betaFeedback

WorksPhysics and Astronomy

Transverse momentum transport and collective-mode localization in dipolar tube arrays

Abstract

We study transverse transport of longitudinal momentum in an array of one-dimensional Bose gases coupled by dipolar interactions without particle tunneling. Within zero-temperature Bogoliubov theory, we characterize this transport using the inter-tube momentum correlator and its squared magnitude, termed amplitude and intensity. In a uniform array, pairs of Bogoliubov quasiparticles propagate ballistically across the tubes, while their dephasing produces a t−3t^{-3} decay of the long-wavelength amplitude. Tube-to-tube density imbalance acts as quenched disorder and localizes the transverse collective modes. At fixed longitudinal momentum, this arrests the growth of the intensity profile, whose time-averaged spatial decay measures the localization length. Equal-time correlations are much less sensitive to disorder and primarily reflect the interaction range: in the uniform array they decay with tube separation as r−7r^{-7} for dipolar coupling, compared with exponential decay for nearest-neighbor coupling. Finally, sufficiently strong inter-tube repulsion destabilizes the longitudinally uniform state toward density modulations staggered between neighboring tubes, at a threshold independent of density imbalance.

Introduction

Two-dimensional optical lattices split a quantum gas into an array of parallel one-dimensional tubes, the standard setting for one-dimensional Bose gases [1, 2], from the quantum Newton’s cradle [3] to quantitative tests of generalized hydrodynamics [4] and the observation of hydrodynamization [5]. With magnetic lanthanides [6, 7] the same geometry acquires a long-range dipolar interaction [8, 9], which breaks integrability by a tunable amount [10, 11] and stabilizes strongly correlated states against collapse [12]. The dipolar interaction also acts between tubes. Single-particle tunneling falls exponentially with lattice depth; the dipolar coupling falls only as the inverse cube of the separation. There is therefore a regime in which the tubes exchange energy and momentum but not particles, as observed in dysprosium bilayers through sympathetic cooling and coupled collective oscillations [13]. Each of the NtN_t tubes then keeps its own particle number and its own condensate phase, a U(1)NtU(1)^{N_t} symmetry instead of the single U(1)U(1) of a coupled superfluid.

With tunneling, an array of tubes undergoes a dimensional crossover, or deconfinement transition, into an anisotropic three-dimensional condensate [14, 15]. Without it, the U(1)NtU(1)^{N_t} structure is that of the sliding phases of coupled XY models and Luttinger liquids [16, 17, 18, 19], in which interchain couplings leave each chain’s phase free and modify its exponents. For dipolar bosons in a planar array, Kollath, Meyer and Giamarchi obtained the phase diagram of this sliding Luttinger liquid, with stripe and checkerboard density-wave order at strong intra-tube correlations and a collapse instability at strong dipolar coupling, and proposed the transfer of center-of-mass momentum between tubes as an experimental probe [20]. Momentum transfer without particle exchange is Coulomb drag in condensed matter [21], including between one-dimensional wires [22]; dipolar drag has been proposed for pairs of pancake-shaped clouds [23] and computed for bilayer Fermi gases [24]. Summed over an extended array, the same process is the transverse transport of longitudinal momentum, the quantity that a shear viscosity summarizes in a fluid, studied in quantum gases down to dilute Bose condensates [25]. Prior work on the dipolar tube array concerns its equilibrium phases, the inter-tube contribution to integrability breaking and thermalization [10], and, in the modeling of experiments, the self-consistent mean-field potential that the array exerts on each tube, the tubes otherwise remaining independent [26]. How longitudinal momentum spreads across many tubes in time has, to our knowledge, not been addressed.

Tube occupancies are fixed at loading and vary from tube to tube, through the envelope of the parent cloud and through fluctuations about it [4, 10, 26]. The collective modes of the array couple through the local densities, so these variations are a quenched random potential for the transverse modes even though the Hamiltonian is uniform. Anderson localization [27] of cold atoms has used externally imposed random or quasiperiodic potentials [28, 29, 30], and Bogoliubov quasiparticles in such potentials have been studied along the tube axis [31, 32, 33]; for bosonic excitations, stability further requires every mode energy to be positive [34]. Here the disorder is intrinsic, it acts across the tubes, and it enters as a random density-dependent rescaling of the coupling matrix. The transverse problem is therefore closer to the harmonic chain with random masses [35, 36] than to a particle in a random potential.

