文献1

426 Vol. 13, No.
2 / June 2021 / Advances in Optics and Photonics Review
Topological photonics in
synthetic dimensions
Eran Lustig1,2 AND Mordechai Segev1,2,3, *
1
Solid State Institute, Technion, Haifa 32000, Israel
2
Physics Department, Technion, Haifa 32000, Israel
3
Electrical Engineering Department, Technion, Haifa 32000, Israel
* Corresponding author: [email protected]
Received December 21, 2020; revised March 30, 2021; accepted March 30, 2021;
published June 1, 2021 (Doc. ID 418074)
Topological photonics is a new and rapidly growing field that deals with topological
phases and topological insulators for light. Recently, the scope of these systems was
expanded dramatically by incorporating non-spatial degrees of freedom. These syn-
thetic dimensions can range from a discrete ladder of cavity modes or Bloch modes of
an array of waveguides to a time-bin division (discrete time steps) in a pulsed system
or even to parameters such as lattice constants. Combining spatial and synthetic dimen-
sions offers the possibility to observe fundamental and exotic phenomena such as
dynamics in four dimensions or higher, long-range interaction with disorder, high-
dimensional nonlinear effects, and more. Here, we review the latest developments in
using non-spatial dimensions as a means to enhance fundamental features of photonic
topological systems, and we attempt to identify the next challenges. c 2021 Optical
Society of America
https://doi.org/10.1364/AOP.418074
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
1.1 Topological Insulators for Electrons . . . . . . . . . . . . . . . . . . 427
1.2 Photonic Topological Insulators . . . . . . . . . . . . . . . . . . . . 429
1.3 Role of Dimensionality and Connectivity in Topological Physics . . . 429
1.4 Historic Development of Synthetic Dimensions in Photonics . . . . . 430
2. Basic Physics of Topology in Synthetic Dimensions . . . . . . . . . . . . . 433
2.1 Topological Phases and Topological Invariants . . . . . . . . . . . . 433
2.2 Richness of Topological Realm . . . . . . . . . . . . . . . . . . . . 435
3. Description and Definition of Non-Spatial Dimensions . . . . . . . . . . . . 436
3.1 Topological Photonics in Parameter Space . . . . . . . . . . . . . . . 437
3.2 Synthetic Space Relying on Modal Ladders . . . . . . . . . . . . . . 441
3.3 Using Time-Bins as Synthetic Dimension . . . . . . . . . . . . . . . 446
4. Future Directions and Vision . . . . . . . . . . . . . . . . . . . . . . . . . 450
Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Disclosures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
References and Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Review Vol. 13, No. 2 / June 2021 / Advances in Optics and Photonics 427
Topological photonics in
synthetic dimensions
Eran Lustig AND Mordechai Segev
1. INTRODUCTION
1.1 Topological Insulators for Electrons
In the past 15 years, topological insulators have been attracting much research interest
in many areas of physics [1–5]. Unlike ordinary materials—conductors, semiconduc-
tors, and insulators—topological insulators can conduct on their edges and insulate
in their bulk, but what makes them so special is that the current on their edge is com-
pletely indifferent to the shape of the material: it is unidirectional (chiral) and robust
against most types of disorder. In fact, these materials are considered to be a phase
of matter of their own, with features dictated by the topological properties of the
conducting and insulating bands that are usually defined on the reciprocal space
[6]. Consequently, the bands of topological insulators are classified according to
their topology—which is an abstract way to count the number of twists, holes, and
other types of geometric features that cannot continuously deform from one to the
other. That is, the allowed states for electrons can be classified through topological
invariants, determined by externally induced fields or internally through interactions
between the particles, and by the structural symmetries in the medium. The first con-
nection between conductance and topology was in the integer quantum Hall effect
(IQH) [7]. In the IQH, a magnetic field applied to a 2D electron gas caused its Hall
conductance to have integer values. It was then realized that the integer values of the
conductance are related to the topology of the allowed states for electrons [6,8,9], and
that the classification of the different bands of states is related to robust edge states
and a topological invariant called the Chern number [10]. In the following years, it
was found that topological insulators can occur even without a net magnetic flux [11]
and, even more surprisingly, without any external fields and without even breaking
time-reversal symmetry. This happens due to the spin of the electrons themselves (the
quantum spin-Hall effect) or via the spin–orbit coupling [4,12,13]. This notion was
considered a major breakthrough some 15 years ago since it meant that the properties
of topological insulators that were thought to be extrinsic can actually be intrinsic in
certain materials. In these intrinsic topological insulators, pairs of edge states related
to different spins propagate in opposite directions, but still each one of them is chiral.
Moreover, these topological edge states do not couple to each other as long as the
time-reversal symmetry is maintained and the gap remains open. In such cases, the
topological protection is dependent, in principle, not only on the gap remaining open
but also on the existence of a symmetry of the system, in this case: time-reversal
symmetry.
These breakthroughs have opened the door for finding many materials with different
topological phases of matter with exotic properties. Due to their unique properties,
topological materials can potentially be utilized in quantum computing [14], spin-
tronics [15], and more. The relation between topological materials and applications
in quantum computing involves additional notions such as fractional Hall states or
Majorana fermions that usually require fermionic interactions and, hence, are beyond
the scope of this review—which deals mostly with photonics. We are now almost four
decades after the discovery of the first topological insulator (the IQH). To understand
the underlying principles, it is instructive to briefly review the basics of topological
428 Vol. 13, No. 2 / June 2021 / Advances in Optics and Photonics Review
insulators. We will do it on an intuitive basis and attempt to view these principles

through the physically observable quantities. In a single sentence, topological insula-
tors are materials that are insulators in the bulk but are perfect conductors on their
edges. The robust conduction on the edge is manifested in the fact that the current
there is lossless even in the presence of disorder or defects, and it does not depend on
the shape of the edge. That is, the current continues to flow without being scattered
into the bulk or being backscattered by local defects, by disorder in the lattice, or by
sharp edges. Due to the lack of backscattering, such a topological edge current is often
viewed as chiral. This property is often called “topologically protected transport,”
and it is illustrated in Fig. 1(a). It is this property that made topological insulators so
important because—apart from the fundamental physics involved—having a mecha-
nism that can bring lossless flow of energy, charges, or information is extremely
important for many applications. The important parameter determining this unique
robustness is the strength of the variations of the potential that would have normally
caused scattering, not the shape nor the position of a particular defect or disorder in
the lattice.
To understand the essence of topological transport, it is instructive to recall the IQH
effect [7]. This phenomenon occurs in semiconductors under low temperature and a
strong magnetic field [Fig. 1(b)]. The Lorentz force opens a bandgap in the dispersion
curve, and the edge states are characterized by a single line crossing the gap in diago-
nal. This oversimplified picture of the quantum Hall effect is sketched in Fig. 1(c),
where the red line marks the dispersion curve of the edge states (in reality, the allowed
states form discrete Landau levels separated by bandgaps), and a is the lattice con-
stant. Notice that there is only one line crossing the gap, without any line crossing
the gap in the opposite direction. The slope of this line provides the group velocity of
any edge excitation (superposition of states on the red line), and it cannot be zero in
any topological insulator. The direction of the magnetic field sets the sign of the slope
(which determines the direction of the edge current), and the strength of the magnetic
field sets the size of the bandgap. Now, if disorder is introduced into this structure, the
bands will be slightly modified, and the slope of the red line will be slightly altered.
But, as long as the strength of the disorder (random variation in the potential) is
smaller than some value, scattering will not couple the edge states to bulk states. The
implication is that disorder weaker than (approximately) the width of the bandgap
will not cause any scattering into the bulk or backscattering [16]. This is the origin of
topologically protected transport in the IQH effect, and its key ingredient is the size
Figure 1
Topological insulators in a nut shell. (a) Topological insulator in two dimensions: a

2D material that is insulating in the bulk but exhibits perfect conduction on the edge.
(b) Simplified sketch of the integer quantum Hall effect, which was the first topologi-
cal insulator. The arrows indicate the cyclotronic movement of electrons induced by
a magnetic field perpendicular to the plane. (c) Simplified dispersion relation for the
quantum Hall effect, with the red line marking the topological edge states.
of the topological bandgap, which determines the degree of topological protection of

transport. More details about the calculation of basic topological invariants are pro-
vided in Section 5. Modern topological insulators rely on the same principles, but the
effects can be caused by a variety of other sources, among them fermionic spin–orbit
coupling, external modulation, and crystal symmetries. These different mechanisms
that can give rise to topological insulators may affect the level and type of protection
of the edge states, making some more limited than others. The full description of all
the different kinds is beyond the scope of this review; therefore, we will mostly use
the simplest kind of topological phases to explain the concepts of topological photon-
ics in synthetic dimensions. However, these concepts apply to all types of topological
insulators. For a thorough derivation on topological insulators, we direct the reader to
one of many reviews that focus on topological insulators and topological phases [17].
1.2 Photonic Topological Insulators

After the discovery of topological insulators, it became evident that topological
phases are not unique to electronic materials. In fact, the topological structure of the
allowed states for electrons giving rise to the robust edge conductance can occur also
for bosonic particles, i.e., photons, phonons, and atoms.
Effectively, the only requirement from a system to present topologically protected
transport is to have a dispersion curve resembling Fig. 1(c), with the edge states
crossing the gap diagonally. Fulfilling this requirement in bosonic systems implies
that topological physics is not unique to fermions and is not in itself a quantum phe-
nomenon. It is a wave phenomenon (because interference is at its heart) that can,
in principle, occur in any wave system—classical or quantum—with a judiciously
engineered potential landscape [18]. The first suggestion for a bosonic topological
insulator was an electromagnetic (EM) system that required breaking time-reversal
symmetry to avoid backscattering [19]. Shortly thereafter, a more concrete idea was
proposed [20]—based on gyro-optics materials where the application of a magnetic
field indeed breaks time-reversal symmetry. This effect is fundamentally weak at
all frequencies above THz; hence, the resultant topological bandgap would be very
small, providing essentially no protection of transport. However, at microwave
frequencies, the effect is strong and opens a large bandgap, and indeed, within a
year, the EM analogue of the IQH effect was demonstrated [21]. At that point, the
challenge was to find a new avenue to bring the concepts of topological insulators
into photonics (optical and near infrared frequencies), without relying on the weak
gyro-optic effects. Several ideas were proposed—ranging from using polarization as
spin in photonic crystals [22] and aperiodic coupled resonators [23] to bianisotropic
metamaterials [24]. Eventually, in 2013, the first photonic topological insulators were
demonstrated [25], and they indeed displayed topological protection of transport
(of light) against defects and disorder. Around the same time, the aperiodic coupled
resonator system was also realized in experiment [26] and a year later demonstrated
topological protection against disorder in the lattice [27]. Within a few years, numer-
ous other EM topological systems were proposed and demonstrated, among them the
topological bianisotropic metamaterials system [28], the so-called “network model”
of strongly coupled resonators [29–31], the crystalline topological insulator [32–36],
and the valley-Hall scheme [37–39]. A contemporary review on photonic topological
insulators can be found in [40,41].
1.3 Role of Dimensionality and Connectivity in Topological Physics

