Professional Documents
Culture Documents
fermionic repulsion
Javier Argüello-Luengo1 , Manfred J. Mark2,3 , Francesca Ferlaino2,3 , Maciej Lewenstein1,4 ,
Luca Barbiero5 , and Sergi Julià-Farré1
1
ICFO - Institut de Ciencies Fotoniques, The Barcelona Institute of Science and Technology, Av. Carl Friedrich Gauss 3, 08860
Castelldefels (Barcelona), Spain
2
Institut für Quantenoptik und Quanteninformation, Österreichische Akademie der Wissenschaften, Technikerstraße 21a, 6020
arXiv:2308.13375v4 [cond-mat.quant-gas] 11 Mar 2024
Innsbruck, Austria
3
Institut für Experimentalphysik, Universität Innsbruck, Technikerstraße 25, 6020 Innsbruck, Austria
4
ICREA, Pg. Lluı́s Companys 23, 08010 Barcelona, Spain
5
Institute for Condensed Matter Physics and Complex Systems, DISAT, Politecnico di Torino, I-10129 Torino, Italy
Thouless pumping represents a power- area [3, 4]. Topological phases are of great in-
ful concept to probe quantized topologi- terest both from a fundamental viewpoint, as
cal invariants in quantum systems. We they are not captured by the standard Ginzburg-
explore this mechanism in a generalized Landau-Wilson theory [5, 6] of spontaneously
Rice-Mele Fermi-Hubbard model charac- symmetry-broken phases and, potentially, at the
terized by the presence of competing on- technological level, as they represent promising
site and intersite interactions. Contrary tools for quantum metrology [7] and quantum
to recent experimental and theoretical re- computation [8]. While noninteracting topologi-
sults, showing a breakdown of quantized cal matter can be accurately described in terms
pumping induced by the onsite repulsion, of the topological invariant and symmetries of
we prove that sufficiently large intersite in- single-particle bands, such classification becomes
teractions allow for an interaction-induced challenging in the presence of interactions [9].
recovery of Thouless pumps. Our analy-
sis further reveals that the occurrence of In this context, Thouless pumps [10, 11] have
stable topological transport at large in- emerged as a powerful tool to characterize the
teractions is connected to the presence topology of interacting one-dimensional systems
of a spontaneous bond-order-wave in the through real-time dynamics, i.e., without rely-
ground-state phase diagram of the model. ing on static groundstate properties. Specifically,
Finally, we discuss a concrete experimental these pumps represent a one-dimensional reduc-
setup based on ultracold magnetic atoms tion of the IQHE, as they exploit the direct cor-
in an optical lattice to realize the newly in- respondence between quantized particle pumping
troduced Thouless pump. Our results pro- and the topological invariant during cyclic adia-
vide a new mechanism to stabilize Thou- batic dynamics around a topological singularity.
less pumps in interacting quantum sys- This scheme provides a natural theoretical and
tems. experimental framework to study the effect of in-
teractions in topological systems undergoing such
adiabatic transport dynamics. From a theoreti-
1 Introduction cal perspective, recent works showed that topo-
logical quantized transport can survive, and even
Since the discovery of the two-dimensional inte- be induced by many-body correlations in inter-
ger quantum Hall effect (IQHE) [1, 2], the classi- acting bosonic systems [12–19]. In contrast, a
fication of phases of matter through global topo- repulsive onsite Hubbard U interaction [20, 21]
logical invariants has become an intense research or electron-phonon coupling [22, 23] lead to a
breakdown of the quantized Thouless pump in
Sergi Julià-Farré: sergi.julia@icfo.eu
fermionic systems. At the experimental level,
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 1
while initially conceived within the realm of solid- (a) (b)
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 2
(BOW) [45, 50–59], in the U − V groundstate The noninteracting limit of Ĥ corresponds to
phase diagram of the extended Fermi-Hubbard the Rice-Mele model [60], or to the static Su-
model. Finally, we also discuss a concrete ex- Schrieffer-Heegger (SSH) model [61] if chiral and
perimental setup and protocol based on mag- inversion symmetries are further imposed, i.e.,
netic atoms in an optical lattice, to highlight with ∆ = 0. The SSH model exhibits symmetry-
that the newly introduced Thouless pumps in the protected topological and trivial insulators for
regime of competing interactions can be observed δ < 0 and δ > 0, respectively [62]. These phases
in state-of-the-art dipolar quantum simulators. are characterized by a global topological invari-
This article is organized as follows. In Sec- ant, the winding number φ, which has an integer-
tions 2 and 3, we describe, respectively, the quantized value φtopo = 2 (φtriv = 0) in the topo-
Hamiltonian model and numerical methods used logical (trivial) sector [63]. In the presence of
to analyze interacting Thouless pumps. In Sec- inversion symmetry, φ can only change at the
tion 4, we present the results of Thouless pumps singularity δ = 0, where a gap closing associ-
in the presence of solely intersite interactions ated with a topological phase transition |∆φ| = 2
(U = 0). In Section 5, we study the compe- takes place.
tition between intersite and onsite interactions.
