Spinor Bose-Einstein condensates

Masahito Uedaa,b , Yuki Kawaguchia

a Department b ERATO

of Physics, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-0033, Japan Macroscopic Quantum Control Project, JST, Tokyo 113-8656, Japan

Abstract An overview of the physics of spinor and dipolar Bose-Einstein condensates (BECs) is presented. Mean-eld ground states, Bogoliubov spectra, and many-body ground and excited states of spinor BECs are discussed. Properties of spin-polarized dipolar BECs and those of spinordipolar BECs are reviewed. Some of the unique features of the vortices in spinor BECs, such as fractional vortices and non-Abelian vortices, are delineated. The symmetry of the order parameter is classied using group theory, and various topological excitations are investigated based on homotopy theory. Some of the more recent developments such as the Kibble-Zurek mechanism of defect formation and the Berezinskii-Kosterlitz-Thouless transition in a spinor BEC are also discussed. Key words: spinor BEC, dipolar BEC, vortices, topological excitations, low-dimensional systems

Contents 1 Introduction 2 General Theory 2.1 Single-particle Hamiltonian . . . . . . . . . . . . . . . . . 2.2 Interaction Hamiltonian . . . . . . . . . . . . . . . . . . . 2.2.1 Symmetry considerations and irreducible operators 2.2.2 Operator relations . . . . . . . . . . . . . . . . . . 2.2.3 Interaction Hamiltonian of spin-1 BEC . . . . . . 2.2.4 Interaction Hamiltonian of spin-2 BEC . . . . . . 2.2.5 Interaction Hamiltonian of spin-3 BEC . . . . . . 5 7 . 7 . 7 . 7 . 8 . 10 . 10 . 11

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

3 Mean-Field Theory of Spinor Condensates 11 3.1 Number-conserving theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3.2 Mean-eld theory of spin-1 BECs . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.3 Mean-eld theory of spin-2 BECs . . . . . . . . . . . . . . . . . . . . . . . . . 17

Email addresses: ueda@phys.s.u-tokyo.ac.jp (Masahito Ueda), yuki@cat.phys.s.u-tokyo.ac.jp (Yuki Kawaguchi)

Preprint submitted to Physics Report

March 16, 2010

4 Dipolar BEC 4.1 Dipoledipole interaction . . . . . . . . . . . . . . . . . 4.1.1 Scattering properties of dipole interaction . . . . 4.1.2 Dipolar systems . . . . . . . . . . . . . . . . . . 4.1.3 Tuning of dipole-dipole interaction . . . . . . . 4.1.4 Numerical method . . . . . . . . . . . . . . . . 4.2 Spin-polarized Dipolar BEC . . . . . . . . . . . . . . . 4.2.1 Equilibrium shape and instability . . . . . . . . 4.2.2 Dipolar collapse . . . . . . . . . . . . . . . . . 4.2.3 Roton-maxon excitation . . . . . . . . . . . . . 4.2.4 Two-dimensional solitons . . . . . . . . . . . . 4.2.5 Supersolid . . . . . . . . . . . . . . . . . . . . 4.2.6 Ferrouid . . . . . . . . . . . . . . . . . . . . . 4.3 Spinor-dipolar BEC . . . . . . . . . . . . . . . . . . . . 4.3.1 Einstein-de Haas eect . . . . . . . . . . . . . . 4.3.2 Ground-state spin textures at zero magnetic eld 4.3.3 Dipole-dipole interaction under a magnetic eld . 5 Bogoliubov Theory 5.1 Spin-1 BECs . . . . . . . . . . 5.1.1 Ferromagnetic phase . . 5.1.2 Antiferromagnetic phase 5.1.3 Domain formation . . . 5.2 Spin-2 BECs . . . . . . . . . . 5.2.1 Ferromagnetic phase . . 5.2.2 Antiferromagnetic phase 5.2.3 Cyclic phase . . . . . . 5.3 Dipolar BEC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

25 25 25 27 27 28 29 30 31 31 32 33 35 35 37 38 40 43 45 46 48 49 50 51 52 54 55 58 59 60 60 64 66 67 68 68 69 70 71 71 73

6 Vortices in Spinor BECs 6.1 Mass and spin supercurrents in spinor BECs 6.2 Spin-1 BEC . . . . . . . . . . . . . . . . . 6.2.1 Ferromagnetic phase . . . . . . . . 6.2.2 Polar phase . . . . . . . . . . . . . 6.2.3 Stability of vortex states . . . . . . 6.3 Spin-2 BEC . . . . . . . . . . . . . . . . . 6.3.1 Ferromagnetic phase . . . . . . . . 6.3.2 Uniaxial nematic phase . . . . . . . 6.3.3 Biaxial nematic phase . . . . . . . 6.3.4 Cyclic phase . . . . . . . . . . . . 6.4 Rotating spinor BEC . . . . . . . . . . . . 6.5 Spin-polarized dipolar BEC . . . . . . . . . 6.6 Non-Abelian vortices . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

7 Topological Excitations 7.1 Symmetry classication of broken symmetry states . . . . . . . . . . . 7.1.1 Order parameter manifold . . . . . . . . . . . . . . . . . . . . 7.1.2 Symmetry of order parameter and stationary states . . . . . . . 7.1.3 Procedure to nd ground states . . . . . . . . . . . . . . . . . . 7.1.4 Symmetry and order parameter structure of spin-2 spinor BECs 7.2 Homotopy theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Classication of topological excitations . . . . . . . . . . . . . 7.2.2 Relative homotopy groups . . . . . . . . . . . . . . . . . . . . 7.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Line defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.2 Point defects . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.3 Action of one type of defect on another . . . . . . . . . . . . . 7.3.4 Skyrmions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Many-Body Theory 8.1 Many-body states of spin-1 BECs . . . . . . . . . . 8.1.1 Eigenspectrum and eigenstates . . . . . . . . 8.1.2 Fragmentation . . . . . . . . . . . . . . . . 8.2 Many-body states of spin-2 BECs . . . . . . . . . . 8.2.1 Eigenspectrum and eigenstates . . . . . . . . 8.2.2 Magnetic response . . . . . . . . . . . . . . 8.2.3 Symmetry considerations on possible phases

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

74 74 74 76 78 80 83 83 86 88 88 89 91 92 94 95 95 97 99 100 102 104

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

9 Special Topics 105 9.1 Quenched BEC: Kibble-Zurek mechanism . . . . . . . . . . . . . . . . . . . . . 105 9.1.1 Instantaneous quench . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 9.1.2 Finite-time quench . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 9.1.3 Numerical simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 9.2 Low-dimensional systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 9.2.1 Berezinskii-Kosterlitz-Thouless transition in a single-component Bose gas 110 9.2.2 2D spinor gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 10 Summary and Future Prospects 115

symbol AF M (r, r ) A aF d F,M f, m f(r) = (f x , fy , fz ) m F(r) = mn fmn n F(r) = mn fmn n m f (r) = mn m fmn n i, j, k M Mz m, n n(r) = m m m p q s = F/|F| t, u m (r) m (r) m m / n , ,

denition annihilation operator for atomic pairs in |F , M pair amplitude of the spin-singlet pair, A00 0 s-wave scattering length of the total spin-F channel unit vector indicating the spin quantization axis total spin of two colliding atoms atomic spin and its projection vector of f f spin matrices spin operator spin expectation value spin expectation value per atom indices for x, y, and z in the coordinate space atomic mass total magnetization of the condensate indices for the magnetic sublevel number density linear Zeeman energy quadratic Zeeman energy unit vector spin along the spin expectation value unit vector satisfying s = t u eld operator order parameter normalized spinor order parameter indices for x, y, and z in the spin space
Table 1: Table of symbols

abbreviation AF BA BEC BN C CSV DDI DW F FL GP MI P PCV SF SS UN

denition antiferromagnetic broken axisymmetry Bose-Einstein condensate biaxial nematic cyclic chiral spin vortex dipole-dipole interaction density wave ferromagnetic ower Gross-Pitaevskii Mott insulator polar polar-core vortex superuid supersolid uniaxial nematic
Table 2: Table of abbreviations

1. Introduction Most Bose-Einstein condensates (BECs) of dilute atomic gases realized thus far have internal degrees of freedom arising from spins. When a BEC is trapped in a magnetic potential, the spin of each atom is oriented along the direction of the local magnetic eld. The spin degrees of freedom are therefore frozen and the BEC is described by a scalar order parameter. On the other hand, when the BEC is conned in an optical trap, the direction of the spin can change dynamically due to the interparticle interaction. Consequently, the order parameter of a spin- f BEC has 2 f + 1 components that can vary over space and time, producing a very rich variety of spin textures. A BEC with spin degrees of freedom is referred to as a spinor BEC. This article provides an overview of the physics of spinor and dipolar BECs. Bose-Einstein condensation is a genuinely quantum-mechanical phase transition in that it occurs without the help of interaction. However, in the case of spinor BECs, several phases are possible below the transition temperature T c , and which phase is realized does depend on the nature of the interaction. With the spin and gauge degrees of freedom, the full symmetry of the system above T c is SO(3) U(1). Below the transition temperature, the symmetry is spontaneously broken in several dierent ways; the number of possible broken symmetry states is four for the spin-1 case, at least ve for the spin-2 case, and the situation has yet to be fully understood for the spin-3 case. This wealth of possible phases makes a spinor BEC a fascinating area of research for quantum gases. The spin is accompanied by a magnetic moment that gives rise to magnetism. Depending on the phases of individual systems, spinor BECs exhibit a rich variety of magnetic phenomena. Moreover, because the systems are superuid, we can expect an interplay between superuidity and magnetism. For example, a ferromagnetic BEC has spin-gauge symmetry; if you rotate the spin locally, the system tries to undo it by generating a supercurrent. The basic properties of the mean-eld spinor condensates are reviewed in Sec. 3. The magnetic moment of the atom leads to the magnetic dipole-dipole interaction. Although the strength of this interaction is the smallest among all the energy scales involved in the system, it plays a pivotal role in producing the spin texturethe spatial variation of the spin direction. The magnetic dipole-dipole interaction plays key roles in determining the static and dynamic properties of spin-polarized dipolar BECs. It also serves as a key parameter in determining the quantum phases in an ultralow magnetic eld (below 10 G). The physics of dipolar BECs is discussed in Sec. 4. The Bose-Einstein condensed system responds to an external perturbation in a very unique manner. Even if the perturbation is weak, the response may be nonperturbative; for example, the phonon velocity depends on the scattering length in a nonanalytic manner. The low-lying excitations of spinor and dipolar BECs are described by the Bogoliubov theory, as discussed in Sec. 5. One of the hallmarks of superuidity manifests itself in its response to an external rotation. In a scalar BEC, the system hosts vortices that are characterized by the quantum of circulation, = h/M, where h is the Planck constant and M, the mass of the atom. The origin of this quantization is the single-valuedness of the order parameter. As mentioned above, however, in the spinor BEC, the gauge degree of freedom is coupled with the spin degrees of freedom. This spin-gauge coupling gives rise to some of the unique features of the spinor BEC. For example, the fundamental unit of circulation can be a rational fraction of , and when two vortices collide, they may not reconnect unlike the case of the scalar BEC. Vortices of spinor BECs are discussed in Sec. 6. 5

The direction of the spin can vary rather exibly over space and time. Nonetheless, the global conguration of the spin texture must satisfy certain topological constraints. This is because the order parameter in each phase belongs to a particular order parameter manifold whose symmetry leads to a conserved quantity called a topological charge. Such a topological constraint determines the nature of topological excitations such as line defects, point defects, skyrmions, and knots. The order parameter can be classied using group theory, and possible topological excitations are determined based on homotopy theory. Elements of group theory and homotopy theory with applications to spinor BECs are reviewed in Sec. 7. Eorts are underway to magnetically shield the system so that the ambient magnetic eld is reduced to below 10 G. In such an ultralow magnetic eld, the spin degeneracy, which is usually lifted by Zeeman eects, sets in signicantly, and we have to confront the question of how the degeneracy can be lifted by minute spin-exchange interactions. It is in this region where the many-body spin correlations dramatically alter the ground states of spinor BECs and their response to an external magnetic eld. The present understanding of this problem is described in Sec. 8. Spinor BECs oer topics that are of broad interest beyond ultracold atomic gases. Prime examples are the dynamics of a quenched BEC and the Berezinskii-Kosteritz-Thouless transition in two dimensions. These subjects are discussed in Sec. 9. It seems almost certain that we will face more interesting questions than have been answered so far. In fact, several new possibilities are emerging both experimentally and theoretically: magnetic crystalization and non-Abelian vortices, to cite only a few examples. Moreover, under an ambient magnetic eld below 10 G, not only do manybody spin correlations become signicant but the magnetic dipole-dipole interaction also plays an equally important role. We therefore have good reasons to expect that new symmetry breakings will occur in this regime. In view of these ongoing developments, it now seems appropriate to consolidate the knowledge that has been accumulated over the past few years. In this paper, we provide an overview of the basics and recent developments on the physics of spinor and dipolar BECs. Some of the remaining and emerging problems are listed in Sec. 10. One of the major topics that is not treated in this review is ctitious spin systems such as a binary mixture of hyperne spin states. Such systems do not possess rotational symmetry in space nor is the ctitious spin associated with the magnetic moment. However, they have some intrinsic interest such as interlaced vortex lattices and vortex molecules caused by the Josephson coupling. A comprehensive review of this subject is given in Ref. [1]. This paper is organized as follows. Section 2 describes the fundamental Hamiltonian of the spinor BEC. Section 3 develops the mean-eld theory of spinor condensates and discusses the ground-state properties of the spin-1, 2, and 3 BECs. Section 4 provides an overview of the dipole BEC and the spinor-dipolar BEC. Section 5 develops the Bogoliubov theory of the spinor and spinor-dipolar BEC. Section 6 discusses various types of vortices that can be created in spinor BECs. In particular, fractional vortices and non-Abelian vortices are discussed. Section 7 examines the topological aspects of spinor BECs. The symmetry of the order parameter is classied based on group theory, and possible topological excitations are investigated using homotopy theory. Section 8 reviews the many-body aspects of spinor BECs. Section 9 discusses the Kibble-Zurek mechanism and the Berezinskii-Kosterlitz-Thouless transition in spinor BECs. Section 10 summarizes the main results of this paper and discusses possible future developments.

2. General Theory 2.1. Single-particle Hamiltonian The fundamental Hamiltonian of a spinor BEC can be constructed quite generally based on the symmetry argument. We consider a system of identical bosons with mass M and spin f that are described by the eld operators m (r), where m = f, f 1, , f denotes the magnetic quantum number. The eld operators are assumed to satisfy the canonical commutation relations n [m (r), (r )] = mn (r r ), n m [m (r), n (r )] = [ (r), (r )] = 0,

(1)

where mn is the Kronecker delta. The noninteracting part of the Hamiltonian comprises the kinetic term, one-body potential U(r), and linear and quadratic Zeeman terms:
f

H0 =

dr
m,n= f

2M

2 + U(r) p(fz )mm + q(fz )mn n ,

(2)

where fz is the z-component of the spin matrix whose matrix elements are given by (fz )mn = mmn , 2 and hence, (fz )mn = m2 mn ; and p = gB B is the product of the Land hyperne g-factor g, e Bohr magneton B = e /2me (me is the electron mass, and e > 0 is the elementary charge), and external magnetic eld B that is assumed to be applied in the z-direction. The coecient q of the quadratic Zeeman term is calculated using the second-order perturbation theory to be q= (gB B)2 , Ehf (3)

where Ehf = Em Ei is the hyperne splitting and it is given by the dierence between the initial (Ei ) and intermediate (Em ) energies. For example, for the case of 87 Rb, g = 1/2 and Ehf 6.8 GHz for the f = 1 hyperne manifold and g = 1/2 and Ehf 6.8 GHz for the f = 2 hyperne manifold. The value of q can be tuned in the negative direction by using a linearly polarized microwave eld due to the AC Stark shift [2, 3]. 2.2. Interaction Hamiltonian 2.2.1. Symmetry considerations and irreducible operators Next, we consider the interaction Hamiltonian. We discuss the case of the magnetic dipoledipole interaction later as it does not conserve the total spin of the system, and hence, it requires a separate treatment. Because the system is very dilute, we consider only the binary interaction. Upon the exchange of two identical particles of spin f , the many-body wave function changes by the phase factor (1)2 f . By the same operation, the spin and orbital parts of the wave function change by (1)F +2 f and (1)L , respectively, where F is the total spin of the two particles and L, is the relative orbital angular momentum between them. To be consistent, (1)2 f must be equal to (1)F +2 f (1)L ; hence, (1)F +L = 1. Thus, F + L must be even, regardless of the statistics of the particles. In the following, we consider only the s-wave scattering (L = 0), and therefore, the total spin of two interacting particles must be even. The interaction Hamiltonian of the binary collision can therefore be written as V=
F =0,2, ,2 f

V (F ) , 7

(4)

 where V (F ) is the interaction Hamiltonian between two bosons whose total spin is F . The interaction V (F ) in Eq. (4) can be constructed from the irreducible operator AF M (r, r ) that annihilates a pair of bosons at positions r and r . The irreducible operator is related to a pair of eld operators via the Clebsch-Gordan coecients F , M| f, m1 ; f, m2 as follows:
f

AF M (r, r ) =
m1 ,m2 = f

F , M| f, m1 ; f, m2 m1 (r)m2 (r ).

(5)

Because the boson eld operators m1 (r) and m2 (r ) commute, AF M (r) vanishes identically for odd F . In fact, when F = 1 and f = 1, the Clebsch-Gordan coecient is given by 1, M|1, m1 ; 1, m2 = (1)1m1 m1 +m2 ,M 2 M,1 (m1 ,1 + m1 ,0 ) + M,0 m1 M,1 (m1 ,0 + m1 ,1 ) .

(6)

Substituting this into Eq. (5), we nd that A1M (r) = 0. Similarly, we can show that AF M (r) = 0 if F is odd. Incidentally, for the case of fermions, we obtain AF M (r) = 0 for even F because the fermion eld operators anticommute. Because the interaction is scalar, it must take the following form: 1 V (F ) = 2
F

dr

dr v(F ) (r, r )
M=F

A M (r, r )AF M (r, r ), F

(7)

where v(F ) (r, r ) describes the dependence of the interaction on the positions of the particles. When the system is dilute and the range of interaction is negligible as compared to the interparticle spacing, the interaction kernel v(F ) may be approximated by the delta function with an eective coupling constant gF : v(F ) (r, r ) = gF (r r ), where gF is related to the s-wave scattering length of the total spin-F channel, aF , as gF = 4 2 aF . M (9) (8)

2.2.2. Operator relations It follows from the completeness relation |F , M F , M| = 1

F ,M

(10)

that
F F =0,2, ,2 f M=F

A M (r, r )AF M (r, r ) =: n(r) (r ) :, n F

(11)

where
f

n(r)
m= f

m (r)m (r) 8

(12)

is the total density operator and :: denotes normal ordering that places annihilation operators to the right of creation operators. Another useful relation can be derived from the composition law of the angular momentum: F1 F2 = 1 1 (F1 + F2 )2 F2 F2 = F2 f ( f + 1), 1 2 2 2 tot (13)

where Ftot = F1 + F2 is the total spin angular momentum vector. Operating this on the completeness relation (10), we have F1 F2 =
F ,M

1 F (F + 1) f ( f + 1) |F , M F , M|. 2

(14)

Applying this formula to the case of f = 1, we obtain

2

F1 F2 =
M=2

|2, M 2, M| 2|0, 0 0, 0| = 1 3|0, 0 0, 0|,

(15)

where we have used the completeness relation (10) to derive the second equality. It follows from this that the following operator identity holds: 3A (r, r )A00 (r, r )+ : F(r) F(r ) :=: n(r) (r ) :, n 00 where F = (F x , Fy , Fz ) is the spin vector operator dened by
f

(16)

F (r) =
m,n= f

m (f )mn (r)n (r) ( = x, y, z).

(17)

Here, (f )mn ( = x, y, z) are the (m, n)-components of spin matrices f . For the case of f = 2, we have F1 F2 = 6|0, 0 0, 0| 3
M

|2, M 2, M| + 4
M

|4, M 4, M|.

(18)

The last term can be eliminated by using the completeness relation (10), giving the following operator identity: 10 1 4 A (r, r )A2M (r, r ) + A (r, r )A00 (r, r ) = : n(r) (r ) : . : F(r) F(r ) : + n 2M 7 7 00 7 M=2 Similarly, for the case of f = 3, we obtain 1 21 : F(r) F(r ) : + A (r, r )A00 (r, r ) 11 11 00 + 9 18 A (r, r )A4M (r, r ) = A (r, r )A2M (r, r ) + : n(r) (r ) : . n 4M 2M 11 M=2 11 M=4 9
2 4 2

(19)

(20)

2.2.3. Interaction Hamiltonian of spin-1 BEC We derive the interaction Hamiltonians for the f = 1 case. For f = 1, the total spin F of two colliding bosons must be 0 or 2, and the corresponding interaction Hamiltonian (7) gives V (0) V (2) = = = g0 2 g2 2 g2 2 drA (r)A00 (r) 00
2

(21)

dr
M=2

A (r)A2M (r) 2M (22)

dr : n2 (r) : A (r)A00 (r) , 00

where AF M (r) AF M (r, r), and Eq. (11) and A1M (r) = 0 are used to derive the last equality. Combining Eqs. (21) and (22), we obtain V= dr g2 g0 g2 A00 (r)A00 (r) . : n2 (r) : + 2 2 (23)

Equation (23) is the interaction Hamiltonian of the spin-1 BEC. We use the operator identity (16) to rewrite Eq. (23) as 1 V= 2 where c0 = g0 + 2g2 g2 g0 , c1 = . 3 3 (25) dr c0 : n2 (r) : +c1 : F2 (r) : , (24)

In Eq. (23), A00 is the spin-singlet pair operator. The explicit form of this operator can be found using the Clebsch-Gordan coecient for F = M = 0: 0, 0| f, m1 ; f, m2 = m1 +m2 ,0 Substituting this into Eq. (5), we obtain A00 (r, r ) = 1
f

(1) f m1 2f + 1

(26)

2 f + 1 m= f

(1) f m m (r)m (r ).

(27)

2.2.4. Interaction Hamiltonian of spin-2 BEC For f = 2, F must be 0, 2, or 4, and the interaction Hamiltonian (7) for F = 4 gives V (4) = = g4 2 g4 2
4

dr
M=4

A (r)A4M (r) 4M
2 M=2

dr : n2 (r) : A (r)A00 (r) 00 10

(r)A2M (r) , A2M

(28)

 where Eq. (11) and A1M = A3M = 0 are used to obtain the last equality. Summing Eqs. (21), (22), and (28), we obtain V = V (0) + V (2) + V (4) g4 g0 g4 g2 g4 = dr : n2 (r) : + A00 (r)A00 (r) + 2 2 2
2 M=2

(r)A2M (r) . A2M

(29)

We can eliminate the last term by using the operator identity (19), obtaining V where c0 = 4g2 + 3g4 g4 g2 7g0 10g2 + 3g4 , c1 = , c2 = . 7 7 7 (31) = 1 2 dr c0 : n2 (r) : +c1 : F2 (r) : +c2 A (r)A00 (r) , 00 (30)

We use the same notations c0 and c1 for both spin-1 and spin-2 cases because there is no fear of confusion. 2.2.5. Interaction Hamiltonian of spin-3 BEC In a similar manner, the interaction Hamiltonian of the spin-3 spinor BEC is given by 2 1 c0 : n2 (r) : +c1 : F2 (r) : +c2 A (r)A00 (r) + c3 A2M (r)A2M (r) , (32) V = dr 00 2 M=2 where c0 = 9g4 + 2g6 g6 g4 11g0 21g4 + 10g6 11g2 18g4 + 7g6 , c1 = , c2 = , c3 = . 11 11 11 11 (33)

3. Mean-Field Theory of Spinor Condensates 3.1. Number-conserving theory The mean-eld theory is usually obtained by replacing the eld operator with its expectation value m . This recipe, though widely used and technically convenient, has one conceptual diculty; that is, it breaks the global U(1) gauge invariance, which implies that the number of atoms is not conserved. However, in reality, the number of atoms is strictly conserved, as are the baryon (proton and neutron) and lepton (electron) numbers. In fact, it is possible to construct the mean-eld theory without breaking the U(1) gauge symmetry [4, 5]. To construct a number-conserving mean-eld theory, we rst expand the eld operator in terms of a complete orthonormal set of basis functions {mi (r)}: m (r) =
i

ami mi (r) (m = f, f 1, , f ),

(34)

where mi (r) describes the basis function for the magnetic quantum number m and spatial mode i, and ami s are the corresponding annihilation operators that satisfy the canonical commutation relations [ mi , a j ] = mn i j , [ mi , an j ] = [ , a j ] = 0. a n a ami n 11 (35)

The basis functions are assumed to satisfy the orthonormality conditions dr (r) m j (r) = i j , mi and the completeness relation mi (r) (r ) = (r r ). mi
i

(36)

(37)

Then, the eld operators (34) satisfy the eld commutation relations (1). In the mean-eld approximation, it is assumed that all Bose-condensed bosons occupy a single spatial mode, say i = 0, and a single spin state specied by a linear superposition of magnetic sublevels. Then, the state vector is given by f N 1 | = m am0 |vac , N! m= f (38)

where |vac denotes the particle vacuum and m s are assumed to satisfy the following normalization condition:
f

|m |2 = 1.
m= f

(39)

It is straightforward to show that m (r) m (r)n (r ) m n (r) (r )k (r ) (r ) where

0 0 0 0

m = (r) =

= 0, ),

(40) (41) (42)

(r)n (r m

= 1

1 (r) (r )k (r ) (r ), n N m

| | and m (r) =

Nm m0 (r).

(43)

As shown in Eq. (40), the expectation value of the eld operator vanishes as it should in a numberconserving theory. Nevertheless, all experimentally observable physical quantities, which are expressed in terms of the correlation function of the eld operators such as (41) and (42), can have nonzero values. These values agree with those obtained by the U(1) symmetry-breaking approach within the factor of 1/N. 3.2. Mean-eld theory of spin-1 BECs We use Eqs. (41) and (42) to evaluate the expectation value of the Hamiltonian H = H0 + V over the state (38) with f = 1, where H0 and V are given in Eqs. (2) and (24). By ignoring the terms of the order of 1/N in Eq. (42), we obtain
1

E[] H

dr
m=1

2M

+ U(r) pm + qm2 m + 12

c0 2 c1 2 n + |F| , 2 2

(44)

where
1

n(r) n(r)

=
m=1

|m (r)|2

(45)

is the particle density and F = (F x , Fy , Fz ) is the spin density vector dened by

1

F (r) F (r)

=
m,n=1

(r)(f )mn n (r) ( = x, y, z). m

(46)

Because the spin-1 matrices are given by 0 1 0 0 1 0 1 0 i 1 f x = 1 0 1 , fy = 1 0 1 , fz = 0 0 2 0 1 0 2 0 1 0 0 0 the components of the spin vector F are explicitly written as 1 F x = 0 + (1 + 1 ) + 0 , 1 0 1 2 i Fy = 0 + (1 1 ) + 0 , 1 0 1 2 2 2 Fz = |1 | |1 | .

0 0 , 1

(47)

(48) (49) (50)

From the energy functional in Eq. (44), we nd that c0 must be nonnegative; otherwise, the system collapses. This condition also guarantees that the Bogoliubov excitation energy corresponding to density uctuations is real, as shown in Sec. 5. It is easy to identify the ground-state magnetism of a uniform system when the Zeeman terms are negligible (p = q = 0). From the energy functional (44), one can see that the ground state is ferromagnetic (|F| = n) when c1 < 0 and antiferromagnetic or polar (|F| = 0) when c1 > 0. Here, the order parameter of the antiferromagnetic phase [6] is given by n/2(1, 0, 1)T and that of the polar state [7] is n(0, 1, 0)T . The former state is obtained by rotating the latter about the y-axis by /2, and therefore, these two states are degenerate in the absence of the magnetic eld. Moreover, all states obtained by rotating these states are degenerate. In the presence of an external magnetic eld, however, the ground state becomes more complicated due to the linear and quadratic Zeeman eects, as shown below. The time evolution of the mean eld is governed by i m (r) E = . t m (r) (51)

Substituting Eq. (44) into the right-hand side gives i

2 2 m = + U(r) pm + qm2 m t 2M 1

+ c0 nm + c1
n=1

F fmn n (m = 1, 0, 1), 13

(52)

which are the multicomponent Gross-Pitaevskii equations (GPEs) that describe the mean-eld properties of spin-1 Bose-Einstein condensates [6, 7]. In a stationary state, we substitute m (r, t) = m (r) e
i

(53)

in Eq. (52), where is the chemical potential, to obtain

2 2 1

2M

+ U(r) pm + qm2 m + c0 nm + c1
n=1

F fmn n = m .

(54)

Writing down the three components m = 1, 0, 1 explicitly, we obtain c1 + U(r) p + q + c0 n + c1 Fz 1 + F 0 = 0, 2 2 2 c1 c1 + U(r) + c0 n 0 + F 1 = 0, F + 1 + 2M 2 2 2M

2 2

(55) (56) (57)

2 2 c1 + U(r) + p + q + c0 n c1 Fz 1 = 0, F + 0 + 2M 2

where F F x iFy . By solving the set of equations (55)(57), we can investigate the properties of the ground and excited states of the spin-1 BEC. Note that because the system is suspended in a vacuum chamber, the projected spin angular momentum in the direction of the external magnetic eld is conserved for a long time ( 1 s) [8]. Because the atomic interaction given by Eq. (7) conserves the total spin of two colliding atoms, the spin dynamics obeying the GPEs conserves the total magnetization. Therefore, when we search for the ground state for a given magnetization Mz = drFz (r), we need to replace E in Eq. (51) with E Mz , where is a Lagrange multiplier. Then, p in Eqs. (55)(57) is replaced with p = p + , which is determined as a function of Mz . Here, we solve Eqs. (55)(57) and investigate the ground-state phase diagram [9, 10] in a parameter space of (q, p). We assume a uniform system for a given number density n. The ground state of a uniform system is obtained if we set the kinetic energy and U(r) to be zero. We also note that c0 n can be absorbed in by dening c0 n. Furthermore, we may assume Fy = 0 without loss of generality, provided that the system is symmetric about the z-axis. Then, by performing a gauge transformation to make 0 real, all 1,0,1 become real. (After obtaining the real solution, we can recover the general complex order parameter by performing a global gauge transformation ei0 and spin rotation e fz z about the z-axis.) Then, Eqs. (55)(57) reduce to (p + q + c1 Fz )1 + c1 (1 + 1 )2 = 0, 0 [ c1 (1 + 1 ) ]0 = 0,
2

(58) (59) (60)

c1 (1 +

1 )2 0

+ (p + q c1 Fz )1 = 0.

The energy per particle is given by 1 (pm + qm2 )| |2 + 1 c n2 + 1 c |F|2 . = m 0 1 n m 2 2 14

(61)

From Eq. (59), we have either (i) 0 = 0 or (ii) = c1 (1 + 1 )2 . In case (i), we have three stationary states: 1 1 = p + q + c0 n + c1 n, 2 2 T 1 1 i1 II : 0, 0, e n , II = p + q + c0 n + c1 n, 2 2 T i p2 n + p/c1 n p/c1 1 , III : e 1 , 0, ei1 , III = q + c0 n 2 2 2 2c1 n
T

I : ei1 n, 0, 0

(62) (63) (64)

where we have recovered the phases of the order parameter with 1 = 0 z . States I and II are fully polarized (Fz = n), and therefore, ferromagnetic, whereas state III has a longitudinal magnetization that depends on p because Fz = p/c1 . State III is often referred to as an antiferromagnetic state. In case (ii), we may solve Eqs. (58) and (60), together with the normalization 2 condition 1 m=1 |m | = n, to obtain the following two stationary states: one, referred to as a polar state, is given by IV : 0, ei0 n, 0
T

IV

1 c0 n, 2

(65)

and it has no magnetization in any direction; the other is given by V : 1 = ei(0 0 = ei0
V z ) q

p 2q

p2 + q2 + 2c1 nq , 2c1 q

(66)

(q2 p2 )(p2 q2 + 2c1 nq) , 4c1 q3

(67) (68)

(p2 + q2 + 2qc1 n)2 1 + c0 n, 2 8c1 nq2

and it exists only when the argument of the square root for every m is nonnegative. State V occurs for c1 < 0 (ferromagnetic case). When p2 q2 + 2|c1 |nq > 0 and q2 > p2 , (69)

all three components and 0 in Eqs. (66) and (67) are nonzero and the magnetization tilts against the external magnetic eld: Fz = p(p2 q2 + 2q|c1 |n) , F+ = eiz 2|c1 |q2 (q2 p2 ){(p2 + 2|c1 |nq)2 q4 } . 2|c1 |q2 (70)

The polar angle [arctan(|F+ |/Fz )] is determined by the interaction and the magnetic eld, whereas the azimuthal angle (z ) is spontaneously chosen in each realization of the system. Because the axial symmetry with respect to the applied magnetic eld axis is broken, this phase is referred to as the broken-axisymmetry phase [10]. Comparing the energies of Eqs. (62)(68), we obtain the phase diagram in a parameter space of (q, p), as shown in Fig. 1 [9]. 15

(a) c1>0
p/c1n p=q+c1n/2

(b) c1=0
p p=q

(c) c1<0
p/|c1|n p=q p2=q22|c1|nq

I III II

p2=2c1nq
1/2

I II

IV
q

I II

V
2

IV
q/|c1|n

IV

q/c1n

p=q p=q+c1n/2

p=q

Figure 1: Phase diagram of spin-1 BEC for (a) c1 > 0, (b) c1 = 0, and (c) c1 < 0. The rotation symmetry about the magnetic eld is broken in the shaded region.

phase (I) (II) (III) (IV) (V) F F AF P BA

order parameter T (ei1 n, 0, 0) (0, 0, ei1 n) (ei1

n+p/c1 i1 2 , 0, e np/c1 2 )

Fz n n
p c1

= c0 n p + q + c1 n p + q + c1 n q 0 q + 2c1 n
p2 q

1 = 2 c0 n

p + q + 1 c1 n 2 p + q + 1 c1 n 2 q 0
(p2 +q2 +2qc1 n)2 8c1 nq2 p2 2c1 n

(0, ei0 n, 0) Eqs. (66), (67)

0
p(p2 +q2 +2qc1 n) 2c1 q2

Table 3: Possible ground-state phases of a spin-1 BEC, where F, AF, P, and BA denote ferromagnetic, antiferromagnetic, polar, and broken-axisymmetry phases, respectively.