All the topological photonics work described above was on 2D topological systems,
but this is more than a mere coincidence [42–44]. Although a large number of fun-
damentally different topological phases and topological insulators do exist also for
three dimensions [2–4,12,17] or more than three dimensions [45–49], these phases
are very hard to realize in photonic systems. This means that, in all our efforts so far
with traditional photonic lattices and structures, we were always only examining the
tip of the iceberg, in terms of the diversity of topological phases with different proper-
ties. The reason is twofold: First, photons propagate at the speed of light; thus, in the
direction of their propagation, it is hard to induce the complex dynamics necessary
for topological phases in high dimensions. The second reason is the complexity of
fabricating 3D topological structures, such as the 3D structures that were recently
used for demonstrating Weyl points [50–52] and the 3D topological insulator in the
microwave regime using magneto-optic effects [53]. 3D fabrication becomes much
more complicated for higher frequencies in the optical regime, where the length
scales are much smaller (implying that higher resolution in fabrication is crucial) and
magneto-optic effects are very weak. Of course, systems of dimensions higher than
three cannot be fabricated directly and require special approaches.
Another aspect in which most photonic topological structures are limited is their
connectivity. Designing a photonic structure that corresponds to a topological phase
usually requires the construction of a lattice, which is a periodic arrangement of
optical elements in space [54–57]. Generally, lattices play an important role in many
topological insulators, as the symmetries and the parameters of the lattice dictate the
topological behavior directly. Since our world has only three spatial dimensions, and
the coupling between sites is mostly by proximity, this limits the number of possible
topological geometries and consequently possible topological phases [11,58], as
much as it does for the topological insulators found in condensed matter. That is, the
possibility of coupling two sites (or elements) that are far apart from one another is
limited in traditional photonic structures, which typically rely on evanescent coupling.
We stress that this problem of connectivity is not limited to high-dimensional systems
nor to photonics; it is a limitation common to all physical systems in real space and
in any dimension. Thus, a possible way to conclude the discussion so far could have
been that the major limiting factor in experimentally realizing lattice models in high
dimensions and with long-range connectivity is the traditional approach in which
lattice sites are identified by their location in space. Luckily, it turns out that this does
not have to be the case: it is possible to utilize non-spatial degrees of freedom such
as spin, momentum, energy, and time as “lattice degrees of freedom” [59–63]. These
non-spatial degrees of freedom are called “synthetic dimensions,” and they are not
restricted by spatial proximity; therefore, they offer new possibilities in experimen-
tally realizing topological phases that have thus far never been observed. Indeed,
recent years have witnessed the use of synthetic dimensions as a means to realize
novel topological phenomena in photonics, and even promising avenues to utilize this
concept for future applications. In this paper, we review the latest advancements on
topological photonics utilizing synthetic dimensions. We explain the fundamental
science underlying the ideas of synthetic dimensions in photonics, and we describe
the current three leading techniques for realizing synthetic dimensions and the recent
experimental efforts. Finally, based on the current state of the research, we map future
directions and possible applications for synthetic dimensions in photonics, and we
suggest some visionary ideas that may open new avenues in untraditional paths.
1.4 Historic Development of Synthetic Dimensions in Photonics

Using non-spatial degrees of freedom as an auxiliary avenue to observe phenomena
that are usually studied in spatial degrees of freedom can be traced back to works
on Anderson localization and quantum chaos from the early 1980s in the theory of
kicked rotors [64–66]. This approach was suggested later on as a means for exper-
iments in cold atoms [67–70] and also for observing high-dimensional topological
phases [71]. In photonics, however, modulating effectively the dielectric properties

of a medium at frequencies comparable to that of the EM field passing through it (as
in the kicked rotor approach) is very challenging for frequencies of THz and higher.
As a consequence, using synthetic dimensions in photonics has evolved in different
avenues, initially relying on a big advantage that photonic systems have on most
other areas: the ability to fabricate and engineer large photonic systems with elaborate
architectures. However, as the technology evolved, temporal modulation has also
been incorporated in implementing synthetic dimensions in photonics. In light of
the many different experimental approaches known today, it would seem that using
synthetic dimensions to study high-dimensional or more complex topological phases
than what the spatial dimensions allow is a multifaced challenge. However, as we
show in Section 5 below, there is a uniting underlying logic behind all the approaches
for synthetic space topological photonics.
As mentioned earlier, the non-spatial degrees of freedom can take several forms in
photonics. Currently, the implementation of synthetic dimensions in photonics can
be cast into three main classes. The first class, which may be called “parametric
synthetic dimensions,” is using a parameter of the Hamiltonian, which, when var-
ied (either continuously or in discrete steps), can constitute a “parametric synthetic
dimension.” However, if the parameter is constant or is varied adiabatically—it lacks
a kinetic term. The absence of a kinetic term means that there is no transport in this
parametric synthetic dimension. Such systems are simpler to implement compared
to other approaches, and they do allow for observing higher dimensional physics in
lower dimensional systems, which made them a useful tool. For example, by mapping
a 1D lattice to a slice in a 2D model, topological Thouless pumping was demonstrated
in photonics experiments [61], and more recently higher dimensional topological
pumps were demonstrated in both optics and cold atoms [72,73]. In fact, a recent
theoretical proposal suggested systems simulating 6D topological pumps [74]. Other
aspects related to topology, such as Fermi arcs and Weyl points, were also demon-
strated by exploiting parameter space [75]. Using parameters as synthetic dimensions
in topological photonics is explained in Section 2 below.
The second class of non-spatial degrees of freedom may be called “eigenstates lad-
der.” It does have a kinetic term in the Hamiltonian that results in propagation along
the synthetic dimension; hence, it is considerably richer because it facilitates transport
in the synthetic dimensions (which the “parametric synthetic dimensions” simply
do not have). In this “eigenstates ladder” class, the synthetic dimensions can be a
series of states such as spin states, momentum states, or a train of frequencies, which
in principle is a ladder of states. The ladder is an uncoupled lattice, and initially the
degree of freedom associated with the ladder does not have a kinetic term, but—with
an additional resonant external modulation—the states become coupled; thus, the
system becomes endowed with an additional kinetic term in the Hamiltonian [60].
Although modulation for photons can be more complex, and certainly adds complex-
ity (compared to the approach that uses a static parameter as a synthetic dimension),
the fact that the modulation is resonant relaxes the conditions for the strength of
the modulation. What makes this approach very appealing for realizing topological
phases is that a spatially dependent modulation phase is equivalent to imprinting an
effective gauge field, which can act on the waves evolving in the system [62]. This
artificial gauge field can be engineered by choosing the modulations scheme. Indeed,
this approach was used in demonstrating the first topological edge states in cold
atoms. There, the synthetic dimensions were spin states, and the coupling between the
states was implemented with an external laser [76,77]. Soon thereafter, similar ideas
of coupling other degrees of freedom in cold atoms, such as optical lattice clocks [78]
and atomic momentum states [79,80], were used in exploring topological phenomena.
The downside of all of these is that the number of spin states and momentum states
is small, and generally excited states have a short lifetime, which altogether limit the
actual size of the synthetic dimension. But a more recent suggestion—of using modes
of a periodically shaken harmonic trap [81]—could, in principle, overcome those
limitations.
Slightly after the suggestion for implementing topological physics by utilizing syn-
thetic dimensions with cold atom systems [51], several groups proposed using this
concept in photonics. These groups suggested the use of a ladder of resonator modes
[82–84], coupled by external modulation, for inducing photonic artificial gauge
fields. And, more recently, photonic waveguide modes coupled by spatial modula-
tion with a spatially varying phase were employed to demonstrate the first photonic
topological insulator utilizing synthetic dimensions [85]. In a similar vein, magnetic
fields in synthetic dimensions were experimentally demonstrated in microwaves [86]
and on a single resonator with two independent mode ladders in optical wavelengths
[87,88]. These ideas, together with suggestions to exploit synthetic dimensions for
generating higher dimensional topological phases [83,89,90] and high-order topo-
logical phases [91], have recently become promising avenues for future research,
together with novel applications employing topologically protected edge transport for
robust unidirectional frequency conversion [92] and mode-locking an array of laser
emitters [93]. We discuss this approach in Section 2 below. Another advantage in
coupling ladders of modes as synthetic dimensions is the ability to couple modes that
are not close to one another. In this way, using synthetic dimensions can be designed
to introduce long-range coupling, which offers a means not only to enrich the topo-
logical phases [94] but also to increase their fundamental dimensionality [95]. In fact,
using long-range connectivity can be a type of synthetic dimension of its own, as was
proposed already in 2013 [63] in the context of demonstrating high-dimensional opti-
cal solitons, and it was recently used to demonstrate experimentally 4D topological
phases in a circuit implementation [96]. Furthermore, coupling synthetic degrees of
freedom does not necessarily require active modulation. A static geometry can also
induce a topological model utilizing synthetic dimensions by transforming not to an
eigenstate basis but to a non-diagonal basis [97–99].
Finally, the third approach for introducing a synthetic dimension has to do with dis-
crete dynamics in a synthetic space of “time-bins.” This approach employs laser
pulses propagating in a periodic cycle. At each period, each pulse splits into sev-
eral consecutive pulses, such that they arrive to the measurement device at different
times—due to different delay for each pulse. Each time slot of arrival (“time-bin”) is
analogous to a location in space. In this approach, the light changes its location (which
is the time-bin) constantly; therefore, this approach is endowed with a kinetic term
and displays transport in the synthetic dimension. Unlike the first two approaches,
here the propagation is only sampled at discrete times, associated with the steps of a
discrete time quantum walk (DTQW). Quantum walks have been known for almost
three decades to be a useful tool for studying a plethora of quantum and wave phe-
nomena [100–104]. In photonics, DTQWs were experimentally realized with short
pulses in fiber loops [105] and allowed the experimental realization of phenom-
ena such as parity–time (PT)-symmetric lattice [59], Kapitza light guiding [106],
time-reversed light [107], optical diametric drive [108], and more [109–112].
A decade ago, along with the growing interest in topological phases of matter, it
was realized that DTQW can exhibit topological phases, not only in 1D but also
in higher dimensions [113]. At first, 1D topological phases and topological edge
states in DTQW were experimentally demonstrated in photonics in several platforms
[114–118]. The 1D topological phases in DTQW are now being extended to include
exotic phenomena such as non-Hermitian topological edge states [119] and obser-
vation of the topological funneling of light in a non-Hermitian topological DTQW
[120]. In parallel, experimental efforts to realize 2D (or higher dimensional) synthetic
lattice models that can exhibit topological phases are ongoing. Since the basic time-
multiplexing system is zero-dimensional, the basic addition of time-bins converts it
to 1D only. Topological systems with higher dimensions can take several forms: they
can either be further extension of time-bins [121,122], or in combination with other
degrees of freedom such as orbital angular momentum [123].
More recently, time-bins were employed to demonstrate unidirectional topological
edge states in two dimensions. This was done both by extending the dimensionality
of the time-bins to the synthetic dimension [123,124] and by using an additional
synthetic dimension of angular momentum modes [125]. This platform is a promising
tool to combine synthetic gauge fields [124], non-Hermitian physics [59], and other
effects for photons to demonstrate exotic photonic topological phases. We focus on
this method and its topological aspects in Section 2 of this review.
2. BASIC PHYSICS OF TOPOLOGY IN SYNTHETIC DIMENSIONS