Section 6 contains the analysis of the connection
between the quantization of Thouless pumps at The effect of interactions in the ground-state
moderately large interactions and the presence of topological properties of Ĥ in the inversion-
a BOW phase in the phase diagram of a parent symmetric case (∆ = 0) has attracted great in-
model. In Section 7, we discuss an experimental terest recently [63–72]. In a nutshell, the ground
proposal with magnetic atoms, and we conclude state of the system remains topologically pro-
in Section 8. tected at δ < 0 for sufficiently small values of U
and V , as these interactions preserve the protect-
ing symmetries of the static SSH model. How-
2 The Rice-Mele extended Fermi- ever, any finite U (V ) has an impact on the bulk-
Hubbard model edge correspondence of the topological phase [65],
and only the edge modes in the charge (spin) sec-
We consider an extended version of the Rice-Mele tor preserve their topological degeneracy. More-
Fermi-Hubbard model, sketched in Fig. 1(a), de- over, large interactions are in general detrimen-
scribing a chain of length L with N spinful tal to the topology. For sufficiently strong V , the
fermions, labeled by σ =↑, ↓. Here we restrict system features a transition to a charge-density-
to the half-filling case where both N and the wave (CDW), characterized by alternating empty
total magnetization Ŝz ≡ i (n̂i,↑ − n̂i,↓ )/2 are
P
and doubly occupied sites. Such a phase breaks
conserved with N = L and Sz = 0. The full the inversion and chiral symmetries that protect
Hamiltonian reads the topological phase, leading to a trivial band
L h i insulator. Concerning onsite interaction, for in-
J − δ(−1)j (ĉ†j,σ ĉj+1,σ + H.c.)+
X
Ĥ = − creasing U the system evolves continuously to-
j=1,σ=↑,↓
X X X wards a MI. While such a state does preserve the
+∆ (−1)j n̂j + U n̂j,↑ n̂j,↓ + V n̂j n̂j+1 . protecting symmetries, the bulk spin gap van-
j i j
ishes in the limit U → ∞, and the degenerate
(1) spin edge states merge with the bulk modes. Fi-
Here J = 1 parametrizes the NN hopping with nally, notice that Ĥ can also exhibit a sponta-
staggered dimerization strength δ, and ∆ is the neously dimerized insulator at δ = 0, the BOW
strength of the staggered onsite energy. Concern- phase, arising at the transition between com-
ing the interactions, U > 0 accounts for an onsite peting orders. For instance, in the extended
repulsive Hubbard term, while V > 0 describes J −U −V Fermi-Hubbard limit of Ĥ, which is be-
the repulsion between fermions in NN sites. Note lieved to capture the physics of several solid-state
that Ĥ faithfully describes a fermionic dipolar chains, a BOW appears in the phase diagram
mixture trapped in a one-dimensional optical lat- for moderate interactions [45, 50–59], in a finite
tice, in the presence of a superlattice giving rise region between the CDW and the MI, prevent-
to the staggered δ and ∆ terms. ing a direct phase transition between these two
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 3
phases 1 . Interestingly, such a BOW phase has taneous bulk current, that is:
the same properties as the trivial and topological Z t(θ)
ground states of the static SSH model, and there- ∆Qθ = J (t′ ) dt′ , (3)
0
fore it can be viewed as an interaction-induced
topological phase [59]. In this work, we show that where
the abovementioned richness of the ground state i X D E
J (t) = Jj ĉ†L/2+j,σ ĉL/2+j+1,σ − H.c. ,
phase diagram of Ĥ in the SSH symmetric case 2 j∈(0,1),σ
(∆ = 0) also leads to a plethora of dynamical ef- (4)
fects in the Rice-Mele regime (∆ ̸= 0), in which is the current operator averaged over a two-site
the adiabatic breaking of the protecting inver- unit cell in the bulk, and Jj = (J ± δ). The con-
sion symmetry can lead to quantized transport nection between charge transport and the pres-
related to the groundstate topology. In particu- ence of a topological invariant comes from the
lar, we study the resulting interacting Thouless fact that the transferred charge at the end of a
pumps, with a focus on the unexplored effect of cycle, ∆Q2π , can be explicitly written in terms
V and its competition with the onsite term U . of the change in the topological invariant across
the topological singularity [73]
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 4
4 Vanishing onsite interaction U (a) 2 ∆0 = J, V = 0
∆0 = J/4, V = 0
1.5 ∆0 = J/4, V = J
Let us first explore the effect of solely nonlocal
∆Qθ
∆0 = J/4, V = 4J
V interactions, i.e., at U = 0, in a Thouless 1
pump. As depicted in Fig. 1(d), the main ef- 0.5
fect of a moderate V term in our model is to
0
suppress the uniform charge distribution in the 0 π 2π
regions θ = π/2 and θ = 3π/2. This enhances θ
the CDW order, and its associated insulating (b) (c) (d)
4
gap, in cooperation with the ∆ term. A di-
rect consequence is that finite values of V can 3 ∆Ei /3
∆Eis /3
∆E/J
allow one to reduce the time period T required 2
∆0 = J/4, V = 4J
for having adiabatic (and thus quantized) Thou- ∆0 = J/4, V = J
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 5
charge currents unless this symmetry is sponta- and ∆0 = J/2 and different values of V . For
neously broken in a thermodynamic phase transi- a relatively small V = 2J, we still observe in
tion. Such a transition to the spontaneous CDW Fig. 3(a) the absence of transport (yellow dashed
implies an unavoidable gap closing leading to a line), showing that U is still the dominant inter-
dramatic breaking of the adiabatic condition. action in the system. The presence of a MI insu-
lator is further confirmed by the strong reduction
of the internal and spin gap observed in Fig. 3(b).