Note that the rotation symmetry about the magnetic-eld axis is broken in state III, despite the fact that the magnetization is parallel to the magnetic eld in this phase. To understand the underlying physics, we introduce the spin nematic tensor dened by N 1 f f + f f n 2 , = x, y, z.
0

(71)

The nematic tensor for the ferromagnetic (I, II) and polar (IV) phases is independent of 0 and z , and given as 1 1 0 0 2 0 0 (72) N I,II = 0 1 0 , N IV = 0 1 0 , 2 0 0 0 0 0 1 respectively. Here, N xx = Nyy indicates that the order parameter has rotational symmetry about 16

the z-axis. On the other hand, the nematic tensor for state III for z = 0 is given by 1 2 (1 + ) 0 0 2 p 1 N III = 0 . (1 ) 0 , = 1 2 c1 n 0 0 1

(73)

Here, N xx Nyy indicates the anisotropy on the xy plane in the spin space. This is the physical origin of the axisymmetry breaking of state III. In fact, it can be shown that o-diagonal elements such as N xy appear when z 0. In general, the symmetry of the order parameter can be visualized by drawing the wave function in spin space. Because an integer-spin state can be described in terms of the spherical harmonics Y m ( s), where s is a unit vector in spin space, the order parameter for a spin- f system f can be expressed in terms of a complex wave function given by ( s) =
m

m Y m ( s). f

(74)

Figure 2 shows the order parameter for states IV. From Fig. 2, one can see that the order parameters for states III and V have the same morphology and that the magnetization is parallel to the magnetic eld in one phase (III) but not in the other (V). Additional symmetry properties are discussed in Sec. 7.1.1.
I
Fz=n

II
Fz=n

III
F//B

IV |F |=0

V
F//B

phase

Figure 2: Order parameter for spin-1 stationary states. Surface plots of |( s)|2 are shown, where the gray scale on the surface represents arg ( s).

3.3. Mean-eld theory of spin-2 BECs We take the expectation value of the Hamiltonian H = H0 + V over the state in Eq. (38) with 0 and V are given in Eqs. (2) and (30): f = 2, where H E[] H =
0 2

dr
m=2

2M

+ U(r) pm + qm2 m +

c0 2 c1 2 c2 2 n + |F| + |A| . 2 2 2

(75)

17

Here, the spin-2 matrices are given by 0 1 0 0 0 2 0 0 0 0 1 0

3 2

0
3 2

0 0
3 2 3 2

fx

0 0 0 1

0 0 0 1 0 0 0

fz

0 0 0 0 0 0 0 0 0 . 0 1 0 0 0 2

0 0 i 0 , fy = 0 0 0 1 0 0

i 0 i 0 0
3 2

0 i 0 i 0
3 2 3 2

0 0 i 0 i
3 2

0 0 0, i 0

(76)

As compared to the energy functional (44) of the spin-1 BEC, the new term c2 |A|2 /2 appears in Eq. (75), where A A00 (r)
0

1 = 22 2 21 1 + 2 0 5

(77)

is the amplitude of the spin-singlet pair. For the spin-1 case, the spin-singlet amplitude is uniquely related to the magnetization by relation (16), and only one of them appears in the energy functional. However, for the spin-2 case, they can change independently, giving rise to new ground-state phases. This can be best illustrated when the system is uniform. In this case, the kinetic-energy term can be ignored and the Hartree term c0 n2 /2 is constant; therefore, the ground state is determined by the remaining terms in Eq. (75). Let us rst consider the ground state in the zero magnetic eld. The phase diagram in this case is determined by the last two terms in Eq. (75), as shown in Fig. 3. When c1 < 0 and c2 > 0, the energy of the system is lowered as magnetization increases. Therefore, the ground state is ferromagnetic, and the order parameter is given by ferro = ( n, 0, 0, 0, 0)T or its rotated state in the spin space. When c1 > 0 and c2 < 0, the energy of the system is lowered as the amplitude of the spin-singlet pair increases. Therefore, the ground state is uniaxial nematic (UN) or biaxial nematic (BN), and the order parameters are given by uniax = (0, 0, n, 0, 0)T (uniaxial) or biax = n/2(1, 0, 0, 0, 1)T (biaxial). While these states are often referred to as polar or antiferromagnetic [4, 11, 5], respectively, here, we adopt the liquid-crystal terminology which implies the symmetry of the order parameter (see the insets in Fig. 3). In this review, we refer to C4 phase discussed below as antiferromagnetic, in order to distinguish it from the UN state. The order parameter n/2(0, 1, 0, 1, 0) also represents the BN state, which is obtained by rotating biax as ieify /2 eifz /4 biax , while uniax cannot be obtaind by rotating biax since the symmetry of the order parameters are dierent. Note that the symmetries of the UN and BN states dier; however, the two states are degenerate at the mean-eld level. Moreover, the superposition of these two states n/2(cos , 0, 2 sin , 0, cos ) ( is an arbitrary real) is also degenerate. It has been shown that zero-point uctuations lift the degeneracy [12, 13]. When both c1 and c2 are positive, neither the ferromagnetic nor UN/BN state is energetically favorable and frustration arises, resulting in a new phase referred to as the cyclic phase [4, 11, 5]. 18

 The order parameter is given by cyclic = n/2(1, 0, i 2, 0, 1)T , which possesses tetrahedral symmetry, as shown in the inset of Fig. 3. In the many-body ground state of the cyclic state, three bosons form a spin-singlet trimer and the boson trimers undergo Bose-Einstein condensation, as shown in Sec. 8.

c2n
cyclic ferromagnetic

uniaxial nematic

c1n

Figure 3: Phase diagram of the spin-2 BEC at zero magnetic eld. In each phase, the prole of the order parameter m ( s) = m m Y2 ( s) is shown.

The time-dependent GPEs for the spin-2 case can be obtained by substituting Eq. (75) in Eq. (51): i
2 2 2 = + U(r) 2p + 4q + c0 n 2c1 Fz 2 t 2M c2 + c1 F 1 + A 2 , 5 2 2 1 i = + U(r) p + q + c0 n c1 Fz 1 t 2M 6 c2 + c1 F 0 + F 2 A 1 , 2 5 2 2 c2 0 6 = + U(r) + c0 n 0 + c1 (F+ 1 + F 1 ) + A , i 0 t 2M 2 5

c2

n=

20 c1

biaxial nematic

(78)

(79) (80)

19

where
F+ = F = 2 1 + 2 + 2 1 2 2 2

6 0 + 1 , 1 0
2

(81) (82)

Fz = 2 |2 | |2 | + |1 | |1 | .

The time-independent GPEs can be obtained by substituting m (r, t) = m (r)eit/ in Eqs. (78) (80). If we assume that the system is uniform (i.e., U(r) = 0), then Eqs. (78)(80) reduce to (4q + 2 0 ) 2 + a = 1 , 2 (83)

a 2 + (4q 2 0 ) = , 2 (84) 1 6 + 2 + (q + 0 ) 1 a = 0 , (85) 1 2 6 a 1 + (q 0 ) + + = 1 , (86) 2 0 2 6 0 a = (r+ 1 + 1 ), (87) 0 2 where c0 n, c1 F , 0 c1 Fz , 0 0 p, and a c2 A/ 5. We may use the degree of gauge transformation (i.e., the U(1) phase) to make 0 real. Furthermore, because the physical properties of the system are invariant under a rotation about the quantization axis (i.e., the z-axis), we may choose the coordinate system to make Fy = 0 such that + = . A mean-eld ground state can be obtained as a solution of Eqs. (83)(87) under such a simplication. Here, we consider the case of = 0, i.e., the case of no transverse magnetization. Then, Eqs. (83)(87) reduce to a 2 + (4q 2 0 ) 2 (q + 0 ) 1 a 1 a 1 + (q 0 ) 1 In this case, the free energy per particle is given by = 1 c0 n c1 2 c2 2 (pm + qm2 ) |m |2 + + Fz + |A| . n m=2 2 2n 2n
2

(4q + 2 0 ) 2 + a = 0, 2

(88) (89) (90) (91) (92)

= 0, = 0,

= 0, ( a) 0 = 0.

(93)

Because Eqs. (88)(92) are decoupled into three parts, the solutions can be classied according to the determinant of the coecient matrix of Eqs. (88) and (89), D2 (4q )2 4 0 |a|2 , 2 and the determinant of the coecient matrix of Eqs. (90) and (91), D1 (q )2 0 |a|2 . 2 20 (95) (94)

 If D1 0 and D2 0, then we have 1 = 1 = 0 and 2 = 2 = 0, so that the solution is the UN state, that is, the order parameter is given by (96) UN: (0, 0, n, 0, 0)T .
1 The chemical potential is determined from (92) to be = 5 c2 n and Fz = 0. The energy 1 1 per particle is found from Eq. (93) to be = 2 (c0 + 5 c2 )n.

If D1 = 0 and D2 magnetization, 1

0, 2 = 2 = 0. From the assumption that there is no transverse 0 leads to 0 = 0. Clearly, F1+ : (0, ei1 n, 0, 0, 0)T (97) (0, 0, 0, ei1 n, 0)T

and F1 : (98)

are the solutions of Eqs. (90) and (91) with Fz = n, = p + q + c1 n and = p + q + 1 1 (c0 + c1 )n, and Fz = n, = p + q + c1 n and = p + q + 2 (c0 + c1 )n, respectively. When 2 1 |p| < |c1 5 c1 |n, a solution with nonzero 1 exists, which is given by C2 : 0, ei1 n + Fz , 0, ei1 2 T n Fz , 0 , 2 (99)

1 where Fz = p/(c1 5 c2 ). The chemical potential and the energy per particle are = 1 q + 5 c2 n and = q + 1 (c0 + 1 c2 )n p2 /[2(c1 1 c2 )n]. At p = 0, this state continuously 2 5 5 becoms the BN state.

If D1 0 and D2 = 0, 1 = 1 = 0. Moreover, if a, we have 0 = 0. In a manner similar to the previous case, Eqs. (88) and (89) have three solutions. Two of them are ferromagnetic and they are given by F2+ : (ei2 n, 0, 0, 0, 0)T (100) with Fz = 2n, = 2p + 4q + 4c1 n and = 2p + 4q + 1 (c0 + 4c1 )n, and by 2 F2 : (0, 0, 0, 0, ei2 n)T (101)

1 with Fz = 2n, = 2p+4q+4c1 n and = 2p+4q+ 1 (c0 +4c1 )n. When |p| < |2c1 10 c2 |n, 2 the other solution is given by

C4 : ei2

Fz 1 n+ , 0, 0, 0, ei2 2 2

T Fz 1 , n 2 2

(102)
1 20 c2 )n].

1 1 where Fz = p/(c1 20 c2 ), = 4q + 1 c2 n and = 4q + 2 (c0 + 1 c2 )n p2 /[2(c1 5 5 This state becomes the BN state at p = 0.

21

 If D1 0 and D2 = 0 and = a, we have 1 = 1 = 0; however, 2 and 0 can be nonzero. They obey |2 |2 + |2 |2 + |0 |2 = n and |2 |2 |2 |2 = 1 Fz , and hence, 2 C2 : ei2 n + Fz /2 2 0 , 0, 0 , 0, ei2 2 T n Fz /2 2 0 . 2

(103)

Because a = (c2 /5) ei(2 +2 )

2 (n 0 )2 Fz /4 + 2 = is real, we have 2 + 2 = 0 0

or . Here, and Fz are determined from Eqs. (88) and (89) to be = 2q 0 /(2q) and 2 Fz = ( 0 + p)/c1 , respectively, where 0 is a real solution of the following equation: 0 + p 0 + 4q[q + 2c1 (n 2 )] 0 + 4pq2 = 0, 3 2 0 and 0 is determined so as to minimize the energy per particle given by 2 1 0 2 p2 c2 = 4q 1 0 + c0 n + + n 2 2c1 n 10n
2

(104)

(n

2 )2 0

2 Fz i(2 +2 ) e + 2 . 0 4

(105)

In particular, when p = 0 and |q| > 2c1 (n 2 ), 0 is determined to be zero, leading to 0 1 Fz = 0. Moreover, when c2 > 0, Eq. (105) is minimized at 2 +2 = and 2 = 2 n+5q/c2 0 1 2 to be = 2q + 2 c0 n 10q /(c2 n). The corresponding order parameter is given by D2 : iei n 10q/c2 , 0, 2 n 10q/c2 , 0, iei 2 T n 10q/c2 , 2 (106)

where we set 2 = . This state continuously becomes the cyclic state at q = 0. On 2 the other hand when c2 < 0, Eq. (105) is minimized at 2 +2 = 0 and the order parameter is obtained to be UN for q > 0 and BN for q < 0. If D1 = D2 = 0, we solve the simultaneous equations D1 = 0 and D2 = 0 to obtain = 5q2 0 2 = 4q2 + |a|2 . 2q (107)

When q 0, a, we nd from Eq. (92) that 0 = 0. Then, = 2 ( 1 + 2 ). We 2 1 may use Eqs. (88)(91) and (107) to show that = 2c1 1 1 2 0 3q 0 + 3q = 2c1 2 1 = 0. 1 0 3q 0 + 3q (108)

Thus, we must, in general, have 2 1 = 1 2 = 0. To be consistent with Eqs. (88)(91), we nd that either 1 = 2 = 0 or 2 = 1 = 0 should hold. In the former case, the order parameter is given by C3+ : i 2 e n + Fz , 0, 0, ei1 3 22 T 2n Fz , 0 3 (109)

with Fz = (p q)/c1 and is given by C3 :

= 2q 1 c0 n (p q)2 /(2c1 n). In the latter case, the solution 2 i 0, e 1 2n + Fz , 0, 0, ei2 3 T n Fz 3

(110)

with Fz = (p + q)/c1 and = 2q 1 c0 n (p + q)2 /(2c1 n). At zero magnetic eld 2 (i.e., p = q 0), these two states become the cyclic state, which are related to cyclic = = n/2(1, 0, i 2, 0, 1)T as 1 0 n 1 0 = i exp ifz exp i f x fy arccos cyclic , 3 4 2 3 2 0 0 f x + fy 3 n 2 1 0 = i exp ifz exp i arccos cyclic . 3 4 3 2 0 1

(111)

(112)

The above results are summarized in Table 4. By comparing the energy of the obtained state, we nd the ground state phase diagram of a spin-2 BEC, as shown in Fig. 4 for the special case of p = 0 and q < 0.

23

state 2n 2n n n 0
nFz /2 2 p c1 c2 /20 1 5 c2 n 1 10 c2 n

Fz 2p + 4q + 4c1 n 2p + 4q + 4c1 n p + q + c1 n p + q + c1 n
1 p + q + 2 c1 n 1 p + q + 2 c1 n

= c0 n 2p + 4q + 2c1 n 2p + 4q + 2c1 n

= 1 c0 n 2

F2+

F2

F1+

F1

UN 4q + 1 c2 n 5
1 10 c2 n (pq)2 2c1 n (p+q)2 2c1 n

order parameter T (ei2 n, 0, 0, 0, 0) (0, 0, 0, 0, ei2 n) (0, ei1 n, 0, 0, 0) (0, 0, 0, ei1 n, 0) (0, 0, ei0 n, 0, 0) 4q + 2q 2q q+
1 10 c2 n

C4
2nFz 3 ,0 pq c1

ei2 2q 2q
1 q + 5 c2 n

n+Fz /2 i2 2 , 0, 0, 0, e

p2 2(c1 c2 /20)n

24
nFz 3 p c1 c2 /5 nFz /22 0 2 p+ 0 c1 p+q c1 nFz 2 ,0

C3+

ei2

n+Fz i1 3 , 0, 0, e

C3

0, ei1

2n+Fz i2 3 , 0, 0, e

C2 2q
0 2 2q

0, ei1

n+Fz i1 2 , 0, e

2 0 n )

p2 2(c1 c2 /5)n

C2

ei2

n+Fz /2+2 0 , 0, ei0 0 , 0, ei2 2

4q(1
c2 + 10n

0 p2 2 2c1 n 2

(n 2 )2 0

2 Fz i(2 +2 20 ) 4 e

+ 2 0

Table 4: Possible states for the ground state of a spin-2 BEC. Each states is associated with a particular symmetry, as identied in Sec. 7.1.1. In the C2 state, 0 is a real solution of Eq. (104), and 2 + 2 20 = for c2 > 0 and 2 + 2 20 = 0 for c2 < 0. At p = 0 the C4 and C2 states continuously become the BN state, and the C3 state becomes the cyclic state. For c2 > 0, the C2 state becomes the D2 state at p = 0 and the cyclic state at p = q = 0.

c2n C3 F2 D2'
10|q|

BN
|q|/2

c2

n=

20 c1

c1n

Figure 4: Ground-state phase diagram of a spin-2 BEC for p = 0 and q < 0, where F2 states and C3 states are degenerate in the region of F2 and C3, respectively. The M state and C state in Ref. [14] correspond to C3 and D2 in this phase diagram, respectively.

4. Dipolar BEC In this section, we consider the dipole-dipole interaction (DDI) that, unlike the s-wave contact interaction, is long-range and anisotropic. The DDI between alkali atoms is a thousand times smaller than the short-range interaction for the background scattering length. However, recent experimental developments, such as the realization of a BEC of 52 Cr atoms [15, 16, 17, 18, 19, 20], a technique for creating cold molecular gases [21, 22, 23], and high-accuracy measurements [24, 25, 26], have enabled us to study several consequences of the DDI. Here, we briey introduce the general properties of the DDI in cold gases (Sec. 4.1), and discuss two important cases: spin-polarized dipolar gases (Sec. 4.2) and spinor dipolar gases (Sec. 4.3). Excellent reviews on dipolar gases have been published by Baranov [27, 28] and by Lahaye et al. [29]. 4.1. Dipoledipole interaction 4.1.1. Scattering properties of dipole interaction The dipole-dipole interaction (DDI) between two dipole moments with relative position r [see Fig. 5 (a)] is given by Vdd (r) = cdd d1 d2 3( d1 r)( d2 r) = cdd r3 d1 Q (r)d2 ,
,

(113)

where d1,2 are unit vectors indicating the directions of dipole moments, r = |r|, r = r/r, and Q (r) (, = x, y, z) is a rank-2 traceless symmetric tensor that can be expressed in terms of 25

m rank-2 spherical harmonics Y2 ( ) as r

Q (r)

3 r r 3 r 2 Y 0 ( ) Y 2 ( ) Y 2 ( ) 3 2 r 2 r 2 r 6 1 2 2 r r iY2 ( ) iY2 ( ) 5 r3 1 1 r r Y2 ( ) Y2 ( )

(114) 2 2 1 1 iY2 ( ) iY2 ( ) r r Y2 ( ) Y2 ( ) r r 2 0 1 1 2 2 r r r r Y2 ( ) + Y2 ( ) + Y2 ( ) iY2 ( ) iY2 ( ) r 3 2 0 1 1 r 2 3 Y2 ( ) r iY2 ( ) iY2 ( ) r (115)

In a special case in which the dipole moments are polarized under an external eld, the DDI takes a simple form: Vdd (r) = cdd Qzz (r) = cdd 1 3 cos2 = r3
0 r 16 Y2 ( ) cdd 3 , 5 r

(116)

where is the angle between the polarization direction and the relative position r [see Fig. 5(b)]. Because Vdd is negative (positive) for = 0 (/2), the DDI interaction favors a head-to-tail conguration [Fig. 5 (c)] as opposed to a side-by-side one [Fig. 5 (d)].

(a) d2 r

(b) d2 r d1

(c)

(d)

d1

Figure 5: Interaction between two dipole moments for (a) an unpolarized case and (b) a polarized case. The dipoledipole interaction is most attractive for (c) the head-to-tail conguration and most repulsive for (d) the side-by-side conguration. The bottom gures below (c) and (d) show characteristic spin textures.

The long-range character ( 1/r3 ) of the DDI becomes prominent in low-energy scattering. When a potential decreases as 1/rn for large r, the phase shift l behaves in the zero-energy limit as k2l+1 if l < (n 3)/2 and as kn2 otherwise [30]. For a potential with n > 3, such as van der Waals potential (n = 6), the main contribution at k 0 is an s-wave channel 0 k, and therefore, the potential can be described with a single parameter, namely, with the s-wave scattering length. On the other hand, when n = 3, all partial waves contribute to the scattering as l k, and therefore, the scattering process is no longer described with a single parameter such as the s-wave scattering length. Moreover, the anisotropy of the DDI induces coupling between partial waves. Because the dipole interaction is described by rank-2 spherical harmonics and it therefore has d-wave symmetry, a partial wave of angular momentum (l, m) is coupled to (l, m + m f ) and (l 2, m + m f ), where m f is the change in the projected total spin of colliding particles and m f = 0 for the polarized case. Although there is no diagonal term for the l = 0 channel, this coupling induces the eective potential for l = 0 that behaves as r6 at large distances. Hence, the dipole interaction generates a short-range interaction as well. Yi and 26

You [31, 32] have shown that away from the shape resonance [33, 34], the eective interaction between two dipoles can be described with the following pseudo-potential (see also [35, 36]): Ve (r) = 4 2 a(d) (r) + Vdd (r), M (117)

where a(d) depends eectively on the strength of the dipole moment. 4.1.2. Dipolar systems There are several systems that undergo long-range and anisotropic DDI. The coecient cdd in Eq. (113) is given by cdd = d2 /(4 0 ) and cdd = 0 d2 /(4) for electric and magnetic dipole moments, respectively, where 0 and 0 are the dielectric constant and magnetic permeability of vacuum, respectively, and d is the magnitude of the dipole moment. The typical order of magnitude of the electric dipole moment is a product of the elementary charge e and the Bohr radius a0 , whereas the magnetic dipole moment of an atom is of the order of the Bohr magneton B . Therefore, the ratio of the energy of the magnetic DDI to that of the electric DDI is given by 0 2 B 2 104 , (ea0 )/ 0 (118)

where = e2 /(4 c) 1/137 is the ne structure constant. Due to the strong electric dipole moments, gases of polar molecules are ideal candidates for dipole-dominant systems. The permanent dipole moments of the lowest 1,3 + states of heteronuclear alkali dimers are calculated to be of the order of 1 Debye ( 3.335 1030 Cm) [37]. However, because the ground state of the molecule is rotationally symmetric and the dipole moment averages to zero, we need to apply a strong electric eld, typically of the order of 104 V/cm, to polarize the dipole moments. Several research groups are now trying to cool molecules using techniques such as buer-gas cooling [21, 38, 39] and Stark deceleration (see [40] for a review). Another way to achieve quantum degeneracy of heteronuclear molecules is to create weakly bound Feshbach molecules from cold atoms and to transfer the molecules to the tightly bound rovibrational ground state [22, 23, 41]. The eects of the DDI have recently been observed in several systems. The Stuttgart group have achieved a BEC of chromium (52 Cr) [15]. The 52 Cr atom has a magnetic dipole moment of 6B , which is six times larger than that of alkali atoms, and therefore, the DDI is 36 times larger than that of alkali atoms. Moreover, the s-wave scattering length of 52 Cr can be decreased by means of Feshbach resonance [42]. Thus, the Stuttgart group realized a dipole-dominant BEC and observed magnetostriction [16, 17, 18] and anisotropic collapse [19, 20]. A dipole-dominant BEC, albeit a very weak one, has also been realized in a BEC of 7 Li by almost quenching the s-wave contact interaction using Feshbach resonance [26]. Fattori et al. observed the decoherence of the interferometer of 39 K due to the dipole interaction [24]. The Berkeley group observed an evidence of the crystallization of spin domains in a BEC of 87 Rb [25]. 4.1.3. Tuning of dipole-dipole interaction Tunability of the DDI is crucial for systematically investigating the properties of the dipolar BEC. The strength of the electric dipole moments can be manipulated by an external electric eld with the magnetic dipole moment kept constant. Giovanazzi et al. [43] proposed a method to control the sign of the dipole interaction as well as its strength by using a rotating eld. This 27

method is applicable to both electric and magnetic dipole interactions. Consider a magnetic dipole moment under a rotating magnetic eld described by sin cos t B(t) = B sin sin t , (119) cos where is the angle between the rotating axis and the dipole moments (see Fig. 6). The angular frequency is set to be much smaller than the Larmor frequency so that the magnetic dipole moment adiabatically follows the direction of B. When is much larger than the trap frequency, the eective dipole interaction is time-averaged over 2/. Substituting d = B/|B| in Eq. (113) and taking its time-average, we obtain the following eective interaction: Vdd = cdd 1 3 cos2 3 cos2 1 . 2 r3 (120)

The last factor in the right-hand side of (120) can take a value between 1/2 to 1. Hence, one can change the sign of the dipole interaction as well as its strength by varying the angle .

d1

' r d2

Figure 6: Two dipoles adiabatically following the external magnetic eld (119) that rotates with angular frequency . The sign and strength of the dipole-dipole interaction (120) can be controlled by varying the angle .

4.1.4. Numerical method In the mean-eld treatment, we need to calculate the eective potential induced by the distribution of the dipole moments d(r), which includes an integral of the form d3 r Q (r r )d (r ).

(121)

While this integral converges, the 1/r divergence of the integrand makes it dicult to numerically carry out the integration in real space. We therefore use the convolution theorem to rewrite (121) as [44] Q (k)d (k)eikr ,
k

(122)

28

where Q (k) dreikr Q (r) and d (k) dreikr d (r). Here, the Fourier components of the kernel Q can be found as follows. First, we note that the spherical harmonics is the eigenstate of the three-dimensional Fourier transformation, i.e., eikr Ylm ( )d r = 4il jl (kr)Ylm ( k), r (123)

where k = |k|, k = k/k, jl (x) is the lth spherical Bessel function and d r denotes the integration of the angular part of r. Because the radial integration is then carried out as
0

j2 (kr) 2 1 r dr = , 3 r3

(124)

we obtain Q (k) dreikr Q (r) = 4 3k k . 3 (125)

For spin-polarized gases, we only require Qzz (k) = 4 (1 3 cos2 k ), 3 (126)

where k is the angle between the polarization direction and the momentum k. In the numerical calculation, we evaluate d (k) from d (r) by means of a standard fast Fourier transform algorithmwe multiply d (k) by Q (k), sum the result over , and perform the inverse Fourier transform. The result is not sensitive to the grid size r in the coordinate space as long as d (1/r) is negligible. However, a small numerical error arises, depending on the system size R, because the discrete Fourier transform presupposes a 3D periodic lattice with a unit cell of linear dimension R. If the Fourier transform is performed in a cubic region of R3 , the accuracy improves if we introduce an infrared cuto to calculate the Fourier transform of the integration kernel as Qcut (k) = |r|<R/2 drQ (r)eikr [45]. 4.2. Spin-polarized Dipolar BEC When the dipole moment of every atom is polarized in the direction of an external eld, say in the z-direction, the condensate is described with a single component order parameter (r) that obeys the GP equation i
2 (r, t) = t 2M 2

g + Vtrap (r) + |(r, t)|2 + dd (r, t) (r, t), 2

(127)

where Vtrap (r) is a trapping potential, g = 4 2 a/M with a being the scattering length for this component, and dd (r, t) = cdd dr 1 3 cos2 |(r , t)|2 |r r |3 (128)

is a mean-eld eective potential due to the DDI. Here, we assume a repulsive short-range interaction a > 0 so that the condensate is always stable in the absence of the DDI, whereas cdd can be either positive or negative due to the tunability discussed in Sec. 4.1.3. 29

4.2.1. Equilibrium shape and instability Due to the anisotropy of the interaction, the equilibrium shape of dipolar BECs is highly nontrivial [46, 47, 45]. The dipole interaction works attractively along the direction of dipole moments and repulsively in the perpendicular direction (see Fig. 5). The attractive part makes a dipole-dominant BEC unstable in a homogeneous system. This instability can be qualitatively understood from the Bogoliubov spectrum in a homogeneous system [46]: =
k k

+ 2ng 1 +

dd (3 cos

k 1) ,

(129)

where n is the number density of atoms and

dd

4cdd 3g

(130)

is a parameter that expresses the relative strength of the DDI against the short-range interaction. Equation (129) shows that for dd > 1, the BEC becomes dynamically unstable because is imaginary for wave numbers below kc = 4gnM( dd 1)/ . When the system is conned in a trapping potential, the quantum pressure due to the connement can stabilize the condensate against the attractive interaction. Santos et al. [48] numerically investigated the equilibrium shapes of pure dipolar gases with a = 0. They considered an axisymmetric trap with trap frequencies r and z in the radial and axial directions, respectively. When the trapping potential is prolate, i.e., when the aspect ratio l = r /z > 1, the dipole interaction is always attractive. Therefore, as in the case of attractive s-wave interaction, there exists a critical number of atoms above which the BEC becomes unstable. On the other hand, if the trap potential is suciently oblate so that l < l 0.43 and if the DDI is not too strong (see the last paragraph of this subsection), the dipole interaction is always positive and the BEC is stable. In the region of l < l < 1, the aspect ratio of the condensate increases with the number of atoms, and nally, it becomes unstable. Yi and You [31, 32] investigated the stability including the eect of repulsive short-range interaction using Gaussian variational ansatz, and obtained results consistent with those of Ref. [48]. In Refs. [32, 44], the low-energy excitation spectrum is calculated using time-dependent variational methods. It is shown that the instabilities of dipolar BECs in prolate (l > 1.29) and oblate (l < l < 0.75) traps originate from dynamical instabilities of the breathing and quadrupole modes, respectively. For the intermediate region, the lowest excitation frequency corresponds to a breathing mode far below the above mentioned critical number of atoms, and to a quadrupolar one near the critical value. In the Thomas-Fermi limit, gn 1, the GP equation (127) has an exact solution [49]. Using the relationship ( 3 r )/r3 = (1/r) 4 (r)/3, the mean-eld dipolar potential in r Eq. (128) can be rewritten as dd (r) = cdd 2 4 (r) + n(r) , 3 z2 (131)

where (r) dr n(r )/|r r |, and therefore, (r) obeys Poissons equation, 2 = 4n(r). Because n(r) is parabolic in the Thomas-Fermi limit, Poissons equation is satised by = a0 +a1 x2 +a2 y2 +a3 z2 +a4 x2 y2 +a5 y2 z2 +a6 z2 x2 +a7 x4 +a8 y4 +a9 z4 , and hence, dd is also quadratic. Therefore, for a harmonically trapped BEC, all terms appearing in the GP equation (127) are quadratic, and as in the simple s-wave case, an inverted parabola for the density prole gives 30

a self-consistent solution of the dipolar hydrodynamic equations. (For an analytic form of the solution, see Refs. [49, 50].) Eberlein et al. [50] pointed out that the potential seen by atoms exhibits a local minimum immediately outside the condensate, i.e., the sum of the trapping and dipole potentials is locally smaller than the chemical potential, causing an instability that brings atoms out from the condensate to ll the dip in the potential. The presence of this instability is in agreement with the biconcave shape of a condensate pointed out by Ronen et al. [51]. By developing a high-precision code for dipolar BECs, Ronen et al. [51] have shown that the same instability causes a dipolar condensate to become unstable in the limit of strong dipole interaction, even though the trap is strongly oblate, in disagreement with the result of Refs. [48, 31, 32]. Interestingly, in a strongly oblate trap, the wave function of a BEC immediately before collapse becomes biconcave, with its maximum density away from the center of the gas. They have also shown that the instability of a strongly oblate dipolar BEC is analogous to the roton-maxon instability reported for 2D dipolar gases (see Sec. 4.2.3). 4.2.2. Dipolar collapse The Stuttgart group observed the magnetostriction of a 52 Cr BEC, where dd = 0.16 for the background scattering length. By reducing the s-wave scattering length using a Feshbach resonance, Lahaye et al. observed a change in the aspect ratio of the condensate [18]. The same group also observed a collapse dynamics due to the DDI [20]. In the experiment, the s-wave scattering length is adiabatically decreased to 30% of the background value, and it is then suddenly decreased below the critical value of the collapse. After a certain hold time, the BEC is released from the trap and the TOF image is taken after expansion. The collapse dynamics proceeds as follows. After a sudden decrease in the scattering length, atoms begin to gather at the center of the trap due to the attractive interaction; with an increase in the atomic density at the trap center, the three-body recombination losses become predominant; as a consequence of the atom loss, the attractive interaction weakens, and the atoms are ejected outward due to the quantum pressure. Figure 7 (a) shows the measured TOF image (upper panels) together with the results of numerical simulation (lower ones). We have numerically solved the GP equation that includes the three-body loss using the loss rate coecient that is determined to best t the measured loss curve (Fig. 3 of Ref. [20]). The excellent agreement between the experiment and the theory demonstrates the validity of the mean-eld description even for dipole-dominant BECs. Moreover, the numerical simulation revealed that the cloverleaf pattern in the TOF image arises due to the creation of a pair of vortex rings, as indicated in Fig. 7 (b). When collapse occurs, the atoms were ejected in the xy plane (vertical direction to the dipole moments), whereas atoms still ow inward along the z direction, giving rise to the circulation. The velocity eld shown in Fig. 7 (c) clearly shows the d-wave nature of the collapse dynamics. 4.2.3. Roton-maxon excitation Santos et al. [52] have shown that the Bogoliubov excitation spectrum exhibits roton-maxon behavior, i.e., the excitation energy has a local maximum and minimum as a function of the momentum q, in a system that is harmonically conned in the direction of the dipole moments (i.e., the z direction) and free in the x and y directions. If in-plane momenta q are much smaller than the inverse size L of the condensate in the z direction, excitations have a 2D character. Because the dipoles are perpendicular to the plane of the trap, particles eciently repel each other and the in-plane excitations are phonons. Then, the DDI increases the sound velocity. For 31

(a)

0 ms

0.1 ms

0.2 ms

0.3 ms

0.4 ms

0.5 ms

(b)

(c)

y y z x z

Figure 7: (a) Series of absorption images of the collapsing condensates for dierent values of hold time (top) and the corresponding results of the numerical simulations obtained without adjustable parameters (bottom). (b) Iso-density surface of an in-trap condensate. The locations of the topological defects are indicated by the red rings. (c) Velocity eld of the atomic ow in the x = 0 plane. Color represents the velocity (red is faster). Reprinted from Ref. [20].

