Fractional vortex molecules and vortex polygonsin a baby Skyrme model

Fractional vortex molecules and vortex polygons
in a baby Skyrme model

Michikazu Kobayashi Department of Basic Science, University of Tokyo, Komaba 3-8-1, Meguro-ku, Tokyo 153-8902, Japan Department of Physics, Kyoto University, Oiwake-cho, Kitashirakawa, Sakyo-ku, Kyoto 606-8502, Japan    Muneto Nitta Department of Physics, and Research and Education Center for Natural Sciences, Keio University, Hiyoshi 4-1-1, Yokohama, Kanagawa 223-8521, Japan
August 24, 2019
Abstract

We construct a molecule of fractional vortices with fractional topological lump charges as a baby Skyrmion with the unit topological lump charge in the anti-ferromagnetic (or XY) baby Skyrme model, that is, an sigma model with a four derivative term and an anti-ferromagnetic or XY-type potential term quadratic in fields. We further construct configurations with topological lump charges and find that bound states of vortex molecules constitute regular polygons with vertices as vortices, where the rotational symmetry in real space is spontaneously broken into a discrete subgroup . We also find metastable and arrayed bound states of fractional vortices for . On the other hand, we find for that the regular polygon is metastable and the arrayed bound state is stable. We calculate binding energies of all configurations.

pacs:
11.27.+d, 14.80.Hv, 12.39.Dc

I Introduction

Vortices are topological solitons present in various physical systems from field theory Manton:2004tk and cosmological models Vilenkin:2000 to condensed matter systems Volovik:2003 . In particular, they play essential roles in condensed matter systems such as superconductors, superfluids, magnetism, quantum Hall states, nematic liquids, optics, and so on. One of the exotic aspects common in condensed matter systems but not familiar in high energy physics and cosmology are vortex molecules, which have been studied in multi-component Bose-Einstein condensates (BECs) Son:2001td ; Kasamatsu:2004 ; Kasamatsu:2005 ; Eto:2012rc ; Cipriani:2013nya ; Cipriani:2013wia ; Eto:2013 , multi-gap superconductors Babaev:2002 ; Goryo:2007 ; Nitta:2010yf , superfluid He (as a double core vortex) Volovik:2003 , and nonlinear optics optics . In the cases of BECs Son:2001td and superconductors Tanaka:2001 , fractional vortices in two different components with fractional topological charges constitute a meson-like bound state with the unit topological charge in total. However, a crucial difference between these two systems is that a repulsion between vortices is exponentially suppressed in superconductors due to the Higgs mechanism in the presence of a gauge field while a repulsion between vortices is polynomially reduced in BECs Eto:2011wp in the absence of a gauge field. Consequently, vortex molecules are stable and visible in BECs because of a balance between the vortex repulsion and the domain wall tension, while in superconductors they can be seen, in principle, only at temperatures above a certain critical temperature by a mechanism similar to the Berezinskii-Kosterlitz-Thouless transition Goryo:2007 ; Nitta:2010yf . However, stable vortex molecules in BECs are global vortices winding around a global symmetry, and consequently their energies are logarithmically divergent with respect to the system size; they are infinitely heavy in infinite space, and thereby they are not very realistic in high energy physics or cosmology.

In this paper, we propose a field theoretical model admitting a vortex molecule with finite energy, motivated by these condensed matter systems. We consider an nonlinear sigma model on the target space in dimensions, described by a unit three-vector of scalar fields with the constraint , which is equivalent to a model. The model admits lumps or sigma model instantons characterized by Polyakov:1975yp as a relative of vortices. We consider a potential term motivated by condensed matter systems admitting vortex molecules. The potential terms make lumps unstable against shrinkage, in general, as can be inferred from the scaling argument Derrick:1964ww , so we also consider a four derivative (Skyrme) term, by which the lumps are stabilized to become baby Skyrmions Piette:1994ug . In the context of baby Skyrmions, the potential terms of the type Piette:1994ug or of the type Kudryavtsev:1997nw ; Weidig:1998ii ; Harland:2007pb ; Nitta:2012kk have already been studied. The latter admits two discrete vacua and a domain wall interpolating between them Abraham:1992vb ; Kudryavtsev:1997nw ; Harland:2007pb ; Nitta:2012kj , and a baby Skyrmion is in the shape of a twisted closed domain wall Weidig:1998ii ; Kobayashi:2013ju . In our previous papers Nitta:2012xq ; Nitta:2012wi ; Kobayashi:2013ju , we considered both types of potential terms in the regime . In this case, a straight domain wall can absorb lumps as sine-Gordon kinks Nitta:2012xq ; Nitta:2012wi ; Kobayashi:2013ju , and a baby Skyrmion is in the form of a domain wall ring with a sine-Gordon kink Kobayashi:2013ju . In condensed matter physics, the quadratic potential admitting two vacua is known as the Ising-type in ferromagnets, so we may refer to this model as the Ising (or ferromagnetic) baby Skyrme model.

