Phase Transitions and Applications of a Few

Model Low-dimensional Optical Lattices

Swapan K Pati

Theoretical Sciences Unit

JNCASR, Bangalore
http://www.jncasr.ac.in/pati/
Bose-Einstein Condensation of Bosons

Experimentally first Bose-Einstein Condensation of 87Rb atoms

was shown by laser cooling down to ≈ 20 nanoK.

Momentum distribution of 87Rb atoms at T > Tc , T = Tc and T < Tc .

(Here peak represents a group of atoms with the same velocity)

M. H. Anderson, J. R. Ensher, M. R. Matthews, C. E. Wieman, and E. A.

Cornell, Science, 269, 198 (1995).
2 / 59
 Bose-Einstein Condensation of Fermions
◮ Fermions can also form BEC by making composite Bosons (pairing of
even number of Fermions).
◮ In an optical lattice one can form composite Bosons from Fermionic
atoms through Magnetic Field Feshbach Resonance
◮ By applying external magnetic field B, one can tune the scattering
length of atoms.

Bound and unbound state of Fermions have different total

electron spin state.

Experimentally BEC of Fermions has been shown for

potassium atoms (40K ).

M. Greiner, C. A. Regal, and D. S. Jin, Nature, 426, 537 (2003)

3 / 59
 Science Vol 322 10 October 2008
A High Phase-Space-Density Gas of Polar Molecules
K.-K. Ni, S. Ospelkaus, M. H. G. de Miranda, ....., P. S. Julienne, D. S. Jin, J. Ye
Eenable explorations of a large class of many-body physics phenomena but also
could be used for quantum information processing. We report on the creation of an
ultracold dense gas of potassium-rubidium (40K87Rb) polar molecules.

Nature, Vol 464 | 29 April 2010

Dipolar collisions of polar molecules in the quantum regime

K.-K. Ni, S. Ospelkaus, D. Wang, ...., J. Ye & D. S. Jin

Nature VO L 5 0 1 26 September 2013

Observation of dipolar spin-exchange interactions with
lattice-confined polar molecules

A p-wave centrifugal
barrier for dipolar
Quantum Zeno effect for polar molecules in a 3D lattice collisions between
Fermionic polar molecules
The lattice depths along X and Z are kept at 40Er, whereas the lattice
depth along Y is reduced to allow tunnelling along the Y direction
(zero & moderate E-field).
 Vol 89, Num13 Physical Review Letters 23 2002
Stefano Giovanazzi, Axel Gorlitz, and Tilman Pfau

Tunability of interaction between permanent dipoles in BECs. Even very weak

magnetic dipole coupling in alkali gases can be used to excite collective
modes.
Physical Review Letters 107, 190401 (2011)
Strongly Dipolar Bose-Einstein Condensate of Dysprosium
Mingwu Lu, Nathaniel Q. Burdick, Seo Ho Youn, and Benjamin L. Lev

We report the Bose-Einstein condensation (BEC) of the most magnetic element, dysprosium.

Nature Physics | VOL 13 January 2017

New frontiers for quantum gases of polar molecules

Steven A. Moses, Jacob P., ....., Deborah S. Jin and Jun Ye
Compared to atoms, molecules possess additional degrees of freedom that can be
exploited in fundamental tests, ultracold chemistry, and engineering new quantum
phases in many-body systems. Here, we review the recent progress in creating and
manipulating ultracold bialkali molecules to study quantum gases of polar molecules.
Outline of this Talk

➢ Introduction to the Optical Lattice Systems

➢ Quantum Phases of Hardcore Bosons in two

Coupled Chains

➢ Triangular dipolar Fermion Lattice: Triplet SF

➢ AB field Breaking of electron-pairs in a SC ring

6 / 59
Outline of this Talk

➢ Introduction to the Optical Lattice Systems

➢ Quantum Phases of Hardcore Bosons in two

Coupled Chains

➢ Triangular dipolar Fermion Lattice: Triplet SF

➢ AB field Breaking of electron-pairs in a SC ring

7 / 59
 Optical Lattices

◮ Optical lattice is an artificial crystal of light created by the

interference patterns of counter propagating laser beams.