q 1/L, excitations acquire a 3D character and the interparticle repulsion is reduced due to the attractive force in the z direction. This decreases the excitation energy with an increase in q. When dd > 1, the excitation energy reaches a minimum (roton) and then begins increasing, and the nature of the excitations continuously become single-particle like. As the dipole interaction becomes stronger, the energy at the roton minimum decreases and then it reaches zero, leading to an instability. The 3D character is essential for the appearance of the roton minimum, which does not appear in the quasi-2D system where the conning potential in the z direction is strong and the BEC has no degrees of freedom in this direction [53]. The roton-maxon spectrum is also predicted for a 1D system with laser-induced DDI [54]. 4.2.4. Two-dimensional solitons It is known that a nonlinear Schr dinger equation with short-range interactions admits soliton o solutions in one dimension, but it does not in higher dimensions. However, Pedri and Santos [55] showed that with a nonlocal DDI, a two-dimensional BEC can have a stable soliton. We consider a dipolar BEC polarized in the z direction, and assume that the system is conned 1 only in the z direction by a harmonic potential Vtrap (r) = 2 M2 z2 , and that it is free in the xy z plane. Then, the DDI is isotropic in the xy plane. We consider the following Gaussian variational ansatz x2 + y2 1 z2 (132) (x, y, z) = exp 2 2, 2 2 3/4 l3/2 L L1/2 2l0 L 2l0 Lz 0 z 32

where l0 /(Mz ) and L and Lz are dimensionless variational parameters that characterize the widths in the xy plane and the z direction, respectively. Using this ansatz, the mean-eld energy is evaluated to give 2 z 1 g + 1 + Lz + 1 E(L , Lz ) = 1 + dd f () , (133) 2 2 2 L l3 4 2 L 2Lz 2 2L z
0

where L /Lz is the aspect ratio and

arctan 2 1, > 1 22 + 1 32 f () 2 . 1 1+ 12 1 (2 1) |2 1| 2 ln 1 12 , < 1

(134)

For xed Lz in the absence of the DDI ( dd = 0), E(L ) monotonically increases or decreases as a function of L , resulting in collapse or expansion. However, in the presence of the DDI, because f () is a monotonic function of with f (0) = 1 and f ( ) = 2, E(L ) may have a minimum. When the trapping potential is strong and Lz = 1, for simplicity, the energy minimum appears for
dd

<1+

l0 < 2 2 aN

dd ,

(135)

where a is the s-wave scattering length and N, the total number of atoms (see Ref. [55]). This condition holds only for dd < 0, which can be achieved using a rotational eld. On the other hand, Tikhonenkov et al. [56] have shown that when dipole moments are polarized in the 2D plane, stable 2D soliton waves can be generated without tuning the dipole interaction. In this case, the soliton is anisotropic and elongated along the direction of polarization. Because a 2D bright soliton in a dipolar gas is stable, the roton instability discussed in the preceding subsection does not cause a collapse of the BEC but creates 2D solitons [57]. If the dipole moments are polarized perpendicular to the 2D plane, these solitons are stable as long as the gas remains 2D. However, if the dipoles are parallel to the 2D plane, the (anisotropic) solitons may become unstable even in 2D if the number of particles per soliton exceeds a critical value. 4.2.5. Supersolid The Bose-Hubbard model with long-range interactions exhibits a rich variety of phases such as the density wave (DW), supersolid (SS), and superuid (SF) phases [58, 59, 60, 61]. While SF and SS phases have a nonzero superuid density, DW and SS phases have a nontrivial crystalline order in which the particle density modulates with a periodicity that is dierent from that of the external potential. Thus, in the SS phase, both diagonal and o-diagonal long-range orders coexist [62, 63, 64]. A dipolar gas in an optical lattice is an ideal system to realize such exotic phases, because the interaction parameters can be experimentally controlled. G ral et al. [65] showed that DW and o SS phases, as well as SF and Mott insulator (MI) phases, can be accomplished in dipolar gases in a 2D optical lattice, and Yi et al. [66] investigated detailed phase diagrams in 2D and 3D optical lattices. A dipolar gas in an optical lattice can be described with an extended Bose-Hubbard model given by HEBH = t
i, j

1 i j (b+ b j + b+ bi ) + U0 2

ni ( i 1) + n
i

1 2

ii Udd ni ( i 1) + n i

1 2

ij Udd ni n j , (136) i j

33

 where bi is the annihilation operator of a particle at the lattice site i; ni = b+ bi , the correspond i ing particle-number operator; and t, the hopping matrix element between the nearest neighbors, and U0 > 0 is the on-site Hubbard repulsion due to the s-wave scattering. Here, t and U can be expressed in terms of the Wannier functions w(r ri ) of the lowest energy band as 2 U = 4a 2 /M dr|w(r)|4 and t = drw (r ri )[ 2M 2 + V0 (r)]w(r r j ), where V0 is the optical lattice potential [67]. The last two terms in Eq. (136) describe the on-side and inter-site DDIs with coupling parameters given by Ui j = cdd dr dr |w(r ri )|2 1 3 cos2 |w(r r j )|2 , |r r |3 (137)

where is the angle between the dipole moment and vector r ri . Here, cdd can take both positive and negative values by using a fast rotating eld (see Sec. 4.1.3). By changing the depth of an ij optical lattice, the transverse connement, and the orientation of the dipoles, t, U0 , and {Udd } can be tuned independently. ij Let us begin by reviewing the case of Udd = 0, i.e., the on-site Bose-Hubbard model. At t = 0 and commensurate lling, i.e., the average number of particles per site is an integer n, the interaction energy is minimized by populating every lattice site with exactly n atoms; therefore, the Mott insulator (MI) phase is realized. The energy cost required to create a particle-hole excitation in the MI phase is U0 (on-site interaction energy), and therefore, the MI state is gapped and incompressible. On the other hand, for nonzero t, the kinetic energy favors particle hopping. When t is suciently large to overcome the interaction-energy cost ( U0 ), the system undergoes a phase transition to the superuid (SF) phase that is gapless and compressible. In the presence of the long-range interaction, the ground state of the extended Bose-Hubbard Hamiltonian (136) in a 2D optical lattice is investigated in Refs. [65, 66] using a variational i approach based on the Gutzwiller ansatz | = i fn |n i , where |n i denotes the state with n=0 i n particles at lattice site i. The coecients fn can be found by minimizing the expectation value |HEBH i ni | under the constraint of a xed chemical potential . In this calculation, a cuto length c is introduced and the interaction between lattice sites with distance larger than c is ignored. When dipole moments are polarized perpendicular to the 2D lattice, the dipole interaction is repulsive (attractive) for cdd > 0 (cdd < 0). If site i is occupied, it is energetically favorable that its neighboring sites are equally populated for negative cdd and less populated for positive cdd . Therefore, for negative cdd , only MI and SF phases appear as in the case of the on-site BoseHubbard model, and local collapse occurs for large |cdd | due to the strong attractive interaction. On the other hand, for positive cdd , the DW phase emerges as an insulating state. According to the Landau theory of the second-order phase transition, phases with distinct symmetry breaking patterns cannot be continuously connected with each other. Therefore, as the hopping increases, the DW state rst changes to the SS state and then to the SF phase. When dipole moments are polarized along the 2D lattice, say in the y-axis, the DDI is anisotropic. The inter-site interaction for cdd > 0 (cdd < 0) is attractive (repulsive) in the y direction and repulsive (attractive) in the x direction. In this case, the particle density along the repulsive direction becomes periodically modulated, whereas it remains constant along the attractive direction, resulting in striped DW and striped SS states. Local collapse occurs for large |cdd | for both positive and negative cdd . In Ref. [66], the phase diagram in a 3D lattice is also investigated, where dipole moments are polarized along the z-axis. For cdd > 0, the DDI is attractive in the z direction and repulsive in the 34

xy direction. Hence, the DW phase forms a checkerboard-type pattern on the xy plane and it is uniform in the z direction. The density pattern in the SS phase reects the same symmetry as the DW phase. On the other hand, for cdd < 0, layers of high-density sites align along the z direction due to the repulsive interaction. Hence, layered DW and SS phases emerge perpendicular to the z-axis. Recently, Menotti et al. [68] studied a polarized dipolar gas in a 2D optical lattice beyond the ground state. They have shown that there exist many metastable states in the insulator region depending on the cuto length c . The number of metastable states and pattern variations increases rapidly with the number of lattice sites. Metastable states also appear in magnetic domains in solid-state ferromagnets, classical ferrofulids, and quantum ferrouids (see the next subsection), and they appear to be a characteristic consequence of long-range interactions. 4.2.6. Ferrouid Let us consider a repulsively interacting two-component BEC, in which atoms in component 1 have a magnetic dipole moment, whereas those in component 2 are nonmagnetic. The m = 3 and 0 states of 52 Cr can be a candidate for such a system. We apply a magnetic-eld gradient in the z direction, so that the two components phase-separate as shown in Fig. 8.
component 2 without magnetic dipole

z B

component 1 with magnetic dipole x interface

Figure 8: Schematic illustration of a two-component BEC system in which atoms in component 1 have a magnetic dipole moment and those in component 2 are nonmagnetic. The magnetic dipole is polarized in the z direction by an external magnetic eld. The two components are separated due to the eld gradient B (z) < 0. Reprinted from Ref. [69].

The equilibrium conguration of this system can be found by solving the two-component GP equation in imaginary time [69]. We found a stable hexagonal pattern of the density distribution of component 1 as shown in Figs. 9 (a) and (b). We also found several metastable states starting from dierent initial states, such as stripe, concentric, and deformed hexagonal patterns, as shown in Figs. 9 (c)(e). The energies for these patterns are almost degenerate; however, each pattern is robust against small perturbations. A striking dierence between the present system and magnetic liquids is that the present system is a superuid that supports a persistent current. We found a metastable state with a nite circulation in component 1 and no circulation in component 2 in a stationary trap. Remarkably, the Rosensweig pattern emerges even in the presence of a superow, indicating that the system simultaneously exhibits a diagonal and o-diagonal long-range order, although the latter is imbedded in the GP theory (see Ref. [69] for more detail). 4.3. Spinor-dipolar BEC Let us now consider a dipolar BEC with spin degrees of freedom. By spinor-dipolar, we imply that the the direction of the spin of the order parameter is not polarized by an external eld 35

(a)
2 z -2

(b)
component 1 component 2 total

y x
0 0.003
2 z 0

(c)

(d)

(e)

y x
0 0.003

Figure 9: (a) Isodensity surface of a hexagonal pattern formed in component 1 at B = 1 G/cm. (b) Column densities for component 1 (left), component 2 (center), and the total system (right). (c)(d) Metastable congurations found at B = 0.3 G/cm, where the isodensity surface (top) and column density (bottom) for component 1 are shown. Reprinted from Ref. [69].

but can vary in space. In such a situation, the system develops a nontrivial spin texture due to the DDI and self-organized magnetic patters may form. Possible quantum phases that arise due to the DDI have been discussed by using the single mode approximation [70] in which all spin components are assumed to share the same spatial mode. The spin ordering and spin waves in Mott states in 1D and 2D optical lattices have also been discussed in Refs. [71, 72, 73]. In the second-quantized form, the magnetic DDI for a spinor BEC is given by cdd Vdd = 2 dr dr

: F (r)Q (r r )F (r ) :,

(138)

where F (r) is dened by Eq. (17) and cdd = 0 (gF B )2 /(4), with gF being the Land g-factor e for the atom. The dipole interaction yields a non-local term in the GP equations. In fact, the GP

36

equations for a spin-1 BEC [Eq. (52)] are given by i

2 2 m = + U(r) pm + qm2 m t 2M 1 1

+ c0 nm + c1
n=1

F fmn n + cdd
n=1

b fmn n ,

(139)

and those for spin-2 BEC [Eqs. (78)(80)] are given by i

2 2 m = + U(r) pm + qm2 m t 2M

+ c0 nm + c1

c2 F fmn n + A + cdd b fmn n m 5 n=2 n=2

(140)

Here, b is the eective dipole eld dened by b dr

Q (r r )F (r ),

(141)

which is nonlocal, and Bdd = cdd b/(gF B ) works as an eective magnetic eld. 4.3.1. Einstein-de Haas eect We rst consider the dynamics of a spinor-dipolar BEC at zero external magnetic eld. Suppose that we prepare a spin polarized BEC in an external eld. We consider what happens if we suddenly turn o the external eld. In the absence of the DDI, a spin polarized BEC is stable because the total spin of the system will then be conserved. However, in the presence of the DDI, the situation drastically changes because the DDI does not conserve the spin angular momentum; it only conserves the total (i.e., spin plus orbital) angular momentum of the system. This can be understood from the fact that the DDI (113) is not invariant under a rotation in spin space due to the spin-orbit coupling term (d r). On the other hand, because Vdd (r) is invariant under simultaneous rotations in spin and coordinate spaces, the DDI conserves the total angular momentum of the system. Therefore, the DDI causes spin relaxation to occur by transferring the angular momentum from the spin to the orbital part; as a consequence, the BEC starts rotating. This is the Einstein-de Haas eect in atomic BECs [74]. We numerically simulated such dynamics for the case of a 52 Cr BEC that is conned in an axisymmetric trap. Figure 10 (a) shows the time evolution of the spin angular momentum Mz = 2 dr 3 dr 3 m=3 m (r) i m (r), m=3 m|m (r)| and the orbital angular momentum Lz = where is the azimuthal angle around the symmetry axis. The initial state is assumed to be spinpolarized in the magnetic sublevel m = 3. We nd that as time progresses, the spin angular momentum decreases with a concomitant increase in the orbital angular momentum, so that the total angular momentum of the system is conserved. In the dynamics, the order parameter is given by m (r, , z) = ei(Jz m) m (r, z), (142)

where (r, , z) are the cylindrical coordinates, m is a complex function of r and z, and Jz is the projected total angular momentum on the z axis (Jz = 3 in the present case). The order 37

parameter (142) is the eigenstate of the projected total angular momentum that satises (fz + z )m (r) = m i m (r) = Jz m (r). (143)

The order parameters for magnetic sublevels m = 3, 2, and 1 at t = 2 are shown in Fig. 10 (b)(d). Each spin component except the m = 3 state is a vortex state with the winding number Jz m, and the BEC as a whole forms a coreless vortex.
[N~]
(a) -3

-2

-1

0 (b)

2 (c)

8 (d)

10

Figure 10: (a) Time evolution of spin and orbital angular momentum due to the Einstein-de Haas eect. (b)(d) Isopycnic surfaces of (b) 3 , (c) 2 , and (d) 1 at t = 2, where the color on the surface represents the phase of the order parameter. Reprinted from Ref. [74].

The physical origin of the spin relaxation is the Larmor precision of atomic spins around the dipole eld given by Eq. (141). Figure 11 (a) shows the dipole eld induced by a spin-polarized BEC in a spherical trap, and Figs. 11 (b) and (c) show the spin textures at t = 2. The whirling patterns in the upper and lower hemispheres exhibit opposite directions, reecting the fact that the xy components of the dipole eld are antisymmetric with respect to the z = 0 plane. In the case of a 52 Cr BEC, the DDI compensates for the kinetic energy to form the spin texture. As discussed in the next section, a spin-1 87 Rb BEC is ferromagnetic, and therefore a spin-polarized state is rather stable. In such a case, an external magnetic eld applied in the opposite direction to the polarization causes a resonance between the kinetic energy and the linear Zeeman energy, leading to the Einstein-de Haas eect [75]. In the case of a spin-1 23 Na BEC, the spin-exchange interaction favors a non-magnetic state, and therefore, it tends to destroy the spin-polarized state. 4.3.2. Ground-state spin textures at zero magnetic eld The DDI is known to create magnetic domains in solid-state ferromagnets. Similar structures are expected to appear in ferromagnetic BECs. Here, we consider ground-state spin structures. 38

(a)
4.1

(b)

z [ m]

(c)

0.25 [mG]

0 0

0.0013
0 0

-4.1 0 x [ m] 4.1 0

1 m

Figure 11: (a) Dipole eld induced by a spin-polarized BEC in a spherical trap, where the color of each arrow denotes the eld strength (see the left gauge) and the solid curve indicates the periphery of the condensate. (b),(c) Spin textures on the (b) z = 2 m plane and (c) z = 2 m plane at t = 2. The length of the arrows represents the magnitude of the spin vector projected on the xy plane, and the color indicates |F|a3 /N with aho being the harmonic oscillator length and ho N, the total number of atoms. Reprinted from Ref. [74].

The mean-eld energy for the DDI can be rewritten as cdd Vdd = 2 = cdd 2 dr dr

F (r)Q (r r )F (r )

F (k)Q (k)F (k)

k

cdd 4 = 2 3

3| k F(k)|2 | F(k)|2 ,
k

(144)

where F(k) is the Fourier transformation of F(r) and Q (k) is given by Eq. (125). When the BEC is fully magnetized, i.e., |F(r)| = f n(r), with f being the atomic spin angular momentum and n(r), the number density, the last term in Eq. (144) makes a constant contribution determined by the density distribution. Then, the dipole-favored structure satises k F(k) = 0 for all k, or in real space, F(r) = 0. This condition is the same as that for solid-state ferromagnets [76], and the corersponding stable domain structure is known as a ux-closure structure. In solid-state ferromagnets, the magnitude of magnetization is constant and the domain structure is constrained by boundary conditions such that the magnetization is parallel to the surface. In contrast, a spatial variation of |F(r)| in proportion to the number density yields a spin texture in trapped condensates. As compared to solid-state ferromagnets, the unique feature of ferromagnetic BECs is that they exhibit the spin-gauge symmetry that generate a supercurrent by developing spin textures (see, Sec. 6.2.1). The supercurrent, dened by Eq. (264), is generated by a spatial variation of the direction of magnetization. Therefore, even in the ground state in a stationary trap, a supercurrent can ow due to a dipole-induced spin texture. 39

Figure. 12 (a) shows the numerically determined phase diagram for a spin-1 ferromagnetic BEC [77] as a function of RTF /sp and RTF /dd , where RTF is the Thomas-Fermi radius; sp = / 2M|c1 |n, the spin healing length; and dd = / 2Mcdd n, the dipole healing length. We nd three phasesower (FL), chiral spin-vortex (CSV), and polar-core vortex (PCV)whose spin congurations are shown in Fig. 12 (b), (c), and (d), respectively. The FL and CSV phases are eigenstates of the projected total angular momentum Jz = 1, whose order parameter is given by Eq. (142). The orbital angular momentum for the order parameter (142) is calculated as Lz = dr
m=0,1

(r) i m

m (r) =

dr
m=0,1

(Jz m)|m (r)|2 .

(145)

Therefore, when Jz = 1, the BEC can have a nonzero orbital angular momentum depending on the population in the m = 0 and 1 states. When the system size is smaller than dd , the kinetic energy, which hinders the spin texture from developing, is dominant, and the spin is almost polarized. Most of the atoms therefore reside in the magnetic sublevel m = 1 and the orbital angular momentum is quite small. As the BEC becomes larger, the spin texture develops to form a ux-closure structure. Consequently, the number of atoms in the m = 0 and 1 magnetic sublevels increases, leading to an increase in the orbital angular momentum. The dierence between FL and CSV phases is the symmetry of spin textures: the spin texture has chirality in the latter phase. Furthermore, an increase in the size of the BEC results in the phase transition to the polar-core vortex state with Jz = 0, because the Jz = 1 state costs a large kinetic energy when the population in the m = 1 state increases. In the Jz = 0 phase, a polar-core vortex appears at the symmetry axis. All three phases in Fig. 12 can be realized in spin-1 87 Rb BECs, which traces on the dotted line shown in Fig. 12 (a). The orbital angular momentum along this line is plotted in Fig. 13 as a function of the total number of atoms N and trap frequency . The orbital angular momentum in the ground state has a signicant value, and it increases up to approximately 0.4 in the CSV phase. For higher-spin BECs, similar spin textures are expected to appear in the ferromagnetic phase, although the mass current depends on the value of the atomic spin and the core structure in the polar-core vortex region is nontrivial. Spin textures in the limit of strong ferromagnetic interaction are discussed in Ref. [78]. On the other hand, even in the case of a non-ferromagnetic phase such as a spin-1 23 Na BEC, a spin texture may appear if the dipole interaction is eectively enhanced in a pancake trap with a large aspect ratio [79]. 4.3.3. Dipole-dipole interaction under a magnetic eld Now, we consider the eect of an external magnetic eld on the DDI. Suppose that a homogeneous external magnetic eld B is applied in the z direction. Because the linear Zeeman term gFz B rotates the atomic spin around the z-axis at the Larmor frequency L = gF B B/ , it is convenient to describe the system in the rotating frame of reference in spin space. The spin vector operators F(rot) in the rotating frame are related to those in the laboratory frame by (rot) F (rot) iFy = eiL t F and Fz = Fz . Then, the dipolar interaction (138) can be x (rot) (rot) F

40

(a)

3
dd

PCV CSV
z

RTF /

1 Rb line 0 0 2 4 6 8

FL
10 12

y x

14

RTF /
(b) PCV phase
x z plane

sp

(c) FL phase
x z plane

(d) CSV phase

x z plane

x y plane

x y plane

x y plane

Figure 12: (a) Phase diagram of a ferromagnetic BEC in a spherical trap. A BEC of spin-1 87 Rb atoms traces on the dotted line. (b)(d) Cross sections of spin textures in the xy (upper) and xz (bottom) planes in the (b) ower phase, (c) chiral spin-vortex phase, and (d) polar-core vortex phase, where the color and length of each arrow indicate |F|a3 /N and ho the spin projection onto the planes. Reprinted from Ref. [77].

rewritten in terms of F(rot) as Vdd = 6 cdd 1 dr dr 5 2 |r r |3 0 Y2 (e) (rot) (rot) (rot) (rot) (rot) (rot) : 4Fz (r)Fz (r ) F+ (r)F (r ) F (r)F+ (r ) 6 1 (rot) (rot) (rot) (rot) + eiL t Y2 (e) F+ (r)Fz (r ) + Fz (r)F+ (r )
1 (rot) (rot) (rot) (rot) eiL t Y2 (e) F (r)Fz (r ) + Fz (r)F (r ) 2 2 (rot) (rot) (rot) (rot) + e2iL t Y2 (e)F+ (r)F+ (r ) + e2iL t Y2 (e)F (r)F (r ) :,

(146)

where e (rr )/|rr |. When the Zeeman energy is much larger than the DDI, i.e., L cdd n, the spin dynamics due to the dipolar interaction is much slower than the Larmor precession. In a large magnetic eld, we may therefore use an eective DDI that is time-averaged over the

41

0.5 0.4

Lz / N

0.3 0.2 0.1 0 0

FL

CSV

PCV

Figure 13: Orbital angular momentum for a spin-1 87 Rb BEC in a spherical trap with trap frequency . Reprinted from Ref. [77].

Larmor precession period 2/L : cdd Vdd = 4 |r r |2 3(z z )2 |r r |5 (rot) (rot) (rot) (rot) : F (r) F (r ) 3Fz (r)Fz (r ) : . dr dr

More generally, an eective DDI under an external eld applied in the B direction is given by cdd Vdd = 4 dr

where the time-averaged integration kernel 1 3( B)2 r Q (r) = ( 3 B B ) r3 (149)

corresponds to Eq. (120). In contrast to Eq. (138), the time-averaged interaction (147) separately conserves the projected total spin angular momentum and the projected relative orbital angular momentum. However, the long-range and anisotropic nature of the dipolar interaction is still maintained in Eq. (147). While the total longitudinal magnetization is conserved in the presence of a magnetic eld, the anisotropic dipole interaction can induce spin textures in the transverse magnetization [80]. Equation (149) provides information about how the dipole moments tend to align in the ground state. Longitudinal magnetization tends to become parallel in the direction of the magnetic eld, whereas it tends to be anti-parallel in the direction perpendicular to the magnetic eld [Figs. 14 (a) and (b)]. On the other hand, transverse magnetization tends to be anti-parallel along the magnetic eld and parallel in the perpendicular direction [Figs. 14 (c) and (d)]. Comparing the dipole energy for longitudinal and transverse magnetizations, the energy for longitudinal magnetization is twice as large (in negative sign) as that for transverse magnetization. The quadratic Zeeman eect also contributes to the ground state. As discussed in Sec. 3, the quadratic Zeeman eect in a spin-1 87 Rb BEC induces transverse magnetization. A stable 42

[rad/s] N2 )

1000

1/5

2000

(147)

dr : F(rot) (r)Q (r r ) F(rot) (r ) :,

(148)

(a)

(b)

(c)

(d)

Figure 14: Alignments of (a)(b) longitudinal and (c)(d) transverse magnetization that are favored by the time-averaged dipole interaction. Due to the Larmor precession, transverse magnetization tends to be anti-parallel along the magnetic eld and parallel perpendicular to the magnetic eld.

spin texture is determined by the balance between the quadratic Zeeman eect and the dipolar eect [81, 82, 83]. For simplicity, we consider a quasi-2D trap where the cloud size in the strongly conned direction is smaller than the spin healing length so that spin texture is uniform in this direction. We also assume that the magnetic eld is applied parallel to the 2D plane and the total magnetization along the magnetic eld is set to zero [25]. When the quadratic Zeeman energy is stronger than the DDI, the magnetization is restricted perpendicular to the external eld. Then, the transverse magnetization forms a helix structure, as in the case of Fig. 14 (c). On the other hand, when the DDI is stronger than the quadratic Zeeman energy, the longitudinal magnetization forms a domain structure to minimize the DDI [Fig. 14 (b)]. However, these ground states dier from the magnetic patterns observed by the Berkeley group [25, 84]. The equilibrium pattern for this system has yet to be understood. A further topic is the rotating spinor dipolar BEC in the presence of an external eld. When the rotating frequency is close to the Larmor precession, the third or fourth line in Eq. (146) remains in the time-averaged dipolar interaction instead of the second line in Eq. (146). For example, when a spin-polarized BEC is rotated, the spin relaxation is expected to occur resonantly with the Larmor frequency. 5. Bogoliubov Theory Quantum and thermal uctuations as well as external perturbations induce excitations from the mean-eld ground state. When the excitations are weak, they can be described by the Bo goliubov theory. We express the eld operator m as a sum of its mean-eld value m and the m: deviation from it m = m + m (m = f, f 1, , f ). (150)

Here, the mean-eld part can be calculated from the Gross-Pitaevskii theory described in Sec. 3. The basic concept of the Bogoliubov theory is to substitute Eq. (150) in the second-quantized Hamiltonian and retain the terms up to the second order in m . The resulting Hamiltonian can be diagonalized by means of the Bogoliubov transformations. The Bogoliubov transformations are canonical transformations that couple particle-like excitations with hole-like ones via a BoseEinstein condensate and they are non-perturbative. The resultant dispersion relation therefore depends on the coupling constant in a non-analytic manner. 43

We consider a spatially uniform system (U(r) = 0). In this case, it is convenient to analyze the system in Fourier space. We obtain the Fourier expansion of the eld operator: 1 m = a km eikr ,
k

(151)

where is the volume of the system. Then, the noninteracting part of the Hamiltonian is written as H0 =
km

pm + qm2 ) km , n

(152)

where k = 2 k2 /2M and n km = a a km . Because the interaction Hamiltonian comprises the km particle density, spin density, and spin-singlet pair operators, it is convenient to introduce the corresponding Fourier-transformed operators: k Fk Ak = = = dr (r)eikr = n
qm

a a q+km qm fmn a a q+kn qm

qmn

(153) (154) (155)

dr F(r)eikr = 2f + 1

drA00 (r)eikr =
qm

(1) f m a qm a qk,m ,

where it is understood that m runs over m = f, f 1, , f . We assume that the BEC occurs in the k = 0 state, and decompose the terms in Eqs. (152) (155) into the k = 0 component and the rest. Then, Eq. (152) is written as H0 =
m

(pm + qm2 ) 0m + n

(
k 0,m

pm + qm2 ) km . n

(156)

As shown below, the interaction Hamiltonians involves quadratic forms of the Fourier-transformed terms in Eqs. (153)(155). They can be decomposed similarly as : k : k
k

= = =

N(N 1) +
k 0

: k :, k : F F k :, k

(157) (158) (159)

: F F k : k
k

: F F0 : + 0
k 0

: A A k : k
k

: A A0 : + 0
k 0 0

: A A k : . k

In Eq. (157), the dominant terms in k q = k: k

0 m

qm

a a q+km are the terms with q = 0 and qm (160)

( a km + a a0m ) = D k + D , a0m km k

0m where D k = m a a km describes the density uctuations from the condensate. Substituting Eq. (160) into Eq. (157), we obtain : k : N(N 1) + k
k k 0

: (2D D k + D k Dk + D D ) : . k k k 44

(161)

 In Eq. (158), the term F0 can be further decomposed into F0 = F +

q 0,mn

fmn a a q+kn , qm

(162)

0m where F = mn fmn a a0n is the zeroth-order term and the last term is of the second order. On the other hand, the term F k 0 can be approximated as Fk
0 mn

fmn ( a kn + a a0n ). a0m km

(163)

This operator describes the spin uctuations from the condensate. Substituting Eqs. (162) and (163) into Eq. (158) and retaining the terms up to the second order, we obtain : F F k : F2 + 2 F k
k q 0,mn

fmn a a q+kn + qm
k 0,mnm n

fmn fm n (164)

(2 a0n a a kn + a a a kn akn + a a0n a a ). a0m km 0m 0m 0n km kn Similarly, Eq. (159) can be approximated as : A k A k : = 4S + S + 2 S +

k

(1)
k 0,m

f m

a km ak,m + H.c. (165)

+4
k 0,mn

(1)m+n a a0,n a a k,n , 0m k,m

where H.c. denotes the Hermitian conjugate of the preceding term and 1 S = (S + ) = 2 (1) f m a0m a0,m .
m

(166)

In the Bogoliubov approximation, we replace a0m by normalization condition: |m |2 = 1

m

Nm , which satises the following

1 N

n km .
k 0,m

(167)

By subtracting the last term, we can construct a number-conserving Bogoliubov theory without introducing the chemical potential. In keeping with the Bogoliubov theory, we retain the last terms in operator form. 5.1. Spin-1 BECs The interaction Hamiltonian for the spin-1 BEC is given by Eq. (23) or (24). In Fourier space, it is expressed as V = 1 2 (c0 + c1 ) : k : c1 A A k , k k
k

(168)

45

where c0 (g0 + 2g2 )/3 and c1 (g2 g0 )/3. Using Eqs. (156), (161), and (159), we obtain the eective Hamiltonian for the spin-1 BEC: 2c1 c0 + c1 N(N 1) S + S (p q) 0,1 + (p + q) 0,1 n n H e = 2 2 + k pm + qm n km c0 + c1 : (2D D k + D k Dk + D D ) : + k k k 2 k 0 c1 f m S + (1) a km ak,m + H.c. k 0m 2c1 (1)m+n a a0n a k,m a k,n . 0m
kmn k 0m

(169)

The rst three terms on the right-hand side were studied in Refs. [85, 4, 86]; the third and fourth lines describe the density and spin uctuations, respectively, and the last line describes the coupling between them. In the following discussions, we shall discuss the case for q = 0 unless otherwise stated. The case for q 0 is discussed in Ref. [10]. 5.1.1. Ferromagnetic phase When c1 < 0, the ground-state phase is ferromagnetic, and we have |1 |2 = 1 1 N n km , 0 = 1 = 0.
k 0,m

(170)

Substituting this into Eq. (169), we obtain 1 H F = (c0 + c1 )n(N 1) pN 2 1 + a k n k1 + (c0 + c1 )n( k1 ak1 + n k1 + H.c.) 2 k 0 +(
k

+ p) k0 + ( n

+ 2p 2c1 n) n k1 ,

(171)

where n = N/. The Hamiltonian (171) shows that the m = 0 and m = 1 modes describe the massive transverse spin and quadrupole spin excitations, respectively, with the dispersion relations given by EF k,0 EF k,1 = =
k k

+ p, + 2p 2c1 n.

(172) (173)

To diagonalize the Hamiltonian of the m = 1 mode, we perform the Bogoliubov transformation: a k = b k cosh b sinh . k 46 (174)

Substituting this into Eq. (171) and requiring that the coecients of the o-diagonal terms such as b k bk vanish, we obtain sinh(2) = gn
2 k (gn)2

, cosh(2) =

k
2 k (gn)2

(175)

where g c0 + c1 = and k
k

4 2 a2 M

(176)

+ (c0 + c1 )n. The diagonalized Hamiltonian then becomes HF = gn 1 (N 1) pN + 2 2 +

k 0 k( k k

+ 2gn)

gn (177)

E F b b k + E F n k0 + E F n k,1 , k,1 k,1 k k,0

where EF k,1 =
k[ k

+ 2(c0 + c1 )n]

(178)

describes the coupled density and longitudinal spin excitations. The sum over k in the rst line of Eq. (177) diverges for large kan artifact that arises due to the use of a delta function that does not correctly describe the short-range behavior of the real potential. For large k, the summand becomes
k( k

+ 2gn)

gn =

(gn)2 + . 2 k

(179)

Subtracting this term, the sum converges, giving 1 2 + 2gn) gn + (gn)2 128 = gn 2 k 15 na3 2 N.

k( k k

(180)

Thus, the ground-state energy of the system is gnN F E0 = 2 1 + 128 gn 15 na3 2 pN.

(181)

It follows from this that the sound velocity is given by c=

F 2 2 E0 = MN 2

g2 n 1 + 8 M

na3 2 ,

(182)

where the leading term is the Bogoliubov phonon velocity and the last term is called the LeeYang correction to it. 47

The Bogoliubov approximation takes into account virtual pair excitations from the meaneld, and therefore, the properties of the ground state are very dierent from those of the mean eld. A mean-eld ground state in the ferromagnetic phase is ( )N a0,1 |vac . N! On the other hand, the Bogoliubov ground state is given by | exp 0 a01 k a a |vac , k1 k1
k x >0,ky ,kz

(183)

(184)

where 2 = N[1 (8/3) na3 /], k = 1 + ck ck (ck + 2) with ck k2 /(8at n), and the t 0 summation is performed over the entire range of ky and kz , but the summation over k x is restricted to the positive side to avoid double counting. In contrast to the single-particle state in Eq. (183), the Bogoliubov ground state is a pair-correlated state. In momentum space, the Bogoliubov ground state exhibits pair correlation having opposite momenta; in real space, it describes the pairwise repulsive interaction. To show this, let us write down the many-body wave function in the coordinate representation: (r1 1 , , rN N ) = vac|1 (r1 ) 1 (r1 )| . Substituting Eq. (184) into this and ignoring the terms of the order of 1/N, we obtain [87]
4 exp 9 (3 8)N

(185)

(r1 1, , rN 1)

na3 / 2

(2NV 2N )1/4

exp

i< j

a2 ri j / , e ri j

(186)

where ri j |ri r j | and (8a2 n)1/2 is the correlation length. Thus, the many-body wave function decays exponentially when any two particles approach closer than the scattering length or the correlation (or healing) length. 5.1.2. Antiferromagnetic phase When c1 > 0, the ground state is antiferromagnetic, and we have 0 = 0, |1 |2 = 1 1 2 2 2N n k,m ,
k 0m

(187)

where p/(c1 n) and the last term is retained in order to construct a number-conserving Bogoliubov theory. Substituting this into Eq. (169), we obtain 1 1 H AF = c0 n(N 1) + c1 n 2 N 1 pN 2 2 1 + ( k + c1 nm,0 ) km + n c0 (D k Dk + D D k ) k 2 k 0 k 0,m +c1 (S k S k + S S k + n|1 1 | k,0 ak,0 ) + H.c. , a k 48 (188)

where Dk Sk =
m

a a km 0m a a km 0m
m

N 2 N 2

1 + a k1 + 2 1 + a k1 2

1 a k,1 2 1 a k,1 . 2

(189) (190)

We note that when 0, the density and spin excitations are coupled because [D k , S ] = /2. k The m = 0 mode is decoupled and its dispersion relation can be calculated in a manner similar to that for the m = 1 mode of the ferromagnetic phase, with the result E AF = k,0
2 k

+ 2c1 n

+ p2 .