We study how longitudinal momentum spreads across an array of tubes coupled by density interactions, and how this dynamics changes when the tube populations are unequal. Within zero-temperature Bogoliubov theory, we examine both the momentum correlator, termed the amplitude, and its squared magnitude, the intensity. In a uniform array the intensity spreads ballistically. Random density imbalance localizes the transverse collective modes and arrests this spreading; at fixed longitudinal momentum, the spatial decay of the time-averaged intensity measures the localization length. Equal-time correlations, by contrast, are only weakly affected by disorder and primarily reflect the interaction range. Section II introduces the model and these observables. Sections III and IV treat the uniform and disordered arrays, respectively. Section V examines the instability at strong inter-tube repulsion, and Sec. VI discusses experimental probes and extensions. Appendix A estimates the experimental time and length scales and the conditions under which tunneling can be neglected (with other Appendices devoted to fleshing out details of calculations).

Models and Bogoliubov theory

Model Hamiltonians

We consider a linear array of NtN_t equidistant Bose tubes with spacing aa Figure 1a, described by bosonic field operators Ψi(x)\Psi_i(x) annihilating a particle of mass MM at position xx along tube ii, with densities n^i=Ψi†Ψi\hat{n}_i=\Psi_i^\dagger\Psi_i. The Hamiltonian is

Array of tubes and dispersion of the uniform array

Figure 1. (a) Array of NtN_t tubes with spacing aa, intra-tube contact interaction UcU_c and inter-tube density coupling Vij(x−x′)V_{ij}(x-x'). (b) Dispersion of the uniform array Eq,k/UcE_{q,k}/U_c against kξk\xi for the dipolar kernel at V=0.3V=0.3: solid, V=0V=0; shaded, the range over transverse qq; dashed, the phonon limit E=n0Uc kE=\sqrt{n_0U_c}\,k and the free-particle limit ϵk+n0Uc\epsilon_k+n_0U_c. Inset: gq(k)g_q(k) across the transverse zone at kξ≈0.6k\xi\approx0.6 (dotted line), against its V=0V=0 value (horizontal line).

H=∑iHi+12∑i≠jHij,(1)\mathcal{H}=\sum_i\mathcal{H}_i+\frac{1}{2}\sum_{i\ne j}\mathcal{H}_{ij}, \tag*{(1)}

where Hi\mathcal{H}_i is the intratube Hamiltonian and Hij\mathcal{H}_{ij} the interaction between distinct tubes,

Hi=∫dx[ℏ22M∂xΨi†∂xΨi+Uc2n^i2(x)],(2)\mathcal{H}_i=\int dx\left[\frac{\hbar^2}{2M}\partial_x\Psi_i^\dagger\partial_x\Psi_i+\frac{U_c}{2}\hat{n}_i^2(x)\right], \tag*{(2)}
Hij=∫dx dx′ Vij(x−x′)n^i(x)n^j(x′),(3)\mathcal{H}_{ij}=\int dx\,dx'\,V_{ij}(x-x')\hat{n}_i(x)\hat{n}_j(x'), \tag*{(3)}

with Uc>0U_c>0 the contact interaction. We consider two inter-tube potentials, nearest-neighbor (n.n.) and dipolar,

Vij(x)={Vδ(x)δ∣i−j∣,1,Vd[x2+a2(i−j)2]−3/2.(4)V_{ij}(x)= \begin{cases} V\delta(x)\delta_{\lvert i-j\rvert,1},\\ V_d\left[x^2+a^2(i-j)^2\right]^{-3/2}. \end{cases} \tag*{(4)}

The dipoles are polarized normal to the plane of the array, so that the dipolar coupling is repulsive and the same for every pair of tubes. The two kernels are normalized to the same nearest-neighbor bond at k→0k\to0,

V=V1(0)=2Vd/a2V=V_1(0)=2V_d/a^2 (see Eq. (14)), and VV denotes this bond for both below.

Single-particle tunneling between tubes, t⊥t_\perp, is set to zero,

t⊥=0;(5)t_\perp=0; \tag*{(5)}

the time and momentum scales below which this is justified in a deep lattice are estimated in Appendix A. We use ℏ=M=1\hbar=M=1, measure energies in units of n0Ucn_0U_c with n0n_0 the mean density, lengths along the tubes in units of the healing length ξ=1/n0Uc\xi=1/\sqrt{n_0U_c}, and set a=1a=1 (for the realization of Appendix A, a≃1.2 ξa\simeq1.2\,\xi). Longitudinal momentum is written kk and transverse quasimomentum qq throughout.