5 Interplay between onsite and nonlo- However, when V becomes larger, the transport
cal interactions is progressively enhanced and there is a revival of
the quantized Thouless pump in a region around
Let us now consider the combined effect of U V ∼ U/2 [blue lines of Fig. 3(a)]. By observ-
and V , which constitutes the primary motivation ing Fig. 1(d), we can explain such interaction-
of this work, due to the possibility of simulat- induced Thouless pumps from the simultaneous
ing this regime of competing interactions with presence of large U and V . This leads to a com-
magnetic atoms in optical lattices. To this aim, petition between the breakdown caused by the
here we focus on the regime of U ≫ ∆0 by fixing MI [20, 21, 39], and the CDW breakdown dis-
U = 8J and ∆0 = J/2. We note that, in the ab- cussed in the previous section. That is, at such
sence of V , this limit is already quite well under- a large U , the V term promotes double occu-
stood [20, 21, 39]: U brings the system into a MI pancies at θ = π/2 and 3π/2, and transport is
which has a dramatic effect in Thouless pumps, recovered. At the same time, in this regime of
as the suppression of double occupancies prevents strong NN repulsion, the U term prevents any
the breaking of inversion symmetry generated by spontaneous CDW in the initial state at θ = 0.
the ∆ term during the pump at θ = π/2 and The absence of a MI or spontaneous CDW is
θ = 3π/2 [see Figs. 1(c)-(d)]. Figure 3 proves how also confirmed in Fig. 3(c), which shows a ro-
this scenario is modified when considering also bust gap during the pump. For an even larger
the presence of V . In particular, Fig. 3(a) and V ≳ 4.5J, U is not able to prevent the emer-
Figs. 3(b)-(d) show, respectively, the pumped gence of the spontaneous CDW order, and the
charge and instantaneous gaps for fixed U = 8J quantized pump progressively disappears again,
as shown in Fig. 3(a). This effect is due to the
appearance of a phase transition to the twofold
(a) 2 V = 2J V = 4.15J
V = 3.5J V = 5J
degenerate CDW, as observed in Fig. 3(d). Note
1.5 V = 4J V = 8J that this CDW breakdown is analogous to the
∆Qθ
V = 8J
2 quantized transport. The latter is expected, as
V = 4J
1 ∆Ei
the tunneling dynamics do not play any role in
V = 2J
∆Eis the large interaction limit. This leads to a phase
0 diagram for Thouless pumps in the U/J − V /J
0 θ 2π 0 θ 2π 0 θ 2π
plane of the Hamiltonian Ĥ in Eq. (1), for fixed
Figure 3: Effect of intersite interactions in Thouless values of δ0 /J = 0.9, ∆0 /J = 0.5, and T J = 50,
pumps at U = 8J. Here we also fix ∆0 = J/2. (a) sketched in Fig. 1(c).
Evolution of the pumped charged for different values of
V , simulated with the iTEBD algorithm. (b)-(c) First Interestingly, one can also understand the in-
excitation energies of Ĥ(θ) during the pumps for (b) teracting Thouless pumps shown in Fig. 3 and
V = 2J, (c) V = 4J, and (d) V = 8J, computed with Fig. 1(c) from the behavior of the topological
exact diagonalization and L = 8. singularity. To this end, in Fig. 4 we sketch
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 6
detecting the spontaneous BOW phase.
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 7
elusive in cold atoms experiments, despite recent picted in Fig. 5 overcomes the main problems
theoretical proposals [77]. that make it difficult to detect the BOW phase
of the extended Fermi-Hubbard model from di-
To illustrate such a relation more explicitly, in
rect measurements. First, with Thouless pumps
Fig. 5 we plot the breaking of the quantization
the topology of the interacting model (1) can be
in Thouless pumps consisting of C consecutive
probed by measuring the average global charge
periodic cycles, defined as δQ ≡ |∆Q2πC | − 2C,
transport in the system, a topological property
for a fixed U = 8J and as a function of V ,
robust to small imperfections such as local dis-
which corresponds to the dashed line in Fig. 1(c).