Figure 1: (a) The target space and the potential. (b) A vortex molecule as one lump. Here, , , and represent and , respectively. The north and south poles and are denoted by and , respectively.

Here, we consider a potential of the XY-type or of anti-ferromagnets, Jaykka:2010bq . The model can be referred as the XY (or anti-ferromagnetic) baby Skyrme model. The vacua characterized by are at the equator of the target space , as in Fig. 1 (a). One lump solution is schematically drawn in Fig. 1 (b), where we chose as the vacuum at the boundary. One can find two separated half lumps (merons) whose centers are mapped to the north and south poles of the target space . These half lumps are separated because the vacua appear between them as indicated by an ellipse in Fig. 1 (b). Therefore, a lump is non-axisymmetric, unlike those for the massless case and the Ising-type case with in which lumps are axisymmetric. There, the vacuum winds (counter)clockwise between the two half lumps so that these half lumps are (anti-)global vortices. While isolated (anti-)global vortices have logarithmically divergent energy in the infinite system size, a pair of global and anti-global vortices have finite energy. They attract each other and collapse in the absence of the Skyrme term, while the Skyrme term forms a vortex molecule. We numerically construct fractional vortex molecules with by a relaxation method. The configuration of , in fact, looks what we expected. For , we find that two molecules face to each other with opposite orientations to constitute a regular square. Since two kinds of constituent vortices are placed at the vertices, the configuration is symmetric. For , we find a regular hexagonal structure of vortex molecules like a benzene. This is symmetric. The configurations with resemble those recently found in a two-component BEC under rotation Cipriani:2013nya . Furthermore, for , we find regular octagonal, decagonal, and dodecagonal structures of fractional vortices with , , symmetries. Therefore, in general, we expect, for the topological charge , a polygon with fractional vortices at vertices with symmetry Jaykka:2010bq . We also find metastable and arrayed bound states of fractional vortices for . These configurations are obtained by squeezing the corresponding stable polygons, and they have slightly higher energies. We also find that the regular polygon is metastable and the arrayed bound state is stable for . Finally, we calculate binding energies of all the configurations.

Conventional lumps in the massless sigma model are invariant under a combination of the rotation of the target space and the space rotation. The lumps spontaneously break the other linear combination of the two symmetries. On the other hand, our solutions spontaneously break the purely space rotation. Note that non-axisymmetric configurations were known before for higher topological charges in the baby Skyrme model Delsate:2012hz , while even one baby Skyrmion is non-axisymmetric in our model. In this regard, our solutions are similar to those in Ref. Jaykka:2011ic , in which non-axisymmetric molecular configurations were found in the model with a more complicated potential. We would like to emphasize that our potential is quite common in anti-ferromagnets and two-component BECs and that it is natural.

A closely related model is a gauged supersymmetric model where a rotation along the axis is gauged Bagger:1982fn . The gauge symmetry induces the potential known as the D-term potential where supersymmetry requires to coincide with the gauge coupling . In this model, a lump is decomposed into two fractional gauged Bogomol’nyi-Prasad-Sommerfield (BPS) vortices, and these two vortices can be placed with arbitrary separation because no force is present between BPS vortices Schroers:1995he ; Nitta:2011um . Each constituent vortex carries half of the lump charge characterized by , as in our model. A set of fractional vortices carrying the total unit charge was also found in supersymmetric gauge theories and sigma models Eto:2009bz .