Symmetry sectors and effective disorder

Both models conserve the particle number N^i=∫dx n^i(x)\hat{N}_i=\int dx\,\hat{n}_i(x) of each tube; equivalently H\mathcal{H} is invariant under Ψi→eiαiΨi\Psi_i\to e^{i\alpha_i}\Psi_i for every ii separately. This U(1)NtU(1)^{N_t} symmetry is the one familiar from sliding phases [16, 17, 18, 20]. The Bogoliubov ansatz of Sec. II C breaks it by picking a pattern of αi\alpha_i; nothing we compute depends on the αi\alpha_i. Translation symmetry along the tubes conserves the total longitudinal momentum of the array,

P^=∑iP^i=∑i∫dk k ai,k†ai,k,(6)\hat{P}=\sum_i\hat{P}_i=\sum_i\int dk\,k\,a_{i,k}^\dagger a_{i,k}, \tag*{(6)}

where P^i\hat{P}_i is the momentum of tube ii and ai,ka_{i,k} its Fourier modes. Unlike N^i\hat{N}_i, the individual P^i\hat{P}_i are not conserved: momentum hops between tubes through the inter-tube interaction, and Mij(t)=⟨P^i(t)P^j(0)⟩cM_{ij}(t)=\langle\hat{P}_i(t)\hat{P}_j(0)\rangle_c, Eq. (24) below, is the object that measures this dynamics.

We distinguish (i) the uniform, or clean, array with ni=n0n_i=n_0, and (ii) disordered arrays with independent tube densities,

⟨ni⟩=n0,Var⁡(ni)=σ2n02,(7)\langle n_i\rangle=n_0,\qquad\operatorname{Var}(n_i)=\sigma^2n_0^2, \tag*{(7)}

drawn from the uniform distribution

ni∈[n0−w,n0+w],w=3 σn0.(8)n_i \in[n_0-w,n_0+w], \qquad w=\sqrt{3}\,\sigma n_0. \tag*{(8)}

Positivity of every density requires

σ<13≃0.577.(9)\sigma< \frac{1}{\sqrt{3}} \simeq0.577. \tag*{(9)}

This represents loading fluctuations in an optical lattice. The localization length of the disordered array is written ξloc\xi_{\mathrm{loc}} and its inverse γ≡1/ξloc\gamma\equiv1/\xi_{\mathrm{loc}}.

Bogoliubov theory

We expand Ψi(x)=nieiαi+δΨi(x)\Psi_i(x)=\sqrt{n_i}e^{i\alpha_i}+\delta\Psi_i(x), retain terms quadratic in the fluctuations, and write ai,ka_{i,k} for the Fourier modes of δΨi\delta\Psi_i along the tube. The resulting Hamiltonian is

H=∫k>0dk ψk†H(k)ψk,(10)\mathcal{H}=\int_{k>0}dk\,\psi_k^\dagger H(k)\psi_k, \tag*{(10)}

where

ψk=(a1,k,…,aNt,k,a1,−k†,…,aNt,−k†)T(11)\psi_k=(a_{1,k},\ldots,a_{N_t,k},a^\dagger_{1,-k},\ldots,a^\dagger_{N_t,-k})^T \tag*{(11)}

is the Nambu spinor and

H(k)=(T^+G^G^G^T^+G^),(12)H(k)= \begin{pmatrix} \hat{T}+\hat{G} & \hat{G}\\ \hat{G} & \hat{T}+\hat{G} \end{pmatrix}, \tag*{(12)}