order or low finite temperature. In contrast, a
Remarkably, we observe perfectly quantized and
direct measurement scheme of the topological na-
cycle-independent transport, i.e., |δQ| = 0, for
ture of the BOW phase would require, e.g., mea-
Thouless pumps in the region V /J ∈ (3.9, 4.3)
suring nonlocal correlators [59, 78] with single-
around the narrow window of the BOW phase
site resolution through a quantum gas micro-
V /J ∈ (4, 4.15). As sketched in Fig. 1(c), the
scope [79–81]. Similarly, even a direct measure-
same behavior is also observed for other values
ment of the BOW local order parameter requires
of U/J. Two important comments are in order
involved probes capable of measuring the bond
at this point, concerning the quantitative behav-
density in-situ, or superlattice modulation spec-
ior of the curves in Fig. 5. In first place, note
troscopy [77]. Second, the spontaneous nature
that the fact that the BOW region does not co-
of the BOW phase, and its very small intrinsic
incide exactly with the region of quantized pumps
spin gap, require very small temperatures and
is expected. This is because, in the adiabatic dy-
complex state-preparation protocols to avoid the
namics considered here, the staggered field δ (∆)
presence of solitonic excitations. On the con-
used to perform the pump prevents the CDW
trary, in Thouless pumps the spin gap is en-
(MI) in a slightly wider range of V values than
hanced due to the presence of the bond guiding
in the ground-phase diagram of the model in the
field δ, which also lifts the groundstate degener-
absence of such guiding fields [45]. In second
acy of the BOW phase, thus preventing the ap-
place, the nonmonotonic behavior of |δQ| after
pearance of solitons. Third, as shown in Fig. 1(c)
the quantized plateau in Fig. 5 is explained by
and Fig. 5, within experimental errors quantized
the fact that there is a phase transition to a lo-
pumps are observed in a much wider parameter
cal CDW order at V /J ≈ 4.75. At this critical
region than the BOW phase.
point, |δQ| exhibits a local maximum due to the
strong deviation from the adiabatic regime in our
finite-time protocol. The latter also leads to dy- 7 Experimental proposal with mag-
namical results that slighlty depend on the bond
dimension χmax . For V larger than the critical
netic atoms
point, the system reenters the adiabatic regime
A prime candidate for an experimental realiza-
due to the insulating nature of the CDW phase.
tion of the Rice-Mele extended Fermi-Hubbard
Inside this CDW phase, increasing V enhances
Hamiltonian (1) are quantum simulators us-
the gap by moving away from the critical point,
ing ultracold lanthanide atoms in optical lat-
thus favoring adiabatic transport, but it also en-
tices [82, 83]. For erbium, the preparation of
hances the amplitude of the CDW spontaneous
a two-component Fermi gas in a lattice with
order. While the competition between these two
long lifetimes and broad interspin Feshbach res-
phenomena initially leads to the counterintuitive
onances for interaction control has been already
decrease of |δQ| for increasing V , we emphasize
demonstrated [84]. To manipulate the control pa-
that in this CDW regime Thouless pumps do not
rameters ∆ and δ of the Thouless pump, a super-
encircle a topological singularity and cannot be
lattice similar to experiments with neutral atoms
associated to any topological invariant. That is,
can be used, see e.g., Refs. [37, 39]. Such a su-
|δQ| is cycle-dependent and not quantized. Fur-
perlattice can also be used to prepare the initial
ther increasing V /J progressively leads to a max-
state e.g., a dimerized chain along the superlat-
imally polarized CDW phase, suppressing any
tice direction. As the proposed scenario is based
hopping (transport) process during the pump.
on single one-dimensional systems, experimental
As a general remark, note that the scheme de- setups need sufficient spacing between individual
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 8
tubes to avoid strong coupling between the tubes Specifically, in configurations with only inter-
due to the long-range character of the dipole- site repulsive interactions V , we showed that a
dipole interaction. This can be realized either by moderate fermionic repulsion favors the presence
choosing a larger lattice spacing perpendicular to of quantized Thouless pumps. We proved this
the superlattice, or by isolating single stripes in phenomenon to be totally induced by this spe-
a two-dimensional lattice plane via the removal cific interaction which tends to increase the in-
of atoms in between using single-site addressing stantaneous energy gaps and thus stabilize the
of a quantum gas microscope. system topology. In contrast, at large nonlocal
As already discussed in Ref. [59], taking real- interactions, we find an expected breakdown of
istic values currently achievable in experiments the transport due to the appearance of a charge-
should allow one to reach the quantized pump density-wave.