◮ Atoms can be trapped by optical dipole traps: Vdip (r ) = −d · E

d = induced dipole moment by oscillating laser field (E).

I. I. Bloch, Nature Physics, 1, 23 (2005).

2. M. Greiner, O. Mandel, T. Esslinger, T. W. Hansch and I. Bloch,
Nature, 415, 39 (2002).
8 / 59
 Optical Lattices

◮Optical lattices are an ideal quantum systems to study various

quantum phases of mater.

◮ In an optical lattice, one can easily manipulate and control

parameters, which gives rise to novel quantum phases…..quantum
magnets, supersolids, superfluids, insulators, topological phases, etc…

9 / 59
 Superfluid and Mott Insulator

Superfluid Mott insulator

In superfluid phase, Bosons form a macroscopic wavefunction with

long-range phase coherence; Well defined, but number of atoms
fluctuate in each lattice site.

In Mott insulator phase, each lattice site is filled with a fixed

number of atoms, but the phase remains uncertain.

10 /
 Low Dimensional Lattices
◮ Low-dimensional quantum systems are quite unique, as quantum
fluctuations destroy the true long range order (LRO).

◮ They either show short range order (SRO) or quasi-LRO.

The Correlation functions decay exponentially -> SRO

Algebraically (nonzero up to the system size) QLRO

◮ For low-dimensional systems, mean-field and Fermi liquid

theories do not yield accurate results.

We have developed finite size DMRG codes for Bosonic,

Fermionic and Spin systems, which give accurate results for
Ground state and its properties, Excited states. We also have
developed tDMRG, which are used to study various t-dependent
phenomena.
11 /
Outline of this Talk

➢ Introduction to the Optical Lattice Systems

➢ Quantum Phases of Hardcore Bosons in two

Coupled Chains

➢ Triangular dipolar Fermion Lattice: Triplet SF

➢ AB field Breaking of electron-pairs in a SC ring

12 /
Quantum phases of hardcore Bosons in
two coupled chains: A DMRG study

13 /
 Introduction
◮ Trefzger et al. have looked at polarized dipolar particles in two
decoupled 2D layers, in the presence of repulsive interactions in the
planes and attractive interactions between the two layers.

◮ They have shown the existence of PSS and PSF phases by solving the
effective extended Bose-Hubbard Hamiltonian in the low-energy
subspace of pairs, by means of a mean-field Gutzwiller approach and
exact diagonalization methods.

C. Trefzger, C. Menotti,et al, Phys. Rev. Lett. 103, 035304

(2009) 14 /
 Dipolar Hardcore Bosons
◮ We have considered hardcore Bosons with dipolar interactions
on two coupled one-dimensional chains.

◮ The dipolar interaction can be defined as

(1 -3cos2(θ))
Vd (r ) =
r3

15 /
 The Model Hamiltonian: Hardcore Boson

◮ The effective Hamiltonian of the system, without taking into

account the interchain hopping, can be written as,

where t (hopping) and V (nearest-neighbour repulsion) are intrachain,

whereas U is interchain attractive term along the rungs.

16 /
 Model Hamiltonian: Spin -1/2
◮ Hardcore Bosonic operators can be mapped onto Spin-1/2 operators
(factors missing; exact only for the GS properties).

◮ The resulting final spin Hamiltonian is equivalent to coupled chain of

spin-1/2 XXZ model with ferromagnetic interchain coupling.

17 /
 Method
We considered spin-1/2 at every site. Varied
DMRG cut-off from 250 to 400.

At times, we have carried out calculations with

Bosonic Hamiltonian and have compared the
two.

We compare our DMRG results with exact

diagonalization up to 28 lattice sites.

18 /
 Phase Diagram: Hardcore Bosons
◮ Small values of U and V → Superfluid Phase (SF ).
◮ Small V and large U → Pair-Superfluid phase (PSF ).

◮ Large values of U and V → Density Wave Phase (DW )

DW
1

0.5 SF

PSF
0
0 0.5 1 1.5 2 2.5 3
U
19 /
 Order Parameters
◮ Superfluid Phase (SF ): Power law decay of two-point correlation
function Cα(r ) and divergence of the correlation length, ξα.