References

  1. [1]T. Stöferle, H. Moritz, C. Schori, M. Köhl, and T. Esslinger, Phys. Rev. Lett. 92, 130403 (2004).
  2. [2]M. A. Cazalilla, R. Citro, T. Giamarchi, E. Orignac, and M. Rigol, Rev. Mod. Phys. 83, 1405 (2011).
  3. [3]T. Kinoshita, T. Wenger, and D. S. Weiss, Nature 440, 900 (2006).
  4. [4]N. Malvania, Y. Zhang, Y. Le, J. Dubail, M. Rigol, and D. S. Weiss, Science 373, 1129 (2021).
  5. [5]Y. Le, Y. Zhang, S. Gopalakrishnan, M. Rigol, and D. S. Weiss, Nature 618, 494 (2023).
  6. [6]M. Lu, N. Q. Burdick, S. H. Youn, and B. L. Lev, Physical Review Letters 107, 190401 (2011).
  7. [7]K. Aikawa, A. Frisch, M. Mark, S. Baier, A. Rietzler, R. Grimm, and F. Ferlaino, Phys. Rev. Lett. 108, 210401 (2012).
  8. [8]T. Lahaye, C. Menotti, L. Santos, M. Lewenstein, and T. Pfau, Reports on Progress in Physics 72, 126401 (2009).
  9. [9]L. Chomaz, I. Ferrier-Barbut, F. Ferlaino, B. Laburthe-Tolra, B. L. Lev, and T. Pfau, Reports on Progress in Physics 86, 026401 (2023).
  10. [10]Y. Tang, W. Kao, K.-Y. Li, S. Seo, K. Mallayya, M. Rigol, S. Gopalakrishnan, and B. L. Lev, Physical Review X 8, 021030 (2018).
  11. [11]K.-Y. Li, Y. Zhang, K. Yang, K.-Y. Lin, S. Gopalakrishnan, M. Rigol, and B. L. Lev, Phys. Rev. A 107, L061302 (2023).
  12. [12]W. Kao, K.-Y. Li, K.-Y. Lin, S. Gopalakrishnan, and B. L. Lev, Science 371, 296 (2021).
  13. [13]L. Du, P. Barral, M. Cantara, J. de Hond, Y.-K. Lu, and W. Ketterle, Science 384, 546 (2024).
  14. [14]A. F. Ho, M. A. Cazalilla, and T. Giamarchi, Phys. Rev. Lett. 92, 130405 (2004).
  15. [15]M. A. Cazalilla, A. F. Ho, and T. Giamarchi, New J. Phys. 8, 158 (2006).
  16. [16]C. S. O'Hern, T. C. Lubensky, and J. Toner, Phys. Rev. Lett. 83, 2745 (1999).
  17. [17]V. J. Emery, E. Fradkin, S. A. Kivelson, and T. C. Lubensky, Phys. Rev. Lett. 85, 2160 (2000).
  18. [18]A. Vishwanath and D. Carpentier, Phys. Rev. Lett. 86, 676 (2001).
  19. [19]R. Mukhopadhyay, C. L. Kane, and T. C. Lubensky, Phys. Rev. B 64, 045120 (2001).
  20. [20]C. Kollath, J. S. Meyer, and T. Giamarchi, Phys. Rev. Lett. 100, 130403 (2008).
  21. [21]B. N. Narozhny and A. Levchenko, Rev. Mod. Phys. 88, 025003 (2016).
  22. [22]V. N. Nazarov and D. V. Averin, Phys. Rev. Lett. 81, 653 (1998).
  23. [23]N. Matveeva, A. Recati, and S. Stringari, Eur. Phys. J. D 65, 219 (2011).
  24. [24]B. Renklioglu, B. Y.-K. Hu, M. Ö. Oktel, and B. Tanatar, Phys. Rev. A 109, 063319 (2024).
  25. [25]E. D. Gust and L. E. Reichl, Physica Scripta T165, 014034 (2015).
  26. [26]Y. Zhang, K. Yang, B. L. Lev, and M. Rigol, Phys. Rev. A 113, 023314 (2026).
  27. [27]P. W. Anderson, Physical Review 109, 1492 (1958).
  28. [28]J. Billy, V. Josse, Z. Zuo, A. Bernard, B. Hambrecht, P. Lugan, D. Clément, L. Sanchez-Palencia, P. Bouyer, and A. Aspect, Nature 453, 891 (2008).
  29. [29]G. Roati, C. D’Errico, L. Fallani, M. Fattori, C. Fort, M. Zaccanti, G. Modugno, M. Modugno, and M. Inguscio, Nature 453, 895 (2008).
  30. [30]L. Sanchez-Palencia and M. Lewenstein, Nature Physics 6, 87 (2010).
  31. [31]N. Bilas and N. Pavloff, Eur. Phys. J. D 40, 387 (2006).
  32. [32]P. Lugan, D. Clément, P. Bouyer, A. Aspect, and L. Sanchez-Palencia, Phys. Rev. Lett. 99, 180402 (2007).
  33. [33]V. Gurarie, G. Refael, and J. T. Chalker, Phys. Rev. Lett. 101, 170407 (2008).
  34. [34]V. Gurarie and J. T. Chalker, Phys. Rev. B 68, 134207 (2003).
  35. [35]F. J. Dyson, Phys. Rev. 92, 1331 (1953).
  36. [36]H. Matsuda and K. Ishii, Prog. Theor. Phys. Suppl. 45, 56 (1970).
  37. [37]T. L. Ho and V. B. Shenoy, Phys. Rev. Lett. 77, 3276 (1996).
  38. [38]J. Sabbatini, W. H. Zurek, and M. J. Davis, Phys. Rev. Lett. 107, 230402 (2011).
  39. [39]R. M. Wilson, C. Ticknor, J. L. Bohn, and E. Timmermans, Phys. Rev. A 86, 033606 (2012).
  40. [40]M. Kappus and F. Wegner, Z. Phys. B 45, 15 (1981).

Paper details

vlnc:W_nd7ej8sp
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 2

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Opus 5.5, Claude Fable 5.1 (bare)

View statement
Looking for related papers…
Contents