regime with e.g., U = 4J and V = 2J [see When further considering a repulsive onsite
Fig. 1(c)] using erbium in a lattice with 532 nm Hubbard U , which is known to destroy the topo-
unit cell (266 nm short lattice spacing), giving a logical transport [20, 21, 39], we still identified a
maximum interaction strength of V /h ≈ 30 Hz. sizeable region where topological transport per-
This can be further increased by using dyspro- sists. Here the nonlocal repulsion counteracts the
sium and shorter wavelength lattices as planned destructive role of a moderately large local repul-
for future experiments [85, 86], reaching interac- sion and vice versa, and therefore the quantized
tion strengths of up to V /h ≈ 200 Hz. To be transport remains stable. Noticeably, this com-
able to observe the revival it would be preferen- peting mechanism allows for a robust quantized
tial to probe the dynamics with and without V . transport at interactions up to one order of mag-
To probe the dynamics without V , one can either nitude larger than the noninteracting energy, i.e.,
rotate the quantization axis to the magic angle, at U/J ∼ 10.
fully eliminating V , or scale U and J up con-
siderably (V ≪ J, U ) to minimize the influence. Interestingly, this stabilization of topological
The regime V ≫ U can also be achieved, since Thouless pumps through competing interactions
U can be tuned independently by means of Fesh- is reminiscent of the ground state properties
bach resonances [87]. The induced transport due of the parent extended Fermi-Hubbard model.
to the Thouless pump mechanism can be directly There, the presence of moderate and compet-
measured using the center-of-mass displacement ing contact and intersite interactions favors the
of the atoms using either high-resolution in-situ appearance of a spontaneous bond-order-wave
imaging (see e.g. [37]) or a quantum gas micro- phase, whose topological properties have been re-
scope. For J = 2π × 100Hz, the adiabatic condi- cently revealed [59]. Based on this fact, we thus
tion T J = 50 used in our numerical simulations expect that quantized topological transport can
would be achieved for realistic pump periods of take place in different strongly correlated models
half a second. We also expect that the averaged where competing interactions give rise to sponta-
center-of-mass displacement is robust against de- neously generated bond-order-wave regimes [17].
fects appearing during the dynamics, making the Beyond our numerical analysis, we have been
lifetime requirements less stringent. further able to propose a realistic experimental
scheme where our findings can be probed. The
latter relies on trapping an ultracold mixture of
8 Conclusions and outlook magnetic atoms in an optical lattice. The ad-
vantage of using such particles comes from the
We investigated the appearance of quantized fact that they allow for a fine-tuning of both
topological transport, i.e., Thouless pumping, contact and intersite interaction, thus providing
in a Rice-Mele extended Fermi-Hubbard model access to simulate the Hamiltonian of Eq. (1)
where fermions are subject to competing inter- in a wide regime of parameters. Importantly,
actions of different ranges. Contrary to previ- note that these configurations where fermions are
ous analysis showing a detrimental effect of on- subject to local and nonlocal repulsions are the
site repulsive interactions to quantized Thouless most common in solid-state platforms. Moreover,
pumps [20, 21, 39], we unveiled that including whilst we focused on repulsive interactions and
nonlocal repulsions leads to novel phenomena. charge transport, our numerical analysis and ex-
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 9
perimental proposal could be adapted to study of the author(s) only and do not necessarily
the role of attractive terms and spin transport in reflect those of the European Union, European
the same model. We, therefore, believe that our Commission, European Climate, Infrastructure
results can have an important impact on a large and Environment Executive Agency (CINEA),
variety of physical systems, thus representing an nor any other granting authority. Neither the
essential step towards a complete understanding European Union nor any granting authority can
of many-body phases of matter characterized by be held responsible for them. The Innsbruck
nontrivial topological properties. group acknowledges support from a QuantERA
grant MAQS from the Austrian Science Fund
Acknowledgments. The authors acknowledge (FWF No. I4391-N). L. B. acknowledges fi-
fruitful discussions with T. Esslinger, Z. Zhu, A. nancial support within the DiQut Grant No.
Dauphin, F. Heidrich-Meisner, A. A. Aligia, and 2022523NA7 funded by European Union – Next
E. Bertok. The DMRG and TEBD calculations Generation EU, PRIN 2022 program (D.D. 104
were performed using the TeNPy Library [49]. - 02/02/2022 Ministero dell’Università e della
ICFO group acknowledges support from: Ricerca).
ERC AdG NOQIA; MICIN/AEI (PGC2018-
0910.13039/501100011033, CEX2019-000910-
S/10.13039/501100011033, Plan National References
FIDEUA PID2019-106901GB-I00, FPI; MICIIN
with funding from European Union NextGen- [1] K. v. Klitzing, G. Dorda, and M. Pepper,
erationEU (PRTR-C17.I1): QUANTERA Phys. Rev. Lett. 45, 494 (1980).
MAQS PCI2019-111828-2); MCIN/AEI/ [2] D. J. Thouless, M. Kohmoto, M. P. Nightin-
10.13039/501100011033 and by the “European gale, and M. den Nijs, Phys. Rev. Lett. 49,
Union NextGeneration EU/PRTR” QUAN- 405 (1982).
