear shapes. Various point symmetries obeyed by the hamiltonian lead then to various types of allowed shapes. In addition, for the time-odd densities we calculate magnetic multipole moments which give information about current distributions in nuclei, i.e., about the "shapes" of matter flow. Again, various conserved symmetries restrict these flow patterns in different ways that are studied in this paper.
Numerous experiments indicate that a great number of nuclei are deformed in their ground states. Interpretation of the corresponding results shows that most often the shapes involved are axially-symmetric. Many realistic calculations, e.g., those based on the nuclear mean-field approximation, reproduce these experimental data. However, the same calculations suggest that the excited nuclear states often correspond to the nucleonic mass distributions that have the so-called triaxial shapes. It thus becomes clear that in a realistic description of the nuclear properties, the spontaneous symmetry breaking leading to the triaxially symmetric objects must be given attention.
The classical point group that contains three mutually perpendicular
symmetry axes of the second order passing through a common point
is denoted D
(cf., e.g.,
Refs. [2,3,4,5]). Attaching to D
the
three mutually perpendicular symmetry planes spanned on the symmetry axes
gives the D
point group
which contains all the spatial symmetries of interest in the present paper.
As it is well known, the classical (single) point groups can be
applied to spinless particles and/or systems of an even number of
fermions, and thus to even-even nuclei. However, for odd fermion
systems and, in particular, in the single-nucleon space, these have
no faithful irreducible representations. There exist two methods to
remedy this problem. One is to extend the notion of the group
representation and to introduce projective or ray representations
[2,6,7]. Another one, which is employed in
the present work, is to enlarge the single groups by adjoining the
rotation through angle
and all its products with the original
group elements, and to double in this way the order of the
group[5].
Physically, the need of such an extension is related to the fact that
in the space of spinors the rotation through angle of necessarily changes the sign of the wave
function of an odd-fermion system.
Since within the group theory a multiplication of
a group element by a number is not defined, the change of sign
must be introduced as an extra group element. The point group
enlarged in this way is called the double point group and usually
denoted with the superscript ``D'' (cf., e.g.,
Refs. [4,5]), although some authors, see, e.g.,
[3], denote single and double point groups by the same
symbols. Here we follow the former convention, and thus the double
group corresponding to single group D
is denoted by D
.
In the case of classical objects the elements of a symmetry point
group are real orthogonal coordinate transformations. In quantum
mechanics it is often of advantage to take into consideration both
the spatial symmetries and the time-reversal operator explicitly
and treat them as elements of a common ensemble of symmetry operators.
Time-reversal symmetry operator (antilinear) and the space
symmetries (linear) have usually been considered separately
(cf., e.g., Ref. [3]). Here we follow Ref.[8]
and add the time-reversal
operator to the set of group elements, thus obtaining new groups denoted
D
and
D
.
Hence, the D
group
is a subgroup of D
composed of its linear elements, and similarly, the
D
group is composed of the linear elements of D
.
Gauge symmetries, which pertain to pairing correlations of nucleons, can be added independently and are not considered in the present study. In particular, neutron-proton correlations are not discussed. Therefore, the isospin degree of freedom is irrelevant in the discussion and can be disregarded to simplify the notation.
In this paper our goal is threefold: First, in Sec. 2, we
present and discuss properties of the single group D
and double
group D
that are appropriate for a description of even and odd
fermion systems, respectively. In particular, we recall the
classification of representations of both groups, and classify
properties of the group elements when they are represented in fermion
Fock space. Second, in Sec. 3, we present explicit symmetry
properties of local densities with respect to the operators in the
D
group. This problem has been solved in particular
applications [9,10]; however, it can be solved in
many different ways, and it is useful to have a systematic approach
which enumerates all available options. Although the local densities
are most important for applications using the local density
approximation (LDA), or those using the Skyrme effective interaction
(see respectively Refs.[11,12] or Ref.[13]
for reviews), they also define general properties of average values
of any local one-body operator. Finally, in Sec. 4, we
discuss symmetries of multipole moments which define the nuclear
shapes and currents, and in Sec. 5 we present conclusions
which can be drawn from our study. In the companion paper[14],
we discuss physical aspects of the symmetry-breaking schemes
pertaining to the point groups in question.