This paper is organized as follows. After our model is explained in Sec. II, we give numerical solutions of vortex molecules in Sec. III. Section IV is devoted to a summary and discussion.

Ii The model

We consider an sigma model in dimensions described by a three vector of scalar fields with a constraint . The Lagrangian of our model is given by

(1)

with . Here, the four derivative (baby Skyrme) term is expressed as

(2)

In this paper, we take the potential term to be Jaykka:2010bq

(3)

This potential is known in anti-ferromagnets and the XY model, while the potential in the form of ferromagnets was studied before in Refs. Weidig:1998ii ; Kudryavtsev:1997nw ; Harland:2007pb ; Nitta:2012kk . The energy density of static configurations is

(4)

Itroducing the projective coordinate of by

(5)

the Lagrangian (1) can be rewritten in the form of the model with potential terms, given by

(6)
(7)

Here, is the Kähler (Fubini-Study) metric of , is its inverse, and are the moment maps (or the Killing potentials) of the isometry generated by . If we gauge the isometry generated by the generator with gauge coupling , the potential (with ) is known as the D-term potential in the supersymmetric gauged model Bagger:1982fn .

Iii Vortex molecules

The topological charge of the lump is given by

(8)

In the presence of the potential, a lump is unstable to shrinking from the Derick’s scaling argument Derrick:1964ww . It can be stabilized in the presence of the baby Skyrme term, resulting in a baby Skyrmion.

Figure 2: Bound states of fractional vortex molecules in the region . The topological charges are from top to bottom. The left and middle panels represent the total energy densities and the topological charge densities , respectively. In the right panels, the arrows denote the scalar fields of the three-vector , where color represents the value of from which one can find whether the constituent fractional vortices are of N or S.
Figure 3: Bound states of fractional vortex molecules in the region . The topological charges are from top to bottom. Details of the plots are the same as those shown in Fig. 2.
Figure 4: Metastable and arrayed bound states of fractional vortex molecules in the region . The topological charges are for the top and bottom. Details of the plots are the same as those shown in Fig. 2.
Figure 5: Stable (arrayed) and metastable (heptagon and the other 3 types of) bound states of fractional vortex molecules for the topological charge . The top and bottom panels of each configuration represent the three-vector and the total energy density , respectively, in the region .

We construct numerical solutions of fractional vortex molecules with the topological lump charge . As the numerical parameters, we fix and . We obtain the stationary state using the relaxation method: introducing the parameter and the -dependence of , and finding the asymptotic solution of the equation

(9)

under the constraint . The detailed numerical procedure is shown in the Appendix A. As the initial state , we give the ansatz for :

(10)

with the monotonically decreasing function satisfying

(11)

The topological charge given by Eq. (8) is invariant for arbitrary .

Our stable solutions are given in Fig. 2 for and Fig. 3 for . For the unit topological charge, , one can find a fractional vortex molecule as we expected; two fractional vortices oppositely wind around the equator of the target space , and their cores are filled by the north and south poles of , as can be seen in the plot of in Fig. 2. The vacua appear between the two half lumps separating them, as can be seen from both the plots of the energy density and topological charge density. Although all of them are (anti-)global vortices having logarithmically divergent energy, a pair them has finite energy. They attract each other, but the Skyrme term prevents the collapse.

For , two vortex molecules face to each other with opposite orientations. Since molecules attract each other with these orientations, they make a bound state to constitute a square. Since the same vortices N or S are placed at each pair of diagonal corners, the configuration is axisymmetric.

For , three vortex molecules with six fractional vortices constitute a hexagon with a axisymmetry. These structures of resemble those in a vortex lattice recently found in a two-component BEC under rotation Cipriani:2013nya .

For , the situation is almost the same; i.e., one finds that four, five, and six vortex molecules with eight, ten, and eleven fractional vortices constitute an octagon, decagon, and dodecagon with , , axisymmetries, respectively.

In general, for the topological charge , we expect fractional vortices to be placed on a circle in a axisymmetric way. In all the cases, one can see that the topological lump charge density is distributed around the fractional vortices, and each of them carries a half lump charge. The rotational symmetry in the - plane is spontaneously broken in all cases to a discrete subgroup .