(191)

The m = 1 modes are coupled, and their dispersion relations can be obtained by solving the Heisenberg equations of motion for the m = 1 modes [6]: E AF = k,
2 + k

2(c0 +c1 )n k

n2 (c0 c1 )2+

4p2 c0 , c1 (192)

In the absence of an external magnetic eld (p = = 0), Eq. (192) reduces to the excitation spectra of the uncoupled density and spin waves: E AF = k,+
k( k

+ 2c0 n), E AF = k,

k( k

+ 2c1 n).

(193)

5.1.3. Domain formation The MIT group observed the formation of metastable spin domains [88]. They rst prepared all atoms in the m = 1 state and transferred half of them to the m = 0 state by irradiating the rf eld. Then, letting the system evolve freely in time while applying a uniform magnetic eld to prevent the m = 1 component from appearing due to the quadratic Zeeman eect, they found that spin domains develop with the two components being aligned alternatively. It was discussed [88, 89, 90] that this phenomenon is due to the imaginary frequencies of the excitation modes. Here, we use the Bogoliubov theory to conrm this hypothesis and to derive a general expression for the dispersion relation. The experimental conditions in eect amount to setting |1 |2 = |0 |2 = 1 1 2 2N ( k1 + n k0 ), 1 = 0. n
k 0

(194)

Substituting this in Eq. (169), we obtain c0 + c1 n(N 1) c1 nN H= 2 8 c0 n 3c1 n c0 n c1 n + + n k1 + k + n k1 k+ 2 4 2 4 k 0 c0 + n( k0 ak0 + h.c.) a 8 c0 + c1 + n( k1 ak1 + a k1 ak0 + a ak0 + h.c.) . a k1 8 49

(195)

The eigenspectrum can be obtained by writing down the equations of motion for ak,1 and ak,0 . Rewriting the resulting equations in terms of A = a k,1 + a and B = a k,0 + a , we obtain k,1 k,0 d2 A = dt2 d2 (i )2 2 B = dt (i )2 Assuming that A e
i

2 k1 2 k0

(p + q)2 A + 2(p + q)( p2 B + 2(p + q)(

i

k1

p q) B, (196)

k0

p)A.

Ekt

and B e =
2 k

Ekt

, we obtain
k

(E k )2

+ (2u + v)
2

3 + v2 4(u2 + 2uv + 2v2 ) 4

3
1 2

2 k

3 +2v (2u + v) k v u + v 4

(197)

where u c0 n/2 and v c1 n/2. For parameters of the MIT experiment, n 1014 cm3 , at 29, and as 26 [91], we nd that v/u 1. In Eq. (197), ignoring higher-order powers of v/u, we obtain (E k )2 =
2 k

+ (2u + v)

2(u + v) k ,

where the plus (minus) sign corresponds to the density (spin) wave. We see that the energy of the spin wave becomes pure imaginary for k < v, implying the formation of spin domains. The corresponding wavelength denes the characteristic length scale of the spin domains as c = 3 . (a2 a0 )n (198)

This result agrees with that of Ref. [90], except for a numerical factor. Using the above parameters, we obtain 18 m; this is in reasonable agreement with the observed value of approximately about 40 m [88]. 5.2. Spin-2 BECs We again assume that the total number of particles is conserved: N BEC +
k 0 m

n k,m = N.

(199)

Then, the one-body part of the Hamiltonian becomes H0 = pN fz +

k 0 m k

p(m fz ) n k,m ,

(200)

where fz
m

m|m |2 .

(201)

50

The interaction Hamiltonian of a spin-2 BEC is given by Eq. (30) and its Fourier transform becomes 1 V= 2 c2 c0 : k : +c1 : F F k : + A A k . k k 5 k (202)

Using Eqs. (156), (161), (164), and (165), we obtain the eective Hamiltonian for the spin-2 BEC: V c0 n c1 n 2 2c2 n 2 + | f| + |s | 2 2 5 4c2 n 2 c1 n| f |2 + |s | n k,m 5 k 0 m c0 n k,m (2 n a a k,m + m n a k,m ak,n + m n a ak,n ) + 2 k 0 mn m k,n N +c1 n
k 0 mn

f fmn a a k,n k,m

fi j fmn (2i n a a k, j + i m a k, j ak,n + j n a a ) k,m k,i k,m k 0 i jmn

c1 n + 2

c2 n s + 5 2c2 n + 5 where

k 0 m

(1) a k,m ak,m + H.c.

m

(1)m+n m n a a k,n , k,m k 0 mn

(203)

f
mn

fmn m n ,

(204) (205)

1 2

(1)m m m .
m

5.2.1. Ferromagnetic phase We assume that only the m = 2 magnetic sublevel is macroscopically occupied. Then, we have Fz = 2 02 = 2(N k 0m n km ), and the one-body part becomes n H0 = 2pN +
k 0,m k

+ (2 m)p n k,m .

(206)

Because the spinor that is normalized to unity is m = m,2 , only the z-component is nonzero in f and s = 0 and Eq. (203) becomes V = 1 g4 nN + n 2 k c0 n k,2 + 2c1 (2 k,2 2 k,0 3 k,1 4 k,2 ) n n n n
0

2 g4 + c2 n k,2 + ( k,2 ak,2 + a a ) , a k,2 k,2 5 2 51

(207)

where g4 c0 + 4c1 . Combining Eqs. (206) and (207), the total Hamiltonian is given by HF = 1 g4 nN 2pN 2 +
k 0

( (
k 0 k

1 + g4 n) n k,2 + g4 n( k,2 ak,2 + a a ) a k,2 k,2 2 + p) k,1 + ( n

k

+ +(

+ 2p 4c1 n) n k,0 + (

+ 3p 6c1 n) n k,1 (208)

+ 4p 8c1 n + 2c2 n/5) n k,2 .

It follows from this Hamiltonian that the m = 2 mode gives the Bogoliubov spectrum
F Em=2,k = 2 k

+ 2g4 n k .

(209)

For the system to be stable, the Bogoliubov excitation energy (209) must be positive. This implies that the s-wave scattering length for the total spin-4 channel must be positive: 4 2 a4 > 0. (210) M This condition is the same as the one required for the ferromagnetic mean eld to be stable, that is, the rst term on the rhs of Eq. (208) being positive. We also note that the Bogoliubov spectrum (209) is independent of the applied magnetic eld and remains massless in its presence. This Goldstone mode is a consequence of the global U(1) gauge invariance due to the conservation of the total number of bosons. On the other hand, all the other modes give single-particle spectra: g4 =
F Em=1,k F Em=0,k F Em=1,k F Em=2,k

= = = =

k k k k

+ p, + 2p 4c1 n, + 3p 6c1 n, + 4p 8c1 n + 2c2 n/5.

(211) (212) (213) (214)

For the system to be stable, the single-particle excitation energies must also be positive. This implies that p > 2c1 n and p > (2c1 c2 /10)n; these conditions are the same as the ones for the ferromagnetic phase to be the lowest-energy mean eld. 5.2.2. Antiferromagnetic phase We consider the situation in which only the m = 2 modes are macroscopically occupied. By requiring that the total magnetization be fz , the normalized spinor is given by 1 fz , 0 = 1 = 0. 2 4 Because of the quantum depletion of the condensate, Fz has to be replaced by (N and the one-body part of the Hamiltonian is therefore given by 2 = ei2 H0 = = pFz + pN fz +
k 0,m

1 fz + , 2 = ei2 2 4

(215)
k 0m

n km ) fz ,

(
k 0,m

pm) km n
k

p(m fz ) n km .

(216)

52

The interaction Hamiltonian can be evaluated by noting that p Hamiltonian is given by H AF = 1 1 (c0 + c2 /5)nN pN fz 2 2 + ( k + g4 n|2 |2 ) k,2 + ( n
k 0

fz (c1 c2/20), and the total

+ g4 n|2 |2 ) k,2 n

g4 n 2 2 ( a k,2 ak,2 + 2 a k,2 ak,2 ) 2 2 k,2 +(c0 4c1 + 2c2 /5)n(2 2 a k,2 ak,2 + 2 2 a a k,2 ) + H.c. + +
k 0 k k

1 + (c1 c2 /5)n + (c1 + c2 /10)n fz n k,1 + 2

1 + (c1 c2 /5)n (c1 + c2 /10)n fz n k,1 2 +2(c1 c2 /5)n(2 2 a k,1 ak,1 + H.c.) c2 n + ( a k,0 ak,0 + H.c.) . ( k c2 n/5) k,0 + n 5 2 2 k 0

(217)

We thus nd that the spectrum is classied into three cases: the coupled m = 2 modes, coupled m = 1 modes, and single m = 0 mode. We rst investigate the eigenspectra of the m = 2 modes. The relevant part of the Hamiltonian is h2 =
k 0

k,2 n k,2 +

k,2 n k,2

g4 n 2 2 ( a k,2 ak,2 + 2 a k,2 ak,2 + H.c.) 2 2 2 + c0 4c1 + c2 n(2 2 a k,2 ak,2 + 2 2 a a k,2 + H.c.) , k,2 5 (218) where k,2 = k + g4 n|2 |2 . Here, the complex numbers 2 can be made real by means of unitary transformations ak,2 ak,2 ei2 . The resulting equations can be diagonalized by writing down the Heisenberg equations of motion for a k,2 + a k,2 , and seeking solutions of the AF form a k,2 + ak,2 exp(iE k,2 t/ ). The resultant eigenspectra read (E AF )2 = k k,2 + g4 n g4 n fz2 8 c2 + 1 c1 4 g4 20
2

fz2 1 4

(219)

For the ground state to be stable, these energies must be positive; this implies the conditions c1 c2 /20 > 0 and c0 + c2 /5 > 0. The former condition is met if the antiferromagnetic phase is the lowest-energy state, whereas the latter condition ensures that the antiferromagnetic phase is mechanically stable, that is, the rst term on the rhs of Eq. (217) is positive. The spectra (219) are massless regardless of the presence of the magnetic eld. This is because they are the Goldstone 53

modes associated with the U(1) gauge symmetry and the relative gauge symmetry, which is the rotational symmetry about the direction of the applied magnetic eld. Next, we investigate the eigenspectra for the coupled m = 1 modes. The relevant part of the Hamiltonian is hAF = 1
k 0 where 2(c1 c2 /5)n2 2 and

k,1 n k,1

k,1 nk,1

+ ( k,1 ak,1 + H.c.) , a

(220)

k,1

+ c1

c2 1 c2 n c1 + n fz . 5 2 10

(221)

The eigenspectrum can be obtained by writing down the Heisenberg equations of motion for a k1 and ak,1 , and seeking solutions of the form exp( iE k,1 t/ ). The eigen matrix is 2 2 and it can be solved straightforwardly, giving E k,1 = c2 1 c1 + n fz 2 10 c2 2 + 2 c1 n + k 5

1 c2 c1 4 5

n2 fz2 .

(222)

The spectra become massive in the presence of a magnetic eld. For the ground state to be stable, the spectra must be positive denite. Hence, we obtain the conditions c1 c2 /20 > p/2n > 0 and c2 < 0; these are in agreement with the conditions for the antiferromagnetic phase to be the lowest-lying one. The spectrum for the m = 0 can be obtained by means of the Bogoliubov transformation: E k,0 =
2 k

2|c2 |n 5

c2 n 10

fz2 .

(223)

This mode becomes massive in the presence of a magnetic eld. At zero magnetic eld, all ve dispersion relations in Eqs. (219), (222), and (223) become gapless and linear, reecting the fact that in the absence of the magnetic eld, the ground state is degenerate with respect to ve continuous variables. Note, however, that only the rst four are Goldstone modes because the continuous symmetry concerning the last mode does not reect the symmetry of the Hamiltonian but the hidden SO(5) symmetry [92]. 5.2.3. Cyclic phase There are several mean-eld solutions for the cyclic phase. Here, we consider the Bogoliubov excitations from the following mean eld: 2 = 1 fz i2 1 e , 1 = 0, 0 = 2 2 1 fz2 /4 i0 e , 2

(224)

54

where 2 + 2 20 = . Because s = 0 and f = 0, the interaction Hamiltonian becomes V c0 n 1 n k,m + pN fz c1 n fz2 2 2 k 0 m c0 + : (D k Dk + D D k + H.c.) : +c1 n fz m k,m n k 2V k 0 k 0 m c1 + fi j fmn ( a0,n a a k, j + a a a k, j ak,n + H.c.) a0,i k,m 0,i 0,m 2V k 0 i jmn + 2c2 5V (1)m+n a a0,n a a k,n . 0,m k,m
k 0 mn

(225)

The interaction Hamiltonian shows that the spectra are divided into two parts: the m = 1 coupled modes and the m = 2 and m = 0 coupled modes. The eigenspectra for the former modes can be found by writing down the Heisenberg equations of motion for ak,1 with the result
C (E1 )2 = 2 k

+ 1

fz2 8

2 2 fz f 32 z 2

3 2 f 16 z

2 f . 64 z

(226)

These excitation energies are always positive semidenite and massive in the presence of a magnetic eld. The eigenspectra for the m = 2 and m = 0 coupled modes can be obtained by writing down the equations of motion for X a k,2 a , Y a k,0 a , and Z a k,2 a k,2 k,0 k,2 . The desired eigenspectra are given by
C E2,0 = k

2c2 n 5

(227)

and 4 + fz2 C (E2,0 )2 = k k + c0 n + c1 n 2 (4 + fz2 )2 (c0 n)2 4 3 fz2 c0 c1 n2 + (c1 n)2 . 4 (228)

The solutions in Eq. (228) are always positive semidenite and massless even in the presence of an external magnetic eld. This is a consequence of the fact that the mean-eld solution is degenerate with respect to at least two continuous variables. The solution in Eq. (227) is massive and independent of the external magnetic eld. 5.3. Dipolar BEC In the presence of a magnetic dipole-dipole interaction (DDI), the Hamiltonian is 2 c0 (r) 2 2 2 2 H = dr + U(r) pm + qm m (r) + n(r) + c1 |F(r)| m 2M 2 m + cdd 2 drdr
,=x,y,z

F (r)Q (r r )F (r ), 55

(229)

where cdd = 0 (gF B )2 /(4) and Q (r) = 3 r r . 3 r (230)

In general, the long-range nature of the DDI yields rich physics in a trapped system, such as nontrivial equilibrium shape and spin textures. The Bogoliubov spectrum in the presence of the DDI has been studied using a Gaussian ansatz and numerical analysis [44, 52, 45, 53]. Here, we discuss the Bogoliubov spectrum for a spinor dipolar BEC under an external eld [93, 94, 82]. We assume that the system is strongly conned in the z-direction and free in other directions so that it is eectively two-dimensional. Consequently, the order parameter is decomposed as follows: (x, y, z) = h(z)(x, y), where h(z) is assumed to be normalized to unity:

(231)

h2 (z)dz = 1.

(232)

Substituting Eq. (231) into Eq. (229), we obtain an eective 2D Hamiltonian: 2 c0 c1 () 2D 2D 2 2 2 2 H = d pm + qm m () + n() + |F()| + Hdd , m 2M 2 2 m where (x, y), c = c

(233)

h4 (z)dz,

(234)

and Hdd is the two-dimensional version of the time-averaged dipolar interaction (149):
2D Hdd =

cdd 2 h4 (z)dz

F ()F ( )
k2D

eik2D ( ) Q(2D) (k2D ).

(235)

Here, the integration kernel Q(2D) (k2D ) is given by Q(2D) (k2D ) = = dkz 2 dkz 2 dzh2 (z) dzh2 (z) dz h2 (z )eikz (zz ) Q (k3D ) dz h2 (z )eikz (zz ) (236)

2 1 3( B k3D )2 ( 3 B B ), (237) 3

where Q (k(3D) ) is the Fourier transform of Eq. (149) and B is a unit vector along the external eld. To proceed with the calculations, we assume that h(z) = 1 (2d2 )
1 4

e 4d2 .

z2

(238)

56

After some algebraic manipulations, we obtain 2 Q(2D) (k2D ) = ( 3 B B )Q(k2D ), 3 2 2 Q(k) =1 3 cos2 B + 3 erfc(kd)kdek d cos2 B sin2 B (k x cos B + ky sin B )2 , (239)

(240)

where B and B specify the direction of the magnetic eld: B = (cos B sin B , sin B sin B , cos B ). With this kernel, the Gross-Pitaevskii equation in the presence of the magnetic DDI can be expressed as follows: i
2 m = t 2M 2

+ U(r) + qm2 + c0 n m + c1
n

F fmn n + cdd
n

b fmn n ,

(241)

where the linear Zeeman term is eliminated by moving into the rotating frame of reference in spin space with the Larmor frequency, and b is an eective dipole eld given by b d
k2D

eik2D ( ) Q(k2D )

( 3 B B )F ( ).

(242)

Note that because the spin and orbital parts are decoupled in the dipole kernel (237), it is not necessary to choose the coordinate axes in real space that coincide with those in the spin space. In what follows, we take the spin quantization axis along the magnetic eld, which is not necessarily parallel to the direction of the strongly conning potential, and their relative direction is given by B = (cos B sin B , sin B sin B , cos B ). In the following discussions, we discuss the ferromagnetic phase (c1 < 0) such as the f = 1 hyperne manifold of 87 Rb. As in this case, we assume that the coecient of the quadratic Zeeman eect is positive: q > 0. When the total longitudinal magnetization is set to zero, the magnetization is transverse due to the quadratic Zeeman eect. In fact, it can be shown that (0) 1 1 q/2 (0) = (1 + q)/2 , (243) 0 (0) 1 q/2 1 gives a stationary solution of Eq. (241) with U(r) = 0, where q= q . 2n(| 1 | + cdd Q(0)) c (244)

The expectation value of the spin vector for this spinor is nonvanishing only for the x-component: F =
mn

((0) ) (f )mn (0) = m n

n(1 q2 ),x .

(245)

To investigate the Bogoliubov excitation, we expand the order parameter from the mean-eld stationary state (243): m = (0) + m
k

u km ei(kr k t/ ) + v ei(kr k t/ km 57

(246)

Substituting Eq. (246) into Eq. (241), we obtain the 6 6 eigenvalue matrix equation: k uk H0 = vk H1 H1 u k , H0 v k (247)

where u k and v k are three-component vectors and H0,1 is the 3 3 matrix. The spin wave mode u, v (1, 0, 1)T decouples from the other two modes and the eigenfrequency is obtained as 2 = k
k

+ q cdd n(1 q)[2Q(k) + Q(0)]

+ cdd n(1 + q)[Q(k) Q(0)] .

(248)

Note here that due to the anisotropy of Q(k), the right-hand side of Eq. (248) becomes negative depending on the direction of k. This result implies that the initial state given by (243) is dynamically unstable. Figures 15 (a)(c) show the distributions of |Im k | (red) and Re k (blue) in momentum space which correspond to the dynamical instability and Landau instability, respectively. For 2 > 0, the sign of k is determined so that the corresponding eigenmode k satises the normalization condition u u k v v k = 1. k k
(a)
0.05

(b)
160 mG, no helix

(c)
200 mG, no helix

kz/2 [m-1]

80 mG, no helix

0.0

0.05

|Im !k| 0 Re !k 0 0.05 0.0

1.4 s-1 0.56 s-1 0.05

|Im !k| 0 Re !k 0

0.44 s-1 0.29 s-1 0

|Im !k|

0.54 s-1

kx/2 [m-1]

Figure 15: Real and imaginary parts of k in Eq. (248) for the case of d = 1 m and n = 2.31014 cm3 . The region with nonzero imaginary part shows the dynamical instability, while the region with Re k < 0 shows the Landau instability. The external magnetic eld is (a) 80 mG, (b) 160 mG, and (c) 200 mG.

The anisotropic distribution of the instability suggests the existence of the periodic pattern in real space. Recently, the Berkeley group observed the crystallization of transverse magnetization in a quasi two-dimensional 87 Rb BEC [25]. In one of the experiment, they started from the initial condition in which the magnetization form a helix and let the system evolve with time. The Bogoliubov spectrum (248) obtained above does exhibit dynamical instabilities that exhibit various patterns such as a helix and stripes. However, it fails to explain the Berkeley experiments in terms of the time and length scale. 6. Vortices in Spinor BECs A scalar BEC can host only one type of vortex, that is, a U(1) vortex or a gauge vortex. However, a spinor BEC can host a much richer variety of vortices. We begin by discussing why this is the case. The properties of a vortex can be characterized by looking at how the order parameter changes along a loop that encircles the vortex. Because the order parameter must be single-valued, it should satisfy (, ) = ( + 2, ), 58 (249)

where is a 2 f + 1-component order parameter; , the azimuthal angle which species the location of the loop; and , a set of coordinates which, together with , completely specify the order parameter. For the case of a scalar BEC whose order parameter is a single complex function = ||ei , the vortex solution to Eq. (249) is (, ) = ||einw , (250)

where nw = 0, 1, 2, is the winding number. If nw 0, the density at the vortex core must vanish. The corresponding superuid velocity is given by vs = 2Mni [ ( )], (251)

where M is the atomic mass and n = ||2 . Substituting Eq. (250) in Eq. (251), we nd that vs = nw M . (252)

Therefore, the circulation of the superuid velocity is quantized in units of = /M. Due to the internal degrees of freedom, spinor BECs allow several solutions other than Eq. (250), depending on the symmetry of the order-parameter manifold. A general order parameter of a spinor BEC is described as n = nei U(, , )0 , (253) where n is the particle density; , the global gauge; U(, , ) = eifz eify eifz , an SO(3) rotation with spin matrices f x,y,z and Euler angles , , and ; and 0 , a representative normalized spinor for the order parameter. Hence, the spacial variation of , , and as well as characterize the structure of a vortex. In the following, we investigate what types of vortices are allowed in each phase. A general classication of topological excitations based on homotopy theory is given in Sec. 7.2. 6.1. Mass and spin supercurrents in spinor BECs The mass current density j(mass) for a spinor BEC is dened as the sum of currents in all spin components: j(mass) = 2Mi ( m ) ( )m . m m
m

(254)

It can be shown from the GPEs that j(mass) satises the continuity equation: n + t j(mass) = 0.
(spin)

(255) as (256)

In a similar manner, we dene the spin current density j j

(spin)

2Mi

(f )mn ( n ) ( )n . m m
mn

59

At zero magnetic eld, j

(spin)

satises the continuity equation for the component of the spin: F + t j

(spin)

= 0.

(257)

In the presence of an external magnetic eld, the spin angular momentum perpendicular to it is not conserved, and Eq. (257) is replaced with the equation of motion of spins given by F + t j
(spin)

p BF

2qn B NB ,

(258)

where B is a unit vector in the direction of the magnetic eld and N is the 3 3 symmetric nematic tensor dened in Eq. (71). The dipole interaction induces an eective magnetic eld and therefore modies the rst term in Eq. (258) The superuid velocity eld for the particle and spins, dened as v(mass) = j(mass) /n and (spin) (spin) v = j /n, respectively, are described in terms of a normalized spinor as v(mass) = v
(spin)

2Mi 2Mi

m ( m ) ( m )m , m (f )mn m ( n ) ( m )n . mn

(259) (260)

In contrast to the case of a scalar BEC, the circulations of Eqs. (259) and (260) are not always quantized, as shown below. 6.2. Spin-1 BEC The rotation matrix for the order parameter of a spin-1 BEC is given by i(+) i e cos2 e2 sin ei() sin2 2 2 ei ei sin sin . cos 2 U(, , ) = 2 i() 2 e ei i(+) 2 sin sin 2 e cos 2 2

(261)

The spin-1 BEC in the absence of an external magnetic eld has two phases: ferromagnetic and polar. Below, we discuss vortices in each phase. 6.2.1. Ferromagnetic phase ferro A representative order parameter of a spin-1 ferromagnetic BEC is 0 = (1, 0, 0)T , where T denotes the matrix transpose. Substituting this and Eq. (261) in Eq. (253), we obtain a general order parameter of a spin-1 ferromagnetic BEC: i e cos2 2 1 (262) ferro = nei() 2 sin . i 2 e sin 2 The direction of the magnetization for this state is given by s F = (sin cos , sin sin , cos ). fn 60 (263)

The linear combination in Eq. (262) reects the spin-gauge symmetry which implies that the rotation in spin space through angle is equivalent to the gauge transformation by = . For simplicity of notation, we set = . We substitute Eq. (262) in Eq. (259), obtaining [7] v(mass,F) = M [ cos ]. (264)

Thus, the spatial variation of the spin gives rise to a supercurrent, and vice versa. In the case of a scalar BEC, the superuid velocity is given by v s = ( /M); it is therefore irrotational, i.e., vs = 0. However, this is not the case with the ferromagnetic BEC. In fact, using ( f a) = ( f ) a + f ( a), we nd that v(mass,F) = M sin ( ). (266) (265)

Introducing unit vectors t 0 = (cos cos , cos sin , sin ) and u0 = ( sin , cos , 0) which satisfy s = t 0 u0 , we obtain = t0 s and sin = u0 s . Then, we can express Eq. (266) in terms of s as v(mass,F) = = = = = M M M M (t0 s ) ( 0 s ) u

(t0 s ) ( 0 s ) u

+ )(t0 s ) ( 0 s ) u s ) (t0 s ) ( 0 s ) u s ). (267)

s ( s

2M

s ( s

This relation is known as the Mermin-Ho relation [95]. v(mass,F) is nonvanishing because of s the contribution from the Berry phase. To see this, we rewrite Eq. (264) as v(mass,F) M (1 cos ) = M ( )

and integrate both sides along a closed contour C. Due to the single-valuedness of the order parameter, the integral of ( ) should be an integral multiple of 2. Hence, we obtain v(mass,F) d
C

(1 cos ) d =
C

h nw (nw : integer). M

(268)

The term C (1 cos ) d gives the Berry phase enclosed by the contour C. We thus conclude that in the ferromagnetic BEC, the circulation alone is not quantized, but the dierence between the circulation and the Berry phase is quantized. 61

As a special case, let us consider the order parameter (262) with = nw and = nw ; then, Eq. (262) changes from = (ei2nw , 0, 0)T to (0, 0, 1)T as changes from 0 to . This implies that the 2nw vortex is topologically unstable. On the other hand, if we take = nw and = (nw + 1), then Eq. (262) changes from = (ei(2nw +1) , 0, 0)T to (0, 0, ei )T as changes from 0 to . Thus, the 2nw + 1 vortex is unstable against the decay to the singly quantized vortex which is stable [7]. We can utilize these properties to create a doubly quantized vortex from a vortex-free state by changing from 0 to as a function of time while keeping = and = [96, 97]. Such a scheme is realized if a spin polarized BEC is prepared in the m = 1 state under a quadrupole eld B(r, , z) = (B (r, z) cos(), B (r, z) sin(), Bz (r, z))T , where (r, , z) are the cylindrical coordinates and B > 0. At t = 0, Bz is set to be positive and much greater than B . As Bz is changed adiabatically from Bz B to Bz B , the atomic spins follow the direction of the local magnetic eld as = and = arctan(B /Bz ), resulting in the transformation of the order parameter from (1, 0, 0)T to (0, 0, e2i )T ; thus, a doubly quantized vortex is imprinted. Using this method, the MIT group have observed multiply quantized vortices [98]. Any axisymmetric conguration of is homotopic to one of the following two types of vortices: a coreless vortex [Fig. 16 (c)] and a polar-core vortex [Fig. 16 (b)]. When = , the vortex is coreless; in particular, the order parameter for the case of = = is given by cos2 i 2 = n e sin , 2 2i 2 e sin 2

coreless

(269)

in which the singularity can be removed by choosing = 0 at r = 0. Let us consider the conguration in which (r = 0, , z) = 0 and (r = r0 , , z) = , where r0 is the radius at the (cylindrical) boundary of the system. Then, the spinor (1, 0, 0)T at the origin with no vortex singularity and (0, 0, e2i )T at the boundary with a doubly quantized vortex attached to the inverted spin conguration. The corresponding superuid velocity is nonsingular at the origin; in fact, v(mass,F) = Mr (1 cos ), (270)

where is the unit vector in the azimuthal direction. The coreless vortex has been observed by the MIT group [99], where a vortex was imprinted using a quadrupolar eld. On the other hand, when = 0 and = , the vortex core has a singularity in the ferromagnetic order, where the order parameter at r is given by polarcore i e cos2 2 1 = n 2 sin . i 2 e sin 2

(271)

Because the m = 1 components have nonzero vorticities, the density in Eq. (271) must vanish at the center of the vortex. In fact, the free-energy analysis shows that the vortex core is lled by a polar state in which only the m = 0 component is occupied; hence, this vortex is called a polar-core vortex. As a consequence, only the spin density vanishes at the vortex core and the number density remains nite. A variational order parameter that describes both the vortex and 62

its core simultaneously is given by ei f (r) cos2 2 = n 1 f 2 (r) cos4 g2 (r) sin4 2 ei g(r) sin2 2 , 2

polarcore

(272)

where f (r) and g(r) are the variational functions subject to f (0) = g(0) = 0 and f () = g() = 1 and it can be obtained so as to minimize the Gross-Pitaevskii energy functional. The spontaneous formation of the polar-core vortices has been observed by the Berkeley group [100].

(a)

(b)

(c)

Figure 16: (a) Order parameter for the spin-1 ferromagnetic phase, where the arrow indicates the direction of the local magnetization. (b) Polar-core vortex given by Eq. (271), where the ferromagnetic order parameter has a singularity at the vortex core which is lled by a polar state. (c) Coreless vortex given by Eq. (269) with (r = 0) = 0 and (r = r0 ) = /2.

To calculate the spin superuid velocity (260), it is convenient to use the Cartesian coordinate in spin space = ( x , y , z )T rather than the spherical coordinate = (1 , 0 , 1 )T . They are related to each other by 1 1 = U i 20 63 0 0 2 1 i . 0

(273)

In the Cartesian coordinate, spin matrices are given by (f ) = i The ferromagnetic order parameter (262) is then given by 1 ei ferro = ( t 0 + iu0 ) = ( t + iu), 2 2

where , , = {x, y, z}.

(274)

where t = cos t 0 sin u0 , u = sin t 0 + cos u0 , and t 0 and u0 are dened below Eq. (266). Substituting Eq. (274) in Eqs. (259) and (260) and using some vector calculus formulae, we obtain v(mass,F) = v
(spin,F)

u t ,

(275) s s .
,

= s v(mass,F)

2M

(276)

The rst term on the right-hand side of Eq. (276) is the ow of spins carried by the particle, and the second term results from the spin ow induced by the gradient of spins that is perpendicular to s. The spin superuid velocity v(spin,F) satises the following relations: v s v
(spin,F) (spin,F)

= s v(mass,F) + v(mass,F) s = v(mass,F) , = s v(mass,F) + = 0,

2M

( s

s) ,

(277) (278)

v s v

(spin,F)

2M

( s s )

s ,

(279) (280)

(spin,F)

where we have used the Mermin-Ho relation (267) to derive Eq. (279). 6.2.2. Polar phase We next consider the order parameter of a spin-1 polar BEC, where 0 = (0, 1, 0)T . A general order parameter of the polar state is given by ei sin 2 polar = nei cos . (281) i e sin 2 The fact that polar does not depend on reects the SO(2) symmetry around the quantization axis d = (sin cos , sin sin , cos ); exp(if d)polar = polar . Substituting Eq. (281) into Eq. (259), we obtain v(mass,P) = M . (282)

Thus, unlike the case of the ferromagnetic phase, the circulation is quantized; however, the unit of quantization is half the usual h/M. To understand this, let us consider a loop that encircles a 64

vortex with a xed radius. Then, each point on the loop is specied by the azimuthal angle . We note that the single-valuedness of the order parameter (281) is met if we take, for example, = = nw /2 and = /2. Then, the order parameter at r is given by 1 n 0 , (283) halfvortex = in 2 e w where nw is an integer. It follows then that the circulation is quantized in units of h/(2M) rather than the usual h/M: v(mass,P) d = h nw . 2M (284)

The underlying physics for the half quantum number is the Z2 symmetry of the order parameter of the polar phase; Eq. (281) is invariant under gauge transformation ( + ) combined with a spin rotation around an axis perpendicular to d; d d. More generally, the singlevaluedness of the order parameter is satised if changes by nw along the closed loop, where nw is even if d( = 2) = d( = 0), and odd if d( = 2) = d( = 0). Thus, the polar phase of a spin-1 BEC can host a half-quantum vortex for the case of nw = 1 [101], which is also referred to as an Alice vortex or Alice string [102]. The dynamic creation of half-quantum vortices was discussed in Ref. [103]. As in the case of the polar-core vortex in the ferromagnetic phase, the atomic density can be nite at the vortex core of the half-quantum vortex by lling the ferromagnetic state [(1, 0, 0)T for the case of Eq. (283)] at the core.

(a)
^ d C2

(b)

0
A B

Figure 17: (a) Order parameter for the polar state, which is invariant under rotation about the C2 axis together with the gauge transformation. (b) Half-quantum vortex of the polar phase of a spin-1 BEC. The order parameter rotates through an angle about the C2 axis dened in (a) as one circumnavigates the vortex. To satisfy the single-valuedness of the order parameter, this round trip must be accompanied by a phase change in (phase dierence between A and B), resulting in a half-quantized vortex.

To calculate the spin current, we again use the Cartesian coordinate in which the polar order parameter is given by polar = ei d. Substituting polar in Eq. (260), we obtain v
(spin,P)

d d .

(285)

65

 If d is restricted in a two-dimensional plane perpendicular to a unit vector a (in spin space), the (spin,P) circulation of a v is quantized: a v
(spin,P)

d =

h nw (nw : integer). 2M

(286)

Here, the unit of the quantization is h/2M due to the spin-gauge Z2 symmetry. The rotation and (spin,P) divergence of the v are calculated to be v
(spin,P)

= =

M M

(d )

(d ),

(287) (288)

(spin,P)

(d

d) .