2.1 Topological Phases and Topological Invariants
The basic notion underlying a phase transition is that physical systems have discon-
tinuous transitions in their properties—upon changing continuously some parameter
of the system. The transition between different phases is traditionally described
in the Landau framework, where an order parameter of the system changes due to
spontaneous symmetry breaking. In recent years, it became evident that a wide class
of materials undergo phase transitions called “topological phase transitions” that
follow a completely different mechanism. In a topological phase transition, the fun-
damental property that changes abruptly is the topology of the eigenfunction space
of the Hamiltonian. The eigenfunctions are directly deduced from the Hamiltonian,
so accordingly the topology is encoded in the Hamiltonian. Topology deals with
continuous deformations of abstract structures, and consequently their classification
should be according to subspaces in which the structures can deform (transform)
from one class of structures into one another without a discontinuity. For example, a
sphere cannot deform continuously to a torus since at a certain point the sphere will
have to go through a discontinuous transition, which punches a hole into it. Thus,
a Hamiltonian with non-trivial topology (such as the toplogy of the torus) cannot
deform continuously to a Hamiltonian with a trivial topology, without its subspace
of eigenfunctions undergoing some discontinuous transition (example, closing of the
bandgap).
When a material (or a system) with non-trivial topology borders a material with differ-
ent topology (which may be either trivial topology or a different non-trivial topology),
the edge between the materials will host edge states with properties different than
those of either of the two bulks. Since the two materials (one of which can be vacuum)
cannot deform into one another continuously, the deformation occurs only on the
edge, which implies that the edge wave functions must be exponentially localized.
The properties of the edge states are unique to the edge and depend on the dimension-
ality, symmetries, and other attributes of the topological system. For concreteness, let
us study a Hamiltonian H(k) describing a spinless boson on a n-dimensional lattice
material, where k is the momentum quantum number, which depends on the space
dimensionality, i.e., k is an n-dimensional vector. The Hamiltonian H associates a
wave function ψ(k) to each k in the Brillouin zone in each band. Thus, the band in a
given Brillouin zone, together with the wave function, forms a topological structure.
If two materials are in different topological phases, then we cannot continuously
deform the Hamiltonian H(k) from one to the other, without closing the gap—which
manifests a discontinuous change in the topological structure. For example, if our
Hamiltonian describes a 1D dimer (diatomic) lattice, i.e., a 1D lattice that has bonds
with strengths alternating between two values (say, u and v, where u is the bond inside
the unit cell and v is the bond between unit cells), then there can be two possible topol-
ogies (Z2 ) [126], where Z2 symbolizes a set of two integers{0,1}. These two possible
topologies are either a trivial topology (u > v), which maps to the topological struc-
ture in Fig. 2(a), or a non-trivial topology (u < v), which is isomorphic to Fig. 2(b). In
Figs. 2(a) and 2(b), the loop maps to the 1D Brillouin zone of the 1D dimer, and the
phase of the eigenstates maps to the arrow that winds around the loop. In Fig. 2(a), the
arrow does not wind around the loop (trivial), and, in Fig. 2(b), it winds once (non-
trivial). The two structures cannot continuously deform to one another and, therefore,
are topologically distinct (a structure that winds twice can continuously deform to
a structure that does not wind). Mathematically, what classifies the two topological
regimes is a topological invariant calculated on the 1D band. In a 1D dimer lattice, the
invariant is called the Zak phase, given by
Z
d
γ =i ψ(k) ψ(k) dk, (1)
k dk
where the integration is carried out on a 1D band in the Brillouin zone and the Zak
phase is equal to 0 or π (indicating the two possible topological phases). As can
be seen in Eq. (1), the Zak phase is more general than the dimer case and can be
calculated for any 1D lattice. The physical meaning is that, if two 1D lattices with dif-
ferent integer Zak phases border each other, this interface will host an exponentially
localized edge state in the gap. From Figs. 2(a) and 2(b) and Eq. (1), we can actually
understand that the Zak phase calculates the geometric phase acquired by the wave
function when it continuously slides along the Brillouin zone.
Interestingly, the picture looks completely different if we look at different dimensions
or different symmetries. For example, for a 2D square lattice with a perpendicular
magnetic field (the IQH) [10] or other lattices such as the extended Haldane model
[11,58], the topology can potentially be classified not to two subclasses (as in the
1D dimer case) but to infinitely many subclasses identified by integers Z. The topo-
logical structures map to structures of 2D vectors on a torus (band in the Brillouin
Figure 2
(a) and (b) Trivial and topological structures of a dimer lattice, respectively, com-
posed of a 1D loop with a vector on each point. The structures are mappings of the
topology of a band in a 1D dimer lattice. There are only two distinct topological
structures (adapted from [127]). (c) Topological structure of a 2D torus with vector
on each point (the projections of the vectors on the plane is presented). This structure
is a mapping of the topology of a band in a 2D Chern insulator. (d) “Periodic table”
of possible topological phases for quadratic Hamiltonians (see description of the
different symmetry classes in [45]).
zone) [Fig. 1(c)]. If we integrate along a loop that encircles the center, we can see that
the vector can wind and accumulate a geometric phase when completing the loop.
The geometric phase accumulated can only come in integer multiples of 2π, and it
is clear from the image why these cannot deform to one another without a disconti-
nuity. Mathematically and more generally, this is calculated by the Chern number.
The Chern number calculation is also an integral on the band, but—since this is a 2D
band—the integral is carried out over a closed path encircling the Brillouin zone,
I
C = hψ(k)|∇k |ψ(k)i dk, (2)
L
where L is a closed path in the 2D Brillouin zone. The possible topologies can be any
integer Z depending on the Hamiltonian, and this maps to different discontinuities
with different windings around them in Fig. 2(c). We note that the Chern number is
not the only topological invariant used to characterize a 2D topological structure;
rather, we give it here only because it is a commonly used one. The choice of topo-
logical invariants depends on the symmetries and dimensionality of the system, and
the Chern number commonly describes the topological structure in models such as
the IQH or the Haldane model, where the non-trivial topology arises from broken
time-reversal symmetry (driven by the magnetic field, in these cases). Of course, one
can go on and describe topological phases in higher dimensions that are described by
other structures.
2.2 Richness of Topological Realm

The fact that different dimensionality and different symmetries can lead to completely
different topologies raises an important question. What are the possible topological
phases for different dimensionalities and Hamiltonian symmetries?
An answer to this question was given in the form of a periodic table for the different
possible topological insulating phases. First, for quadratic Hamiltonians, the pos-
sible phases were organized according to dimensionality (up to eight dimensions)
and non-spatial symmetries, as shown in Fig. 2(d) [45,128]. The different classes
(CL) in Fig. 2(d) are composed of different combinations of time-reversal sym-
metry (2), particle-hole symmetry (4), and chiral symmetry (5), and the symbols
(Z2 , Z, 2Z, 0) refer to the possible classification for each symmetry class at a certain
dimensionality—for example, the IQH model just described is in two dimensions in
class A (it is defined with no symmetries), so its topological structures can be classi-
fied to an infinite number of topologically distinct phases (for each integer Z). Later
on, this table was extended to include also spatial symmetries [129–132]. Current
theoretical and experimental efforts are dedicated toward finding topological phases
in non-Hermitian systems [133–141] and extending the periodic table [46,142] to
include them, likewise with systems of interacting particles [143–145] and nonlinear
optics [146,147].
Interesting topological phases in high dimensions also occur not only for insulators
but also for semimetals, such as the 5D generalization of the topological Weyl semi-
metal and more [49]. Of course, dimensionality changes dramatically the properties of
the edge states. Just as plain examples, 3D topological insulators include 2D surface
currents as edge states [12], while 4D topological insulators can obtain 3D volume
edge states [96]. These are of course just examples of the many differences that come
about to the different symmetries and topological structures that are made possible by
different dimensions.
Since we live in a 3D world, most systems are naturally limited to two or three spa-
tial dimensions. Hence, it became evident that, in order to experimentally study and
fully understand high-dimensional topological phases or complex topological phases

that cannot be realized in simple lattices, new experimental approaches should be
developed. This was the reason underlying the very large progress on topological
physics in synthetic dimensions.
3. DESCRIPTION AND DEFINITION OF NON-SPATIAL DIMENSIONS

An avenue for realizing physical systems with high dimensionality and new forms of
lattice connectivity is to use non-spatial degrees of freedom as additional dimensions
to the physical system, thus extending the reach of experimental platforms. Examples
of non-spatial dimensions are sequences (either discrete or continuous) such as time,
energy, spatial frequency, spin, or simply some parameter of the system. Thus, if α is
a non-spatial variable associated with the wave function, it has a quantum momentum
kα attached to it, and this pair (α and kα ) can serve as a non-spatial dimension. If the
non-spatial dimension α describes a periodic structures, kα is a quasi-momentum. It is
customary to refer to this non-spatial degree of freedom as “synthetic.” We note that
even if α is a constant parameter, we can still use a notion of an auxiliary Hamiltonian
in which α is a variable, and the real Hamiltonian will be a subspace of this auxiliary
Hamiltonian. As mentioned, the topology is usually defined on the Brillouin zone, so
defining a coordinate kα is a notion that exists in almost all approaches of topology in
synthetic dimensions. An example of an exception to this, which we will not discuss
in detail in this review, is the family of topological phases defined in non-periodic
systems and is not calculated on the bands in Brillouin zone—for example, quasicrys-
tal, amorphous, or fractal topological systems [56,57,148]. In this case, the synthetic
dimensions will only be α, and there is no need for a notion of kα . In such systems,
the Chern number is defined on the real-space structure of the system [rather than the
momentum space definition of Eq. (1)], in the thermodynamic limit of a very large
system (see discussion in [57]).
The idea is to choose the non-spatial dimension wisely, such that the Hamiltonian
H(k, kα ) will describe an extended “Brillouin zone” on the extended momentum
space with n + 1 dimensions (or more generally n + m dimensions, with m being the
number of synthetic dimensions). Thus, the eigenstates of H(k, kα ) and the extended
Brillouin zone can have topological phases associated with more than n dimensions,
which consequently extends the ability of experiments to reach uncharted territories,
some predicted by theory and some that were not even theorized.
One might ask at this point, where can this extra dimensionality come from? The
answer to this question differs from approach to approach, and even sometimes for
different techniques within the same approach. In the case of a synthetic dimension,
which is a true internal degree of freedom, such as the polarization of a photon, the
answer is straightforward—and it is that the extra dimensionality comes from the con-
version of an internal degree of freedom to a spatial degree of freedom. Furthermore,
we need to remember that k represents a Bloch lattice momentum—for example, if
we have a lattice of ring resonators, then the momentum k only represents a discrete
lattice of the position of each ring, but the ring itself has inner spatial complexity (and,
therefore, frequency modes) that can be converted (as we will see in Section 3.2) into
a ladder of discrete states acting as a spatial axis. This discussion is intended to make
clear the obvious reasoning that a synthetic dimension is never “added” to a physical
system. That is why, if a constant parameter is used as a synthetic dimension, it will
only give access to a small part of the synthetic system, at a given time.
After formulating the Hamiltonian as H(k, kα ), with one or more quantum momen-
tum variables that are synthetic (not a momentum of a real spatial axis), the
topological classification and topological invariants are calculated based on this
extended Brillouin zone. For example, the Chern number [Eq. (2)] in a system with
one real spatial dimension k x and a synthetic space dimension kα is calculated in the
following way:
I
C= ψ( k̃) ∇k̃ ψ( k̃) d k̃), (3)
L̃
where L̃ is a loop on the extended Brillouin zone of k̃ = (k x , kα ). Since the implica-

tions of a non-trivial Chern number in the Brillouin zone are edge states in the space
coordinate reciprocal to k̃, when using a synthetic space kα , the edge states reside on
the edge of the synthetic parameter α, which may yield very interesting dynamics
when viewed in the complete real-space experimental setup. These special dynamics
will be discussed in Section 3.2.
In the next sections, we describe the three leading approaches in constructing syn-
thetic dimensions in topological photonics. As it turns out, while achieving higher
dimensionality was the main motivation to use synthetic dimensions at first, it is now
clear that the use of non-spatial dimensions is much more than “just” adding dimen-
sionality. Rather, using synthetic dimensions had led to exploring new phenomena
such as robust unidirectional transport in non-spatial dimensions, non-local inter-
actions, long-range connectivity, and more. Some of these new ideas were shown to
be promising tools not only for basic science but actually also for future technology.
3.1 Topological Photonics in Parameter Space

We begin by describing a synthetic dimension in the form of a static parameter, i.e., a
parameter that does not add a kinetic term to the Hamiltonian. In this approach, we
treat a static parameter as an additional quantum momentum kα , and with it we obtain
an extended Brillouin zone and higher dimensional topological phases. This method
allows for observing phenomena in dimensions higher than the physical dimension of
the system they are constructed in, but the lack of kinetic term implies that there is no
transport along the synthetic dimensions.
A very elegant example of the idea of synthetic dimensions in parameter space is
the use of the 1D Andre–Aubry (AA) model [149]. This is an aperiodic 1D lattice
model with non-interacting particles, but it represents a 2D lattice if one of its param-
eters is treated as the momentum quantum number. As such, the AA model allows
for experimentally observing a phenomenon related to 2D topological effects in a
1D lattice—namely, the quantum Hall topological pumping [61]. Extending this
approach to a 2D lattice, which is the vector product of two 1D AA lattices, has facili-
tated observing experimentally 4D topological phenomena in both photonics and cold
atoms [72,73], and suggestions for observing 6D topological phenomena are now
awaiting to be realized in the lab [74,150]. We will use these examples to explain the
ideas underlying synthetic dimensions in parameter space.
The AA model is given by the Hamiltonian,
X
λc n† c n cos(2πbn + φ) + tc n† c n+1 + h.c ,

H(n) = (4)
n
where c n (c n† ) is the annihilation (creation) operator related to site n, h.c. is the

Hermitian conjugate, and λ, t, φ, b are all parameters of the Hamiltonian setting the
strength of the coupling and on-site energy in each site in the 1D lattice. Physically,
the parameter φ relates to the spatial variation of the coupling in some prescribed
fashion—it is a parameter of the Hamiltonian. If we treat the parameter φ as a quan-
tum number representing the momentum in some auxiliary second dimension m, that
is, if φ ≡ km , then H(kn , km ) becomes the Hamiltonian of a 2D lattice (n, m) with
constant magnetic flux b, where one of its dimensions m has been Fourier transformed
to the momentum coordinate km . Substituting instead of φ, km gives
X
H(n, km ) = λc n† c n cos(2πbn + km ) + tc n† c n+1 + h.c .