TERA DYNAMITE PCI2022-132919 within [3] M. Z. Hasan and C. L. Kane, Rev. Mod.
the QuantERA II Programme that has received Phys. 82, 3045 (2010).
funding from the European Union’s Horizon 2020 [4] C.-K. Chiu, J. C. Y. Teo, A. P. Schnyder,
research and innovation programme under Grant and S. Ryu, Rev. Mod. Phys. 88, 035005
Agreement No 101017733Proyectos de I+D+I (2016).
“Retos Colaboración” QUSPIN RTC2019- [5] L. D. Landau, E. M. Lifshitz, and
007196-7); Fundació Cellex; Fundació Mir-Puig; M. Pitaevskii, Statistical Physics
Generalitat de Catalunya (European Social (Butterworth-Heinemann, New York,
Fund FEDER and CERCA program, AGAUR 1999).
Grant No. 2021 SGR 01452, QuantumCAT [6] K. G. Wilson and J. Kogut, Phys. Rep. 12,
U16-011424, co-funded by ERDF Operational 75 (1974).
Program of Catalonia 2014-2020); Barcelona [7] K. von Klitzing, Nat. Phys. 13, 198 (2017).
Supercomputing Center MareNostrum (FI-2023-
[8] C. Nayak, S. H. Simon, A. Stern, M. Freed-
1-0013); EU (PASQuanS2.1, 101113690); EU
man, and S. Das Sarma, Rev. Mod. Phys.
Horizon 2020 FET-OPEN OPTOlogic (Grant
80, 1083 (2008).
No 899794); EU Horizon Europe Program
[9] S. Rachel, Rep. Prog. Phys. 81, 116501
(Grant Agreement 101080086 — NeQST), Na-
(2018).
tional Science Centre, Poland (Symfonia Grant
No. 2016/20/W/ST4/00314); ICFO Internal [10] D. J. Thouless, Phys. Rev. B 27, 6083
“QuantumGaudi” project; European Union’s (1983).
Horizon 2020 research and innovation pro- [11] Q. Niu and D. J. Thouless, Journal of
gram under the Marie-Sklodowska-Curie grant Physics A: Mathematical and General 17,
agreement No 101029393 (STREDCH) and 2453 (1984).
No 847648 (“La Caixa” Junior Leaders fellow- [12] E. Berg, M. Levin, and E. Altman, Phys.
ships ID100010434: LCF/BQ/PI19/11690013, Rev. Lett. 106, 110405 (2011).
LCF/BQ/PI20/11760031,LCF/BQ/PR20/ [13] S. Greschner, S. Mondal, and T. Mishra,
11770012, LCF/BQ/PR21/11840013). Views Phys. Rev. A 101, 053630 (2020).
and opinions expressed are, however, those [14] A. Hayward, C. Schweizer, M. Lohse,
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 10
M. Aidelsburger, and F. Heidrich-Meisner, [31] T. Ozawa, H. M. Price, A. Amo, N. Gold-
Phys. Rev. B 98, 245148 (2018). man, M. Hafezi, L. Lu, M. C. Rechtsman,
[15] S. Mondal, S. Greschner, L. Santos, and D. Schuster, J. Simon, O. Zilberberg, and
T. Mishra, Phys. Rev. A 104, 013315 (2021). I. Carusotto, Rev. Mod. Phys. 91, 015006
[16] L. Lin, Y. Ke, and C. Lee, Phys. Rev. A (2019).
101, 023620 (2020). [32] Y. E. Kraus, Y. Lahini, Z. Ringel,
[17] S. Mondal, A. Padhan, and T. Mishra, M. Verbin, and O. Zilberberg, Phys. Rev.
Phys. Rev. B 106, L201106 (2022). Lett. 109, 106402 (2012).
[18] Y. Kuno and Y. Hatsugai, Phys. Rev. Res. [33] A. Cerjan, M. Wang, S. Huang, K. P. Chen,
2, 042024 (2020). and M. C. Rechtsman, Light: Science & Ap-
plications 9, 178 (2020).
[19] A. Padhan, S. Mondal, S. Vishveshwara,
and T. Mishra, “Interacting bosons on [34] M. Jürgensen, S. Mukherjee, and M. C.
a Su-Schrieffer-Heeger ladder: Topological Rechtsman, Nature 596, 63 (2021).
phases and Thouless pumping,” (2023), [35] M. Jürgensen, S. Mukherjee, C. Jörg, and
arXiv:2306.09325 [cond-mat.quant-gas] . M. C. Rechtsman, Nat. Phys. 19, 420 (2023).
[20] M. Nakagawa, T. Yoshida, R. Peters, and [36] M. Lohse, C. Schweizer, O. Zilberberg,
N. Kawakami, Phys. Rev. B 98, 115147 M. Aidelsburger, and I. Bloch, Nat. Phys.
(2018). 12, 350 (2016).
[37] S. Nakajima, T. Tomita, S. Taie, T. Ichi-
[21] E. Bertok, F. Heidrich-Meisner, and A. A.
nose, H. Ozawa, L. Wang, M. Troyer, and
Aligia, Phys. Rev. B 106, 045141 (2022).