Note that at zero magnetic eld, the continuity equation (257) becomes F (spin,P) = nv t = = M M
i=x,y,z d i (ni d )

(i n)d (i d ) + nd (i i d )

i=x,y,z

F = t M

d i (ni d).
i=x,y,z

(289)

This result implies that i i (ni d) has to be either zero or parallel to d in a stationary d texture. Otherwise, the local magnetization would develop with time and the order parameter would no longer belong to a manifold for the polar state. The d-eld around the half-quantum vortex of = (cos nw /2, sin nw /2, 0), which satises i i (ni d) = n2 d/(4r2 ) if n(r, ) is Eq. (283) is d w axisymmetric, i.e., e n = 0, where e is the unit vector in the azimuthal direction. Therefore, the half-quantum vortex can be a stationary state. 6.2.3. Stability of vortex states The stability of a vortex state can be studied by using the multicomponent GP equations. An important feature of the spinor BEC is that each of the components can accommodate different vortices (see, for example, Eq. (281)); however, this gives rise to relative phases between spinor components, and from the energetics point of view, certain conditions must be met. From Eq. (44), we nd that only the term c1 |F|2 /2 depends explicitly on the relative phases. Expressing the term in terms of components, we obtain c1 2 | f| 2 = c1 (|1 |2 |1 |2 )2 + 2|0 |2 (|1 |2 + |1 |2 ) 2 +4|2 1 1 | cos (1 + 1 20 ) , 0

(290)

where m arg(m ) is the phase of the m-th component. To minimize the energy, 1 + 1 20 must be n, where n must be even if c1 < 0 and odd if c1 > 0. Substituting m = m + m , where 66

 is the azimuthal angle, we must have 1 + 1 20 = n and 1 + 1 20 = 0. Assuming that the maximum vorticity is 1, we nd that the following three vortex states are allowed [104]: (1 , 0 , 1 ) = (1, 1, 1), (1, 0, 1), (1, 1/2, 0), (291)

where the last state is allowed only if 0 = 0, and therefore, it is described as (1, none, 0). The (1, 1, 1) state is the usual singly quantized vortex state with an empty vortex core, where the order parameter is given by ei (|1 |ei1 , |0 |ei0 , |1 |ei1 )T . (292)

The (1, 0, 1) state has a vortex in the m = 1 component and an anti-vortex in the m = 1 component with the vortex core lled by the m = 0 state, where the order parameter of this state is given by (|1 |ei( +) , |0 |, |1 |ei( +) )T . (293)

This is the polar-core vortex in the ferromagnetic phase or an integer spin vortex in the polar phase. The (1, none, 0) state corresponds to the Alice vortex [Eq. (283)] whose order parameter is given by (|1 |ei(1 +) , 0, |1 |ei1 )T , (294)

where the vortex core is lled by the m = 1 component. Except for the (1,1,1) state, the vortex core is lled by a non-vortex component which plays the role of a pinning potential, stabilizing the vortex state in the absence of an external potential [104]. 6.3. Spin-2 BEC The rotation matrix is U(, , ) = 2i(+) 4 C e i(+2) 3 2e C S 6e2i C 2 S 2 i(2) 3 2e CS 2i() 4 e S 2i 2 2 6e C S
1 4 (1 3 i 8e

2ei(2+)C 3 S 2ei(+)C 2 S 2
3 i 8e

2ei(2)CS 3 2ei()C 2 S 2
3 i 8e

sin 2

sin 2
3

+ 3 cos 2)

sin 2

2ei()C 2 S 2 2e
i(2)

CS

3 i e sin 2 8 2i 2 2

2ei(+)C 2 S 2 2ei(2+)C 3 S

6e C S

e2i() S 4 2ei(2)CS 3 2i 2 2 6e C S , 2ei(+2)C 3 S 2i(+) 4 e C (295)

where C cos and S sin . The spin-2 BEC has four mean-eld ground-state phases in 2 2 the absence of the magnetic eld: ferromagnetic, uniaxial nematic, biaxial nematic, and cyclic. The uniaxial and biaxial nematic phases are degenerate at the mean-eld level (75) [105]. The uniaxial and biaxial nematic phases are also referred to as the antiferromagnetic phase in Refs. [4, 5] and the polar phase in Ref. [11].

67

6.3.1. Ferromagnetic phase ferro The representative state of the ferromagnetic phase is 0 = (1, 0, 0, 0, 0)T . Substituting this and Eq. (295) in Eq. (253), we obtain e2i cos4 2 i 2e cos3 sin 2 2 i(2) ferro 6 cos2 sin2 . = ne (296) 2 2 i 2e cos sin3 2 2 e2i sin4 2 Similar to the spin-1 case [see Eq. (262)], the order parameter has the spin-gauge symmetry with now replaced by 2 because the magnitude of the spin is 2 rather than 1. The superuid velocity is similarly obtained as v(mass,F) = M [( 2) 2 cos ], (297)

whose circulation is not quantized as in the case of the spin-1 ferromagnetic BEC. Applying the discussions about the stability of the vortex states for the ferromagnetic spin-1 BEC, it can be shown that the 4nw + mw vortex (mw = 0, 1, 2, 3) is unstable against the decay to the mw vortex. The relationships for the spin superuid velocity corresponding to Eqs. (277)(280) change to the following ones: v s v
(spin,F)

= 2 s v(mass,F) + v(mass,F) s = 2 v(mass,F) , = 2 s v(mass,F) + = v(mass,F) =

2M

( s

s) ,

(298) (299)

(spin,F)

v s v

(spin,F)

(2 s s )

(300) (301)

(spin,F)

s ( s

s ).

6.3.2. Uniaxial nematic phase uniax The representative state of the spin-2 uniaxial nematic phase is 0 = (0, 0, 1, 0, 0)T and substituting this and Eq. (295) in Eq. (253), we obtain the general order parameter of this phase: e2i sin2 2ei sin cos 6n i 2 uniax 2 (302) = e 3 (3 cos 1) . 4 2ei sin cos e2i sin2 Similar to the case of the spin-1 polar phase, the spin degrees of freedom do not contribute to the superuid velocity that only depends on the spatial variation of the gauge angle: v(mass,uniax) = 68 M . (303)

This is a general property for the non-magnetized phase in spinor BECs. The shape of the uniaxial order parameter in the spin space is shown in Fig. 18 (a). As one can see from Fig. 18 (a), the uniaxial nematic phase has a SO(2) symmetry around the direction of d = (cos sin , sin sin , cos ). The uniaxial nematic phase also has Z2 symmetry as in the case of the spin-1 polar phase. However, in the present case, Z2 symmetry is independent of the gauge angle : uniax is invariant under a rotation around an axis perpendicular to d. Hence, although the uniaxial nematic phase can accommodate a vortex, as shown in Fig. 18 (b), the mass circulation for this vortex is zero. We call it a 0-1/2 vortex, where 0 and 1/2 indicate the gauge transformation and rotation angle in spin space around the vortex, respectively. An example of the order parameter around the 1/2-spin vortex is given as i e 0 6n spinvortex 2/3 , (304) = 4 0 i e where is an azimuthal angle around the vortex. The spin superuid velocity for the uniaxial nematic phase is calculated to be v
(spin,uniax)

3 M

d d .

(305)

If d is restricted in a plane perpendicular to a, the circulation of the spin superuid velocity (spin,uniax) is quantized in terms of a unit of 3h/M. a v Because the spin and gauge is completely decoupled in this phase, the mass circulation is quantized in terms of a unit of /M. An example of such a vortex is (0, 0, ei , 0, 0)T .

(a)
C2

U(1)

(b)

0
Figure 18: (a) Order parameter for the spin-2 uniaxial nematic phase, which has SO(2) symmetry corresponding to rotations about d. The uniaxial nematic phase is also invariant under a rotation about the C2 axis, i.e., this phase has a Z2 symmetry. (b) The order parameter conguration for the 0-1/2 vortex, around which the order parameter rotates by about the C2 axis in (a).

6.3.3. Biaxial nematic phase biax The representative state of the spin-2 biaxial nematic phase is 0 = (1/ 2, 0, 0, 0, 1/ 2)T and substituting this and Eq. (295) in Eq. (253), we obtain the general order parameter of this 69

phase: 2i e 1 1 sin2 cos 2 i cos sin 2 2 i e sin (cos cos 2 i sin 2) n i 2 3 e 2 sin cos 2 2 ei sin (cos cos 2 + i sin 2) 2i e 1 1 sin2 cos 2 + i cos sin 2 2 .

biax =

(306)

The superuid velocity only depends on the spatial variation of the gauge angle: v(mass,biax) = M . (307)

The prole of biax is depicted in Fig. 19 (a). When we rotate the order parameter by /2 about the C4 axis in Fig. 19 (a) as we circumnavigate a vortex, the order parameter changes its sign, as shown in Fig. 19 (b) (see, the phase dierence between A and B). Therefore, this vortex has a mass circulation of h/(2M). We call this vortex the 1/2-1/4 vortex, where 1/2 and 1/4 denote the gauge transformation and the spinrotation angle, respectively, around the vortex. If we rotate the order parameter by about the C2 axis around a vortex, then it goes back to the initial state, as shown in Fig. 19 (c). In this case, the vortex has no mass circulation and it is referred to as a 0-1/2 vortex. On the other hand, if we rotate the order parameter by about the C2 axis, the sign of the order parameter is reversed [Fig. 19 (d)]. This vortex has a mass circulation of h/(2M), and it is called a 1/2-1/2 vortex. Because the roations about C4 , C2 , and C2 are non-commutable, these vortices obey the non-Abelian algebra, which is discussed in Sec. 6.6. The vortices shown in Fig. 19 are the simplest examples, and in general, the order parameter can vary in a more complex manner. However, in order to meet the single-valuedness, the mass circulation in the biaxial nematic phase has to be quantized in terms of a unit of h/(2M). The spin superuid velocity is very complicated, and its circulation is not quantized. 6.3.4. Cyclic phase cyclic A representative state of the spin-2 cyclic phase is 0 = (1/2, 0, i/ 2, 0, 1/2). Figure 20 m cyclic (a) shows the prole of the order parameter of the cyclic phase = m Y2 m . One can clearly see that it has a three-fold symmetry about the (1, 1, 1) axis [C3 axis in Fig. 20 (a)]. This implies that the system has a one-third vortex [106]. The rotation matrix about the (1, 1, 1) axis through angle is given by ei e
i
f x +fy +fz 3

cyclic 0 .

(308)

For = 2/3, this reduces to e4i/3 ei 0 ; the single-valuedness of the order parameter is met by the gauge transformation of +2/3. Because this is one-third of the usual 2, the cyclic cyclic phase can possess a one-third vortex. Similarly, for = 4/3, Eq. (308) reduces to e2i/3 ei 0 ; the single-valuedness of the order parameter is met by the choice of = 4/3. Thus, the cyclic phase also has a two-third vortex. The cyclic phase can also accommodate the 1/2-spin vortex, around which the order parameter rotates by about the direction of one of the lobes [Fig. 20 (c)]. The vortices in the cyclic phase also obey the non-Abelian algebra (see Sec. 6.6). 70

cyclic

(a)

C4

C2 C'2

(b)

(c)

0
B

(d)

A B
Figure 19: (a) Order parameter for the spin-2 biaxial nematic phase, which has one symmetry axis of the fourth order (C4 ) and two axes of the second order (C2 and C2 ). (b)(d) The conguration of the order parameter around (b) a 1/2-1/4 vortex, (c) a 0-1/2 vortex, and (d) a 1/2-1/2 vortex.

6.4. Rotating spinor BEC The stationary state of a rotating BEC can be obtained by minimizing F = E Lz , where the rotation axis is assumed to be in the z-direction. In the case of a spin-1 BEC, we have 2 2 c0 2 c1 2 (309) F= dr + U(r) z m + n + |F| . m 2M 2 2 m The stationary state is obtained by requiring that F = m m , m (310)

where m is the chemical potential for the m-th component that is determined so as to conserve the total number of particles N = dr m |m |2 , total magnetization Mz = dr(|1 |2 |1 |2 ), and total projected angular momentum Lz = dr i z i , where z = i /. The vortex i lattice formation in the presence of an external rotation is discussed in Ref. [107]. The vortex phases of spin-2 BECs under rotation is discussed in Ref. [108]. 6.5. Spin-polarized dipolar BEC The dipole interaction is long-range and anisotropic, and it has signicant eects on the ground state of rotating BECs [109, 110, 111, 112, 113]. Let us assume that all the dipoles of the atoms are polarized in the same direction. Then, the interaction is described by 1 3 cos2 V(r) = c0 (r) + cdd , r3 71 (311)

(a) 2/3 4/3 0 (b)

C2 C3
y

(c)

A B
Figure 20: Order parameter of the cyclic phase (1, 0, i/ 2, 0, 1)T , where the blue, yellow, and green colors denote m cyclic ] = 0, 2/3, and 4/3, respectively. The order parameter has a two-fold symmetry about the x, y, and z arg[ m Y2 m axes and a three-fold symmetry about the (1, 1, 1) axes. (b) The 1/3-1/3 vortex, where the order parameter is rotated by 2/3 about the C3 axis from A to B in a clockwise direction, results in the gauge transformation of 2/3. (c) The 0-1/2 vortex, around which the order parameter is rotated by about the y-axis.

where the rst and second terms on the right-hand side are the s-wave contact interaction and the dipole interaction, respectively, and is the angle between the dipole moment and the relative coordinate r between two atoms. Thus, depending on the direction of the polarization, the dipole interaction can be either attractive or repulsive. This anisotropy of the interaction leads to the deformation of the vortex core and the structure of the vortex lattice. Yi and Pu [112] considered a situation in which a system in an axisymmetric harmonic trap is rotated about the z-direction, and the dipoles are polarized in the x-direction in the frame of reference corotating with the system, and found that the vortex core is deformed into an elliptic shape. This is because the dipoles undergo repulsive and attractive interaction in the x- and y-directions, respectively. Cooper et al. [109] and Zhang and Zhai [111] considered a situation in which the interparticle interaction is so weak that the lowest Landau level approximation holds, and found that with an increase in the strength of the dipole interaction, the vortex lattice undergoes structural changes from triangular to square and to stripe congurations. Such changes were not found in Ref. [112], and possible reasons for the discrepancy are discussed in Ref. [113]. In the aforementioned work, the dipole interaction is treated as a perturbation to the s-wave contact interaction. Very recently, the Rice group [26] used a shallow zero crossing in the wing of a Feshbach resonance of 7 Li in the |F = 1, mF = 1 state to decrease the scattering length to as less as 0.01aB . This opens up several new possibilities on the dipole-interaction-dominated BEC. An interesting question is how the degeneracy of a fast rotating BEC is lifted by a weak but genuine dipole interaction and what is the corrsponding many-body ground state.

72

6.6. Non-Abelian vortices The topological charge of a vortex determines how the order parameter changes as one circumnavigates the vortex. For example, if the phase of the order parameter of a U(1) vortex changes as ein , the topological charge of this vortex is said to be n. One important feature of a spinor BEC is that the order parameter has a geometrical shape in spin space, as illustrated in Figs. 1620. Take, for example, the polar phase. This phase has two generators ei f d and eiC2 , where the former describes the rotation about the symmetry axis and the latter describes the rotation about the perpendicular axis followed by the gauge transformation by [see Fig. 17(a)]. These two operations commute with each other; thus, vortices of the polar phase of a spin-1 BEC are Abelian. Similarly, the vortices in the ferromagnetic phase are also Abelian because there is only one symmetry axis set by the magnetization. Therefore, the spin-1 BEC can host only Abelian vortices. It is predicted [114] that the cyclic phase of a spin-2 BEC can host non-Abelian vortices. To understand this, we note that the order parameter of the cyclic phase is invariant under the following 12 elements the tetrahedral group T [106]: 1, I x = eifx , Iy = eify , Iz = eifz , of = e2i/3 e2i(fx +fy +fz )/3 3 , C 2 , I xC, IyC, IzC, I xC 2 , IyC 2 , and IzC 2 , where f is the component C of the spin-2 matrices ( = x, y, z) [see, Fig. 20]. Furthermore, these elements are classied into four conjugacy classes: (I) integer vortices: {1}; (II) 1/2 - spin vortices: {I x , Iy , Iz }; (III) 1/3 vortices: {C, I xC, IyC, IzC}; (IV) 2/3 vortices: {C 2 , I xC 2 , IyC 2 , IzC 2 }. Topological charges in the same conjugacy class transform into one another under the global gauge and/or spin rotation. The order parameter of a 1/2-spin vortex can be constructed from that of a non-vortex state nonvortex as 1/2vortex = eifz nonvortex , (312)

where is an azimuthal angle around a vortex line. The non-Abelian characteristics of the vortices manifest themselves most dramatically in the collision dynamics. In general, when two vortices collide, they reconnect themselves, pass through [Fig. 21(d)], or form a rung that bridges the two vortices [Fig. 21(b),(c)]. When two Abelian vortices collide, all these three cases are possible, and one of them occurs depending on the kinematic parameters and initial conditions. However, when two non-Abelian vortices collide, only a rung can be formed; however, reconnection and passing through are topologically forbidden because the corresponding generators do not commute with each other. In fact, the nonzero commutator of the two generators gives the generator of the rung vortex. Figure 22 illustrates a typical rung formation. When a core of 1/3 vortex in the cyclic phase is lled with the ferromagnetic state, which is possible for a certain parameter set, it is possible to observe such dynamics of vortex lines by using a phase-contrast imaging technique that can detect local magnetization.

73

(a)
B A

(b)
B AB

(c)
B

A BA-1 ABA-1

(d)
B A

BAB-1

ABA-1 A

ABA-1

Figure 21: Collision dynamics of two vortices. (a) Initial conguration, where A or B represents an operation that characterizes the corresponding vortex (a set of spin rotations and gauge transformations for the case of a cyclic BEC). The vortex on the right end, which is connected to B, is identied as ABA1 . The conguration in (a) is topologically equivalent to (b) and (c), where a rung is formed. If the vortices are a pair of vortex and anti-vortex, i.e., A = B1 , the rung in (b) disappears, giving rise to reconnection, whereas the rung in (c) corresponds to a doubly quantized vortex that costs a large kinetic energy. If A and B are commutative, passing through is also possible because the congurations of (a) and (d) will then be topologically equivalent. However, when A and B are not commutative, the collision always results in the formation of a rung.

(a)

(b)

(c)

Figure 22: Numerical simulation for the collision dynamics of non-Abelian vortices.

7. Topological Excitations When a system undergoes Bose-Einstein condensation, some symmetries of the original Hamiltonian are spontaneously broken. The classication of the symmetry breaking can be carried out systematically using a group-theoretic method. The symmetry of the order parameter determines the types of possible topological excitations. In this section, we present a brief overview of some basic notions and theoretical tools for investigating symmetries of the order parameter and topological excitations. 7.1. Symmetry classication of broken symmetry states 7.1.1. Order parameter manifold The order parameter for a spin- f BEC is described with a set of 2 f + 1 complex amplitudes = {m }, and the order parameter manifold M is denoted as M = C2 f +1 . In this section, we treat a normalized spinor = / with its order parameter manifold given by M = U(2 f + 1) = U(1) SU(2 f + 1) . Z2 f +1 (313)

Consider a group G of transformations that act on M while leaving the mean-eld energy functional invariant. We assume that the total spin and number of particles of the system are 74

conserved. Then, the energy functional is invariant under SO(3) rotations in spin space and U(1) gauge transformations. The full symmetry group G for a spinor BEC is therefore given by G = SO(3) F U(1) , (314)

where subscripts F and denote the spin and gauge, respectively. Any element g G can be represented as g = ei eifz eify eifz , where , , and are Euler angles [see Fig. 23 (a)] and f x,y,z are (2 f + 1) (2 f + 1) spin matrices. When an external magnetic eld is applied, e.g. in the z-direction, G reduces to GB = U(1)Fz U(1) (315)

because the system will remain invariant only under rotations about the z-axis apart from the gauge transformation. Any element g GB is represented as g = ei eifz . For each M, the isotropy group (or the little group) G is dened as the subgroup of G that leaves invariant: G = {g G | g = }. For example, spin-1 polar phase polar = (0, 1, 0)T is invariant under a spin rotation about the z-axis, eifz polar = polar , and therefore, G polar = U(1)Fz . Next, we dene the orbit G() of by G() = {g | g G}, i.e., G() is the set of all points obtained as a result of letting all transformations in G act on . Because elements of G do not change the mean-eld energy, the orbit G() constitutes a degenerate space in M. For the case of a spin-1 polar BEC, a general element of the orbit G( polar ) is given by ei sin 2 0 = ei eifz eify eifz 1 = ei cos . i e sin 0
2

g polar

(316)

The orbit is thus parametrized by and d (cos sin , sin sin , cos ) that species the (a three-dimensional unit vector) is a unit sphere which is quantization axis. The manifold of d referred to as S 2 , the two-sphere [see Fig. 23 (b)], whereas the manifold of is U(1). Therefore, we conclude that the order parameter manifold for the spin-1 polar state is given by Rpolar S 2 U(1) . F
(a) z (b) d

y x
Figure 23: (a) Euler rotation with Euler angles (, , ). (b) Two-sphere S 2 which is the manifold of a topological space of a three-dimensional unit vector d.

Mathematically, such a manifold is dened as a left coset of G . For a subgroup H G, the left cosets of H are dened as G/H = {gH | g G}. The coset space is identical to the orbit of 75

H, and it forms a group if H is a normal subgroup of G, i.e., gHg1 = H for g G. The coset space G/H is equivalent to G/H if and only if H and H are conjugate, i.e., H = gHg1 . On the other hand, one can prove that the isotropy groups of all points on the same orbit are conjugate, i.e., Gg = gG g1 for g G. Therefore, the coset space does not depend on the location in the orbit. We use H for a representation of the conjugate class of the isotropy groups gG g1 , and dene the order parameter manifold as R = G/H. (317)

Here, H represents the remaining symmetry of the ordered state, and R describes the broken symmetry, or a manifold of the degenerate space. We next discuss the discrete symmetry of the order parameter manifold. The spin-1 polar phase has spin-gauge coupled discrete symmetry. In fact, polar = (0, 1, 0)T is invariant under a -rotation about an axis perpendicular to the z-axis followed by the gauge transformation of ei . This can also be understood from the fact that Eq. (316) is invariant under the simultaneous transformations of (, d) ( + , d). A group for such an operation {1, ei eifx } is isomorphic i if x 2 to Z2 = {0, 1} because (e e ) = 1. Therefore, the correct isotropy group of polar is the dihedral group ( D )Fz , U(1)Fz (Z2 ) F, , and the order parameter manifold is given by Rpolar = SO(3) F U(1) U(1)Fz (Z2 ) F, S 2 U(1) F , (Z2 ) F, (318)

where U(1)Fz (Z2 ) F, implies that when the nontrivial element of Z2 acts on an element g U(1), g changes to g1 . In the case of a spin-1 ferromagnetic phase, the order parameter ferro = (1, 0, 0)T is invariant under a spin rotation about the z-axis followed by a gauge transformation of the same amount as the rotation angle; ei eifz ferro = ferro . This is a spin-gauge coupled U(1) symmetry denoted as H = U(1)Fz + , where the subscript Fz + is shown to indicate the nature of the coupling. A general form of the spin-1 ferromagnetic order parameter is given by i e cos2 1 2 1 = ei eifz eify eifz 0 = ei() 2 sin . i 2 0 e sin
2

g ferro

(319)

Note that the linear combination in Eq. (319) manifestly exhibits a spin-gauge symmetry, i.e., the equivalence between phase change and spin rotation. Individual congurations of ferro are completely specied by the entire range of Euler angles (, , ), and therefore, the order parameter manifold is given by Rferro = SO(3) F U(1) U(1)Fz + SO(3) F, . (320)

7.1.2. Symmetry of order parameter and stationary states There exists a close relationship between the symmetry of an order parameter and a stationary state of the free-energy functional. To illustrate this, let us consider a smooth real function f (x2 + y2 ) on manifold R2 . The set of operations that leave f invariant constitute the group G = SO(2) of rotations about the z-axis, where the origin x = y = 0 is the only point that has 76

SO(2) symmetry as an isotropy group, whereas any other point traces a circle, which is called an orbit, because every element of G acts on that point (see Fig. 24); evidently, the isotropy group of each point on a circle consists only of the identity element of G. The union of all orbits that have mutually conjugate isotropy groups constitutes what is referred to as a stratum. In the present example, all orbits of the concentric circles constitute one stratum, and the origin forms the other isolated stratum because the isotropy group of the origin, which is SO(2), is dierent from those of other neighboring points. Now, the fact that the origin is an extremum (maximum, minimum, or saddle point) of the function f (x2 + y2 ) suggests that an orbit that constitutes an isolated stratum gives a stationary orbit of a real smooth function, as illustrated in Fig. 24.
(a) (b)

Figure 24: (a) Orbits of f (x2 + y2 ) on a manifold R2 . The union of all concentric circles constitute one stratum, and the origin forms an isolated stratum. (b) The origin x = y = 0 is isolated in its stratum, and it is an extremum of any real smooth function f (x2 + y2 ).

Michel generalized this idea by substituting R2 with an arbitrary manifold M, and SO(2) with an arbitrary compact Lie group G [115]. Michel showed that if the orbit G() is isolated in its stratum S (), i.e., if there exists a neighborhood U of G() such that U S () = G(), then G() is a stationary orbit of a smooth real function on M that is invariant under G. Here, stratum S () is the union of all orbits that share the same type of isotropy group, that is, and belong to the same stratum if and only if G and G are conjugate. Michels theorem has been utilized to nd the ground states of superuid He-3 [116], pwave [117, 118] and d-wave [105] superconductors, and spinor BECs [119, 120]. In the case of a spin-1 BEC, we can uniquely determine the orbits of polar and ferromagnetic phases from their isotropy groups. For example, ferro = (1, 0, 0)T and polar = (0, 1, 0)T are the only solutions of ei eifz = and eifz = , respectively, indicating that stratum S ( ferro,polar ) consists of G( ferro,polar ) alone. Therefore, the ferromagnetic and polar states are always extrema of the mean-eld energy functional. The order parameter for such states does not depend on parameters that appear in the energy functional, and these states are called inert states. An inert state is robust against a change in the interaction parameters, implying that it is a stationary point of each term in the energy functional. In superuid He-3, A, A1 and B phases are inert states. It is possible, however, that the energy functional has other orbits of extrema that depend explicitly on the coecients involved in the functional. The corresponding states are called non-inert states. Michel showed that the gradient of a G-invariant function is tangential to the stratum, implying that if we choose one stratum and nd an orbit in it that minimizes the energy functional, then the orbit is stationary in the entire manifold. In contrast to inert states, the order parameters of non-inert states are determined by the competition among several terms in the energy functional. In the case of spin-1 spinor BEC, non-inert states appear in the presence of an external magnetic eld, where the entire symmetry is reduced to GB = U(1)Fz U(1) . The mean-eld energy 77

per particle for a spin-1 BEC in a uniform system is given by [] = where f =

mn m fmn n

1 1 c0 n + c1 n| f |2 p 2 2

m|m |2 + q
m m

m2 |m |2 ,

(321)

is the spin expectation value per particle. Consider an order parameter AF = (cos , 0, sin )T , (322)

where 0 < < and /2. Dierent values of lead to dierent orbits, whereas the relative phase between 1 and the global phase may vary in each orbit. The rotational symmetry about the magnetic eld is broken, and the isotropy group of AF is given by H AF = (C2 ) F, = {1, ei eifz } GB . Substituting AF in Eq. (321) and minimizing with respect to , we obtain the following stationary state: 1 AF = 2 1 p , 0, c1 n 1+ p c1 n
T

p < 1, c1 n

(323)

and the corresponding energy functional [ AF ] = 1 p2 c0 n + q. 2 4c1 n (324)

The AF state corresponds to the spin part of the A2 phase in superuid He-3 where the magnetic eld changes the population of up-spin and down-spin Cooper pairs. The energies for polar and ferromagnetic states are given by [ polar ] = 1 c0 n, 2 1 [ ferro ] = (c0 + c1 )n p + q. 2 (325) (326)

Comparing Eqs. (324), (325), and (326), we can obtain the phase diagram shown in Fig. 1. While symmetry considerations are helpful to nd stationary states, they elude those states that have no remaining symmetry. As an example, let us consider a broken-axisymmetry state [10] whose C2 symmetry axis is not parallel to the external eld. This state cannot be obtained from symmetry considerations because GB is completely broken. Nonetheless, we can obtain a state that has the same symmetry (as a subgroup of G) as the real ground state from the symmetry consideration. 7.1.3. Procedure to nd ground states According to the discussions in the preceding subsection, the ground state that has remaining symmetries can be found by the following procedure [121]: 1. List all subgroups H of G; 2. for each H, nd an order parameter invariant under h H; 3. if a non-inert state is obtained, then nd the parameters that appear in the order parameter so as to minimize the mean-eld energy; 4. nd the lowest-energy one among the obtained states. 78

Here, we present a list of the subgroups of SO(3) and show how to carry out step 2 of the above procedure. The subgroups of SO(3) are well known and they are classied into continuous and discrete groups, as listed below. Continuous symmetries The only continuous subgroup of SO(3) is U(1). Let us choose the z-axis as the symmetry axis. An innitesimal transformation h H can be expressed as a combination of spin rotation and gauge transformation: h = ei eifz z 1 + i ifz z , (327)

where z and are innitesimal real values. The order parameter that is invariant under h is an eigenstate of fz . Conversely, the eigenstate of fz with eigenvalue m has U(1) symmetry and it is invariant under a spin rotation eifz accompanied with gauge transformation eim . For a spin- f system, there are 2 f + 1 states that have the U(1) symmetry. We denote this symmetry as U(1)Fz +m , where the subscript is shown to explicitly indicate the symmetry between the spin and the gauge degrees of freedom. The combined symmetry for m 0 states is called continuous spin-gauge symmetry, and it relates spin textures to the superuid current, as discussed in Sec. 6. Discrete symmetries The elements of the point groups of SO(3) can be described with a combination of the following generators: Cn : the cyclic group of rotations about the z-axis through angle 2k/n with k = 1, , n 1. The group is isomorphic to Zn and it has generators {Cnz }. Dn : the dihedral group generated by Cnz and an additional rotation through about an orthogonal axis. The group is isomorphic to Zn Z2 and it has generators {Cnz , C2x }. T : the point group of the tetrahedron whose generators are given by {C2z , C3,x+y+z }. O : the point group of the octahedron whose generators are given by {C4z , C3,x+y+z , C2,x+y }. Y : the point group of the icosahedron composed of 12 vefold axes, 20 threefold axes, and 30 twofold axes. It is shown [119] that a state with icosahedron symmetry does not exist for f 4. Here, we specify the symmetry axis for convenience. The generator Cn denotes a 2/n rotation around the direction of . In the present case, Cn can be expressed in terms of the spin operator f as Cn exp i 2 f , n || (328)

which is described by a (2 f + 1) (2 f + 1) matrix. Combined with a gauge transformation, the state invariant under the transformation h = Cn ei can be obtained by solving the eigenvalue equation Cn = ei . (329)

If the eigenvalue is 1, the invariance of the eigenstate is satised by a spin rotation alone, and therefore, the gauge symmetry is completely broken. On the other hand, the eigenstate 79

with an eigenvalue ei 1 is invariant under a simultaneous discrete transformation in spin and gauge. Such a symmetry is called discrete spin-gauge symmetry. What we need to do is to nd simultaneous eigenstates of a set of generators for all discrete symmetry groups. The results for the spin-1 BEC are summarized in Table 5 and the structure of each order parameter is shown in Fig. 25 (a). phase F+ F P AF H U(1)Fz + U(1)Fz ( D )Fz , (C2 )Fz , generators {e e {e {e
i ifz

(1, 0, 0) (0, 0, 1) (0, 1, 0)

1 2 1 2 (c0 + c1 )n 1 2 (c0 + c1 )n 1 2 c0 n p2 1 2 c0 n 2c1 n

} }

p+q + p+q +q

i ifz

ifz i

, e C2x }
i

{e C2z }

1+

p c1 n , 0,

p c1 n

Table 5: Symmetry and order parameter structure for the spin-1 spinor BEC, where H is the isotropy group, is a representative order parameter, and is the corresponding energy of the system.

7.1.4. Symmetry and order parameter structure of spin-2 spinor BECs It is straightforward to carry out the procedure for spin-2 spinor BEC. The mean-eld energy for a spin-2 spinor BEC per particle is given by [] = 1 1 1 c0 n + c1 n| f |2 + c2 n|A|2 p 2 2 2 m|m |2 + q
m m

m2 |m |2 ,

(330)

1 where A = (22 2 21 1 + 0 )/ 5 is the spin-singlet pair amplitude. We rst consider stationary states that have isotropy groups H G. Among them, the state with H GB can survive as a stationary state in the presence of an external eld (p 0 and q 0). Continuous group U(1): There are ve states that have continuous symmetries: F2+ : (1, 0, 0, 0, 0)T , HF2+ = U(1)Fz +2 : {ei2 eifz }, F2+ = 2p + 4q + 2c1 n, }, F2 = 2p + 4q + 2c1 n, 1 F1+ : (0, 1, 0, 0, 0)T , HF1+ = U(1)Fz + : {ei eifz }, F1+ = p + q + c1 n, 2 1 F1 : (0, 0, 0, 1, 0)T , HF1 = U(1)Fz : {ei eifz }, F1 = p + q + c1 n, 2 1 c2 n, UN : (0, 0, 1, 0, 0)T , HUN = ( D )Fz : {eifz , eiC2x }, UN = 10
T

(331) (332) (333) (334) (335)

F2 : (0, 0, 0, 0, 1) , HF2 = U(1)Fz 2 : {e

i2 ifz

where we indicate the generators of the isotropy group in curly brackets { } and = c0 n/2. These states are stationary in the presence of an external magnetic eld, where the isotropy group HUN is reduced to U(1)Fz . 80

(a) spin 1
(i) F U(1)Sz (ii) P ei C2 U(1)Sz (iii) AF ei C2

phase

(b) spin 2
(i) F2 U(1)Sz2 (ii) F1 U(1)Sz (iii) UN C2 (iv) C U(1)Sz ei2/3 C3 C2

(v) BA ei C4 C2

(vi) D2 C2

C2

(vii) D2'

C2

C2 C2 C2 (viii) C4 ei C4 (ix) C3 ei2/3 C3 (x) C2 ei C2 (xi) C2' C2

Figure 25: Order parameters for (a) f = 1 and (b) f = 2 spinor BECs. Shown are the surface plots of the amplitude | m m Y f,m ( )|, where Y f,m ( ) is a rank- f spherical harmonics in spin space. The gray scale on the surface represents s s the phase. The continuous and discrete symmetry axes are indicated along with the gauge transformation factors.