To investigate the stability of vortex polygons, we also choose a randomly placed lump solution

(12)

as initial states for the relaxation. Here, the projective coordinate is defined in Eq. (5), and , , and are the real random numbers. The initial state (12) indicates that single vortex molecules are placed at with the angle , having the topological charge .

For , any initial state of Eq. (12) relaxes to vortex polygons as shown in Figs. 2 and 3, while, for , several initial states relax to metastable states that are different from the vortex decagon and dodecagon, i.e., vortex array states as in Fig. 4, the energy of which is larger than vortex polygon states.

We also find five (meta)stable bound states for as in Fig. 5. Unlike the cases with , the arrayed bound state is at the absolute minimum. The regular polygon and the other three bound state are metastable at local minima.

1 29.04 8.840 8.829 46.71
2 53.35 16.96 16.96 87.26 6.153
3 78.45 25.17 25.17 128.8 11.33
4 103.8 33.44 33.44 170.7 16.14
5 (decagon) 129.3 41.73 41.73 212.7 20.79
5 (array) 129.8 41.68 41.68 213.1 20.42
6 (dodecagon) 154.8 50.03 50.03 254.9 25.35
6 (array) 155.2 49.90 49.90 255.0 25.22
7 (heptagon) 180.4 58.34 58.34 297.1 29.86
7 (array) 180.7 58.14 58.13 297.0 29.97
7 (type 1) 181.3 58.29 58.28 297.9 29.09
7 (type 2) 181.2 58.22 58.22 297.6 29.31
7 (type 3) 181.6 58.32 58.32 298.3 28.68
Table 1: Gradient energy , Skyrme energy , potential energy , total energy , and binding energy for vortex molecules.

Finally, we show in Table 1 the gradient energy , the Skyrme energy , the potential energy , the total energy , and the binding energy between vortex molecules, defined by

(13)

respectively. In our choice of numerical parameters, the gradient energy is dominant in the total energy. Total energy deviates from the linear relation, i.e., , and the difference between and corresponds to the binding energy between molecules, which dominates in about % of the total energy.

Iv Summary and Discussion

We have constructed stable and metastable configurations of the fractional vortex molecules as lumps (baby Skyrmions) in the XY (or anti-ferromagnetic) baby Skyrme model which has the anti-ferromagnetic (XY) potential . We have found that, for the unit charge , two fractional vortices whose centers are filled by the north and south poles of the target space are placed within a certain distance. We have found that, for and , bound states of two, three, four, five, and six vortex molecules constitute quadrangle, hexagonal, octagonal, decagonal, and dodecagonal vortex configurations, respectively. At least up to this topological number, symmetric vortex molecule configurations appear for the topological charge . Our configurations are all non-axisymmetric, and they spontaneously break the rotational symmetry in the - plane. While all vortex polygons are stable and at global minima, we have also found metastable and arrayed bound states of fractional vortices for , which are obtained by squeezing the corresponding stable polygons and have slightly higher energies. We also find for that the arrayed bound state is at the absolute minimum and the regular polygon together with the other three bound state is metastable at a local minimum, unlike the cases with Finally, we have calculated the binding energies of all the configurations.

As denoted in the Introduction, similar configurations of vortex molecules are present in condensed matter systems, such as two component BECs, described by two condensations (scalar fields) and with the internal coherent (Josephson) coupling c.c. Son:2001td ; Kasamatsu:2004 ; Kasamatsu:2005 ; Eto:2012rc ; Cipriani:2013nya , where a four derivative term is not present. In these cases, the vortex molecules are global vortices, that is, they wind around a global symmetry. If we gauge the common phase of the two components, they become semi-local vortices. If we send the scalar coupling and gauge coupling to infinity with keeping the internal coherent coupling, we obtain our model except for the Skyrme term where the Josephson coupling c.c reduces to , which we did not consider in this paper. We expect that semi-local vortices, keeping the couplings, can make a stable vortex molecule if one properly adds a four derivative term.