Y. Takahashi, Nat. Phys. 12, 296 (2016).
[22] S. Mondal, E. Bertok, and F. Heidrich-
[38] J. Minguzzi, Z. Zhu, K. Sandholzer, A.-S.
Meisner, Phys. Rev. B 106, 235118 (2022).
Walter, K. Viebahn, and T. Esslinger, Phys.
[23] S. Mondal, E. Bertok, and F. Heidrich- Rev. Lett. 129, 053201 (2022).
Meisner, Phys. Rev. B 107, 239903 (2023).
[39] A.-S. Walter, Z. Zhu, M. Gächter, J. Min-
[24] R. P. Feynman, Int. J. Theor. Phys. 21, 467 guzzi, S. Roschinski, K. Sandholzer,
(1982). K. Viebahn, and T. Esslinger, Nat. Phys.
[25] J. I. Cirac and P. Zoller, Nat. Phys. 8, 264 19, 1471 (2023).
(2012). [40] K. Viebahn, A.-S. Walter, E. Bertok,
[26] I. M. Georgescu, S. Ashhab, and F. Nori, Z. Zhu, M. Gächter, A. A. Aligia,
Rev. Mod. Phys. 86, 153 (2014). F. Heidrich-Meisner, and T. Esslinger,
[27] A. J. Daley, I. Bloch, C. Kokail, S. Flanni- “Interaction-induced charge pumping in a
gan, N. Pearson, M. Troyer, and P. Zoller, topological many-body system,” (2023),
Nature 607, 667 (2022). arXiv:2308.03756 [cond-mat.quant-gas] .
[28] E. Altman, K. R. Brown, G. Carleo, L. D. [41] M. Lewenstein, A. Sanpera, and V. Ahufin-
Carr, E. Demler, C. Chin, B. DeMarco, ger, Ultracold Atoms in Optical Lattices:
S. E. Economou, M. A. Eriksson, K.- Simulating Quantum many-body systems,
M. C. Fu, M. Greiner, K. R. Hazzard, Vol. 54 (Oxford University Press, Oxford,
R. G. Hulet, A. J. Kollár, B. L. Lev, 2012).
M. D. Lukin, R. Ma, X. Mi, S. Misra, [42] I. Bloch, J. Dalibard, and W. Zwerger, Rev.
C. Monroe, K. Murch, Z. Nazario, K.-K. Mod. Phys. 80, 885 (2008).
Ni, A. C. Potter, P. Roushan, M. Saffman, [43] P. Sompet, S. Hirthe, D. Bourgund,
M. Schleier-Smith, I. Siddiqi, R. Simmonds, T. Chalopin, J. Bibo, J. Koepsell, P. Bo-
M. Singh, I. Spielman, K. Temme, D. S. jović, R. Verresen, F. Pollmann, G. Sa-
Weiss, J. Vučković, V. Vuletić, J. Ye, and lomon, C. Gross, T. A. Hilker, and I. Bloch,
M. Zwierlein, PRX Quantum 2, 017003 Nature 606, 484 (2022).
(2021). [44] J. Léonard, S. Kim, J. Kwan, P. Segura,
[29] N. R. Cooper, J. Dalibard, and I. B. Spiel- F. Grusdt, C. Repellin, N. Goldman, and
man, Rev. Mod. Phys. 91, 015005 (2019). M. Greiner, Nature 619, 495 (2023).
[30] R. Citro and M. Aidelsburger, Nat. Rev. [45] S. Ejima and S. Nishimoto, Phys. Rev. Lett.
Phys. 5, 87 (2023). 99, 216403 (2007).
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 11
[46] T. Lahaye, C. Menotti, L. Santos, M. Lewen- [67] B.-T. Ye, L.-Z. Mu, and H. Fan, Phys. Rev.
stein, and T. Pfau, Rep. Prog. Phys. 72, B 94, 165167 (2016).
126401 (2009). [68] B. Sbierski and C. Karrasch, Phys. Rev. B
[47] L. Chomaz, I. Ferrier-Barbut, F. Ferlaino, 98, 165101 (2018).
B. Laburthe-Tolra, B. L. Lev, and T. Pfau, [69] L. Barbiero, L. Santos, and N. Goldman,
Reports on Progress in Physics 86, 026401 Phys. Rev. B 97, 201115 (2018).
(2022). [70] N. H. Le, A. J. Fisher, N. J. Curson, and
[48] U. Schollwöck, Ann. Phys. 326, 96 (2011). E. Ginossar, npj Quantum Inf. 6, 24 (2020).
[49] J. Hauschild and F. Pollmann, SciPost Phys. [71] Y.-T. Lin, D. M. Kennes, M. Pletyukhov,
Lect. Notes , 5 (2018). C. S. Weber, H. Schoeller, and V. Meden,
[50] M. Nakamura, J. Phys. Soc. Japan 68, 3123 Phys. Rev. B 102, 085122 (2020).