Tetrahedron group T: We solve the simultaneous eigenstate of C2z and C3,x+y+z , whose matrix forms are given by 1 6 2i 1 2i 1 0 0 0 0 0 1 0 0 0 2 2i 0 2i 2 1 0 0 1 0 0 , C 6 0 = C2z = 2 0 6 . (336) 3,x+y+z 4 0 0 0 1 0 2 2i 0 2i 2 0 0 0 0 1 6 2i 1 1 2i The only solution is C = 1 (1, 0, i 2, 0, 1)T (up to the phase factor). Here, C satises 2 C2z C = C and C3,x+y+z C = ei2/3 C . Therefore, the generators for the tetrahedron group as a subgroup of G are C2z and ei2/3C3,x+y+z : 1 C : (1, 0, i 2, 0, 1)T , HC = T F, : {C2z , ei2/3C3,x+y+z }, C = 2q. (337) 2 This state is called a cyclic state. Cyclic states need not be stationary in the presence of an external eld because T GB . 81

 Dihedral groups Dn : In a manner similar to the tetrahedron group, we nd 1 c2 n BN : (1, 0, 0, 0, 1)T , HBN = (D4 )Fz , : {eiC4z , C2x }, BN = 4q + , (338) 10 2 T cos cos c2 n D2 : , 0, sin , 0, , , HD2 = (D2 ) F : {C2z , C2x }, D2 = 4q cos 2 + 10 2 2 (339) T 10q 10q 10q 1 , D2 : 1 , 0, i 2 1 + , 0, 1 2 c2 n c2 n c2 n HD2 = (D2 ) F : {C2z , C2x }, D2 = 2q 10q2 . c2 n (340)

There are two energy minima D2 and D2 in the order parameter manifold with an isotropy group D2 . When p = q = 0, UN, BN, and D2 states are all degenerate, and in D2 state can take any arbitrary real value, while D2 phase goes to the cyclic state. In the presence of an external eld, however, BN, D2 and D2 states need not be stationary because Dn GB . Cyclic groups Cn : All states that have cyclic symmetries are non-inert states. Here, we have chosen the order parameter to minimize the mean-eld energy. T 1 p/2 p/2 , C4 : 1 + , 0, 0, 0, 1 (c1 c2 /20)n (c1 c2 /20)n 2 HC4 = (C4 )Fz , : {eiC4z }, C4 = 4q + 1 C3+ : 3 pq 1+ , 0, 0, c1 n pq 2 ,0 c1 n 1 p2 c2 n , 10 2(c1 c2 /20)n
T

(341)

, (p q)2 2c1 n (342)

HC3+ = (C3 )Fz , : {ei2/3C3z }, C3+ = 2q 1 C3 : 0, 3 2+ p+q , 0, 0, c1 n 1 p+q c1 n

T

, (p + q)2 2c1 n
T

HC3 = (C3 )Fz , : {ei2/3C3z }, C3 = 2q 1 C2 : 0, 2 1+ p , 0, (c1 c2 /5)n 1

(343) , (344)

p ,0 (c1 c2 /5)n

HC2 = (C2 )Fz , : {eiC2z }, C2 = q +

1 p2 c2 n , 10 2(c1 c2 /5)n

1 C2 : (a, 0, b, 0, c)T , HC2 = (C2 )Fz : {C2z }. (345) 2 The order parameter and energy for C2 state are the same as those of C2 state in Table 4. The obtained results are the same as those listed in Table 4. The structure of each order parameter is shown in Fig. 25 (b). 82

7.2. Homotopy theory 7.2.1. Classication of topological excitations Bose-Einstein condensates can accommodate topological excitations such as vortices, monopoles, and Skyrmions. These topological excitations are diverse in their physical properties but have one thing in common; they can move freely in space and time without changing their characteristics that are distinguished by topological charges. The topological charges take on discrete values and have very distinct characteristics independently of the material properties. It is these material-independent universal characteristics and robustness to external perturbations that make topological excitations unique. The classication of topological excitations is best made by homotopy theory [122, 123]. This theory describes what types of topological excitations are allowed in what order-parameter manifold. For example, 2 (U(1)) = 0 implies that monopoles, which are characterized by the second homotopy group 2 , are not topologically stable in systems described by scalar order parameters. This theory also describes what happens if two defects coexist and how they coalesce or disintegrate. By way of introduction, we rst consider a BEC described by a scalar order parameter (r) = |(r)|ei(r) . To classify line defects such as vortices, we take a loop in the condensate and consider a map from every point r on the loop to the phase (r) of the order parameter : r (r) arg(r). (346)

If the loop encircles no vortex (e.g., loop A in Fig. 26), the image of the map covers only a part of the unit circle, as shown in the right-hand side of Fig. 26. If the loop encircles a vortex (e.g., loop B), the image covers the entire unit circle at least once. If it covers the circle n times, it is said that the winding number of the vortex is n. The crucial observation here is that the winding number does not change if the loop deforms or moves in space as long as it does not cross the vortex. This property can be used to classify loops according to their winding number.

vortex r A B S1 P

Figure 26: Mapping from a loop in real space onto the order-parameter manifold S 1 (unit circle) according to the correspondence (r) : r arg(r). This mapping denes the rst homotopy group 1 (S 1 ).

If two loops a and b are continuously transformable without crossing singularities of the order parameter, they are said to be homotopic to each other and written as a b . The homotopic relationship is an equivalent one in that it is symmetric (i.e., a a ), reexive (i.e., if a b , 83

then b a ), and transitive (i.e., if a b and b c , then a c ). By this equivalence relationship, all loops are classied into equivalent classes called homotopy classes: [ 1 ], [ 2 ], [ 3 ], , (347)

where i (i = 1, 2, 3, ) is an arbitrarily chosen loop from the homotopy class [ i ] because all loops belonging to the same class are equivalent and continuously deformable to each other. Mathematically, a loop is dened as a mapping from I = [0, 1] to a topological space such that (0) = (1) = x0 , where x0 is called a base point. If two loops a and b share the same base point x0 and they are continuously deformable to each other, they are said to be homotopic at x0 . If the base point is not shared, they are said to be freely homotopic. The constant loop c is dened as the one such that c(t) = x0 for t [0, 1]. The inverse of loop is dened as 1 (t) (1 t) for t [0, 1]. The product of two homotopy classes [ 1 ] and [ 2 ] is dened as [ 1] [ 2] = [
1

2 ],

(348)

where 1 2 denotes the product of two loops in which 1 is rst traversed and then 2 is traversed. Members of [ 1 2 ] that are homotopic to 1 2 need not return to the base point x0 en route to the terminal point (see the dashed curve in Fig. 27).
1

x0
Figure 27: The product of two loops 1 2 in which 1 is rst traversed and then 2 is traversed. Loops homotopic to 1 2 need not return to x0 en route to the terminal point, as shown for the dashed curve.

With the denition of the product (348), a set of homotopy classes form a group. In fact, they satisfy the associative law ([ 1 ] [ 2 ]) [ 3 ] = [ 1 ] ([ 2 ] [ 3 ]); (349)

those loops that are homotopic to the constant loop c constitute the identity element [c] such that [c] [ ] = [ ] [c] for any [ ]; nally, the inverse [
1

(350)

] of [ ] is the homotopy class that consists of inverse loops of [ ] so that [

1

][ ]=[ ][ 84

] = [c].

(351)

The group dened above is called the fundamental group or the rst homotopy group and it is dened as 1 (M, x0 ), where x0 is the base point of all loops. In many situations in physics, the choice of the base point is not important, and we shall omit it and write 1 (M, x0 ) simply as 1 (M). The fundamental group is said to be Abelian if any two elements commute (i.e., [m] [n] = [n] [m] for [m] and [n]) and non-Abelian if some elements do not commute. For the case of U(1) vortices, the fundamental group is Abelian and is simply addition. For example, if two singly quantized U(1) vortices coalesce, a doubly quantized vortex results. This can be described as [1] [1] = [2]. Thus, the fundamental group is the additive group of integers Z: 1 (S 1 ) Z. (352)

Point defects such as monopoles can be characterized by considering a sphere that covers the object (see Fig. 28). We consider a map from each point r on to a point m in the order parameter space M: m : M. (353)

We classify elements of such a map by regarding as equivalent any two elements that can transform into each other in a continuous manner. For the case of the monopole, we consider M to be a two-dimensional sphere or 2-sphere S 2 , as shown in the right-hand side of Fig. 28. In this case, if the map (353) wraps S 2 n times, we have 2 (S 2 ) [n].

r O m(r)

S2

Figure 28: Mapping from a sphere in real space onto the order-parameter manifold M according to the correspondence (r) : r M. This mapping denes the second homotopy group 2 (M). Here, we take M to be S 2 a two-dimensional unit sphere (2-sphere).

Skyrmions and knots are classied by the third homotopy group. We assume that the order parameter becomes uniform at spatial innity. Then, the three-dimensional space is compacted into a three-dimensional sphere S 3 . To help envisage it, imagine a two-dimensional plane and identify all innite points. Then, the two-dimensional plane is compacted into a two-dimensional sphere S 2 . By considering the map from S 3 to the order parameter space M and by identifying maps that can be continuously transformed into each other without crossing any singularity of the order parameter space as belonging to the same equivalent class, we can introduce elements of the third homotopy group 3 (M). Higher homotopy groups can be dened in a similar manner. Moreover, the zeroth homotopy 0 (M) classies the domain walls and gives the number of disconnected domains; in particular, 85

0 (M) = 0 implies that the order-parameter manifold is connected. Using homotopy theory, one can elicit insights into the topology of complicated structures, which is dicult to obtain intuitively. Here, we list some useful formulas of homotopy theory: Z if m = n 1; 0 if m n 1; m (354) n (S ) 0 if m = 1 and n 2; Z if n = 3 and n = 2 (Hopf charge); Z if n = m + 1 4 or n = m + 2 4, 2 where 0 implies that trivial congurations (i.e., those contractible to a point) exist and Z2 = {0, 1} is the two-element group. Let G be a Lie group and H be its subgroup. If G is connected (i.e., 0 (G) = 0) and simply connected (i.e., 1 (G) = 0), the following isomorphism holds: 1 (G/H) Similarly, if 2 (G) = 1 (G) = 0, then 2 (G/H) 1 (H). (356) 0 (H). (355)

As a corollary, when G = SU(2) and H = U(1), we obtain 2 (SU(2)/U(1)) 1 (U(1)) Z. (357)

Some order-parameter spaces R and their rst, second, and third homotopy groups are summarized in Table 6. The topological objects classied by homotopy groups are summarized in Table 7. For example, in the case of ferromagnetic BECs, 1 (SO(3)) Z2 implies that there exist two types of linear defects that are singular and nonsingular; 2 (SO(3)) 0 implies the absence of point defects and 2D Skyrmions; and 3 (SO(3)) Z implies the presence of a nonsingular soliton-like object such as a Shankar monopole (Skyrmion) [124, 125]. Here, we dene a Skyrmion as a nonsingular point object characterized by a 2 charge in two dimensions and by a 3 charge in three dimensions. In the polar phase of a spin-1 condensate, a 2 Skyrmion takes a conguration in which the director points in the positive z-direction at the center, ares out in the radial direction, and eventually points in the negative z-direction as one goes away from the center. On the other hand, a 3 Skyrmion is exemplied by a knot soliton, as discussed in Sec. 7.3.4. In contrast, a monopole is a singular point object in three dimensions (a singular point object in two dimensions is called a vortex). The charge of the monopole is characterized by a 2 charge; examples include a hedgehog (t Hooft-Polyakov monopole) in a polar condensate (see Sec. 7.3.2). Note, however, that the 2 monopole cannot be created in a ferromagnetic condensate because its order-parameter manifold is SO(3), which does not give a non-trivial 2 charge; in this case, the monopole must be attached by a Dirac string (see Sec. 7.3.2). 7.2.2. Relative homotopy groups To fully understand broken-symmetry systems, it is important to take into account not only the global boundary conditions but also various constraints imposed on the order-parameter eld. Relative homotopy groups provide a useful tool for characterizing eects when the orderparameter manifold changes its character on a surface or when it is constrained by boundary 86

scalar BEC Heisenberg spin nematics biaxial nematics ferromagnetic BEC spin-1 polar BEC cyclic BEC

R U(1) S2 RP2

S 2 /Z2 RP3 F,

1 Z 0 Z2 Q Z2 Z T

2 0 Z Z 0 0 Z 0

3 0 Z Z Z Z Z Z

SU(2)/Q SO(3) F,

[S 2 U(1) ]/(Z2 ) F, F [SO(3) F U(1) ]/T F,

Table 6: List of homotopy groups. Subscripts F and denote the manifolds for the spin and gauge degrees of freedom, respectively. Q and T respectively denote the quaternion group and the binary tetrahedral group, the latter being a subgroup of SU(2) R.

n 0 1 2 3

defects domain walls vortices monopoles

solitons dark solitons nonsingular domain walls 2D Skyrmions Skyrmions, knots

Table 7: Topological objects (defects and solitons) described by homotopy groups.

conditions. Relative homotopy groups can also be used to discuss properties of a defect that is itself regarded as a singularity in the order-parameter eld in ordinary homotopy theory. Finally, the relative homotopy group is useful to calculate the absolute homotopy group via the exact sequence of homomorphisms. As an illustrative example, let us again consider a ferromagnetic BEC. Suppose that the system is contained in a cylinder of radius r0 and impose a boundary condition that the spin points in the positive z-direction at r = 0 and outward or downward on the wall at r = r0 , where r is the radial coordinate. Then, the spin texture would look like cos2 (r) 2 i coreless (r, , z) = n e sin (r) . (358) 2 2i 2 (r) e sin 2 Because the direction of spin is s(r, , z) = (e x cos + ey sin ) sin (r) + ez cos (r), the above boundary conditions are met if we choose (0) = 0 and (r0 ) = /2 or (r0 ) = . The topological objects corresponding to (r0 ) = /2 and (r0 ) = are referred to as the Mermin-Ho vortex and Anderson-Toulouse vortex, respectively. It can be shown that the circulation at the wall is 1 for the Mermin-Ho vortex and 2 for the Anderson-Toulouse vortex. They are good examples of relative homotopy because both of them are not stable in free space and they can be stabilized only if the boundary conditions on the wall are imposed by some means. In general, the nth relative homotopy group n (R, R) consists of the elements of n (R) from are subtracted. We consider an n-sphere in R and let it expand into R by which those of n (R) continuous deformation. Such an expansion may be regarded as a one-to-one mapping of n (R) 87

 into a subgroup of n (R). Let Im(n (R) n (R)) be the image of this homomorphism. Then, by the denition given above, the relative homotopy n (R, R) is a quotient group of Im(n (R) n (R)) in n (R): n (R, R) n (R) . Im(n (R) n (R)) (359)

The relative rst homotopy group 1 (R, R) can cope with planar defects with nontrivial boundaries. A well-known example is a planar soliton (Maki domain wall) in the A phase of superuid helium-3, where the order-parameter manifold is R = (S 2 SO(3))/Z2 within a dipole healing length from the domain wall and R = SO(3) outside this layer due to dipole locking. The domain wall may be regarded as a linear defect as seen from the edge of the wall. We take the axis perpendicular to the domain wall as the z-axis. Then, the order-parameter eld denes a mapping from the z-axis into the contour in R whose end points belong to R. Because 1 (SO(3)) Z2 , such contours are also classied as Z2 , of which the nontrivial one corresponds to the Maki domain wall. The relative second homotopy group 1 (R, R) allows the stabilization of a point object subject to a boundary condition on a sphere enclosing it. Another example is the boojuma point defect that appears in the A phase of superuid helium-3 when it is contained in a hard-wall sphere. Because 2 (SO(3)) = 0, the bulk superuid A phase does not support a topologically stable point defect. However, because the l vector must be perpendicular to the surface, a point singularity for the l-vector appears on the surface that is connected to a nonsingular vortex texture in the bulk. 7.3. Examples 7.3.1. Line defects The line defects are characterized by the rst homotopy group (or the fundamental group) 1 (M). For the case of s-wave superconductors, liquid helium-4, and spin-polarized or spinless gaseous BECs, the order parameter is scalar and the order-parameter manifold is M = U(1). The fundamental group is therefore the additive group of integers: 1 (U(1)) Z. (360)

For the case of a spin-1 BEC, the ground state can be ferromagnetic or polar. The orderparameter manifold of the ferromagnetic phase is SO(3) and the fundamental group is 1 (SO(3)) = Z2 = {0, 1}, the two-element group. The spin conguration that corresponds to class 0 is continuously transformable to a uniform conguration, whereas the spin conguration that corresponds to class 1 describes a singly quantized vortex that is topologically stable. Because 1 + 1 = 0 in Z2 , the coalescence of two singly quantized vortices is homotopic to a uniform conguration. Conversely, there is a conguration called the Anderson-Toulouse vortex that is nonsingular at the origin but has circulation of two away from the origin. Next, let us consider the polar phase of a spin-1 BEC. The order-parameter manifold is Rpolar = [S 2 U(1) ]/(Z2 ) F, . As we have discussed in Sec. 7.1.1, the Z2 symmetry arises F from the fact that the order parameter given by Eq. (316) is invariant under the simultaneous transformation of (, d) ( + , d). Because G = S 2 U(1) is not simply connected, to calpolar 2 culate 1 (R ), we lift G to G = S R and use Eq. (355). Then, the points that are identical 88

 to (, d) in G are ( + n, d) for even n and ( + n, d) for odd n, and those points constitute an additive group of integers n. Hence, we have 1 (Rpolar ) Z F, . (361)

As in the case of a scalar BEC, vortices in the polar phase are classied by integers. However, due to the Z2 discrete symmetry, the minimum unit of circulation is one-half of the usual value of h/M, as discussed in Sec. 6.2.2. In a similar manner, we may expect that the uniaxial, biaxial, and cyclic phases in spin-2 spinor BECs can support various types of vortices due to the discrete symmetry, as discussed in Sec. 6. Some of them have fractional circulations, and some others have no mass circulation. The topological charge can be expressed algebraically as a functional of the order-parameter eld. The topological charge corresponding to 1 (S 1 ) can be expressed as a line integral of the velocity eld. 7.3.2. Point defects Point defects are characterized by the second homotopy group. Because 2 (U(1)) = 0, scalar BECs have no topologically stable point defects. When the order-parameter manifold is S 2 , it can host monopoles because 2 (S 2 ) Z. Here, we consider the situation shown in Fig. 28, where m(r) is a mapping from S 2 to S 2 . The topological charge N2 of the point defect can be calculated as follows. By denition, N2 gives the number of times the vector m(r) shown in Fig. 28 wraps S 2 . Expressing the components of this vector in spherical coordinate as m x = sin cos , my = sin sin , and mz = cos , we can calculate N2 as N2 = 1 4 d

(, ) , (, )

(362)

where d sin dd, with and are the spherical coordinates of r, and the last term on the right-hand side is the Jacobian of the transformation of the coordinates. The fact that the integrand is the Jacobian explicitly indicates that N2 gives the degree of mapping from S 2 in real space to S 2 in order-parameter space. The right-hand side of Eq. (362) can be directly expressed in terms of m as N2 = where j= 1 2
i jk mi (

1 4

ddm

m m 1 = 4

j dS,

(363)

m j mk ).

(364)

We can use Gauss law to rewrite the right-hand side of Eq. (363) as a volume integral: N2 = where nm (r) = 1 j(r) 4 89 (366) nm dr, (365)

gives the density distribution of point singularities. t-Hooft-Polyakov monopole (hedgehog) Next, we consider the polar phase. The second homotopy group is given by 2 (Rpolar ) = Z, and therefore, it can accommodate point defects. To investigate their nature, we write the order parameter in the following form: polar = d x + id n i y e 2dz . 2 d x + idy

(367)

By setting = 0, we obtain the spherical monopole called t-Hooft-Polyakov monopole or hedgehog: r d(r) = . |r| (368)

It follows from Eq. (363) that the topological charge of the hedgehog is N2 = 1. Moreover, j dened in Eq. (364) is related to the circulation of spin currents given by Eq. (287) as 2 (spin,P) d v = j, M (369)

which implies that the surface integral of the spin circulation is quantized in terms of 4h/M; 4h (spin,P) d v dS = N2 . M (370)

Alice ring The monopole in the spin-1 polar phase is energetically unstable against deformation into an Alice ring [126]. The Alice ring is a combined object of two defects characterized by 1 and 2 (see Fig. 29). Viewed far from the origin, it appears to be a point defect (N2 = 1); however, along the ring C on which the order parameter is singular, it appears to be a continuous distribution of line defects (N1 = 1). This topological object is called the Alice ring, and it has been predicted to be realized in an optically trapped spin-1 23 Na BEC [126]. Dirac monopole The magnetic monopole was originally envisaged as a magnetic analogue of the quantized electric charge: B = 4g(r), (371)

where g denotes the strength of the magnetic monopole. The solution of this equation is easily found to be B = gr/r3 and the corresponding vector potential is given by A= g g(1 + cos ) (y, x, 0) = e , r(r z) r sin 90 (372)

Figure 29: Alice ring comprised of continuously distributed half-quantum vortices along a contour C. Far from the origin, it appears to be a monopole.

where (r, , ) are the polar coordinates and e is the unit vector in the -direction. The vector potential (372) reproduces the magnetic eld (371), except on the positive z-axis along which the magnetic eld exhibits a singularity called the Dirac string: rot A = g r 4g(x)(y)(z)ez , r3 (373)

where ez is the unit vector along the z-direction. The Dirac monopole can be created in the ferromagnetic phase of a spin-1 BEC [127]. Substituting , , and in Eq. (319), we obtain 2i cos2 2 e 1 i (374) Dirac Dirac = e sin 2 n 2 sin 2 . The corresponding superuid velocity v(mass) = ( /M)Im( ) [dened by Eq. (259)] is the same as the vector potential (372) of the Dirac monopole with the identication g = /M. It is interesting to note that the order parameter (374) exhibits a doubly quantized vortex only along the positive z-axis. One can say that because the rst homotopy group of SO(3) is Z2 , which allows only a singly quantized vortex called a polar-core vortex as a stable topological object, the doubly quantized vortex must terminate at the magnetic monopole. In fact, the singular Dirac monopole can deform continuously to a nonsingular spin texture if we take the following parametrization: 2i e cos2 2 i = e sin , (r, , z) = (1 t) + t. 2 2 sin 2

Dirac

(375)

We can see that as t increases from 0 to 1, the order parameter continuously changes from the Dirac monopole at t = 0 to a nonsingular texture = (0, 0, 1)T at t = 1. 7.3.3. Action of one type of defect on another When two dierent types of topological objects coexist, there exists an action of one object on the other as they move relative to each other. As an example, consider a spin-1 polar phase in which two monopoles with charge 1 merge in the presence of a half-quantum vortex. As a 91

monopole turns around the vortex, it changes sign. Thus, if they merge along line 1, the orderparameter eld must deform, as shown in Fig. 30 (b), and the combined object has the topological charge of two. On the other hand, when they merge along line 2, the order-parameter manifold can deform into the one shown in Fig. 30 (c), and the topological charge is zero, indicating that the two monopoles annihilate in pairs. This example illustrates that the charge of the monopole changes its sign as it moves around the half-quantum vortex along curve 2.
half-quantum vortex

(a)

(b)

(c)

1 A 2 B

Figure 30: (a) Coalescence of two monopoles with charge one in the presence of a half-quantum vortex. (b) If they merge along the dashed curve 1 shown in Fig. (a), the topological charge of the combined object is two. (c) If they merge along the dashed curve 2, the topological charge of the combined object is zero.

7.3.4. Skyrmions Shankar Skyrmion The third homotopy group characterizes topological objects called Skyrmions that extend over the entire three-dimensional space. A prime example of this is the so-called Shankar monopole [124, 125], although it is actually not a monopole but a Skyrmion. The topology of the Shankar Skyrmion is 3 (SO(3)) Z, which is supported by the ferromagnetic phase of a spin-1 BEC. The order parameter of the ferromagnetic BEC is characterized by the direction of the local spin s and the superuid phase that can be parametrized in terms of two unit vectors such as t and u that form a triad with s: s = t u. Concrete forms of s and t are introduced in Sec. 6.2.1. The class one element of 3 (SO(3)) Z may be realized by rotating the triad at position r about the direction [125, 124, 128, 129] r = f (r)N3 , N3 Z (376) r through an angle f (r)N3 , where f (0) = 2 and f () = 0. It follows from the last condition that the triad is uniform in spatial innity, and thus, the three-dimensional space is compacted to S 3 . The case of n = 1 is shown in Fig. 31. The topological charge of the third homotopy group can be introduced in a manner similar to that of the second homotopy group. Now, the mapped vector has a four-component vector that is normalized to unity. One such representation is r f (r) f (r) sin , n4 = cos . r 2 2 The invariant of such a mapping is again given by the Jacobian of the transformations: n= N3 = 1 122 dr
i jk n i n j n k n ,

(377)

(378)

92

Figure 31: Shankar monopole with charge one.

where i jk and are the completely antisymmetric tensors of the third and fourth orders, respectively, and the Roman letters run over x, y, z and the Greek letters run over 1, 2, 3, 4. Knot soliton The third homotopy group also classies knot solitons. Knots dier from other topological excitations such as vortices, monopoles, and Shankar monopoles in that they are classied by a linking number while others are classied by winding numbers. Knots are characterized by mappings from a three-dimensional sphere S 3 to S 2 . As in the case of Shankar monopoles, the S 3 domain is prepared by imposing a boundary condition that the order parameter takes on the same value in every direction at spatial innity. Here, we consider the spin-1 polar phase [130]. 2 The order parameter manifold of the polar phase is given by Rpolar (S S U(1) )/(Z2 )S, , of which neither U(1) nor Z2 symmetry contributes to the homotopy groups in spaces higher than one dimension; hence, we nd that 3 (Rpolar ) 3 (S 2 ) Z. The associated integer topological charge Q is known as the Hopf charge, and it is given by Q= 1 42 d3 x
i jk Fi j Ak ,

(379)

where Fi j = i A j j Ai = d (i d j d) [131]. Note that the domain (r) is three-dimensional, while the target space ( d) is two-dimensional. Consequently, the preimage of a point on the target S 2 constitutes a closed loop in S 3 , and the Hopf charge is interpreted as the linking number of these loops: if the d eld has Hopf charge Q, two loops corresponding to the preimages of any two distinct points on the target S 2 will be linked Q times [see Fig. 32 (a)]. Figure 32 (b) shows 93

 an example of the d eld of a polar BEC with Hopf charge 11 . Knots can be created by manipulating an external magnetic eld. In the presence of an external magnetic eld, the linear Zeeman eect causes the Larmor precession of d, whereas d tends to become parallel to the magnetic eld because of the quadratic Zeeman eect. Suppose that we prepare an optically trapped BEC in the m = 0 state, i.e., d = (0, 0, 1)T , by applying a uniform magnetic eld in the z-direction. Then, we suddenly turn o the uniform eld and switch on a quadrupole eld. Because of the linear Zeeman eect, d starts rotating around the local magnetic eld, and therefore, the d eld winds as a function of t, resulting in the formation of knots. Figure 33 shows the creation dynamics of knots in an optical trap subject to the quadrupole eld, where the upper panels show the snapshots of the preimages of d = and d = z and z the bottom panels show cross sections of the density for m = 1 (bottom) components on the xy plane. The density pattern in m = 1 components is characteristic of knots; a double-ring pattern appears corresponding to one knot. As the d eld winds in the dynamics, the number of rings increases. This prediction can be tested by the Stern-Gerlach experiment. Topological objects are conjectured to play important roles in the formation of our universe. Among them, knot solitons have recently attracted the interest of cosmologists since Faddeev and Niemi suggested that knots might exist as stable solitons in a three-dimensional classical eld theory [131].

(a)

(b)

d
S2

preimage

Figure 32: (a) Preimages of two distinct points on S 2 form a link. (b) Spin conguration of a knot with Hopf charge 1 in a polar BEC, where arrows inicate the d eld of the polar phase. The solid and dashed lines trace the point where d points to x and z, respectively, and form a link.

8. Many-Body Theory In spinor BECs, the mean-eld energy is much larger than the spin-exchange interaction. Therefore, we rst determine the spatial dependence of the condensate wave function independently of the spin degrees of freedom, and then, we consider the spin state by assuming that the obtained spatial wave function is shared by all spin states. This approximation is called the
1 Technically speaking, the conguration in Fig. 32 (b) is an unknot of a pair of rings with linking number 1, because the preimage of one point on S 2 forms a simple unknotted ring.

94

(a) 0.5 TL

(b) 1.1 TL
d = (1, 0, 0)T

(c) 2.2 TL

(d) 3.3 TL

d = (0, 0, 1)T

Figure 33: Dynamics of the creation of knots in a spherical optical trap under a quadrupole magnetic eld. Snapshots of the preimages of d = (0, 0, 1)T and d = (1, 0, 0)T (top), and cross sections of the density for m = 1 components on the xy plane (bottom).

single-mode approximation. In this section, we examine the many-body spin states by using this approximation. Suppose that the coordinate part of the ground-state wave function 0 (r) is independent of the spin state. Then, 0 (r) is the lowest-energy solution of the GP equation 2 + Utrap + c0 (N 1)|0 |2 0 = 0 , 2M
2

(380)

where Utrap is the trapping potential. Let 0 and 1 be the rst- and second-lowest eigenvalues of Eq. (380), respectively. The criterion for the validity of the single-mode assumption is that the rst excited energy 1 0 is much larger than other energy scales: 1 0 |p|, |c1 |N/V e , |c2 |N/V e , (381)

where V e ( dr|0 |4 )1 is the eective volume of the system. When this condition (381) is satised, the eld operator m (m = f, f 1, , f ) may be approximated as m (r) am 0 (r), where am is the annihilation operator of the bosons that have magnetic quantum number m and coordinate wave function 0 . 8.1. Many-body states of spin-1 BECs 8.1.1. Eigenspectrum and eigenstates As explained above, the eld operator of a spin-1 BEC in the single-mode approximation can be expressed as [85, 4, 86] m (r) = am 0 (r) (m = 1, 0, 1). 95 (382)

For the sake of simplicity, let us consider the case of zero magnetic eld (i.e., p = q = 0). Substituting Eq. (382) in Eq. (24) and replacing m a am and m,n a a an am with N and N(N m m n 1), respectively, we obtain H=N where
1

dr 0

2M

+ U(r) +

c0 (N 1)|0 |2 0 + c1 : F2 :, 2

(383)

F=
m,n=1

fmn a an m

(384)

and c1 (c1 /2) dr|0 |4 ; 0 is determined by Eq. (381) subject to the normalization condition |0 |2 dr = 1. Substituting the solution of Eq. (381) back into (383), we obtain H = N c0 N(N 1) + c1 : F2 :, where c0 (c0 /2) dr|0 |4 . From Eq. (16), we nd that the total spin operator F satises : F2 := N(N 1) 3 S S , where 1 S ( 2 2 1 a1 ) a a0 3 is the spin-singlet pair operator. We use Eq. (387) to rewrite Eq. (386) as H = N c0 N(N 1) + c1 N(N 1) 3S S . (389) (388) (387) (386) (385)

Because N and S S commute, the eigenvalue problem reduces to nding their simultaneous eigenstates. Let |0 be the state in which there is no spin-singlet pair, i.e., S |0 = 0. Using the commutation relation 2 S , (S )k = k (2N + 2k + 1)(S )k1 , (390) 3 we obtain 2 S S (S )k |0 = k (2N 2k + 1)(S )k |0 . 3 We then dene the k-pair state |k : |k (S )k |0 . 0|S k (S )k |0 (392) (391)

It follows from the commutation relation (390) that 2 S S |k = k (2N 2k + 1)|k . 3 96 (393)

8.1.2. Fragmentation Next, we consider a state |k, d in which 2k atoms form spin-singlet pairs and the remaining N 2k atoms reside in the m = 0 state along the d direction (|d| = 1). Dening ad = d x a x + dy ay + dz az d x + idy d x idy a1 + dz a0 + a1 , 2 2

(394)

the corresponding eigenstate and eigenenergy are given by ad |k, d = Zm 2 ( )N2k (S )k |0 , Ek = N c0 N(N 1) + c1 [N(N 1)2k(2N 2k + 1)],
1

(395) (396)

where Zm is the normalization constant. In the absence of an external magnetic eld, all magnetic sublevels are degenerate. If we take the quantization axis along the z-direction, we can specify the eigenstate in terms of the total spin F, magnetic quantum number Fz , and number of spin-singlet pairs k: |k, F, Fz = Z 2 (S )k (F )FFz ( )F |vac , a1
1

(397)

where F = N 2k, and F

m,n

(f )mn a an = m

2( 0 a1 + a a0 ) a 1

(398)

is the lowering operator of the magnetic quantum number. Next, we consider a situation in which all F atoms points in the z-direction in addition to the k spin-singlet pairs, i.e., Fz = F in Eq. (397), and seek the number of the atoms that are present in the m = 0 state. We introduce the basis state |n1 , n0 , n1 in which the magnetic sublevels m = 1, 0, 1 are occupied by n1 , n0 , and n1 atoms, respectively. Then, |vac = |0, 0, 0 and a a0 0 = = vac| 1 S k a a0 (S )k ( )F |vac aF 0 a1 F k k F vac| S (S ) ( ) |vac a a
1 1

F, 0, 0|S k a a0 (S )k |F, 0, 0 0 . F, 0, 0|S k (S )k |F, 0, 0

(399)

Substituting a a0 = a0 a 1 and using the fact that a commutes with S , we have 0 0 0 a a0 = 0 We use Eq. (390) to rewrite S (S )k = = [S , (S )k ] + (S )k S 2 k (2N + 2k + 1)(S )k1 + (S )k S . 3 97 F, 1, 0|S k (S )k |F, 1, 0 1. F, 0, 0|S k (S )k |F, 0, 0 (400)

(401)

 Noting that S |F, 1, 0 = S |F, 0, 0 = 0 and N|F, 1, 0 = (F + 1)|F, 1, 0 , we have a a0 0 = = = (2F + 2k + 3) F, 1, 0|S k1 (S )k1 |F, 1, 0 1 (2F + 2k + 1) F, 0, 0|S k1 (S )k1 |F, 0, 0 (2F + 2k + 3)(2F + 2k + 1) (2F + 5) 1 (2F + 2k + 1)(2F + 2k 1) (2F + 3) 2k NF = . 2F + 3 2F + 3
1 m=1

Because a a1 a a1 = F and 1 1 n1 n0 n1

a am = N, we obtain m (402) (403) (404)

a a1 = 1

NF + F 2 + N + 2F , 2F + 3 NF a a0 = 0 , 2F + 3 (N F)(F + 1) a a1 = 1 . 2F + 3

When all atoms form spin-singlet pairs, i.e., F = 0, Eqs. (402)(404) show that all magnetic sublevels are equally populated: a a1 = a a0 = a a1 = 1 0 1 N . 3 (405)

That is, the condensate is fragmented in the sense that more than one single-particle state is macroscopically occupied [132, 4, 86, 133]. This is in sharp contrast with the mean-eld result that predicts that for c1 > 0, the ground state is either polar ( 0 = N) or antiferromagnetic n ( 1 = n1 = N ). This discrepancy originates from the mean-eld assumption that there exists one n 2 and only one BEC. However, such an assumption cannot be justied when the system possesses certain exact symmetries such as rotational symmetry in the present case. Equation (403) shows that n0 decreases rapidly with an increase in F. In fact, when F = O( N), we have n0 = O( N). This implies that although mean-eld results break down at zero magnetic eld, the validity of mean-eld theory quickly recovers as the magnetic eld increases. This aords an example of why fragmented BECs are fragile against symmetry-breaking perturbations. Finally, we note an interesting relationship between a mean-eld state and the many-body spin-singlet state. A mean-eld state is one in which all particles occupy the same single-particle state: ( )N ad |vac , N! (406)

where a d is dened in Eq. (394). Substituting d x = sin cos , dy = sin sin , and dz = cos in Eq. (394), we obtain N 1 a1 sin ei + a cos + a1 sin ei |vac . 0 |, = N! 2 2 98 (407)

We calculate the equal-weighted superposition state of |, over all solid angles d = sin dd. |sym = =
2 1 sin d d |, 4 0 0 0 if N is odd; 2 N 1 (N+1) N! (a0 2a1 a1 ) 2 |vac if N is even.