(5)
n
Next, we Fourier transform km in our synthetic space Hamiltonian of Eq. (5) and
obtain the 2D Harper model, describing a square lattice with perpendicular magnetic
flux b,
X
H(n, m) = λe i2πbn c n,m
† †
c n,m+1 + tc n,m c n+1,m . (6)
n
The Brillouin zone of our fictitious 2D lattice is a surface in the coordinates (kn , km ),
and due to the magnetic flux b it has a topological invariant—the Chern number
[Eq. (2)].
As mentioned earlier, it is clear that 2D photonic systems are easier to fabricate and
to control experimentally than 3D ones, but this is not the only reason why using
synthetic dimensions increases the experimental possibilities. For example, in the AA
model just described, the magnetic field is now included in the parameter b, which
represents part of the prescribed spatial variation in the coupling constant. Thus, while
effective magnetic gauge fields are very hard to implement for photons in the physical
world because photons are non-interacting particles (and, thus, are not affected by
free-space EM fields), this model allows us to effectively do this simply by changing
the distances between lattice sites.
In the AA model, it is possible to calculate the Chern number in synthetic dimensions
with the following expression:
I
C= ψ( k̃ )|∇k̃ |ψ( k̃) d k̃, (7)
L̃
where k̃ = (kn , km ) = (νn , φ), and L̃ is a closed path encircling the extended Brillouin
zone in synthetic space.
Importantly, the reason to use synthetic dimensions is to observe physical phenomena
that are hard or impossible to realize in a real-space lattice. One such phenomenon,
relevant to the Harper model, is called quantum pumping, and it is deeply related to
the Chern number and edge states. Quantum pumps are the quantized transformation
of wave packets (quantum particles) through an adiabatic process [151,152]. For
photons, the process results in the transformation of all the light from one edge to the
other if the process is adiabatic, but disorder can cause some backscattering [153].
Due to the fact that this approach allows for demonstrating 2D physics with a mag-
netic field on a 1D lattice with just changing distances between sites, it was possible to
experimentally observe, for the first time in photonics, topological pumping of edge
states [55,61]. We will now explain how this was demonstrated experimentally in
more detail. The experiment was done in a transparent dielectric medium with index
of refraction n 0 + 1n(E r ), where 1n(E
r ) represents small variations in the refractive
index: |1n(E r )| n 0 . In this system, the light is propagating in the medium along
the optical axis z, with the linearly polarized field E (E r ) = ψ(E r )e −ik0 z e iωt (where
ω
k0 = c n 0 ). We assume the paraxial approximation, which means that light only
deviates by small angles from the optical axis: | ∂ψ | 2k0 | ∂ψ
2
∂z 2 ∂z
|. Substituting these
expressions into the Helmholtz equation yields the following propagation equation,
known as the paraxial wave equation:
∂ψ 1 2 k0
i = ∇⊥ ψ + 1n(E
r ) · ψ. (8)
∂z 2k0 n0
This equation is equivalent to the Schrodinger equation with two spatial dimensions,
where the optical axis z plays the role of time, nk00 1n(E r ) acts as the potential V (E
r ),
1 2
and 2k0 ∇⊥ is the kinetic term. To observe topological effects in this optical system,
one can use a lattice potential 1n(E r ), which is formed by an array of evanescently
coupled waveguides. Thus, by fabricating a lattice with distances that correspond
to the coupling in Eq. (4) [Fig. 3(a)], the wave dynamics is the same as the temporal
evolution of a single electron in an AA lattice. Scanning φ (which is equivalent to the
momentum in the synthetic space) is translated to scanning the 2D Brillouin zone with
a 1D line. This scanning reveals the unidirectional evolution of edge states associated
with the 2D Harper model system [Fig. 3(b)]. Thus, by adiabatically changing (in
z) the coupling along the propagation axis (equivalent to adiabatically varying the
hopping parameter in time), the photons are transferred (“pumped”) between different
regions of the topological band structure [Fig. 3(b)] and, consequently, move from
one edge of the lattice to the other edge, through the bulk of the lattice [Fig. 3(c)].
The topological pumping (also known as Thouless pumping) is a phenomena that is
not restricted to one type of topological lattice and also not to a single dimension. In
fact, by extending this idea to a 2D physical lattice, which is the vector product of two
1D AA spatial lattices, one can obtain a representation of 4D quantum Hall physics,
described by the Hamiltonian:
X
† †
H= tx c n,m c n+1,m + t y c n,m c n,m+1 + h.c, (9)
n,m
where ti = t˜i + λi cos(2π bi i + φi ), with i = {x , y }. In Eq. (9), each AA model is

written in the off diagonal form [not in the diagonal form of Eq. (5)], but both of
these representations map to the Harper model [55,61]. Here, φi represents two syn-
thetic momenta in two different independent axes {kz , kw }, bi represents the synthetic
magnetic fields in two planes x − z and y − w, and t˜ι , λi are constant parameters.
Thus, the Hamiltonian has four quantum momentums—two real (k x , k y ) and two
synthetic—and it maps to four dimensions. Figure 4 shows a cross section scan of
the 4D Brillion zone along the line φx = φ y , which is equivalent to kz = kw . The 4D
Hamiltonian has a first-order Chern number in each 2D plane, which accounts for the
Figure 3
(a) (b) (c)
(a)–(c) 1D Andre–Aubry lattice exhibiting the physics of 2D integer quantum Hall

effect [61]. (a) Illustration of the 1D waveguide array with adiabatically varying
coupling along the propagation axis. Light is transferred (“pumped”) from one edge
to the other through the bulk of the lattice. (b) Band structure in parameter space, with
φ being the index that changes adiabatically in the propagation axis z [61]. (c) Light
transferred from edge to edge in the system described in (a) and (b).
multiple edge states in Fig. 4, but also a second-order Chern number in the 4D space,
which corresponds to corner states. The explicit calculation of the 4D Chern number
in the synthetic dimension is
∂ Pj ∂ Pj ∂ Pj ∂ Pj

1
Z
Cj =− 2 εµξρσ Tr P j d 4 k, (10)
8π ∂kµ ∂kξ ∂kρ ∂kσ
where P j (k) is the projector onto the subspace below gap j spanned by the extended
Brillouin zone k̃ = {kn , km , kk , kl }. The second Chern number (as the first Chern num-
ber) is related to quantized pumping and edge states.
In this fashion, even though the system describes a plane in the 4D hyper-volume
at each moment, 4D topological pumping can be observed experimentally with a
2D paraxial photonic waveguide array [72,73] at each “instant” of the propagation
along the propagation axis. The light has a 2D band structure, which is a 2D plane in
the 4D band structure in parameter space. Consequently, by changing the coupling
distances, it is possible to topologically pump light along a line in this 4D parameter
space (Fig. 4).
This example is just one out of several experiments relying on synthetic dimensions
in parameter space. In this vein, treating a static parameter as the quantum momentum
was also used to experimentally demonstrate type-III Weyl semimetal [154], and
other parameters (such as magnetic field) were used to show photonic Weyl points
due to broken time-reversal symmetry [51] and exceptional surfaces [155]. While
the experiment described above is based on a space that is a hybrid real-synthetic
space, it is also possible to have a full synthetic parameter space. For example, it was
demonstrated that specific inner layer parameters, p and q of a 1D photonic crystal,
can form a 3D parameter space together with the wavenumber k. This parameter space
has experimentally measurable Weyl points and edge states [75]. A similar approach
was also used to observe exceptional cones in 4D parameter space [156].
To summarize, parametric synthetic dimensions represent a useful tool for demon-
strating topological phases in high dimensions, with the main advantage being that,
unlike other methods, here the complexity does not scale up with the dimensionality
Figure 4
2D lattice exhibiting the physics of 4D integer quantum Hall effect [72]: eigen-
energies of the system for φx = φ y . The orange and red wedges in the dispersion curve
are edge states (there are several due to different momentum in k x − k y ). The black
curves are corner states, which are not wedges in the dispersion curve.
of the topological phenomena as much as it does in other approaches. When a param-

eter represents a dimension, the additional experimental complexity is much smaller
in this parametric method than when using the other methods for adding dimensional-
ity [157]. On the other hand, a synthetic dimension lacking a kinetic term means that
transport is fundamentally restricted to a lower dimensional subspace of the synthetic
space. In the next section, we describe a different method for synthetic dimensions,
in which all the dynamics is available at each instant and support transport in all
dimensions, spatial and synthetic alike.
3.2 Synthetic Space Relying on Modal Ladders

In the previous method described above, the extension of dimensionality was based on
a parameter that played the role of a synthetic dimension. Using a parameter p as the
quantum momentum k implies that the Fourier transform space of p is the kinetic term
∂
∂ p̃
, where p̃ is the “spatial” coordinate of the momentum p. However, static parame-
ters stay constant under Fourier transform, which means that the Hamiltonian in syn-
thetic space usually maps to a slice of a higher dimensional Hamiltonian.
On the other hand, if p—the synthetic dimension—is not a fixed parameter of the
system but a quantity that can take different values depending on the wave function
such as frequency, spin polarization, and momentum, then it can fully map to a spa-
tial coordinate. The reason is that a wave can change these properties dynamically
just as if it changes its location in space. Naturally, a wave does not change such
quantities (as spin or energy) as it changes its location in space, so the technique
we are about to describe focuses on engineering systems that make these quantities
behave as spatial axes. For example, the quantity that we will now use to explain this
concept is the energy of the wave function and, specifically, the energy ladder of a
wave in a harmonic potential. We will now show how the dimensionality of an n-
dimensional lattice of harmonic potentials can be increased by using the energy ladder
of each harmonic potential. Furthermore, we will show how, by properly modulat-
ing the Hamiltonian in the time-domain, topological edge states emerge and the full
dynamics of edge states in the high dimension can be observed.
For the sake of demonstration, we will use the Hamiltonian of a harmonic potential as
in [81], but the principles apply to any kind of bounded potential for any wave (cavity,
ring resonator, waveguide structure, etc.) [82–86,158,159]. Let us first look at a sin-
gle potential well. The Hamiltonian for a harmonic potential can be obtained for pho-
tons by using quadratures, for example, for photons in a cavity or in waveguides with
the paraxial equations, as will be shown in this review. The Hamiltonian for such har-
monic potential is
∞
p̂ 2 1 X
H0 = x + Mω2 x̂ 2 = ω λ|λihλ|, (11)
2M 2 E =0
where p̂ x is the momentum operator in x̂ , M is a mass term, and ω is a fixed param-

eter. |λi represents the equally spaced eigen-energies, which will be used for forming
the synthetic dimension. For H0 , this ladder (1D lattice) of energies is completely
decoupled. In order to couple the energies, we need to add external modulation:
Ĥtot = Ĥ0 + V̂ (t) = Ĥ0 + κ x̂ cos(ω D t + φ), (12)
where ω D = ω − 1 with 1 ω and φ is a constant phase of the modulation. Now, the

state basis |λi is no longer an eigenbasis of Htot since it is the eigenstate of the unper-
turbed Hamiltonian Ĥ0 . However, we would still like to describe Htot in the basis of
|λi. The reason is that |λi is our synthetic coordinate, and we deliberately couple the
ladder of energies so that it will not be a “good” quantum number, but rather a wave
will travel along the λ coordinate as it travels in space, i.e., we would like Htot to have
a kinetic term. As in every basis change, the transition to the basis λ can be done by
a unitary transformation. When transforming q Htot with Û = exp(itω D λ̂) and using
λ
the rotating wave approximation, ω D κ 8Mω , we obtain the following effective
Hamiltonian:
X X
Heff = 1 λ|λihλ| + J λ (|λ − 1ihλ|e iφ + h.c), (13)
λ λ
q
λ
where J λ is the effective coupling between modes and is equal to J λ = κ 8Mω . The
effective Hamiltonian obtained is that of a 1D lattice of modes [Fig. 5(a)] with an
effective electric field 1 in the direction of the synthetic axis, and a phase term in the
coupling: φ. At this stage, the effective electric field, which can be tuned by changing
the size of 1, affects the dynamics along the “modal dimension,” but the phase φ does
not since it can be gauged out. According to the gauge theory, a local phase degree
of freedom of the wave function is equivalent to gauge fields (electric and magnetic
fields) acting on the particles represented by the wave function—in our case, photons.
At this point, the phase φ represents a global phase of the wave function and, thus, has
no impact on the dynamics.
However, this will become completely different upon introducing more harmonic
traps to the system, with a phase difference between one another. Having a phase
φ(Er ) that depends on the location in space (of the harmonic trap) introduces a local
gauge degree of freedom, which translates to gauge fields according the gauge theory.
Indeed, by choosing different φ(E r ), it is possible to induce effective EM fields in the
hybrid modal-spatial space. To see this in a simple form in our derivation, we couple
such resonators with each other via proximity, in a spatial lattice with n dimensions.
The coupling is resonant (hence, efficient) when phase matching is satisfied; that is,
each mode is coupled to the same mode in its neighboring sites.
The idea is then to have an n + 1 dimensional lattice, where the extra dimension is
the ladder of modes associated with each site in the spatial lattice. Formally, this is
achieved by a basis transformation of the Hamiltonian of the n-dimensional lattice
of harmonic oscillators Hn D,tot with the unitary transformation, V = ⊗s U , which
Figure 5
Mode ladder of a harmonic potential. The coupling between modes is achieved by