(1999). [72] A. Montorsi, U. Bhattacharya, D. González-
[51] M. Nakamura, Phys. Rev. B 61, 16377 Cuadra, M. Lewenstein, G. Palumbo, and
(2000). L. Barbiero, Phys. Rev. B 106, L241115
[52] E. Jeckelmann, Phys. Rev. Lett. 89, 236401 (2022).
(2002). [73] D. J. Thouless, M. Kohmoto, M. P. Nightin-
[53] P. Sengupta, A. W. Sandvik, and D. K. gale, and M. den Nijs, Phys. Rev. Lett. 49,
Campbell, Phys. Rev. B 65, 155113 (2002). 405 (1982).
[54] A. W. Sandvik, L. Balents, and D. K. [74] S. R. White, Phys. Rev. Lett. 69, 2863
Campbell, Phys. Rev. Lett. 92, 236401 (1992).
(2004). [75] R. Orús and G. Vidal, Phys. Rev. B 78,
[55] Y. Z. Zhang, Phys. Rev. Lett. 92, 246404 155117 (2008).
(2004). [76] J. A. Marks, M. Schüler, J. C. Budich, and
[56] K.-M. Tam, S.-W. Tsai, and D. K. Camp- T. P. Devereaux, Phys. Rev. B 103, 035112
bell, Phys. Rev. Lett. 96, 036408 (2006). (2021).
[57] S. Glocke, A. Klümper, and J. Sirker, Phys. [77] K. Loida, J.-S. Bernier, R. Citro,
Rev. B 76, 155121 (2007). E. Orignac, and C. Kollath, Phys.
[58] M. Di Dio, L. Barbiero, A. Recati, and Rev. Lett. 119, 230403 (2017).
M. Dalmonte, Phys. Rev. A 90, 063608 [78] L. Barbiero, A. Montorsi, and
(2014). M. Roncaglia, Phys. Rev. B 88, 035109
[59] S. Julià-Farré, D. González-Cuadra, (2013).
A. Patscheider, M. J. Mark, F. Fer- [79] W. S. Bakr, J. I. Gillen, A. Peng, S. Fölling,
laino, M. Lewenstein, L. Barbiero, and and M. Greiner, Nature 462, 74 (2009).
A. Dauphin, Phys. Rev. Res. 4, L032005 [80] M. Endres, M. Cheneau, T. Fukuhara,
(2022). C. Weitenberg, P. Schauß, C. Gross,
[60] M. J. Rice and E. J. Mele, Phys. Rev. Lett. L. Mazza, M. C. Bañuls, L. Pollet, I. Bloch,
49, 1455 (1982). and S. Kuhr, Science 334, 200 (2011).
[61] W. P. Su, J. R. Schrieffer, and A. J. Heeger, [81] T. A. Hilker, G. Salomon, F. Grusdt, A. Om-
Phys. Rev. Lett. 42, 1698 (1979). ran, M. Boll, E. Demler, I. Bloch, and
[62] S. Ryu, A. P. Schnyder, A. Furusaki, and C. Gross, Science 357, 484 (2017).
A. W. W. Ludwig, New J. Phys. 12, 065010 [82] A. Patscheider, B. Zhu, L. Chomaz, D. Pet-
(2010). ter, S. Baier, A.-M. Rey, F. Ferlaino, and
[63] S. R. Manmana, A. M. Essin, R. M. Noack, M. J. Mark, Phys. Rev. Research 2, 023050
and V. Gurarie, Phys. Rev. B 86, 205119 (2020).
(2012). [83] L. Su, A. Douglas, M. Szurek, R. Groth,
[64] V. Gurarie, Phys. Rev. B 83, 085426 (2011). S. F. Ozturk, A. Krahn, A. H. Hébert, G. A.
[65] T. Yoshida, R. Peters, S. Fujimoto, and Phelps, S. Ebadi, S. Dickerson, F. Ferlaino,
N. Kawakami, Phys. Rev. Lett. 112, 196404 O. Marković, and M. Greiner, Nature 622,
(2014). 724 (2023).
[66] D. Wang, S. Xu, Y. Wang, and C. Wu, [84] S. Baier, D. Petter, J. H. Becher,
Phys. Rev. B 91, 115118 (2015). A. Patscheider, G. Natale, L. Chomaz, M. J.
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 12
Mark, and F. Ferlaino, Phys. Rev. Lett. 121,
093602 (2018).
[85] J. Fraxanet, D. González-Cuadra, T. Pfau,
M. Lewenstein, T. Langen, and L. Barbiero,
Phys. Rev. Lett. 128, 043402 (2022).
[86] M. Sohmen, M. J. Mark, M. Greiner, and
F. Ferlaino, SciPost Phys. 15, 182 (2023).
[87] A. D. Lange, K. Pilch, A. Prantner, F. Fer-
laino, B. Engeser, H.-C. Nägerl, R. Grimm,
and C. Chin, Phys. Rev. A 79, 013622
(2009).
Accepted in Quantum 2024-03-04, click title to verify. Published under CC-BY 4.0. 13