(408)

This result shows that the spin-singlet state is the superposition state of a ferromagnetic state |, over all angles of magnetization. Conversely, the mean-eld state |, may be interpreted as a broken symmetry state of the spin-singlet state with respect to the direction of quantization (, ). 8.2. Many-body states of spin-2 BECs Next, we discuss the many-body states of spin-2 BECs in the single-mode approximation. Substituting m (r) = am (r) in Eq. (30), we obtain the spin-dependent part of the Hamiltonian as c1 2c2 H= : F 2 : + e S+ S pFz , e 2V 5V where F
mn

(409)

fmn a an m

(410)

is the spin vector operator, Fz

m

m am am

(411)

is its z-component, and 1 S = (S + ) 2 (1)m am am

m

(412)

are the creation (S+ ) and annihilation (S ) operators of a spin-singlet pair. + , when applied to a vacuum, creates a pair of bosons in the spin-singlet While the operator S state, the pair should not be regarded as a single composite boson because S+ does not satisfy together with Sz (2N + 5)/4 the commutation relations of bosons. In fact, the operators S satisfy the S U(1, 1) commutation relations: [Sz , S ] = S , [S+ , S ] = 2Sz . (413)

Here, the minus sign in the last equation is the only dierence from the usual spin commutation relations. This dierence, however, leads to some important consequences. In particular, the Casimir operator S2 , which commutes with S and Sz , does not take the usual form of the squared sum of spin components but instead takes the form S2 S+ S + S2 Sz . z 99 (414)

8.2.1. Eigenspectrum and eigenstates The eigenstates are classied according to the eigenvalues of the Casimir operator (414). Because S+ S = S2 Sz S2 is positive semidenite, Sz must have a minimum value Smin . z z z (2N + 5)/4, Smin can be expressed in terms of a non-negative integer N0 as Recalling that S z Smin = (2N0 + 5)/4. Let | be the eigenvector corresponding to this minimum eigenvalue; then, z S | = 0. It follows that S2 | = Smin (Smin 1)| ; z z (415)

hence, the eigenvalue of S2 is given by = S (S 1) with S = (2N0 + 5)/4. + on | increases the eigenvalue of Sz by 1 and that of N increases by The operation of S 2, as observed from the commutation relations (413). Therefore, the allowed combinations of eigenvalues S (S 1) and S z for S2 and Sz , respectively, are given by S = (2N0 + 5)/4 (N0 = 0, 1, 2, . . .) and S z = S + NS (NS = 0, 1, 2, . . .), where N0 and NS satisfy N = 2NS + N0 , (417) (418) (416)

and we may therefore interpret NS as the number of spin-singlet pairs and N0 the number of remaining bosons. To nd the exact energy eigenvalues of Hamiltonian (409), we rst note that S and Sz commute with F : [S , F ] = 0, [Sz , F ] = 0. (419)

The energy eigenstates can be classied according to quantum numbers N0 and NS , total spin F, and magnetic quantum number Fz . We denote the eigenstates as |N0 , NS , F, Fz ; , where = 1, 2, . . . , gN0 ,F labels orthonormal degenerate states with gN0 ,F provided in Ref. [5]. The rst commutator in Eq. (413) implies that S plays the role of changing the eigenvalue of Sz by 1 by creating or annihilating a spin-singlet pair. The eigensystem can be constructed by rst preparing the eigenstates that do not involve spin-singlet pairs, |N0 , 0, F, Fz ; , N and then, by generating new eigenstates through successive operations of S + S on it: |N0 , NS , F, Fz ; = where | N S + S |N0 , 0, F, Fz ; N S + S |N0 , 0, F, Fz ; , (421) (420)

| , and the orthnormality condition is N0 , NS , F, Fz ; |N0 , NS , F, Fz ; = , . (422)

To write down the eigenstates (421) explicitly, we need to nd the matrix elements of S + , of which only the nonvanishing ones are N0 , NS + 1, F, Fz ; |S + |N0 , NS , F, Fz ; = 100 (NS + 1)(NS + N0 + 5/2). (423)

Finally, the energy eigenvalue for the state |N0 , NS , F, Fz ; is given by E= c1 c2 [F(F + 1) 6N] + NS (N + N0 + 3) pFz , 2V e 5V e (424)

where the relationship 2NS + N0 = N is used. The total spin F can, in general, take integer values in the range 0 F 2N0 . However, there are some forbidden values [5]. That is, F = 1, 2, 5, 2N0 1 are not allowed when N0 = 3k (k Z), and F = 0, 1, 3, 2N0 1 are forbidden when N0 = 3k 1. We prove this at the end of this subsection. To gain a physical insight into the nature of the energy eigenstates |N0 , NS , F, Fz ; , it is useful to express them in terms of some building blocks. In fact, we can express them in terms of one-, two-, and three-boson creation operators. Let A(n) be the operator such that when applied f to the vacuum state, it creates n bosons that have a total spin F = f and magnetic quantum number Fz = f . While the choice of a complete set of operators for the building blocks is not unique, we choose the following ve operators for constructing the eigenstates: A(1) 2 A(2) 0 A(2) 2 A(3) 0 = a2 = 1 2 a a a S+ [( 0 )2 2 1 a1 + 2 2 a2 ] = 5 10 1 = [2 2 a 3( )2 ] a2 0 a1 14 1 = [ 2( )3 3 2 1 a a + 3 3( )2 a a0 a 0 1 a1 2 210 2 +3 3 2 ( 1 ) 6 2 2 a0 a2 ] a a a 1 = [( )3 6 2 a1 a0 + 2( )2 a ]. a a2 1 a1 20 (425) (426) (427)

(428) (429)

A(3) 3

Note that A(2) and A(3) vanish identically because of the Bose symmetry, and that the operators 1 1 (n) commute with F+ . Af We next consider a set B of unnormalized states: |n12 , n20 , n22 , n30 , n33 ( )n12 (A(2) )n20 (A(2) )n22 (A(3) )n30 (A(3) )n33 |vac , a2 0 2 0 3 (430) with n12 , n20 , n22 , n30 = 0, 1, 2, . . . , and n33 = 0, 1. The total number of bosons and the total spin of the state (430) are N = n12 + 2(n20 + n22 ) + 3(n30 + n33 ) and F = Fz = 2(n12 + n22 ) + 3n33 , respectively. For given N and F, n33 is uniquely determined by the parity of F: n33 = F mod 2. (431)

It can be shown [5] that B forms a nonorthogonal complete basis set of the subspace H(Fz =F) in which the magnetic quantum number Fz is equal to the total spin F. It can also be shown [5] that the energy eigenstates can be represented as a2 (F )F (A(2) )n20 P(NS =0) ( )n12 (A(2) )n22 (A(3) )n30 (A(3) )n33 |vac , 3 0 2 0 101 (432)

where n12 , n20 , n22 , n30 = 0, 1, 2, . . . , , n33 = 0, 1, and F = 0, 1, . . . , 2F. These parameters are related to {N0 , NS , F, Fz } as N0 NS F Fz = = = = n12 + 2n22 + 3n30 + 3n33 , n20 , 2n12 + 2n22 + 3n33 , F F, (433) (434) (435) (436)

and the corresponding eigenenergy is given by Eq. (424). Note that the states dened in (432) are unnormalized, and the states having the same energy (i.e., those belonging to the same set of parameter values {N0 , NS , F, Fz }) are nonorthogonal. It might be tempting to envisage a physical picture that the system, as in the case of 4 He, to be made up of nn f composite bosons whose creation operator is given by A(n) . However, this is f (n) an oversimplication because the operator A f does not obey the boson commutation relation. Moreover, the projection operator P(NS =0) in (432) imposes many-body spin correlations such that the spin correlation between any two bosons must have a vanishing spin-singlet component. Note that two bosons with independently uctuating spins generally have a nonzero overlap with the spin-singlet state. The many-body spin correlations of the energy eigenstates are thus far more complicated than what a naive picture of composite bosons suggests. On the other hand, as long as quantities such as the number of bosons, magnetization, and energy are concerned, the above simplied picture is quite helpful. As an illustration, we provide an explanation for the existence of forbidden values for the total spin F. For example, to construct a state with F = 0 or F = 3, composite particles with total spin 2 must be avoided, namely, n12 = n22 = 0. Then, we have N0 = 3(n30 + n33 ), implying that F = 0 or F = 3 is only possible when N0 = 3k(k Z). For a state with F = 2 or F = 5, we have n12 + n22 = 1 and N0 = 1 + n22 + 3(n30 + n33 ), implying that N0 3k(k Z). The above simplied picture is also helpful when we consider the magnetic response discussed below. 8.2.2. Magnetic response Here, we consider how the ground state and the magnetization Fz respond to the applied magnetic eld p. From Eq. (424), we see that the minimum energy states always satisfy Fz = F when p > 0. The problem thus reduces to minimizing the function E(Fz , N s ) = c1 c2 [Fz (Fz + 1) 6N] + NS (2N 2NS + 3) pFz . e 2V 5V e (437)

Here, we only consider the case of c2 < 0. Detailed investigations of this problem can be found in Ref. [5]. In this case, it is convenient to introduce a new parameter c1 c1 c2 , 20 (438)

and consider the energy as a function of Fz and l 2N0 Fz : E(Fz , l) = c1 V e 1 c2 Fz + p+ e e c1 2 2V 8V c2 l(l + 2F + 6) + const. 40V e 102
2

(439)

While the averaged slope Fz /p V e /c1 coincides with that in MFT, the oset term c2 /(8V e ) in Eq. (439) makes a qualitative distinction from MFT, that is, the onset of the magnetization displaces from p = 0 to p = |c2 |/(8V e ). Note that the slope V e /c1 and the oset |c2 |/(8V e ) are determined by independent parameters. A typical behavior of the magnetic response when |c2 | c1 is shown in Fig. 34.

Figure 34: Typical dependence of the ground-state magnetization on the applied magnetic-eld strength, for c2 < 0 and c1 > 0.

We now calculate the Zeeman-level populations of the ground states for c2 < 0. In MFT, the lowest-energy states for c2 < 0 have vanishing population in the m = 0, 1 levels. In contrast, the exact ground states derived in the preceding subsection, (A(2) )NS ( )n12 (A(2) )n22 (A(3) )n33 |vac a2 0 2 3 with n22 = 0, 1 and n33 = 0, 1, have nonzero populations in the m = 0, 1 levels. The exact forms for the averaged population a am are calculated as follows. The above ground states have the m form (A(2) )NS | (S+ )NS | , with | being a state with a xed number (s n12 + 2n22 + 3n33 ) 0 of bosons satisfying S | = 0. The average Zeeman population for the ground states, a am m |(S )NS a am (S+ )NS | m , |(S )NS (S+ )NS | NS ( a am s + 5/2 m (440)

is then simply related to the average Zeeman populations for the state | as a am = a am m m
0

+ a am m

+ 1),

(441)

where a am 0 | am | / | . This formula implies that when NS m am s, the Zeeman populations of the ground states are sensitive to the form of | . With this formula, it is a straightforward task to calculate the average Zeeman-level populations for four types of ground states, (A(2) )NS ( )n12 (A(2) )n22 (A(3) )n33 |vac a2 0 2 3 103 (442)

with n22 = 0, 1 and n33 = 0, 1. A striking feature appears in the leading terms under the condition 1 n12 NS . The results are summarized as a a1 a a1 NS (1 + n33 )/n12 1 1 and a a0 NS (1 + 2n22 )/n12 . 0 (443) (444)

These results show that the population of each magnetic sublevel depends very sensitively on the spin correlations. As a consequence, a minor change in the magnetization might lead to a major change in the population. Such dramatic changes originate from bosonic stimulations caused by the term a a in A(2) or the term ( )2 a in A(3) . 2 0 a2 1 2 3 8.2.3. Symmetry considerations on possible phases The possible ground states of spinor BECs can be inferred based on symmetry arguments without solving the GPEs. The spin-2 BEC has ve internal degrees of freedom corresponding to magnetic quantum numbers m = 2, 1, 0, 1, 2. Let am s be their amplitudes. Under rotation of the coordinate system, they must transform so as to guarantee that the linear combination
2

=
m=2

m am Y2 (, )

(445)

m is scalar, where Y2 (, ) is the spherical harmonic function of rank 2 with and being the polar and azimuthal angles, respectively, in the spherical coordinate system. Let

k = (k x , ky , kz ) = (sin cos , sin sin , cos )

(446)

be the unit vector in the (, ) direction. In terms of the unit vector k, we can rewrite Eq. (445) as = 15 T k M k, 8 i(a1 + a1 ) 2 2 3 a0 a1 + a1 (448) (447)

where superscript T denotes the transpose, and a + a 2a 2 2 3 0 1 i(a2 a2 ) 2 a1 + a1 xy xz 1 xx xy yy yz 3 xz yz zz i(a2 a2 ) a2 a2 .

2 3 a0

i(a1 + a1 )

The order parameter is thus characterized by a 3 3 traceless symmetric matrix TrM = 0 with unit normalization Tr(M M) = 1. 104 (449)

Let us now consider the eigenvalues of this order parameter matrix: det E 1 3M = E 3 3S E + A = 0, 3 3 (450)

where 3 is added in front of the matrix for the sake of convenience, and S and T are the amplitudes of the spin-singlet pair and spin-singlet trio, respectively: S T = =
i, j,k

1 2

2j = i
i, j

1 2 1 1 + 2 2 , 2 0

(451)

i jk xi y j zk

9(2 2 + 2 2 ) + 60 (2 + 31 1 + 62 2 ). 1 1 0 3 Substituting S = xy and A = 2 3(x + y3 ), Eq. (450) reduces to = E 3 + x3 + y3 3Exy = 0, and hence, the solutions are E = (x + y), (x + 2 y), (2 x + y), where = e2i/3 and x= 1 S = A + A2 48S 3 1 y 48 6
1 3

(452)

(453)

(454)

(455)

We can infer from this that the spin-singlet pair and the spin-singlet trio of atoms form building blocks of the ground state of the spin-2 BEC. It is noteworthy that this fact arises solely from the fact that the order parameter matrix of the spin-2 BEC is traceless symmetric. 9. Special Topics 9.1. Quenched BEC: Kibble-Zurek mechanism Topological excitations can be spontaneously created when a system undergoes a phase transition in a nite time. The basic concept is that after the phase transition domains of the new phase emerge at causally disconnected places, and therefore, the order parameters of the new phase at dierent locations are not correlated. When the order parameters grow to overlap, they may be able to adjust dynamically so that they are connected smoothly; if they cannot, singularities should develop in the order parameter space, giving rise to topological excitations. Such a scenario of topological defect formation was rst discussed by Kibble [134] in the context of cosmic-string and monopole formations in the early Universe, and an experimental test for the Kibble scenario in condensed matter systems was proposed by Zurek [135] (see Ref. [136] for a review). Bose-Einstein condensates of dilute atomic vapor oer an ideal opportunity for studying the Kibble-Zurek mechanism because the temperature, strength of interaction, and external parameters such as magnetic eld and trapping potentials can be changed in a time shorter than the characteristic time scale of topological defect formation. In fact, scalar vortices [137, 138] and 105

spin vortices [100, 25] have been observed to emerge spontaneously upon both temperature and magnetic-eld quenches. Topological defect formation via the Kibble-Zurek mechanism can occur in a ferromagnetic phase of a spinor BEC, where the phase transition is triggered by a sudden change in a magnetic eld (magnetic eld quenching) [100, 139, 140, 141]. The ground-state phase of a spin-1 ferromagnetic BEC is shown in Fig. 1, where p and q are the coecients of the linear and quadratic Zeeman eects, respectively. When we consider a type of magnetic-eld quench for which the total magnetization of the system is conserved, the quench occurs along a constant-p line of the phase diagram. As in the experiment described in Ref. [100], we consider the case of p = 0 that implies that the total magnetization of the system is kept to be zero. Then, the state of the system can be in one of the following two phases. When q is larger than the critical value qc = 2|c1 |n, with n being the number density, the ground state is the polar phase polar = (0, n, 0) which is nonmagnetic. When q < qc , the ground state is the broken-axisymmetry (BA) phase in which the system develops transverse magnetization. The ground state for q < qc is the broken-antisymmetry (BA) phase with the order parameter given by i e 1 q/2 (456) BA = n (1 + q)/2 , i e 1 q/2 where q q/qc . The BA phase has transverse magnetization F+ F x + iFy = ei 1 q2 which breaks the U(1) symmetry corresponding to the direction of the transverse magnetization. By rapidly quenching the magnetic eld from above to below qc , local transverse magnetization develops in random directions, and thus, spin vortices are spontaneously created. To analyze such a situation, let us assume that the initial state is the polar phase (0, n, 0) and expand the eld operator as 1 m = 1 nm,0 + a k,m ei(kr c0 nt) .
k

(457)

Substituting this into the Hamiltonian and keeping the terms up to the second order in a k,m with k 0, we obtain the Bogoliubov Hamiltonian for the polar phase. We are interested in how the modes m = 1 grow with time. The Heisenberg equations of motion for the modes are [139] i d a k,1 = ( dt
k

+ q + c1 n) k,1 + c1 n k, 1 , a a

(458)

with the solutions given by a k,1 (t) = cos i where Ek =

2 k2

E kt

+ q + c1 n E kt sin a k,1 (0) Ek (459)

c1 n E kt sin ak, 1 (0), Ek

2M

+q

2 k2

2M 106

+ q + 2c1 n

(460)

gives the Bogoliubov spectrum. If E k has an imaginary part, the corresponding mode is dynamically unstable and it grows exponentially. Because c1 < 0 and q > 0 for spin-1 87 Rb atoms, exponential growth occurs for q < qc , in agreement with the phase boundary. 9.1.1. Instantaneous quench We can use the solution (459) to investigate how the transverse magnetization develops with time. Because 0 n, the magnetization operator F+ = F = F x + iFy is written as F+ = 2n( (r) + 1 (r)). (461) 1 It follows that the correlation function of the transverse magnetization is given by F+ (r, t)F (r , t) n qc 2 k<k qc q
c

e
k

|E k |t+ik(rr )

(462)

The exponential terms describe dynamical instabilities of which the dominant term is the one for which the imaginary part of E k is maximal. Let kmu be the most unstable wave number for which |ImE k | is maximal. It follows from Eq. (460) that kmu = 0 for qc /2 < q < qc and kmu = 2M qc q 2 2 (463)

1 for q < qc /2. Thus, for q < qc /2, the typical size of magnetic domains is kmu ; for qc /2 < q < qc , it is determined by the width of the distribution of ImE k . In any case, the size of the spin domains grows as the nal value of q increases, as experimentally observed in Ref. [3]. The time scale for the magnetization is given by

2|Ekmu |

(464)

which is / 4q(qc q) for qc /2 < q < qc and /qc for q < qc /2. 9.1.2. Finite-time quench Next, we consider the case in which the magnetic eld is quenched linearly in a nite time Q : q(t) = qc 1 t . Q (465)

In this case, the Bogoliubov excitation energy (460) also varies with time and the spin correlation function is estimated to be F+ (r, t)F (r , t) d k exp 2
0 2 exp f (t) |r r |2 /Q , t

|E k (t )|t dt + ik (r r ) (466)

where Q is the correlation length and f (t) determines the growth rate of magnetization. In two dimensions, we have [139] 16 2 t(Q t) Q = M2 107
1 4

(467)

and Q qc f (t) = 2 1 tan t Q t t t 1 Q Q 2t 1 Q . (468)

The time scale for magnetization is determined by f (t) = 1. Solving this equation by assuming that t Q , we obtain tQ Substituting this into Eq. (467), we obtain Q
4
1 6 3 Q . 1

Q q2 c
2

1 3

(469)

M 3 qc

(470)

The same power law was obtained in Ref. [140]. 9.1.3. Numerical simulations The topological defect formation caused by a quantum phase transition can be best illustrated by numerical simulations. Figure 35 shows the time evolution following an instantaneous quench of the magnetic eld from q > qc to q = 0 (Fig. 35(a)) and q = qc /2. Here, we assume that the system is a two-dimensional disk with a hard wall at radius 100 m and the potential is assumed to be at inside the wall. The initial state is considered to be a stationary state of the GP equations and the following quench dynamics is obtained by using the time-dependent GP equations. To trigger the dynamical instabilities that cause defect formation, we assume small uctuations in the initial amplitudes of the m = 1 components according to a k,1 (0) = + i, (471)

where and are random variables whose amplitudes x follow the normal distribution p(x) = 2/ exp(2x2 ). Figure 35 shows snapshots of the amplitude |F+ (r)| and the phase argF+ (r) of the transverse magnetization after the quench. We observe that many holes emerge spontaneously after 100 ms. These holes represent topological excitations called polar-core vortices in which the m = 1 components have vortices with the cores lled by the m = 0 components. By counting the number of vortices in the nal state, we can determine the number of spin vortices. Alternatively, we can calculate the same numberthe spin winding number wfrom the algebraic relation w= 1 2 1 F F+ F+ F d = 2 2 2i|F+ | d , (472)

C(R)

where C(R) is a circle of radius R and w is a count of the number of rotations of the spin vector in the xy plane along C(R). Figure 36 shows the R dependence of the ensemble average of w2 (R). We note that w2 (R) avg is proportional to R for large R in agreement with the KZ theory, whereas it is proportional to R2 for small R. 108

(a) q = 0

t = 0 ms

t = 50 ms

t = 75 ms

t = 100 ms

t = 200 ms
1

|F+|

arg|F+|

(b) q = qc / 2

t = 0 ms

t = 50 ms

t = 75 ms

t = 100 ms

t = 200 ms

|F+|

arg|F+|

Figure 35: Spontaneous magnetization following the quench from q > qc to q < qc . Shown are time developments of the magnetization |F+ | (upper) and its direction arg |F+ | (lower) for (a) q = 0 and (b) q = qc /2. The black circles in (b) indicate a topological defect. Reprinted from Ref. [139].

If spin vortices can be generated randomly, w2 (R) avg should be proportional to the area of the system and hence to R2 . The KZ scaling law R that we observe for large R arises from the fact that the number of vortices matches that of anti-vortices inside the disk due to spin conservation. In other words, for small q, magnetic domains are aligned in such a manner as to cancel the local spin when averaged over the spin correlation length. Thus, over a greater length scale, the magnetic domains can be created independently. This is the underlying physics that makes the spin conservation compatible with the KZ postulate of independent defect creation at a long distance and the KZ scaling law valid for 2D systems. 9.2. Low-dimensional systems It had been conjectured from the work of Peierls and Landau in the 1930s [142, 143] that two-dimensional systems with continuous order parameters could not have a conventional longrange order, and this was conrmed by Mermin and Wagner [144] and by Hohenberg [145]. Long wavelength uctuations are more important in one and two dimensions than in three dimensions and deviations from the mean-eld theory are much greater. However, there still exists a type of quasi-long-range order in the low-temperature phase in two-dimensional systems that is characterized by a power-law decay of the order parameter correlation function. Such a phase 109

10

sl
avg

e op

2
q=0 q=qc/2

w2

slo
0.1

pe

=1

Figure 36: R dependence of the variance of the spin winding number w(R) for instantaneous quench of the magnetic eld to q = 0 and to q = qc /2. The dashed and dotted lines are respectively proportional to R and R2 . Reprinted from Ref. [139].

transition is known as the Berezinskii-Kosterlitz-Thouless (BKT) transition. In this subsection, we rst explain the BKT transition for a single-component Bose gas, and then, we extend it to a spinor gas. 9.2.1. Berezinskii-Kosterlitz-Thouless transition in a single-component Bose gas We start from a Ginzburg-Landau energy functional H[] = d2 r J | |2 + ||2 + ||4 , 2 2 4 (473)

H which gives an equilibrium probability for as Peq [] exp[ kB T ]. Here, > 0 and changes its sign from positive to negative as the temperature decreases from above to below the critical temperature T c , so that the equilibrium value of changes from = 0 at T > T c to || = 0 ||/ at T < T c . In three dimensions, the one-body correlation function (r)(0) remains nite at large distance below T c , exhibiting a long-range order. In constrast, above T c , the correlation function decays exponentially:

where + = J/|| (T T c )1/2 except in the immediate vicinity of T c (see e.g. Ref. [146]). In this case, the order is short-ranged. In two dimensions, the eect of uctuations is much more drastic. We rst assume that || takes its equilibrium value everywhere and consider only phase uctuations. Substituting (r) = 0 ei(r) in Eq. (473), we have H= 1 2 K0 | |2 d2 r = 110 1 K0 2 k 2 k , k
k

0.01 1 10 R [m] 100

(r)(0)

er/+ for r , r

(474)

(475)

where K0 J0 |0 |2 and k is the Fourier component of (r) satisfying k = . Because the k gradient of induces a superuid current as vs = ( /M) , the coecient K0 is related to the superuid density 0 as s 0 = s so that H is rewritten as H= 1 2 d2 r 0 |vs |2 . s (477) M 2 K0
2

(476)

Because Eq. (477) is quadratic in k , we can calculate the correlation function as (r)(0) = |0 |2 ei{(r)(0)} kB T 2 = |0 | exp {1 cos(k r)} . 2 K0 k k In a two-dimensional system, the summation of k is replaced by an integral as kB T 1 kB T {1 cos(k r)} = 2 K0 k (2)2 K0 = kB T 2K0 dk dk k
2 0

(478) (479)

d[1 cos(kr cos )]

(480) (481)

1 J0 (kr) , k

where J0 (z) is the zeroth Bessel function. For k 1/r, we can ignore the rapidly oscillating function J0 (kr), whereas the contribution from k < 1/r is small. We also introduce a cuto wavelength corresponding to the coherence length, and neglect spatial structures smaller than . Then, the correlation function is calculated as 1 kB T dk r 2 (r)(0) = |0 | exp , (482) 2K k 0 r1 where = kB T . 2K0 (483)

Thus, phase uctuations destroy the true long-range order; however, the quasi long-range order characterized by a power-law behavior remains. Next, we consider the eect of uctuations on the amplitude ||. A crucial eect arises when || passes through zero. If |(r)| is nite everywhere in a singly connected region, the phase of the order parameter is dened as a single-valued function. On the other hand, if = 0 at r = ri , can be a multi-valued function. The points {ri } correspond to vortices around which varies by an integral multiple of 2. The phase transition for the appearance of a quasi-long-range order can be captured by a simple discussion of vortex nucleation. When a single vortex passes between two distinct points, the relative phase between these two points changes by 2, implying that free vortices destroy the phase coherence. Hence, the quasi-long-range order is destroyed when free vortices are thermally excited. 111

The free energy for a single vortex can be calculated as follows. The circulation of the superuid velocity around the vortex is quantized as vs d = h/M. Assuming axisymmetry around the vortex, we have vs = e /(Mr), where (r, ) is the polar coordinate around the vortex and e is a unit vector in the direction of . Then, the energy cost required to create a single vortex is given by Ev = Ec + 1 2
R 2

rdr
0

d 0 |vs |2 s

(484) (485)

= Ec + K0 ln

R ,

where R is the radius of the system; , the radius of the vortex core; and Ec , the vortex core energy arising from the region of r < . On the other hand, because the vortex can freely move in the area of R2 , the entropy is given by S = kB ln(R2 /2 ). Then, we obtain the free energy of a free vortex as F = Ev T S = (K0 2kB T ) ln R + Ec . (486)

One can immediately see from this free energy form that the phase transition occurs at T c0 = K0 , 2kB (487)

above which free vortices are generated and destroy the phase coherence. From Eqs. (483) and (487), the exponent must satisfy 1/4. A more careful argument requires the interaction between vortices to be considered. If there exists a pair of a vortex (charge 1) and an anti-vortex (charge 1) at distance d, the velocity elds generated by these vortices at a distance larger than d cancel each other out. As a result, the vortex pair has a nite energy of d Epair (d) = 2Ec + 2K0 ln , (488)

which indicates that a vortex and an anti-vortex attract each other and form a pair at low temperature. Because vortex pairs do not destroy phase coherence (see Fig. 37), the superuid phase transition can be interpreted as a dissociation of vortex pairs. If there exist many vortex pairs, the attractive interaction between a vortex and an anti-vortex is weakened by the screening eect. Kosterlitz and Thouless [147] analyzed this phase transition by mapping the system to a two-dimensional Coulomb gas of unit charge q = K0 . They introduced a dielectric constant that renormalizes K0 to K = K0 /, and constructed the renormalization-group ow equations. Their main result is that the superuid density jumps at the transition temperature that satises the universal relation:
2 s (T c ) 2 K(T c ) = 2 = . kB T c M kB T c

(489)

This universal behavior in superuid helium was observed by Bishop and Reppy [148]. With regard to cold atom systems, the BKT transition of a single-component Bose gas was observed in Ref. [149] 112

(a)

(b)

Figure 37: Local phase of the order parameter around (a) a single vortex and (b) a vortex pair, where the direction of each arrow indicates a phase that varies between 0 and 2. The phase around a single vortex varies, causing a velocity eld proportional to 1/r, whereas that away from the vortex pair can be uniform.

9.2.2. 2D spinor gases We now consider the case of a spinor. As discussed above, vortices play a crucial role in the superuid phase transition in two-dimensional systems. When a system acquires spin degrees of freedom, the BKT phase transition becomes highly nontrivial because the nature of the vortices depends on the order parameter manifold, as discussed in Sec. 7. It is well known that though the XY spin model exhibits a BKT phase transition where the order parameter manifold U(1) is the same as that of a single-component Bose gas, the BKT transition does not occur for the Heisenberg spin model at nite temperature [147]. This is because the order parameter manifold for the Hisenberg model, which is isomorphic to S 2 , does not accommodate a topologically stable vortex because 1 (S 2 ) = 0. As discussed in Sec. 7, the order parameter manifold for the polar phase of a spin-1 BEC can accommodate half-quantum vortices, and therefore, the BKT transition occurs due to the binding of half-quantum vortex pairs. On the other hand, if the condensate accommodates several types of vortices, the BKT transition is expected to occur in several stages. Here, we discuss two BKT transitions of ferromagnetic spinor gases that occur under an external magnetic eld. For simplicity, we neglect the trapping potential and assume an innite system. The energy functional for a spin-1 spinor gas under an external magnetic eld B in the z-direction is given by 2 H= d2 r (490) | m |2 + c0 n2 + c1 |F|2 + q |1 |2 + |1 |2 , 2M
m=0,1

where n = m |m |2 and F = mn (f)mn n are the number density and spin density, rem spectively, and q is the quadratic Zeeman energy proportional to B2 . For spin-1 BECs, q = (B B)2 /(4hf ). The linear Zeeman term simply causes Larmor precession due to the spin conservation, and therefore, it is omitted by going onto a rotating frame of reference in the spin space. We consider the ferromagnetic interaction, c1 < 0. At zero temperature, the ground state phase is the same as that we have obtained for a three-dimensional system in Sec. 3. Here, we assume that the total longitudinal magnetization is conserved to be zero. Then, the order

113

parameter for q > qc 2|c1 |n0 is the polar phase whose order parameter is given by 0 i polar = n0 e 1 , 0 whereas for q < qc , the broken-axisymmetry phase appears: i ei 1 q n0 e 2(1 + q) , BA = 2 i e 1q

(491)

(492)

where n0 is the absolute square of the order parameter at zero temperature and q q/qc . For q > qc , there exists only one type of vortex; a conventional phase vortex. Therefore, the BKT transition for q > qc is the same as that in single-component gases. Unbinding of phase vortex pairs destroys the coherence of the U(1) phase that appears in the correlation function of the m = 0 component: (r)0 (0) r , 0 (493)

where is given by Eq. (483) with renormalized K instead of K0 = 2 n0 /M. The universal jump is given by Eq. (489). On the other hand, for q < qc , there exist two types of vortices, i.e., a phase vortex and a spin vortex. The phase vortex is the same as that in the polar state, around which the U(1) phase changes by integer multiples of 2. The spin vortex is dened as a vortex around which the direction of transverse magnetization (i.e., ) varies by integer multiple of 2. Here, the transverse magnetization appears at q < qc and it is given by F+ = F x + iFy = 2 0 + 1 = n0 0 1
2 2

1 q2 ei ,

(494) (495)

Fz = |1 | |1 | = 0.