modulation. (b) A 1D chain of equally spaced harmonic traps form a 2D synthetic
space with one spatial dimension and one modal dimension. The modulation in each
site couples the modal degree of freedom, and the phase difference results in magnetic
flux. (c) and (d) Examples of realizations of the concept of using a ladder of modes
as a synthetic dimension: (c) waveguide array structure with array modes coupled by
oscillations in the propagation direction, thereby forming a synthetic dimension [85];
(d) an optical resonator that forms a ladder of modes coupled by temporal modulation,
hence forming a synthetic dimension in frequency space.
is the tensor product of U with itself over the spatial dimension s . The resulting
Hamiltonian is H(n+1)D,eff = V † Hn D,tot V .
This recipe, which can be applied to any kind of arrangement of sites with inner
modes, results in the following Hamiltonian for an n + 1 dimensional lattice:
X
H(n+1)D,eff = 1s λ|λ, s ihλ, s | + J λ (|λ − 1, s ihλ, s |e iφs + h.c)
λ,s
+ J s ,λ (|λ, s − 1ihλ, s | + h.c), (14)
where J s ,λ is the spatial coupling between mode λ in sites s and s − 1, and is depen-
dent on the geometry of the lattice, and φs , 1s are the phase term φ and frequency bias
term 1, which are dependent on the modulation phase and frequency at each site s .
Importantly φs , 1s form effective gauge fields (magnetic and electric) in the synthetic
dimension, which can be controlled by adjusting a different modulation in each site.
For example, a 1D system s = {1, . . . m}, with φs = s π2 φ and 1s = 0, corresponds to
a 2D lattice with a perpendicular magnetic field with flux π/2 in each unit cell, and 0
electric field since 1s = 0 [Fig. 5(b)]. With these parameters, Eq. (14) becomes the
Hamiltonian of the IQH, which is a 2D model that is now realized on a 1D spatial
lattice. This idea was demonstrated experimentally in two different platforms in pho-
tonics (spatial modes of an array of waveguides and frequency modes of a temporally
modulated ring resonator [85,88]), in microwaves [86] and also, before that, in cold
atoms with spin modes [76,77] and momentum modes [79].
We will now explain how this was realized in photonics and what the implications are
of such realizations on physical photonic systems. We go back to the physical system
of light propagating in a paraxial waveguide array (as discussed in Section 3.1). In
order to have synthetic dimensions in modal space, the first step is to have a confined
structure with a ladder of modes [examples in Figs. 5(c) and 5(d)]. This confined
structure plays the role of the harmonic trap as in the example above. In direct anal-
ogy to the harmonic trap example, it is possible to create a harmonic potential by
changing the index refraction 1ε(E r ) in a dielectric material. Conveniently, instead of
encoding a continuous harmonic potential in the permeability 1ε(E r ), it is possible to
have a potential made of discrete (single-mode) waveguides at varying distances, so
as to obtain the equally spaced modal structure of a harmonic potential. This lattice
is called a J x lattice, and in this lattice the distances between adjacent waveguides
increase as a function of their distance from the lattice center [Fig. 6(a)] [160]. As
a harmonic potential, the J x lattice has equally spaced modes that function as the
synthetic space. In order to couple the modes, an external perturbation is needed
[Eq. (12)]. The perturbation itself can take many forms. One option is to oscillate the
waveguide spatially at exactly the longitudinal spatial frequency that separates the
different modes [Fig. 6(b)]. Thus, this J x structure of a waveguide lattice functions
as a harmonic trap, and it can be placed near other identical J x lattices and obtain a
Hamiltonian as in Eq. (14). For example, placing the J x lattices near each other in
a row such that they are evanescently coupled [Fig. 6(c)] corresponds to a 2D syn-
thetic space with two dimensions: one spatial dimension (the location of the J x lattice
along x axis) and one synthetic dimension (the mode of each J x lattice). In the 1D
row of different J x lattices, each J x lattice oscillates at a different phase according to
φs = s π2 φ, where s counts the J x lattice along the x axis. A phase shift in φs that breaks
the z-reversal symmetry in the paraxial equation is equivalent to an effective magnetic
field [see Eq. (14)] and induces the formation of edge states in the synthetic space
lattice. Importantly, just as for the Harmonic traps, it is possible also to place the J x
photonic topological insulators in a geometry of a 2D lattice (for example, a diamond
lattice) and obtain lattices in three dimensions [85,161] with a 2D lattice geometry.
To understand the dynamics of the propagation of light on the edge of a space with
one spatial dimension and one synthetic dimension, we describe a wave packet that
propagates along the edge state in synthetic dimension of the waveguide structure
described.
In practice, the direct meaning of this in photonics is that we can design pho-
tonic systems with a topologically protected transport that is not on the spatial
edge of the lattice. Rather, the protected transport can be in the bulk of the lattice
instead of the edge [85], in the frequency dimension [84,88], or in both dimensions
simultaneously—for example, in a spiral motion [89]. To understand what it means,
let us analyze the propagation of light in an edge state in the modal dimension, as
illustrated by Fig. 6(c).
A single edge state in the synthetic space of Fig. 6 encircles the perimeter of the 2D
synthetic structure shown in Fig. 6(c). Thus, a wave packet propagating along the
edge in this synthetic space comprises a superposition of edge states and takes the
following course. In Stage A in Fig. 6(c), the wave packet propagates to the left in
the lowest mode of each J x lattice; a snapshot of the propagating wave packet in real
space is given in Fig. 6(d). Note that the lowest mode is the edge in synthetic space,
but not the edge in real space, as seen in Fig. 6(d). In fact, at this point the synthetic
space edge state is extended on the entire real-space lattice. This, however, does not
change the fact that its transport is topologically protected [162], since the edge state
is still a state crossing the gap exactly as an edge state in the IQH [see Fig. 1(c)]. When
this state reaches the leftmost J x lattice of the array, it does not scatter backward but
Figure 6
(a) J x waveguide array with equally spaced spatial modes. The modes are separated
by a longitudinal spatial frequency . (b) The same J x lattice as in (a), but here it
oscillates in z, at a frequency equal to . The oscillation couples the modes, thereby
creating a 1D lattice in synthetic space. A wavepacket initially populating the lowest
mode is efficiently transferred to the second mode by the modulation, then to the third
mode, etc. (c) A row of J x lattices (columns) and above it the 2D picture in synthetic
dimensions, where the horizontal axis is the location of every J x lattice (column) and
the vertical axis is the mode in each J x lattice. The arrows indicate the directionality
of the topological edge state in the synthetic dimension. (d) and (e) The real-space
intensity of the light when propagating in the structure: (d) is on the bottom edge [see
A in (c)], so the wavepacket populates a single lower mode (extended) on many J x
lattices. (e) The wavepacket at instant B in the propagation—on the leftmost edge.
In this case, the light is in a coherent superposition of modes on the leftmost J x lat-
tice (concentrated due to interference between the modes). The frame of the lattice
has a wave shape since the columns oscillate at a different phase along the vertical
real-space dimension (adapted from [85]).
rather concentrates entirely on the leftmost J x and starts ascending in the modes unidi-
rectionally (Stage B). At this point, the wave packet is in a superposition of modes that
appears as a concentrated wave packet [Fig. 6(e)]. After ascending to the top mode,
the light propagates to the right, while remaining strictly in the top mode at Stage C.
Finally, the light descends to the lowest mode on the rightmost J x lattice. In doing
that, the wave packet has completed a full cycle of transport around the synthetic
space edge.
As mentioned earlier, robust topological propagation in non-spatial dimensions is
by itself interesting even before considering the dimensionality increase. In these
systems, the light performs a robust unidirectional ascent (or descent) on the modal
ladder, which can potentially be used in future applications. One example of a poten-
tial application of synthetic space systems relying on modal ladders is the recent
proposal for a mode-locked topological insulator laser [93]. In short, a topological
insulator laser is an array of emitters (typically semiconductor lasers) judiciously
positioned on a topological photonic platform, so as to force locking of all the emit-
ters associated with the topological mode, making them act as a single high-power
coherent laser source [41,93,139,140,163–167].
The topological photonic platform can be any of the 2D platforms described above,
where each site is replaced by a laser emitter (e.g., a microring laser), with pumping
provided preferentially to the edge mode. What makes the emitters lock together
is two features of topological edge modes: (1) the non-zero velocity, implying that
the light in the edge mode must travel between all emitters associated with the edge
mode and (2) the topological robustness, which facilitates coupling between emitters
despite some inevitable level of imperfections or disorder. Thus far, however, all the
settings in which topological insulator lasers were demonstrated were real-space topo-
logical platforms, and they were geared toward coherent locking of many emitters to
a single frequency. But a recent proposal has targeted a different goal: locking many
emitters in a train of equally spaced frequencies for the purpose of generating mode-
locked laser pulses from a large array of semiconductor lasers [93]. The necessity of
locking a train of frequencies naturally suggested the use of synthetic dimensions,
with the frequency eigenmodes of the laser resonators acting as the modal dimension.
Let us describe that system in some detail. Considering a row of coupled ring res-
onators [83,84], we can treat each ring resonator in the same way we did with the
harmonic trap. The ring resonator also has a ladder of equally spaced resonance
frequencies that can be coupled by electro-optic phase modulation [Fig. 7(a)]. This
turns each ring resonator into a 1D lattice of modes. Then, as before, a row of ring
resonators represents a 2D lattice with one spatial dimension—the location of each
ring—and one synthetic dimension—the frequency mode of each ring. This princi-
ple was recently demonstrated on a single ring with two ladders (each frequency is
actually degenerate to two modes—clockwise and anticlockwise), and it was shown
experimentally that this structure can have synthetic magnetic fields [88] and even
exhibit non-Hermitian physics [168]. Each ring can be made to lase by adding satu-
rable gain, and together with the electro-optical modulation a ring laser can be made
to mode-lock and emit a train of mode-locked pulses. However, if we would like
several such rings to mode-lock together, it is virtually impossible to synchronize all
the ring resonators to lock on the same train of equally spaced frequencies. This is
simply because coupling just two rings causes the splitting of each mode, resulting
in two sets of equally spaced frequencies. In this vein, having N rings creates N sets
of frequencies, and it becomes virtually impossible to get all rings to synchronize on
the same set of frequencies, as required for coherent emission of mode-locked pulses.
In fact, the largest number of semiconductor emitters that was ever made to mutually
mode-lock is 10 [169]. Here is where the concept of topology in synthetic dimensions

gets into play. If the lasing mode is an edge state engulfing several rings, it may force
them to mode-lock together without dephasing due to the topological protection. Such
structure was suggested in [93] and has an elaborated pattern for the gain in each ring
[Fig. 7(b)].
Another avenue where synthetic dimensions can have a large impact is in the world
of nonlinear optic systems [81,170,171]. Since experimentally studying interacting
systems in dimensions higher than three is physically impossible (at least for the time
being), facilitating the ability to do so in synthetic space photonics can have major
impact on fundamental issues. Nonlinear physics in high dimensions is one of those
fields [170,172]. As a pioneering example, in [170], it was proposed to have ring
resonators with χ (3) nonlinearity and modulated group velocity dispersion (GVD)
[Fig. 7(c)]. In this case, the χ (3) nonlinearity maps to local interactions in synthetic
space, allowing to simulate interacting quantum many-body systems resembling the
local interactions of electronic condensed matter systems.
The experimental demonstration of nonlinear frequency conversion to produce high-
dimensional physics with gauge fields in nonlinear waveguides is another promising
realization of the modal dimension experimental approach [173], and quantum inter-
actions of photons on nonlinear frequency combs were also suggested in the context
of four wave mixing [171].
Other suggestions for utilizing the modal dimension approach focus on extending
the dimensionality of topological systems. Some of these suggestions were never
demonstrated experimentally in any system. As mentioned, by using both the extra
lattice dimension [83] and the long-range coupling [174,175] in the modal dimension
approach in the linear and nonlinear regimes, it is possible to experimentally study
high-dimensional topological phases and topological phases that are hard to realize
in ordinary lattices. Several suggestions for models, such as 3D Weyl points [90], 3D
PT symmetric models [53], Haldane model [175]. Likewise, high-order topological
insulators [91], evolution invariant synthetic space topological insulators [98], and
more, are waiting to be realized experimentally with the modal dimensions approach.
3.3 Using Time-Bins as Synthetic Dimension