◮ Density Wave Phase (DW ): Non-zero values of order parameter OG (L).

◮ Pair-superfluid phase (PSF ): Power law decay of four-point correlation

function P(r ).

20 /
 SF to DW Transition

◮ Cα(r ) shows power law behavior for V < 1.1.

◮ In SF phase, L/ξα is constant
L for all lengths.
◮ SF to DW transition at V = 1.1 ± 0.05 for U = 0.5.
21 /
 SF To DW phase transition (Berezinskii-KT type)

◮ Transition between SF to DW here is BKT type.

◮ KC (L) can be obtained by fitting the C1(r ) with
◮ The transition point is where Kc = 1/2.
\

r
22 /
 PSF to DW Transition

◮For U = 2, P(r ) shows power law behavior upto V = 0.4

± 0.08.

23 /
 Conclusion

◮ Found various phases of hardcore Bosons in two coupled

chains: interchain attraction and intrachain repulsion.
◮ Lower values of U and V → Superfluid Phase (SF ).
◮ Lower values of V and large values of U → Pair-
superfluid phase (PSF ).
◮ Larger values of U and V → DW Phase of Dimers.

B. Pandey, S. Sinha, and SKP, Phys. Rev. B 91, 214432

(2015).
24 /
Outline of this Talk

➢ Introduction to the Optical Lattice Systems

➢ Quantum Phases of Hardcore Bosons in two

Coupled Chains

➢ Triangular dipolar Fermion Lattice: Triplet SF

➢ AB field Breaking of electron-pairs in a SC ring

25 /
Triplet Superfluidity on a Triangular Lattice
with Dipolar Fermions

26 /
 Introduction

◮ At low-T, liquid 3He forms p-wave SFs, where 3He atoms

(or quasi particles) form pairs in spin triplet state.
◮ Chromium based quasi-1D superconductors and strontium
based oxide, Sr2RuO4, are considered to be good
candidates for triplet pairing.

◮ Unconventional SF, particularly with triplet pairing is of

huge interest, due to its intrinsic connection to
topological phases.
27 /
 Introduction
◮ Recent progress in dipolar Fermi gas gives opportunity
to realize Fermionic superfluids with triplet pairing.

◮ In presence of attractive head to tail arrangement of

dipolar interactions, the one and two dimensional dipolar
fermions become unstable and they undergo either col-
lapse or phase separation.

◮ To overcome these difficulties, bilayer system has been

proposed, where dipoles are aligned perpendicular to
the layers, giving more stable paired phases.

A. C. Potter, et al., Phys. Rev. Lett. 105, 220406 (2010)

28 /
 Model Hamiltonian
◮ We consider dipolar Fermions on a two-leg triangular ladder.
◮ The effective Hamiltonian of the system can be written as:

Fermionic DMRG with max=300 to 600.

29 /
 SDW → TSF → CDW Transition
We consider a fixed value of U = 2 and vary Va.
◮ For lower values of Va → SDW phase.

◮ For moderate values of Va → TSF-phase (S = 1 and sz = 0).

◮ At large values of Va → CDW-phase.

30 /
Correlation Functions
 Effect of attractive interaction: Va

◮ For Va < 1.7 → S(r) dominates.

◮ For 1.7 < Va < 2.5 → P(r) dominates.

◮ For Va > 2.5 → C(r) dominates.

32 /
 Effect of interchain hopping t⊥: On SDW Phase

r
◮ The SDW-phase is unstable in presence of t⊥.
◮ Transition to TSF phase at t⊥ ∼ 0.1 (U = 2;Va = 1.6).
33 /
 Effect of Three-body Interaction: W
◮ Due to triangular geometry and dipolar interactions, an
additional three-body interaction term appears in each of the
triangular plaquette. H = Hold + Wi ni ni +1ni +2

With increase in W, correlation function, P(r) increases

Further increase leads to a phase-separated state
34 /
 Effect of Three-body Interaction: W

◮ With increase in W (>0.6), periodic modulation in C (r ) appears

◮ Coexistence of periodic modulation in C (r ) and algebraic decay
of P(r ) shows Fermionic supersolid (TSF+DW) phase for 0.6 <
W < 1.6 ± 0.1 .
35 /
 Conclusion
◮ Quantum phases of dipolar Fermions, at half filling,
on a triangular lattice has been studied in details.