The eects of these vortices are independent because the energy for these uctuations are decoupled as H= where K0 K0
sp p

1 2

d2 r K0 | |2 + K0 | |2 ,

sp

(496)

n0 , M 2 n0 (1 q) . 2M

(497) (498)

Similar to the discussion for Eq. (486), a rough estimation for the transition temperature for the appearance of a free phase vortex and a free spin vortex are given by
p T c0

T c0

sp

K0 2 n0 = = , 2kB 2MkB sp K0 2 n0 (1 q) 1 q p = = = T c0 , 2kB 4MkB 2 114

(499) (500)

respectively. Hence, as the temperature decreases, phase vortex pairs rst bind and then spin vortices form pairs at lower temperature. More proper treatment following the Kosterlitz-Thoulesss renormalization method gives the relation between the universal jump and the transition temperature: K p (T c ) K sp (T c ) 2 p = sp = . kB T c kB T c
p sp

(501)

The phase transitions related to the binding of the phase and spin vortices can be observed as an appearance of the quasi-long-range order in the correlation function of m = 0 and m = 1 components, respectively, which are given by (r)0 (0) ei((r)(0)) rp , 0 (r)1 (0) ei((r)(0)) e 1 with p = kB T , 2K p sp = kB T . 2K sp (504)
i((r)(0))

(502) rp sp , (503)

The binding of spin vortices also contributes to the ferromagnetic long-range order F+ (r)F (0) ei{(r)(0)} rsp . (505)

Figure 38 summarizes the above results. Depending on the temperature and the external magnetic eld, there exist three ordered phases: (I) quasi-long-range order arises only in the m = 0 component, where the gas has no local magnetization; (II) a quasi-long-range order arises only in the m = 0 component, where the gas is locally magnetized, but with no long-range order of magnetization; (III) a quasi-long-range order appears in every m component, where the transverse magnetization also exhibits the quasi-long-range order. Here, we discuss the phase diagram at low magnetic elds. In the above system, spin vortices are stable because the quadratic Zeeman eect suppresses the uctuation of longitudinal magnetization. As the external eld decreases, however, the contribution of such uctuations becomes larger and induces coupling between the spin and the phase vortices. Then, both transition temperatures are expected to be suppressed. At zero magnetic eld, the quasi-long-range order does not appear at nite temperature as in the case of the Hisenberg spin system. Understanding the detailed behavior in the low-magnetic-eld region requires further investigation. 10. Summary and Future Prospects In the present paper, we have reviewed the basic knowledge concerning spinor Bose-Einstein condensates (BECs) that has been accumulated thus far. The fundamental characteristics of spinor BECs are rotational invariance, coupling between the spin and the gauge degrees of freedom, and magnetism arising from the magnetic moment of the spin. The rotational invariance and gauge invariance alone uniquely determine the microscopic Hamiltonian of the spinor BEC, as discussed in Sec. 2. The mean-eld theory described in Sec. 3 discribes the ground-state phases and the dynamics of spinor BECs. In a scalar BEC, the measurements of the collective modes are found to agree with the theoretical values with an accuracy better than 1% (see Ref. [150] for a comprehensive review). In the near future, 115

T
normal

Tcphase I Tcspin

II

III qc

Figure 38: Schematic phase diagram of a 2D ferromagnetic Bose gas. (I) a quasi-long-range order appears only in the m = 0 component with no local magnetization; (II) a quasi-long-range order arises only in the m = 0 component, where the gas is locally magnetized, but without any long-range order of the magnetization; (III) a quasi-long-range order appears in every m component, where the transverse magnetization also exhibits a quasi-long-range order. In the shaded region, the uctuation of the longitudinal magnetization is expected to destroy the quasi-long-range order.

experimental tasks to be carried out include investigating the extent to which the predictions of the Bogoliubov theory described in Sec. 5 can be veried experimentally. The topological excitations described in Sec. 7 are of cross-disciplinary interest; in particular, the non-Abelian nature of vortices in the nematic and cyclic phases of spin-2 BECs will have fundamental implications on quantum turbulence and other subelds of physics. The topological excitations are usually considered within a given order parameter manifold. However, as exemplied in Fig. 33 in Sec. 7.3.4, the order-parameter manifold can change dynamically from one to another, and during a transition, the topological charge is no longer conserved. It is an interesting theoretical question to investigate whether, if any, a more general topologically invariant quantity can be dened in such a general situation. The relative homotopy theory described in Sec. 7.2.2 may be used as a guiding theoretical framework. A closely related subject is vortex nucleation dynamics and vortex lattice formation. Vortices are singularities of the order parameter. In a scalar BEC, the only possible singularity is the density hole in the condensate. However, the spinor BEC has several dierent phases and the choice of creating a density hole is energetically most costly. Thus, in spinor BECs, the vortex core is lled by a state that does not belong to the order parameter manifold of the condensate. As discussed in Sec. 6, the spin-1 polar BEC has a ferromagnetic vortex core, and the spin-1 ferromagnetic BEC can host a polar-core vortex. For spin-2 and higher-spin BECs, more than two possible phases can ll the vortex core. Because each phase has its own internal spin structure, an interesting question arises as to how the order parameter of the BEC smoothly connects the spin structure of the vortex core [151]. The many-body quantum correlations of spinor BECs described in Sec. 8 are fragile against an external magnetic eld because the bosonic stimulation favors the mean eld. The spinordipolar phenomena described in Sec. 4.3 are also vulnerable against a magnetic eld because 116

the Larmor precession averages out the part of the dipole interaction that couples the dierent total spin components. It is noteworthy that both many-body spin correlations and spinor-dipolar eects become signicant as the external magnetic eld is reduced down to approximately 10 G. We have good reasons to expect that further symmetry breaking occurs in such an environment of ultralow magnetic eld. From the viewpoint of fundamental physics, the topological defect formation based on the Kibble-Zurek mechanism discussed in Sec. 9.1 merits further experimental and theoretical study. In particular, this eect combined with the spinor Berezinskii-Kosterlitz-Thouless transition in two dimensions described in Sec. 9.2.1 raises many interesting questions. In fact, the gauge, spin, and dipole-dipole interactions set very dierent energy scales that, we can conceive, lead to topological phase transitions at dierent energy and temperature scales. All the discussions presented in this review are restricted to absolute zero. However, experiments on spinor BECs have been carried out at temperatures much higher than the typical energy scale of the spin-exchange interaction. Nevertheless, once the system undergoes Bose-Einstein condensation, all the condensed particles respond to an external perturbation in exactly the same manner. When the number of condensate particles is of the order of 105 , the collective energy for the dipole interaction amounts to 1 K for 87 Rb and 100 K for 52 Cr. Therefore, for temperatures below 0.1 K, the dynamics of the system can be understood, at least qualitatively, based on the zero-temperature theory. Nonetheless, it is likely that the precise measurements of collective modes are sensitive to nite-temperature eects. Moreover, when the long-time dynamics is considered, the thermalization process is expected to play a crucial role in determining the magnetic structure of the system. In fact, very recent experiments on the 87 Rb f = 1 BEC show evidence of magnetic ordering [25] as a consequence of thermalization. To fully understand such phenomena and to explore new types of temperature-quenched symmetry-breaking phase transitions, it is obvious that we need to develop theoretical tools to investigate the dynamics of spinor-dipolar physics at nite temperatures. Acknowledgements We acknowledge the fruitful collaborations with Hiroki Saito and Michikazu Kobayashi. MU acknowledges the Aspen Center for Physics, where part of this work was carried out. References
[1] K. Kasamatsu, M. Tsubota, M. Ueda, Vortices in multicomponent Bose-Einstein condensates, International Journal of Modern Physics B 19 (2005) 1835. [2] F. Gerbier, A. Widera, S. F lling, O. Mandel, I. Bloch, Resonant control of spin dynamics in ultracold quantum o gases by microwave dressing, Phys. Rev. A 73 (2006) 041602. [3] S. R. Leslie, J. Guzman, M. Vengalattore, J. D. Sau, M. L. Cohen, D. M. Stamper-Kurn, Amplication of uctuations in a spinor Bose-Einstein condensate, Phys. Rev. A 79 (2009) 043631. [4] M. Koashi, M. Ueda, Exact Eigenstates and Magnetic Response of Spin-1 and Spin-2 Bose-Einstein Condensates, Phys. Rev. Lett. 63 (2000) 013601. [5] M. Ueda, M. Koashi, Theory of spin-2 Bose-Einstein condensates: Spin correlations, magnetic response, and excitation spectra, Phys. Rev. A 65 (2002) 063602. [6] T. Ohmi, K. Machida, Bose-Einstein Condensation with Internal Degrees of Freedom in Alkali Atom Gases, J. Phys. Soc. Jpn. 67 (1998) 1822. [7] T.-L. Ho, Spinor Bose Condensates in Optical Trap, Phys. Rev. Lett. 81 (1998) 742. [8] M.-S. Chang, C. D. Hamley, M. D. Barrett, J. A. Sauer, K. M. Fortier, W. Zhang, L. You, M. S. Chapman, Observation of Spinor Dynamics in Optically Trapped 87Rb Bose-Einstein Condensates, Phys. Rev. Lett. 92 (2004) 140403.

117

[9] J. Stenger, S. Inouye, D. M. Stamper-Kurn, H.-J. Miesner, A. P. Chikkatur, W. Ketterle, Spin domains in groundstate Bose-Einstein condensates, Nature 396 (1998) 345. [10] K. Murata, H. Saito, M. Ueda, Broken-axisymmetry phase of a spin-1 ferromagnetic Bose-Einstein condensate, Phys. Rev. A 75 (2007) 013607. [11] C. V. Ciobanu, S.-K. Yip, T.-L. Ho, Phase diagrams of F=2 spinor Bose-Einstein condensates, Phys. Rev. A 61 (2000) 033607. [12] J. L. Song, G. W. Semeno, F. Zhou, Uniaxial and Biaxial Spin Nematic Phases Induced by Quantum Fluctuations, Phys. Rev. Lett. 98 (2007) 160408. [13] A. M. Turner, R. Barnett, E. Demler, A. Vishwanath, Nematic Order by Disorder in Spin-2 Bose-Einstein Condensates, Phys. Rev. Lett. 98 (2007) 190404. [14] H. Saito, M. Ueda, Diagnostics for the ground-state phase of a spin-2 Bose-Einstein condensate, Phys. Rev. A 72 (2005) 053628. [15] A. Griesmaier, J. Werner, S. Hensler, J. Stuhler, T. Pfau, Bose-Einstein Condensation of Chromium, Phys. Rev. Lett. 94 (2005) 160401. [16] J. Stuhler, A. Griesmaier, T. Koch, M. Fattori, T. Pfau, S. Giovanazzi, P. Pedri, L. Santos, Observation of DipoleDipole Interaction in a Degenerate Quantum Gas, Phys. Rev. Lett. 95 (2005) 150406. [17] S. Giovanazzi, P. Pedri, L. Santos, A. Griesmaier, M. Fattori, T. Koch, J. Stuhler, T. Pfau, Expansion dynamics of a dipolar Bose-Einstein condensate, Phys. Rev. A 74 (2006) 013621. [18] T. Lahaye, T. Koch, B. Fr hlich, M. Fattori, J. Metz, A. Griesmaier, S. Giovanazzi, T. Pfau, Strong dipolar eects o in a quantum ferrouid, Nature 448 (2007) 672. [19] T. Koch, T. Lahaye, J. Metz, B. Fr hlich, A. Griesmaier, T. Pfau, Stabilization of a purely dipolar quantum gas o against collapse, Nature Physics 4 (2008) 218. [20] T. Lahaye, J. Metz, B. Fr hlich, T. Koch, M. Meister, A. Griesmaier, T. Pfau, H. Saito, Y. Kawaguchi, M. Ueda, o d-Wave Collapse and Explosion of a Dipolar Bose-Einstein Condensate, Phys. Rev. Lett. 101 (2008) 080401. [21] J. M. Doyle, B. Friedrich, J. Kim, D. Patterson, Buer-gas loading of atoms and molecules into a magnetic trap, Phys. Rev. A 52 (1995) R2515. [22] K.-K. Ni, S. Ospelkaus, M. H. G. de Miranda, A. Peer, B. Neyenhuis, J. J. Zirbel, S. Kotochigova, P. S. Julienne, D. S. Jin, J. Ye, A High Phase-Space-Density Gas of Polar Molecules, Science 322 (2008) 321. [23] J. Deiglmayr, A. Grochola, M. Repp, K. M rtlbauer, C. Gl ck, J. Lange, O. Dulieu, R. Wester, M. Weidem ller, o u u Formation of Ultracold Polar Molecules in the Rovibrational Ground State, Phys. Rev. Lett. 101 (2008) 133004. [24] M. Fattori, G. Roati, B. Deissler, C. DErrico, M. Zaccanti, M. Jona-Lasinio, L. Santos, M. Inguscio, G. Modugno, Magnetic Dipolar Interaction in a Bose-Einstein Condensate Atomic Interferometer, Phys. Rev. Lett. 101 (2008) 190405. [25] M. Vengalattore, S. R. Leslie, J. Guzman, D. M. Stamper-Kurn, Spontaneously Modulated Spin Textures in a Dipolar Spinor Bose-Einstein Condensate, Phys. Rev. Lett. 100 (2008) 170403. [26] S. E. Pollack, D. Dries, M. Junker, Y. P. Chen, T. A. Corcovilos, R. G. Hulet, Extreme Tunability of Interactions in a 7 Li Bose-Einstein Condensate, Phys. Rev. Lett. 102 (2009) 090402. [27] M. Baranov, L. Dobrek, K. G ral, L. Santos, M. Lewenstein, Ultracold Dipolar Gases - a Challenge for Experio ments and Theory, Physica Scripta. T102 (2002) 74. [28] M. A. Baranov, Theoretical progress in many-body physics with ultracold dipolar gases, Physics Reports 464 (2008) 71. [29] T. Lahaye, C. Menotti, L. Santos, M. Lewenstein, T. Pfau, The physics of dipolar bosonic quantum gases, arXiv (2009) 0905.0386. [30] L. D. Landau, E. M. Lifshitz, Quantum Mechanics (Non-Relativistic Theory) 3rd edition, ButterworthHeinemann, 1981. [31] S. Yi, L. You, Trapped atomic condensates with anisotropic interactions, Phys. Rev. A 61 (2000) 041604(R). [32] S. Yi, L. You, Trapped condensates of atoms with dipole interactions, Phys. Rev. A 63 (2001) 053607. [33] M. Marinescu, L. You, Controlling Atom-Atom Interaction at Ultralow Temperatures by dc Electric Fields, Phys. Rev. Lett. 81 (1998) 4596. [34] B. Deb, L. You, Low-energy atomic collision with dipole interactions, Phys. Rev. A 64 (2001) 022717. [35] A. Derevianko, Anisotropic pseudopotential for polarized dilute quantum gases, Phys. Rev. A 67 (2003) 033607. [36] A. Derevianko, Erratum: Anisotropic pseudopotential for polarized dilute quantum gases [Phys. Rev. A 67, 033607 (2003)], Phys. Rev. A 72 (2005) 039901. [37] M. Aymar, O. Dulieu, Calculation of accurate permanent dipole moments of the lowest 1,3Sigma+ states of heteronuclear alkali dimers using extended basis sets, Journal of Chemical Physics 122 (2005) 204302. [38] J. D. Weinstein, R. deCarvalho, T. Guillet, B. Friedrich, J. M. Doyle, Magnetic trapping of calcium monohydride molecules at millikelvin temperatures, Nature 395 (1998) 148. [39] D. Egorov, T. Lahaye, W. Sch llkopf, B. Friedrich, J. M. Doyle, Buer-gas cooling of atomic and molecular o beams, Phys. Rev. A 66 (2002) 043401.

118

[40] S. Y. T. van de Meerakker, H. L. Bethlem, G. Meijer, Taming molecular beams, Nature Physics 4 (2008) 595. [41] K. Aikawa, D. Akamatsu, J. Kobayashi, M. Ueda, T. Kishimoto, S. Inouye, Toward the production of quantum degenerate bosonic polar molecules, 41 K87 Rb, New J. Phys. 11 (2009) 055035. [42] J. Werner, A. Griesmaier, S. Hensler, J. Stuhler, T. Pfau, A. Simoni, E. Tiesinga, Observation of Feshbach Resonances in an Ultracold Gas of 52 Cr, Phys. Rev. Lett. 94 (2005) 183201. [43] S. Giovanazzi, A. G rlitz, T. Pfau, Tuning the Dipolar Interaction in Quantum Gases, Phys. Rev. Lett. 89 (2002) o 130401. [44] K. G ral, L. Santos, Ground state and elementary excitations of single and binary Bose-Einstein condensates of o trapped dipolar gases, Phys. Rev. A 66 (2002) 023613. [45] S. Ronen, D. C. E. Bortolotti, J. L. Bohn, Bogoliubov modes of a dipolar condensate in a cylindrical trap, Phys. Rev. A 74 (2006) 013623. [46] K. G ral, K. Rzazewski, T. Pfau, Bose-Einstein condensation with magnetic dipole-dipole forces, Phys. Rev. A o 61 (2000) 051601(R). [47] J.-P. Martikainen, M. Mackie, K.-A. Suominen, Comment on Bose-Einstein condensation with magnetic dipoledipole forces, Phys. Rev. A 64 (2001) 037601. [48] L. Santos, G. V. Shlyapnikov, P. Zoller, M. Lewenstein, Bose-Einstein Condensation in Trapped Dipolar Gases, Phys. Rev. Lett. 85 (2000) 1791. [49] D. H. J. ODell, S. Giovanazzi, C. Eberlein, Exact Hydrodynamics of a Trapped Dipolar Bose-Einstein Condensate, Phys. Rev. Lett. 92 (2004) 250401. [50] C. Eberlein, S. Giovanazzi, D. H. J. ODell, Exact solution of the Thomas-Fermi equation for a trapped BoseEinstein condensate with dipole-dipole interactions, Phys. Rev. A 71 (2005) 033618. [51] S. Ronen, D. C. E. Bortolotti, J. L. Bohn, Radial and Angular Rotons in Trapped Dipolar Gases, Phys. Rev. Lett. 98 (2007) 030406. [52] L. Santos, G. V. Shlyapnikov, M. Lewenstein, Roton-Maxon Spectrum and Stability of Trapped Dipolar BoseEinstein Condensates, Phys. Rev. Lett. 90 (2003) 250403. [53] U. R. Fischer, Stability of quasi-two-dimensional Bose-Einstein condensates with dominant dipole-dipole interactions, Phys. Rev. A 73 (2006) 031602(R). [54] D. H. J. ODell, S. Giovanazzi, G. Kurizki, Rotons in Gaseous Bose-Einstein Condensates Irradiated by a Laser, Phys. Rev. Lett. 90 (2003) 110402. [55] P. Pedri, L. Santos, Two-Dimensional Bright Solitons in Dipolar Bose-Einstein Condensates, Phys. Rev. Lett. 95 (2005) 200404. [56] I. Tikhonenkov, B. A. Malomed, A. Vardi, Anisotropic Solitons in Dipolar Bose-Einstein Condensates, Phys. Rev. Lett. 100 (2008) 090406. [57] R. Nath, P. Pedri, L. Santos, Phonon Instability with Respect to Soliton Formation in Two-Dimensional Dipolar Bose-Einstein Condensates, Phys. Rev. Lett. 102 (2009) 050401. [58] H. Matsuda, T. Tsuneto, O-Diagonal Long-Range Order in Solids, Supplement of the Progress of Theoretical Physics 46 (1970) 411. [59] K.-S. Liu, M. E. Fisher, Quantum lattice gas and the existence of a supersolid, Journal of Low Temperature Physics 10 (1973) 655. [60] C. Bruder, R. Fazio, G. Sch n, SuperconductorMott-insulator transition in Bose systems with nite-range intero actions, Phys. Rev. B 47 (1993) 342. [61] A. van Otterlo, K.-H. Wagenblast, Coexistence of diagonal and o-diagonal long-range order: A Monte Carlo study, Phys. Rev. Lett. 72 (1994) 3598. [62] A. F. Andreev, I. M. Lifshitz, Quantum theory of crystal defects, Sov. Phys. JETP 29 (1969) 1107. [63] G. V. Chester, Speculations on Bose-Einstein Condensation and Quantum Crystals, Phys. Rev. A 2 (1970) 256. [64] A. J. Leggett, Can a Solid Be Superuid?, Phys. Rev. Lett. 25 (1970) 1543. [65] K. G ral, L. Santos, M. Lewenstein, Quantum Phases of Dipolar Bosons in Optical Lattices, Phys. Rev. Lett. 88 o (2002) 170406. [66] S. Yi, T. Li, C. P. Sun, Novel Quantum Phases of Dipolar Bose Gases in Optical Lattices, Phys. Rev. Lett. 98 (2007) 260405. [67] D. Jaksch, C. Bruder, J. I. Cirac, C. W. Gardiner, P. Zoller, Cold Bosonic Atoms in Optical Lattices, Phys. Rev. Lett. 81 (1998) 3108. [68] C. Menotti, C. Trefzger, M. Lewenstein, Metastable States of a Gas of Dipolar Bosons in a 2D Optical Lattice, Phys. Rev. Lett. 98 (2007) 235301. [69] H. Saito, Y. Kawaguchi, M. Ueda, Ferrouidity in a two-component dipolar Bose-Einstein Condensate, arXiv (2009) 0812.0278. [70] S. Yi, L. You, H. Pu, Quantum Phases of Dipolar Spinor Condensates, Phys. Rev. Lett. 93 (2004) 040403. [71] H. Pu, W. Zhang, P. Meystre, Ferromagnetism in a Lattice of Bose-Einstein Condensates, Phys. Rev. Lett. 87 (2001) 140405.

119

[72] K. Gross, C. P. Search, H. Pu, W. Zhang, P. Meystre, Magnetism in a lattice of spinor Bose-Einstein condensates, Phys. Rev. A 66 (2002) 033603. [73] W. Zhang, H. Pu, C. Search, P. Meystre, Spin Waves in a Bose-Einstein Condensed Atomic Spin Chain, Phys. Rev. Lett. 88 (2002) 060401. [74] Y. Kawaguchi, H. Saito, M. Ueda, Einstein-de Haas Eect in Dipolar Bose-Einstein Condensates, Phys. Rev. Lett. 96 (2006) 080405. [75] K. Gawryluk, M. Brewczyk, K. Bongs, M. Gajda, Resonant Einsteinde Haas Eect in a Rubidium Condensat, Phys. Rev. Lett. 99 (2007) 130401. [76] L. D. Landau, E. M. Lifshitz, Electrodynamics of Continuous Media, 2nd edition, Butterworth-Heinemann, 1984. [77] Y. Kawaguchi, H. Saito, M. Ueda, Spontaneous Circulation in Ground-State Spinor Dipolar Bose-Einstein Condensates, Phys. Rev. Lett. 97 (2006) 130404. [78] M. Takahashi, S. Ghosh, T. Mizushima, K. Machida, Spinor Dipolar Bose-Einstein Condensates: Classical Spin Approach, Phys. Rev. Lett. 98 (2007) 260403. [79] S. Yi, H. Pu, Spontaneous Spin Textures in Dipolar Spinor Condensates, Phys. Rev. Lett. 97 (2006) 020401. [80] Y. Kawaguchi, H. Saito, M. Ueda, Can Spinor Dipolar Eects be Observed in Bose-Einstein Condensates?, Phys. Rev. Lett. 98 (2007) 110406. [81] J. Zhang, T.-L. HO, Spontaneous Vortex Lattices in Quasi 2D Dipolar Spinor Condensates, arXiv:0908.1593 . [82] Y. Kawaguchi, K. Kudo, H. Saito, M. Ueda, Magnetic crystallization of a ferromagnetic Bose-Einstein condensate, arXiv:0909.0565 . [83] J. A. Kj ll, A. M. Essin, J. E. Moore, Magnetic phase diagram of a spin-1 condensate in two dimensions with a dipole interaction, arXiv:0909.1751 . [84] M. Vengalattore, S. R. Leslie, J. Guzman, D. M. Stamper-Kurn, Crystalline Magnetic Order in a Dipolar Quantum Fluid, arXiv:0901.3800 . [85] C. K. Law, H. Pu, N. P. Bigelow, Quantum Spins Mixing in Spinor Bose-Einstein Condensates, Phys. Rev. Lett. 81 (1998) 5257. [86] T.-L. Ho, S. K. Yip, Fragmented and Single Condensate Ground States of Spin-1 Bose Gas, Phys. Rev. Lett. 84 (2000) 4031. [87] M. Ueda, Many-Body Theory of Dilute Bose-Einstein Condensates with Internal Degrees of Freedom, Phys. Rev. A 63 (2000) 013601. [88] H.-J. Miesner, D. M. Stamper-Kurn, J. Stenger, S. Inouye, A. P. Chikkatur, W. Ketterle, Observation of Metastable States in Spinor Bose-Einstein Condensates, Phys. Rev. Lett. 82 (1999) 2228. [89] E. V. Goldstein, P. Meystre, Quasiparticle instabilities in multicomponent atomic condensates, Phys. Rev. A 55 (1997) 2935. [90] E. Mueller, G. Baym, Finite-temperature collapse of a Bose gas with attractive interactions, Phys. Rev. A 62 (1999) 053605. [91] J. J. P. Burke, C. H. Greene, J. L. Bohn, Multichannel Cold Collisions: Simple Dependences on Energy and Magnetic Field, Phys. Rev. Lett. 81 (1998) 3355. [92] S. Uchino, M. Kobayashi, M. Ueda, Bogoliubov Theory and Lee-Huang-Yang Correction in Spin-1 and Spin-2 Bose-Einstein Condensates in the Presence of the Quadratic Zeeman Eect, arXiv:0912.0355 . [93] R. W. Cherng, E. Demler, Roton softening and supersolidity in Rb spinor condensates, arXiv:0806.1991 . [94] A. Lamacraft, Long-wavelength spin dynamics of ferromagnetic condensates, Phys. Rev. A 77 (2008) 063622. [95] N. D. Mermin, T.-L. Ho, Circulation and Angular Momentum in the A Phase of Superuid Helium-3, Phys. Rev. Lett. 36 (1976) 594. [96] M. Nakahara, T. Isoshima, K. Machida, S.-i. Ogawa, T. Ohmi, A simple method to create a vortex in Bose-Einstein condensate of alkali atoms, Physica B: Condensed Matter 284288 (2000) 17. [97] T. Isoshima, M. Nakahara, T. Ohmi, K. Machida, Creation of a persistent current and vortex in a Bose-Einstein condensate of alkali-metal atoms, Phys. Rev. A 61 (2000) 063610. [98] A. E. Leanhardt, A. G rlitz, A. P. Chikkatur, D. Kielpinski, Y. Shin, D. E. Pritchard, W. Ketterle, Imprinting o Vortices in a Bose-Einstein Condensate using Topological Phases, Phys. Rev. Lett. 89 (2002) 190403. [99] A. E. Leanhardt, Y. Shin, D. Kielpinski, D. E. Pritchard, W. Ketterle, Coreless Vortex Formation in a Spinor Bose-Einstein Condensate, Phys. Rev. Lett. 90 (2003) 140403. [100] L. E. Sadler, J. M. Higbie, S. R. Leslie, M. Vengalattore, D. M. Stamper-Kurn, Spontaneous symmetry breaking in a quenched ferromagnetic spinor Bose-Einstein condensate, Nature 443 (2006) 312. [101] F. Zhou, Spin Correlation and Discrete Symmetry in Spinor Bose-Einstein Condensates, Phys. Rev. Lett. 87 (2001) 080401. [102] U. Leonhardt, G. E. Volovik, How to Create an Alice String (Half-Quantum Vortex) in a Vector Bose-Einstein Condensate, JETP Lett. 72 (2000) 46. [103] B. Li, X.-F. Zhang, Y.-Q. Li, Y. Chen, W. M. Liu, Dynamical Creation of Fractionalized Vortices and Vortex Lattices, Phys. Rev. Lett. 101 (2008) 010402.

120

[104] T. Isoshima, K. Machida, T. Ohmi, Quantum Vortex in a Spinor Bose-Einstein Condensate, J. Phys. Soc. Jpn. 70 (2001) 1604. [105] N. D. Mermin, d-wave pairing near the transition temperature, Phys. Rev. A 9 (1974) 868. [106] G. W. Semeno, F. Zhou, Discrete Symmetries and 1/3-Quantum Vortices in Condensates of F=2 Cold Atoms, Phys. Rev. Lett. 98 (2007) 100401. [107] T. Mizushima, N. Kobayashi, K. Machida, Coreless and singular vortex lattices in rotating spinor Bose-Einstein condensates, Phys. Rev. A 70 (2004) 043613. [108] W. V. Pogosov, R. Kawate, T. Mizushima, K. Machida, Vortex structure in spinor F=2 Bose-Einstein condensates, Phys. Rev. A 72 (2005) 063605. [109] N. R. Cooper, E. H. Rezayi, S. H. Simon, Vortex Lattices in Rotating Atomic Bose Gases with Dipolar Interactions, Phys. Rev. Lett. 95 (2005) 200402. [110] E. H. Rezayi, N. Read, N. R. Cooper, Incompressible Liquid State of Rapidly Rotating Bosons at Filling Factor 3/2, Phys. Rev. Lett. 95 (2005) 160404. [111] J. Zhang, H. Zhai, Vortex Lattices in Planar Bose-Einstein Condensates with Dipolar Interactions, Phys. Rev. Lett. 95 (2005) 200403. [112] S. Yi, H. Pu, Vortex structures in dipolar condensates, Phys. Rev. A 73 (2006) 061602(R). [113] S. Komineas, N. R. Cooper, Vortex latties in Bose-Einstein condensates with dipolar interactions beyond the weak-interaction limit, Phys. Rev. A 75 (2007) 023623. [114] M. Kobayashi, Y. Kawaguchi, M. Nitta, M. Ueda, Collision Dynamics and Rung Formation of Non-Abelian Vortices, arXiv:0810.5441 . [115] L. Michel, Symmetry defects and broken symmetry. Congurations Hidden symmetry, Reviews of Modern Physics 52 (1980) 617. [116] C. Bruder, D. Vollhardt, Symmetry and stationary points of a free energy: The case of superuid He3, Phys. Rev. B 34 (1986) 131. [117] G. E. Volovik, L. P. Gorkov, Superconducting classes in heavy-fermion systems, Sov. Phys. JETP 61 (1985) 843. [118] M. Ozaki, K. Machida, T. Ohmi, On p-Wave Pairing Superconductivity under Cubic Symmetry, Progress of Theoretical Physics 74 (1985) 221. [119] S.-K. Yip, Symmetry and inert states of spin Bose-Einstein condensates, Phys. Rev. A 75 (2007) 023625. [120] H. M kel , K.-A. Suominen, Inert States of Spin-S Systems, Phys. Rev. Lett. 99 (2007) 190408. a a [121] Y. Kawaguchi, M. Ueda, in preparation . [122] N. D. Mermin, The topological theory of defects in ordered media, Reviews of Modern Physics 51 (1979) 591. [123] V. P. Mineev, Topologically Stable Defects and Solitons in Ordered Media, Harwood Academic Publishers, Australia, 1998, 1998. [124] R. Shankar, Applications of topology to the study of ordered systems, J. Phys. 38 (1977) 1405. [125] G. E. Volovik, V. P. Mineev, Particle-like solitons in superuid 3 He phases, Sov. Phys. JETP 46 (1977) 401. [126] J. Ruostekoski, J. R. Anglin, Monopole Core Instability and Alice Rings in Spinor Bose-Einstein Condensates, Phys. Rev. Lett. 91 (2003) 190402. [127] C. M. Savage, J. Ruostekoski, Dirac monopoles and dipoles in ferromagnetic spinor Bose-Einstein condensates, Phys. Rev. A 68 (2003) 043604. [128] U. A. Khawaja, H. Stoof, Skyrmions in a ferromagnetic Bose-Einstein condensate, Nature 411 (2001) 918. [129] U. A. Khawaja, H. Stoof, Skyrmion physics in Bose-Einstein ferromagnets, Phys. Rev. A 64 (2001) 043612. [130] Y. Kawaguchi, M. Nitta, M. Ueda, Knots in a Spinor Bose-Einstein Condensate, Phys. Rev. Lett. 100 (2008) 180403. [131] L. Faddeev, A. J. Niemi, Stable knot-like structures in classical eld theory, Nature 387 (1997) 58. [132] P. N. eres, D. S. James, Particle vs. pair condensation in attactive Bose liquids, J. Phys. (Paris) 43 (1982) 1133. [133] E. J. Mueller, T.-L. Ho, M. Ueda, G. Baym, Fragmentation of Bose-Einstein condensates, Phys. Rev. A 74 (2006) 033612. [134] T. W. B. Kibble, Topology of cosmic domains and strings, J. Phys. A 9 (1976) 1387. [135] W. H. Zurek, Cosmological experiments in superuid helium?, Nature 317 (1985) 505. [136] W. H. Zurek, Cosmological experiments in condensed matter systems, Physics Reports 276 (1996) 177. [137] D. Scherer, C. Weiler, T. Neely, B. Anderson, Vortex Formation by Merging of Multiple Trapped BECs, Phys. Rev. Lett. 98 (2007) 110402. [138] C. Weiler, T. Neely, D. Scherer, A. Bradley, M. Davis, , B. Anderson, Spontaneous vortices in the formation of BECs, Nature 455 (2008) 948. [139] H. Saito, Y. Kawaguchi, M. Ueda, Kibble-Zurek mechanism in a quenched ferromagnetic Bose-Einstein condensate, Phys. Rev. A 76 (2007) 043617. [140] A. Lamacraft, Quantum Quenches in a Spinor Condensate, Phys. Rev. Lett. 98 (2007) 160404. [141] M. Uhlmann, R. Sch tzhold, U. R. Fischer, Vortex Quantum Creation and Winding Number Scaling in a Quenched u Spinor Bose Gas, Phys. Rev. Lett. 99 (2007) 120407.

121

[142] R. Peierls, Quelques propri t s typiques des crops solides, Ann. Inst. Henri Poincar 5 (1935) 177. ee e [143] V. L. Landau, Zur Theorie der phasenumwandlungen I, Phys. Z. Sowjetunion 11 (1937) 26. [144] N. D. Mermin, H. Wagner, Absence of Ferromagnetism or Antiferromagnetism in One- or Two-Dimensional Isotropic Heisenberg Models, Phys. Rev. Lett. 17 (1966) 1133. [145] P. C. Hohenberg, Existence of Long-Range Order in One and Two Dimensions, Phys. Rev. 158 (1967) 383. [146] L. D. Landau, E. M. Lifshitz, Statistical Physics, 3rd edition part 1, Butterworth-Heinemann, 1980. [147] J. M. Kosterlitz, D. J. Thouless, Ordering, metastability and phase transitions in two-dimensional systems, J. Phys. C: Solid State Phys. 6 (1973) 1181. [148] D. J. Bishop, J. D. Reppy, Study of the Superuid Transition in Two-Dimensional 4 He Films, Phys. Rev. Lett. 40 (1978) 1727. [149] Z. Hadzibabic, P. Kr ger, M. C. B. Battelier, J. Dalibard, Berezinskii2013Kosterlitz2013Thouless crossover in a u trapped atomic gas, Nature 441 (2006) 1118. [150] F. Dalfovo, S. Giorgini, L. P. Pitaevskii, S. Stringari, Theory of Bose-Einstein condensation in trapped gases, Reviews of Modern Physics 71 (1999) 463. [151] M. Kobayashi, Y. Kawaguchi, M. Ueda, Vortex Tiling in a Spin-2 Spinor Bose-Einstein Condensate, arXiv:0907.3716v1 .

122