In the previous approaches, the synthetic dimension was either a parameter of the
Hamiltonian (such as the coupling strength) or a variable of the wave function such
as spin or frequency, which were exploited for generating a modal ladder acting as a
Figure 7
(a) Edge state in the synthetic modal space of a row of phase modulated ring res-
onators. (b) A chain of modulated ring resonators with selective saturable gain that
forms an edge state. This causes a topological edge state to lase given robust proper-
ties to the lasing (adapted from [93]). (c) Temporally modulated ring resonator with
χ 3 nonlinearity and spatially modulated GVD can support Hamiltonians with local
interactions in synthetic space (adapted from [170]).
synthetic dimension. In the third approach, the synthetic dimension is a coordinate—

specifically, the only coordinate that is not spatial: time. However, time usually
already has a role in topological systems—it is the evolution parameter. Hence, if time
were to be used as a synthetic dimension, it would be necessary to somehow exchange
between time and space [176]. But how can this be used to increase dimensionality?
A popular method to do just that is by time-bin encoding. To understand this concept,
consider an arbitrary wave function ψ0 in N dimensions that undergoes a periodic
modulation every time period T. The evolution of ψ0 after time T can be described by
the unitary operation U , which satisfies ψnT = U n ψ0 , where n is the number of peri-
ods that the wave function went through. The operator U corresponds to an effective
Floquet Hamiltonian according to
U n = e −i Heff nT , (15)
where Heff is a static effective Hamiltonian. Notice that the full temporal dynamics
of Heff coincide with the dynamics ψnT at the beginning of each period. The Floquet
Hamiltonian is known to exhibit topological phases in a variety of lattice systems
[25,29,177], but here we will not assume any underlying spatial lattice whatsoever.
Since analyzing the dynamics of this system requires sampling it every period T,
the remaining time on the time axis remains unused. The idea then is to utilize the
“unused time” as a spatial degree of freedom. If each time-bin represents a different
location in space, adding a delay 1t (−1t) to the information expected to arrive at
time nT means that the information has moved one step to the right (or left).
On the basis of the time-bin lattice with an additional degree of freedom of the
Hamiltonian (such as polarization), it is possible to construct a system with a topologi-
cal band structure, where the Brillouin zone is the reciprocal space to the time-bins,
and the additional degree of freedom allows for the complexity required for the
non-trivial topology. To understand this, consider a typical system composed of two
coupled fiber loops with different length x ± 1x , as shown in Fig. 8(a). A light pulse
is launched into the system and completes encircling each fiber in times T ± 1T,
depending on which fiber it travels in. After completing n periods, the pulse can be
in one of the following times: t = nT + p1T, p = {−n, −n + 1, . . . , n}. Thus, each
value of p is a time-bin, and it is a synthetic dimension that represents a location
in space. Each time 1T is regarded as one step in the motion in which the photon
can move left or right in this synthetic space [Fig. 8(b)]. In principle, additional
synthetic space dimensions can be obtained by using several time scales. That is,
adding another bin splitting to obtain two different paths with temporal difference
1T̃, which is larger than 2n1T, makes it possible to have another synthetic spatial
dimension. The pulses in the system have a spatial dimensionality of the number of
time scales added, but in order to have a sense of a geometric phase involved in topo-
logical phases, another degree of freedom is added to the spatial one. In the system
described in Fig. 8, it is the polarization of the pulse. The polarization is described
by a 2D vector and can accumulate a phase and, thus, a geometric phase. In this set-
ting, topological edge states can be obtained by coupling the polarization degree of
freedom to the translation degree of freedom. This operation can be done simply by
using a polarization-dependent beam splitter, which can be described by the following
operator:
X
Tx = |x + 1ihx |⊗| ↑ih↑ |+| x − 1x ih| ⊗ |↓ih↓| , (16)
x
where |x i is the synthetic position degree of freedom and | ↑i (| ↓i) is the spin up
(down). The polarization here is analogous to the geometric phase of the wave func-
tion, which is responsible for the topological effects described in previous sections.
To obtain a topological phase in 1D, we split Tx to two operators, T↑ (T↓ ), which move
the light one site right (left) if the spin is up (down). Then, by adding two rotations
between the two splitted parts of Tx we obtain
U1D = T↑ R(θ2 )T↓ R(θ1 ), (17)
where R(θ) is polarization rotation by angle θ . The DTQW related to this operator
[Fig. 8(c)] has a topological phase diagram with two possible topological phases
[Fig. 8(d)]. The topological phases are characterized by the Zak phase from Eq. (1)
of the effective Hamiltonian [178]. The implication of a boundary in the DTQW
between different regions with different θ2 [Fig. 8(e)], and with different topologi-
cal phases, can be seen in the left panel [Fig. 8(f)]. Exciting the edge state causes
localization, where the wave remains concentrated on the edge without coupling to
neighboring sites. This is contrary to the right panel in which the two regions have the
same topological phase (but different θ2 ), and in this case no localization occurs. As
for the system described in previous sections, the studies on these synthetic dimen-
sions can be extended both in dimensionality and in fundamental properties such as
hermiticity and nonlinearity. Moreover, by considering an additional time scale—it
is possible to add the same spin dependence also in y : Ty and to consider the cyclic
process:
Figure 8
(a) Illustration of two coupled fiber loops of different lengths. Different output arrival
times correspond to different locations on a 1D lattice in space. (b) Steps of the
discrete time quantum walks (DTQWs) corresponding to the two-fiber setting in
(a) (adapted from [179]). (c) DTQW coupling of the polarization degree of freedom to
the spatial (time-bins) degree of freedom in a DTQW process [113]. (d) Topological
phase diagram corresponding to different rotation angles θ1 and θ2 of the process in
(c). Here, gray regions correspond to non-trivial topology with Rudner number 1
[113]. (e) A boundary with edge states forming by changing θ2 [113]. (f) Edge propa-
gation of two regions in the DTQW. In the left panel, the two regions are in a different
topological phase inducing a localized state on the edge. In the right panel, the two
regions have distinct θ2 but same topological phase, so no localization occurs [113].
U = Ty R(θ2 )Tx R(θ1 ). (18)
The process described by this equation also has topological phases, and, since it is
a 2D model, the topological properties are different from those of the 1D model. In
this Floquet system, the topological invariant is the Rudner number [180]. Just like
the Chern number, the Rudner number predicts the existence of unidirectional robust
edge states in the gap. As in the 1D case, one can have a phase diagram as of a function
of θ1 and θ2 [Fig. 9(a)]. The edge state at the boundary between different values of θ1
is illustrated in Fig. 9(b), and the band diagram. The reason that the Chern number is
not used here is because the Chern number gives for each band the difference between
the number of edge states bellow the band and above the band. Since in this type of
Floquet Hamiltonians the difference is zero (but edge states still exist) [see Figs. 9(c)
and 9(d)], the Rudner number, which takes into account the temporal dependence of
the Hamiltonian, is the proper quantity to use.
The propagation of light in this system is simulated in Fig. 9(e). The light bypasses
two corners along the edge, demonstrating the topological robustness of the
propagation [125].
Figure 9
(a)–(e) 2D topology in DTQW [125]. (a) Topological phase diagram of the 2D

DTQW for different values of θ1 and θ2 . Each color corresponds to a different Rudner
number. (b) Illustration of the unidirectional edge state propagation in the spatial
synthetic space. (c) and (d) Band structure with edge states for different boundaries.
(e) Propagation around an edge with two corners at different stages in the evolution.
Figure 10
(a) Non-Hermitian fiber loop setup. (b) PT-symmetric fiber loop setup. The purple
color on the fiber loops indicates that the gain and loss are alternating on every other
round trip [111].
As in previous approaches discussed, beyond extending the dimensionality of the

models, it is also possible to study non-Hermitian Hamiltonians with the DTQW.
This can be done by adding gain and loss to the fibers [Fig. 10(a)], and, if added in
an alternating fashion, it can also simulate a PT-symmetric lattice [Fig. 10(b)] [59].
Furthermore, in recent works, even solitons [111] and synthetic gauge fields [124]
were studied, which shows the wide possibilities of these systems in studying novel
physics using synthetic dimensions.
4. FUTURE DIRECTIONS AND VISION

In this review, we explained the basic physics of topological phases in synthetic
dimensions and described the three different approaches for using synthetic dimen-
sions in photonics, supporting topological phases and topological edge states. All
of these approaches have already demonstrated the important experimental mile-
stone of having 2D topological phases and topological edge states. These efforts
are important for establishing this field experimentally, but we are now entering the
most exciting stage of this field: experimentally studying topological phases that
were never observed in any other experimental platform, whether if they are with
high dimensionality, non-linear or non-Hermitian. Naturally, this is the most exciting
stage, where synthetic dimensions can present opportunities where no one has gone
before. As has already happened with “ordinary” photonic topological insulators
based on spatial lattices, utilizing synthetic dimensions in experiments can lead to
new theoretical and technological ideas, which are already starting to appear. The
paradigm of robust control over the evolution of photons in synthetic dimensions
offers new possibilities, as the propagation is no longer confined to spatial edges of
the system. Another avenue where exciting research lies ahead is in simulating quan-
tum interacting many-body models. Such models in high dimensions, with or without
non-Hermiticity, are of fundamental value, and photonics presents an unprecedented
opportunity to study them in the lab.
FUNDING
Air Force Office of Scientific Research; Israel Science Foundation.
ACKNOWLEDGMENT
Eran Lustig is supported by the Adams Fellowship Program of the Israel Academy of
Sciences and Humanities.
DISCLOSURES
The authors declare no conflicts of interest.
REFERENCES AND NOTES

1. C. L. Kane and E. J. Mele, “Quantum spin Hall effect in graphene,” Phys. Rev.
Lett. 95, 226801 (2005).
2. B. A. Bernevig, T. L. Hughes, and S.-C. Zhang, “Quantum spin Hall effect and
topological phase transition in HgTe quantum wells,” Science 314, 1757–1761
(2006).
3. M. König, S. Wiedmann, C. Brüne, A. Roth, H. Buhmann, L. W. Molenkamp,
X.-L. Qi, and S.-C. Zhang, “Quantum spin Hall insulator state in HgTe quantum
wells,” Science 318, 766–770 (2007).
4. D. Hsieh, D. Qian, L. Wray, Y. Xia, Y. S. Hor, R. J. Cava, and M. Z. Hasan, “A

topological Dirac insulator in a quantum spin Hall phase,” Nature 452, 970–974
(2008).
5. H. Lin, L. A. Wray, Y. Xia, S. Xu, S. Jia, R. J. Cava, A. Bansil, and M. Z.
Hasan, “Half-Heusler ternary compounds as new multifunctional experimental
platforms for topological quantum phenomena,” Nat. Mater. 9, 546–549 (2010).
6. D. Thouless, M. Kohmoto, M. Nightingale, and M. den Nijs, “Quantized Hall
conductance in a two-dimensional periodic potential,” Phys. Rev. Lett. 49,
405–408 (1982).
7. K. V. Klitzing, G. Dorda, and M. Pepper, “New method for high-accuracy deter-
mination of the fine-structure constant based on quantized Hall resistance,” Phys.
Rev. Lett. 45, 494–497 (1980).
8. B. I. Halperin, “Quantized Hall conductance, current-carrying edge states, and
the existence of extended states in a two-dimensional disordered potential,”
Phys. Rev. B 25, 2185–2190 (1982).
9. A. H. MacDonald and P. Středa, “Quantized Hall effect and edge currents,”
Phys. Rev. B 29, 1616–1619 (1984).
10. Y. Hatsugai, “Chern number and edge states in the integer quantum Hall effect,”
Phys. Rev. Lett. 71, 3697–3700 (1993).
11. F. D. M. Haldane, “Model for a quantum Hall effect without Landau levels:
condensed-matter realization of the ‘parity anomaly’,” Phys. Rev. Lett. 61,
2015–2018 (1988).
12. L. Fu, C. L. Kane, and E. J. Mele, “Topological insulators in three dimensions,”
Phys. Rev. Lett. 98, 106803 (2007).
13. J. E. Moore and L. Balents, “Topological invariants of time-reversal-invariant
band structures,” Phys. Rev. B 75, 121306 (2007).
14. A. Y. Kitaev, “Fault-tolerant quantum computation by anyons,” Ann. Phys. (N.
Y.) 303, 2–30 (2003).
15. A. R. Mellnik, J. S. Lee, A. Richardella, J. L. Grab, P. J. Mintun, M. H. Fischer,
A. Vaezi, A. Manchon, E.-A. Kim, N. Samarth, and D. C. Ralph, “Spin-transfer
torque generated by a topological insulator,” Nature 511, 449–451 (2014).
16. Magnetic disorder breaks this topological protection.
17. M. Z. Hasan and C. L. Kane, “Colloquium: topological insulators,” Rev. Mod.
Phys. 82, 3045–3067 (2010).
18. Topological insulators are not restricted to classical wave physics only. Rather,
there are examples of topological insulators that fundamentally rely on inter-
actions such as the fractional quantum Hall systems. However, the fundamental
concepts of topological insulators do not require quantum interactions: they rely
on wave phenomena such as interference and geometrical phases.
19. F. D. M. Haldane and S. Raghu, “Possible realization of directional optical
waveguides in photonic crystals with broken time-reversal symmetry,” Phys.
Rev. Lett. 100, 013904 (2008).
20. Z. Wang, Y. D. Chong, J. D. Joannopoulos, and M. Soljačić, “Reflection-free
one-way edge modes in a gyromagnetic photonic crystal,” Phys. Rev. Lett. 100,
013905 (2008).
21. Z. Wang, Y. Chong, J. D. Joannopoulos, and M. Soljačić, “Observation of uni-
directional backscattering-immune topological electromagnetic states,” Nature
461, 772–775 (2009).
22. R. O. Umucalılar and I. Carusotto, “Artificial gauge field for photons in coupled
cavity arrays,” Phys. Rev. A 84, 043804 (2011).
23. M. Hafezi, E. A. Demler, M. D. Lukin, and J. M. Taylor, “Robust optical delay
lines with topological protection,” Nat. Phys. 7, 907–912 (2011).
24. A. B. Khanikaev, S. H. Mousavi, W.-K. Tse, M. Kargarian, A. H. MacDonald,