In the model with a ferromagnetic potential, a Q-lump solution is known Leese:1991hr , in which a Nambu-Goldstone mode associated with the spontaneously broken internal symmetry of the rotation in the - plane in the target space is rotating in time. In our case, there is a Nambu-Goldstone mode associated with the spontaneously broken rotational symmetry in real space, instead of an internal Nambu-Goldstone mode. Consequently, we may have a Q-lump in our case as a spinning molecule.

If we promote our configuration linearly in dimensions, it becomes a bound state of two cosmic strings. As usual, the solution breaks translational symmetries in two transverse directions, resulting in two translational zero modes which propagate along the string. One added feature of our solutions is the existence of a Nambu-Goldstone mode associated with the spontaneously broken rotational symmetry along the string, resulting in a twisting wave propagating along the string. This is known as a “twiston” in two-gap superconductors Tanaka:2007 . In cosmology, our solutions can be regarded as some exotic cosmic strings with an internal structure. For instance, it is an interesting question whether or not two such strings reconnect to each other when they collide.

We have found that, for the topological charge , two kinds of vortices are placed at vertices of a regular polygon, on which a symmetry acts. In this regard, symmetric vortex configurations were studied in the Abelian-Higgs model Arthur:1995eh . This is equivalent to vortices on the orbifold studied recently Kimura:2011wh . Vortex polygons have also been studied in hydrodynamics for a long time fluid . Vortex polygons with less than seven vortices as vertices are shown to be stable, while those with more than seven are unstable. One example realized in nature is a vortex hexagon found by Cassini in the north poles of Saturn saturn . In our case too, we have found vortex polygons up to to be stable, which may be interesting compared with hydrodynamics.

In this paper, we have found a two vortex molecule, that is, a vortex dimer with the unit topological charge in the model with the anti-ferromagnetic potential. Stable three or vortex molecules, that is, vortex trimers or -omers, are present in three or component BECs Eto:2012rc ; Eto:2013 . Since two components with a constraint divided by imply the model, the same procedure for components yields a model with a certain potential term. While a generalization of the anti-ferromagnetic type potential was considered before AlvarezGaume:1983ab and was found to admit parallel multiple domain walls Gauntlett:2000ib or domain wall junctions or networks Eto:2005cp ; Eto:2006pg , depending on parameters in the potential, a generalization of the ferromagnetic (XY-type) of the potential has not been studied thus far in the model. Skyrme-like terms in the model were studied before without Ferreira:2008nn ; Liu:2009rz and with Bergshoeff:1984wb ; Eto:2012qda ; Adam:2011hj supersymmetry. A or generalization of the baby Skyrme model with an anti-ferromagnetic potential will admit vortex trimers or -omers, respectively.

Note added

After the paper was published, vortex polygons have been studied in a baby Skyrme model with a different potential Winyard:2013ada .

Acknowledgements

M. N. thanks Minoru Eto for collaborations which motivated this work. This work is supported in part by Grant-in-Aid for Scientific Research (Grants No. 22740219 (M.K.) and No. 23740198 and No. 25400268 (M.N.)) and the work of M. N. is also supported in part by the “Topological Quantum Phenomena” Grant-in-Aid for Scientific Research on Innovative Areas (No. 23103515 and No. 25103720) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

Appendix A Detailed numerical procedure

Equation (9) can be solved as the steepest descent method:

(14)

where we omit the spatial dependence of . is defined as

(15)

Here, the Lagrange multiplier is fixed to satisfy .

For the space, to approximately consider the infinite space, we use the following scaling transformation:

(16)

for , and consider the dependence of on instead of , where is the scaling parameter. We use the square with the grid points. On the -th grid point in the -direction, the value of is defined as

(17)

where is the value of at . We omit the -dependence on here. For or , which corresponds to infinity, the value of is fixed to the ground state:

(18)

To calculate the spatial derivative of , we use the spectral collocation method. We expand in the Chebyshev polynomials:

(19)

where is the -th coefficient of the Chebyshev expansion in the -direction. The first and second spatial derivatives and can be calculated as

(20)

Coefficients , and satisfy the following recurrence relations:

(21)

Equations (19) and (20) can be calculated by the fast Fourier transform algorithm.

In this paper, we fix , , .

References

Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
290084
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description