◮ In presence of moderate values of repulsive onsite

interaction and attractive intersite interactions, the
Fermions form spin triplet superfluid phase.

◮ In presence of three body interaction, a Fermionic

supersolid phase appears, where both TSF and CDW
coexist.

B. Pandey and SKP: Phys. Rev. B 95, 085105 (2017)

36 /
Outline of this Talk

➢ Introduction to the Optical Lattice Systems

➢ Quantum Phases of Hardcore Bosons in two

Coupled Chains

➢ Triangular dipolar Fermion Lattice: Triplet SF

➢ AB field Breaking of electron-pairs in a SC ring

37 /
Breakdown of Electron-pairs in a
Superconducting ring: Effect of Electric Field

38 /
 Introduction
➢ In a recent experiment, on exposure to photon flux, breaking of electron-
pairs (Cooper-pairs) of a superconducting Aluminium thin film is shown [1].
➢ In a superconductor, electron can form pairs by coupling with lattice
vibrations (electron-phonon interaction) [BCS theory] .
➢ Electron can also form local pairs (real space pairing) via short-range
attractive potential. (Polaronic or coupling between electrons and excitons)
[2].
➢ The electron pairing can be recognized by flux quantization[3] and by off-
diagonal long-range order [4].
➢ At low enough temperature, superconductors are condensate of electron
pairs which are sensitive towards the external perturbations.

1. P. J. de Visser et al., Nature Communications, 5, 3130 (2014).

2. R. Micnas, et al, Rev. Mod. Phys., 62, 113 (1990).
3. L. Onsagar, Phys. Rev. Lett. 7, 50 (1961).
4. C. N. Yang, Rev. Mod. Phys. 34, 694 (1962).
39 /
 Model Hamiltonian
◮ We consider a finite ring in presence of time dependent
Aharonov Bohm flux, φ(t) = eFLt (AB-flux).
◮ We model the system within attractive Hubbard model U.

40 /
 Method

◮ Exact diagonalization method to get ground state

wavefunction,|ψ(0)>, at t = 0.

◮ Crank-Nicolson’s algorithm to obtain |ψ(t)>.

◮ Using |ψ(t)>, we calculate E(t), J(t) and other

time dependent quantities.

41 /
 Flux Quantization
◮ Depending on the state of the ring, the presence of AB-flux
gives rise to different periodicity of energy (E ).
◮ For non-interacting case, the periodicity is equal to the
extended AB-period, φ(t) = Lφ0.
◮ For finite values of |U|, due to pairing of electrons, the
periodicity becomes halved, φ(t) = Lφ0/2.

U = -2.0, F = 0.0005
U = 0.0, F = 0.0005

0 3 6 9 12
FLt
42 /
 Effect of Electric Field for Different U
◮ The current density J(t) as a function of the applied AB-flux.

43 /
 Effect of F on Pair-Correlation Function
◮ Time evolution of the pair-correlation function is given as:

◮ For large values of F , P(r , t) shows short range behavior.

44 /
 Effect of F on Average Current J
◮ The time-averaged current

With integration over quarter of the extended AB period (φ(t) = L/4)

45 /
 Conclusion

◮ The time-dependent non-equilibrium properties of

superconducting ring has been studied.

◮ The applied field breaks the electron-pairs driving

the system from superconducting to metallic phase.

◮ The required strength of the applied field depends on

the strength of the attractive interaction potential.

B. Pandey, S. Dutta and SKP, J. Phys.: Condens. Matter 28,

195601 (2016).

46 /
 Acknowledgements: Thank you for your attention
Geetanjali, Exciton, Onsagar (Ours),
Sampige, Booruga, Taavare, Mallige (TUE-CMS Clusters)
SERB and J C Bose National Fellowship; KIST and EU for Funding

Collaborators (in this talk)

Postdoc@Paris-Sud

Prof. Subhasis Sinha

(IISER/Kolkata)

Dr. Sudipta Dutta

(IISER/Tirupati)