and G. Shvets, “Photonic topological insulators,” Nat. Mater. 12, 233–239
(2012).
25. M. C. Rechtsman, J. M. Zeuner, Y. Plotnik, Y. Lumer, D. Podolsky, F. Dreisow,
S. Nolte, M. Segev, and A. Szameit, “Photonic Floquet topological insulators,”
Nature 496, 196–200 (2013).
26. M. Hafezi, S. Mittal, J. Fan, A. Migdall, and J. M. Taylor, “Imaging topological
edge states in silicon photonics,” Nat. Photonics 7, 1001–1005 (2013).
27. S. Mittal, J. Fan, S. Faez, A. Migdall, J. M. Taylor, and M. Hafezi,
“Topologically robust transport of photons in a synthetic gauge field,” Phys.
Rev. Lett. 113, 087403 (2014).
28. W.-J. Chen, S.-J. Jiang, X.-D. Chen, B. Zhu, L. Zhou, J.-W. Dong, and C. T.
Chan, “Experimental realization of photonic topological insulator in a uniaxial
metacrystal waveguide,” Nat. Commun. 5, 5782 (2014).
29. G. Q. Liang and Y. D. Chong, “Optical resonator analog of a two-dimensional
topological insulator,” Phys. Rev. Lett. 110, 203904 (2013).
30. J. Nichols, X. Gao, S. Lee, T. L. Meyer, J. W. Freeland, V. Lauter, D. Yi, J. Liu,
D. Haskel, J. R. Petrie, E.-J. Guo, A. Herklotz, D. Lee, T. Z. Ward, G. Eres, M. R.
Fitzsimmons, and H. N. Lee, “Emerging magnetism and anomalous Hall effect
in iridate–manganite heterostructures,” Nat. Commun. 7, 12721 (2016).
31. L. J. Maczewsky, J. M. Zeuner, S. Nolte, and A. Szameit, “Observation of
photonic anomalous Floquet topological insulators,” Nat. Commun. 8, 13756
(2017).
32. L.-H. Wu and X. Hu, “Scheme for achieving a topological photonic crystal by
using dielectric material,” Phys. Rev. Lett. 114, 223901 (2015).
33. S. Yves, R. Fleury, T. Berthelot, M. Fink, F. Lemoult, and G. Lerosey,
“Crystalline metamaterials for topological properties at subwavelength scales,”
Nat. Commun. 8, 16023 (2017).
34. Y. Yang, Y. F. Xu, T. Xu, H.-X. Wang, J.-H. Jiang, X. Hu, and Z. H. Hang,
“Visualization of a unidirectional electromagnetic waveguide using topological
photonic crystals made of dielectric materials,” Phys. Rev. Lett. 120, 217401
(2018).
35. M. A. Gorlach, X. Ni, D. A. Smirnova, D. Korobkin, D. Zhirihin, A. P.
Slobozhanyuk, P. A. Belov, A. Alù, and A. B. Khanikaev, “Far-field probing
of leaky topological states in all-dielectric metasurfaces,” Nat. Commun. 9, 909
(2018).
36. S. Barik, A. Karasahin, C. Flower, T. Cai, H. Miyake, W. DeGottardi, M. Hafezi,
and E. Waks, “A topological quantum optics interface,” Science 359, 666–668
(2018).
37. T. Ma and G. Shvets, “All-Si valley-Hall photonic topological insulator,” New J.
Phys. 18, 25012 (2016).
38. Y. Kang, X. Ni, X. Cheng, A. B. Khanikaev, and A. Z. Genack, “Pseudo-spin–
valley coupled edge states in a photonic topological insulator,” Nat. Commun. 9,
3029 (2018).
39. M. I. Shalaev, W. Walasik, A. Tsukernik, Y. Xu, and N. M. Litchinitser, “Robust
topologically protected transport in photonic crystals at telecommunication
wavelengths,” Nat. Nanotechnol. 14, 31–34 (2019).
40. T. Ozawa, H. M. Price, A. Amo, N. Goldman, M. Hafezi, L. Lu, M. C.
Rechtsman, D. Schuster, J. Simon, O. Zilberberg, and I. Carusotto, “Topological
photonics,” Rev. Mod. Phys. 91, 015006 (2019).
41. M. Segev and M. A. Bandres, “Topological photonics: where do we go from
here?” Nanophotonics 10, 425–434 (2020).
42. A. B. Khanikaev and G. Shvets, “Two-dimensional topological photonics,” Nat.

Photonics 11, 763–773 (2017).
43. X.-C. Sun, C. He, X.-P. Liu, M.-H. Lu, S.-N. Zhu, and Y.-F. Chen, “Two-
dimensional topological photonic systems,” Prog. Quantum Electron. 55, 52–73
(2017).
44. There is also considerable literature on 1D topological photonic systems,
mostly relying on the 1D binary lattice known as the SSH model [N. Malkova, I.
Hromada, X. Wang, G. Bryant, and Z. Chen, “Observation of optical Shockley-
like surface states in photonic superlattices,” Opt. Lett. 34, 1633–1635 (2009) ].
These 1D systems are endowed with zero-dimensional edge states, but because
their low dimensionality are not topological insulators, they are therefore
generally not discussed in this review.
45. A. Kitaev, “Periodic table for topological insulators and superconductors,” AIP
Conf. Proc. 1134, 22–30 (2009).
46. Z. Gong, Y. Ashida, K. Kawabata, K. Takasan, S. Higashikawa, and M. Ueda,
“Topological phases of non-Hermitian systems,” Phys. Rev. X 8, 031079
(2018).
47. B. Lian and S.-C. Zhang, “Weyl semimetal and topological phase transition in
five dimensions,” Phys. Rev. B 95, 235106 (2017).
48. X.-L. Qi, T. L. Hughes, and S.-C. Zhang, “Topological field theory of
time-reversal invariant insulators,” Phys. Rev. B 78, 195424 (2008).
49. B. Lian and S.-C. Zhang, “Five-dimensional generalization of the topological
Weyl semimetal,” Phys. Rev. B 94, 041105 (2016).
50. L. Lu, Z. Wang, D. Ye, L. Ran, L. Fu, J. D. Joannopoulos, and M. Soljačić,
“Experimental observation of Weyl points,” Science 349, 622–624 (2015).
51. D. Wang, B. Yang, W. Gao, H. Jia, Q. Yang, X. Chen, M. Wei, C. Liu, M.
Navarro-Cía, J. Han, W. Zhang, and S. Zhang, “Photonic Weyl points due to
broken time-reversal symmetry in magnetized semiconductor,” Nat. Phys. 15,
1150–1155 (2019).
52. J. Noh, S. Huang, D. Leykem, Y. D. Chong, K. P. Chen, and M. C. Rechtsman,
“Experimental observation of optical Weyl points and Fermi arc-like surface
states,” Nat. Phys. 13, 611–617 (2017).
53. E. Lustig, Y. Plotnik, Z. Yang, and M. Segev, “3D parity time symmetry in
2D photonic lattices utilizing artificial gauge fields in synthetic dimensions,”
in Conference on Lasers and Electro-Optics, OSA Technical Digest (Optical
Society of Ameria, 2019), paper FTu4B.1.
54. Note that, beyond periodic systems, topological phases (without topologically
protected edge transport) in 1D quasicrystals were demonstrated experimentally
in [51] and [57]. More recently, Floquet topological insulators were predicted to
occur in photonic quasicrystalline and fractal lattices (in [52] and [53]).
55. M. Verbin, O. Zilberberg, Y. Lahini, Y. E. Kraus, and Y. Silberberg,
“Topological pumping over a photonic Fibonacci quasicrystal,” Phys. Rev.
B 91, 064201 (2015).
56. M. A. Bandres, M. C. Rechtsman, and M. Segev, “Topological photonic qua-
sicrystals: fractal topological spectrum and protected transport,” Phys. Rev. X 6,
011016 (2016).
57. Z. Yang, E. Lustig, Y. Lumer, and M. Segev, “Photonic Floquet topological
insulators in a fractal lattice,” Light Sci. Appl. 9, 128 (2020).
58. D. Sticlet and F. Piéchon, “Distant-neighbor hopping in graphene and Haldane
models,” Phys. Rev. B 87, 115402 (2013).
59. A. Regensburger, C. Bersch, M.-A. Miri, G. Onishchukov, D. N.
Christodoulides, and U. Peschel, “Parity–time synthetic photonic lattices
(supplement),” Nature 488, 167–171 (2012).
60. O. Boada, A. Celi, J. I. Latorre, and M. Lewenstein, “Quantum simulation of an

extra dimension,” Phys. Rev. Lett. 108, 133001 (2012).
61. Y. E. Kraus, Y. Lahini, Z. Ringel, M. Verbin, and O. Zilberberg, “Topological
states and adiabatic pumping in quasicrystals,” Phys. Rev. Lett. 109, 106402
(2012).
62. A. Celi, P. Massignan, J. Ruseckas, N. Goldman, I. B. Spielman, G. Juzeliūnas,
and M. Lewenstein, “Synthetic gauge fields in synthetic dimensions,” Phys. Rev.
Lett. 112, 043001 (2014).
63. D. Jukić and H. Buljan, “Four-dimensional photonic lattices and discrete tesser-
act solitons,” Phys. Rev. A 87, 013814 (2013).
64. S. Fishman, D. R. Grempel, and R. E. Prange, “Chaos, quantum recurrences, and
Anderson localization,” Phys. Rev. Lett. 49, 509–512 (1982).
65. D. R. Grempel, R. E. Prange, and S. Fishman, “Quantum dynamics of a noninte-
grable system,” Phys. Rev. A 29, 1639–1647 (1984).
66. G. Casati, I. Guarneri, and D. L. Shepelyansky, “Anderson transition in a one-
dimensional system with three incommensurate frequencies,” Phys. Rev. Lett.
62, 345–348 (1989).
67. J. Chabé, G. Lemarié, B. Grémaud, D. Delande, P. Szriftgiser, and J. C. Garreau,
“Experimental observation of the Anderson metal-insulator transition with
atomic matter waves,” Phys. Rev. Lett. 101, 255702 (2008).
68. R. Gommers, S. Denisov, and F. Renzoni, “Quasiperiodically driven ratchets for
cold atoms,” Phys. Rev. Lett. 96, 240604 (2006).
69. B. G. Klappauf, W. H. Oskay, D. A. Steck, and M. G. Raizen, “Observation of
noise and dissipation effects on dynamical localization,” Phys. Rev. Lett. 81,
1203–1206 (1998).
70. H. Ammann, R. Gray, I. Shvarchuck, and N. Christensen, “Quantum delta-
kicked rotor: experimental observation of decoherence,” Phys. Rev. Lett. 80,
4111–4115 (1998).
71. J. M. Edge, J. Tworzydło, and C. W. J. Beenakker, “Metallic phase of the
quantum Hall effect in four-dimensional space,” Phys. Rev. Lett. 109, 135701
(2012).
72. O. Zilberberg, S. Huang, J. Guglielmon, M. Wang, K. P. Chen, Y. E. Kraus, and
M. C. Rechtsman, “Photonic topological boundary pumping as a probe of 4D
quantum Hall physics,” Nature 553, 59–62 (2018).
73. M. Lohse, C. Schweizer, H. M. Price, O. Zilberberg, and I. Bloch, “Exploring 4D
quantum Hall physics with a 2D topological charge pump,” Nature 553, 55–58
(2018).
74. I. Petrides, H. M. Price, and O. Zilberberg, “Six-dimensional quantum Hall effect
and three-dimensional topological pumps,” Phys. Rev. B 98, 125431 (2018).
75. Q. Wang, M. Xiao, H. Liu, S. Zhu, and C. T. Chan, “Optical interface states pro-
tected by synthetic Weyl points,” Phys. Rev. X 7, 031032 (2017).
76. B. K. Stuhl, H.-I.-I. Lu, L. M. Aycock, D. Genkina, and I. B. Spielman,
“Visualizing edge states with an atomic Bose gas in the quantum Hall regime,”
Science 349, 1514–1518 (2015).
77. M. Mancini, G. Pagano, G. Cappellini, L. Livi, M. Rider, J. Catani, C. Sias, P.
Zoller, M. Inguscio, M. Dalmonte, and L. Fallani, “Observation of chiral edge
states with neutral fermions in synthetic Hall ribbons,” Science 349, 1510–1513
(2015).
78. S. Kolkowitz, S. L. Bromley, T. Bothwell, M. L. Wall, G. E. Marti, A. P. Koller,
X. Zhang, A. M. Rey, and J. Ye, “Spin–orbit-coupled fermions in an optical lat-
tice clock,” Nature 542, 66–70 (2016).
79. F. A. An, E. J. Meier, and B. Gadway, “Direct observation of chiral currents and
magnetic reflection in atomic flux lattices,” Sci. Adv. 3, e1602685 (2017).
80. H. Cai, J. Liu, J. Wu, Y. He, S.-Y. Zhu, J.-X. Zhang, and D.-W. Wang,
“Experimental observation of momentum-space chiral edge currents in
room-temperature atoms,” Phys. Rev. Lett. 122, 023601 (2019).
81. H. M. Price, T. Ozawa, and N. Goldman, “Synthetic dimensions for cold atoms
from shaking a harmonic trap,” Phys. Rev. A 95, 023607 (2017).
82. X.-W. Luo, X. Zhou, C.-F. Li, J.-S. Xu, G.-C. Guo, and Z.-W. Zhou, “Quantum
simulation of 2D topological physics in a 1D array of optical cavities,” Nat.
Commun. 6, 7704 (2015).
83. T. Ozawa, H. M. Price, N. Goldman, O. Zilberberg, and I. Carusotto, “Synthetic
dimensions in integrated photonics: from optical isolation to four-dimensional
quantum Hall physics,” Phys. Rev. A 93, 043827 (2016).
84. L. Yuan, Y. Shi, and S. Fan, “Photonic gauge potential in a system with a
synthetic frequency dimension,” Opt. Lett. 41, 741–744 (2016).
85. E. Lustig, S. Weimann, Y. Plotnik, Y. Lumer, M. A. Bandres, A. Szameit, and
M. Segev, “Photonic topological insulator in synthetic dimensions,” Nature 567,
356–360 (2019).
86. C. W. Peterson, W. A. Benalcazar, M. Lin, T. L. Hughes, and G. Bahl, “Strong
nonreciprocity in modulated resonator chains through synthetic electric and
magnetic fields,” Phys. Rev. Lett. 123, 063901 (2019).
87. A. Dutt, M. Minkov, Q. Lin, L. Yuan, D. A. B. Miller, and S. Fan, “Experimental
band structure spectroscopy along a synthetic dimension,” Nat. Commun. 10,
3122 (2019).
88. A. Dutt, Q. Lin, L. Yuan, M. Minkov, M. Xiao, and S. Fan, “A single photonic
cavity with two independent physical synthetic dimensions,” Science 367, 59–64
(2020).
89. Q. Lin, X.-Q. Sun, M. Xiao, S.-C. Zhang, and S. Fan, “A three-dimensional pho-
tonic topological insulator using a two-dimensional ring resonator lattice with a
synthetic frequency dimension,” Sci. Adv. 4, eaat2774 (2018).
90. Q. Lin, M. Xiao, L. Yuan, and S. Fan, “Photonic Weyl point in a two-
dimensional resonator lattice with a synthetic frequency dimension,” Nat.
Commun. 7, 13731 (2016).
91. A. Dutt, M. Minkov, I. A. D. Williamson, and S. Fan, “Higher-order topological
insulators in synthetic dimensions,” Light Sci. Appl. 9, 131 (2020).
92. L. Yuan, Y. Shi, and S. Fan, “Achieving the gauge potential for the photon in a
synthetic space,” in Conference on Lasers and Electro-Optics (Optical Society
of America, 2016), paper SF2E.5.
93. Z. Yang, E. Lustig, G. Harari, Y. Plotnik, Y. Lumer, M. A. Bandres, and M.
Segev, “Mode-locked topological insulator laser utilizing synthetic dimensions,”
Phys. Rev. X 10, 011059 (2020).
94. B. A. Bell, K. Wang, A. S. Solntsev, D. N. Neshev, A. A. Sukhorukov, and B. J.
Eggleton, “Spectral photonic lattices with complex long-range coupling,” Optica
4, 1433–1436 (2017).
95. L. Yuan, Q. Lin, M. Xiao, and S. Fan, “Synthetic dimension in photonics,”
Optica 5, 1396–1405 (2018).
96. Y. Wang, H. M. Price, B. Zhang, and Y. D. Chong, “Circuit implementation of a
four-dimensional topological insulator,” Nat. Commun. 11, 2356 (2020).
97. L. Nemirovsky, M. Cohen, E. Lustig, and M. Segev, “Magnetic gauge field for
photons in synthetic dimensions by a propagation-invariant photonic structure,”
Society of America, 2009), paper FW3D.7.
98. L. Nemirovsky, M. Cohen, Y. Lumer, E. Lustig, and M. Segev, “Topological
evolution-invariant photonic structures in synthetic dimensions,” in Conference
on Lasers and Electro-Optics, OSA Technical Digest (Optical Society of

America, 2020), paper FW4A.1.
99. T. Ozawa, “Artificial magnetic field for synthetic quantum matter without
dynamical modulation,” Phys. Rev. A 103, 33318 (2009).
100. Y. Aharonov, L. Davidovich, and N. Zagury, “Quantum random walks,” Phys.
Rev. A 48, 1687–1690 (1993).
101. E. Farhi and S. Gutmann, “Quantum computation and decision trees,” Phys. Rev.
A 58, 915–928 (1998).
102. R. J. Sension, “Quantum path to photosynthesis,” Nature 446, 740–741 (2007).
103. S. Godoy and S. Fujita, “A quantum random-walk model for tunneling diffusion
in a 1D lattice. A quantum correction to Fick’s law,” J. Chem. Phys. 97, 5148–
5154 (1992).
104. T. Oka, N. Konno, R. Arita, and H. Aoki, “Breakdown of an electric-field driven
system: a mapping to a quantum walk,” Phys. Rev. Lett. 94, 100602 (2005).
105. A. Schreiber, K. N. Cassemiro, V. Potoček, A. Gábris, P. J. Mosley, E.
Andersson, I. Jex, and C. Silberhorn, “Photons walking the line: a quantum
walk with adjustable coin operations,” Phys. Rev. Lett. 104, 50502 (2010).
106. A. L. M. Muniz, A. Alberucci, C. P. Jisha, M. Monika, S. Nolte, R. Morandotti,
and U. Peschel, “Kapitza light guiding in photonic mesh lattice,” Opt. Lett. 44,
6013–6016 (2019).
107. M. Wimmer and U. Peschel, “Observation of time reversed light propagation by
an exchange of eigenstates,” Sci. Rep. 8, 2125 (2018).
108. M. Wimmer, A. Regensburger, C. Bersch, M.-A. Miri, S. Batz, G. Onishchukov,
D. N. Christodoulides, and U. Peschel, “Optical diametric drive acceleration
through action–reaction symmetry breaking,” Nat. Phys. 9, 780–784 (2013).
109. M. Wimmer, M.-A. Miri, D. Christodoulides, and U. Peschel, “Observation of
Bloch oscillations in complex PT-symmetric photonic lattices,” Sci. Rep. 5,
17760 (2015).
110. I. D. Vatnik, A. Tikan, G. Onishchukov, D. V. Churkin, and A. A. Sukhorukov,
“Anderson localization in synthetic photonic lattices,” Sci. Rep. 7, 4301 (2017).
111. M. Wimmer, A. Regensburger, M.-A. Miri, C. Bersch, D. N. Christodoulides,
and U. Peschel, “Observation of optical solitons in PT-symmetric lattices,” Nat.
Commun. 6, 7782 (2015).
112. A. Dikopoltsev, M. Kremer, H. H. Sheinfux, S. Weidemann, A. Szameit, and
M. Segev, “Observation of Anderson localization by virtual transitions,” in
Conference on Lasers and Electro-Optics (Optical Society of America, 2020),
pp. 1–2.
113. T. Kitagawa, M. S. Rudner, E. Berg, and E. Demler, “Exploring topological
phases with quantum walks,” Phys. Rev. A 82, 033429 (2010).
114. T. Kitagawa, M. A. Broome, A. Fedrizzi, M. S. Rudner, E. Berg, I. Kassal, A.
Aspuru-Guzik, E. Demler, and A. G. White, “Observation of topologically
protected bound states in photonic quantum walks,” Nat. Commun. 3, 882
(2012).
115. F. Cardano, F. Massa, H. Qassim, E. Karimi, S. Slussarenko, D. Paparo, C. de
Lisio, F. Sciarrino, E. Santamato, R. W. Boyd, and L. Marrucci, “Quantum
walks and wavepacket dynamics on a lattice with twisted photons,” Sci. Adv. 1,
e1500087 (2015).
116. F. Cardano, M. Maffei, F. Massa, B. Piccirillo, C. de Lisio, G. De Filippis, V.
Cataudella, E. Santamato, and L. Marrucci, “Statistical moments of quantum-
walk dynamics reveal topological quantum transitions,” Nat. Commun. 7, 11439
(2016).
117. S. Barkhofen, T. Nitsche, F. Elster, L. Lorz, A. Gábris, I. Jex, and C. Silberhorn,

“Measuring topological invariants in disordered discrete-time quantum walks,”
Phys. Rev. A 96, 033846 (2017).
118. F. Cardano, A. D’Errico, A. Dauphin, M. Maffei, B. Piccirillo, C. de Lisio, G.
De Filippis, V. Cataudella, E. Santamato, L. Marrucci, M. Lewenstein, and
P. Massignan, “Detection of Zak phases and topological invariants in a chiral
quantum walk of twisted photons,” Nat. Commun. 8, 15516 (2017).
119. L. Xiao, X. Zhan, Z. H. Bian, K. K. Wang, X. Zhang, X. P. Wang, J. Li, K.
Mochizuki, D. Kim, N. Kawakami, W. Yi, H. Obuse, B. C. Sanders, and P. Xue,
“Observation of topological edge states in parity–time-symmetric quantum
walks,” Nat. Phys. 13, 1117–1123 (2017).
120. S. Weidemann, M. Kremer, T. Helbig, T. Hofmann, A. Stegmaier, M. Greiter,
R. Thomale, and A. Szameit, “Topological funneling of light,” Science 368,
311–314 (2020).
121. A. Schreiber, A. Gábris, P. P. Rohde, K. Laiho, M. Štefaňák, V. Potoček,
C. Hamilton, I. Jex, and C. Silberhorn, “A 2D quantum walk simulation of
two-particle dynamics,” Science 336, 55–58 (2012).
122. T. Nitsche, F. Elster, J. Novotný, A. Gábris, I. Jex, S. Barkhofen, and C.
Silberhorn, “Quantum walks with dynamical control: graph engineering, initial
state preparation and state transfer,” New J. Phys. 18, 063017 (2016).
123. B. Wang, T. Chen, and X. Zhang, “Experimental observation of topologically
protected bound states with vanishing Chern numbers in a two-dimensional
quantum walk,” Phys. Rev. Lett. 121, 100501 (2018).
124. H. Chalabi, S. Barik, S. Mittal, T. E. Murphy, M. Hafezi, and E. Waks,
“Synthetic gauge field for two-dimensional time-multiplexed quantum random
walks,” Phys. Rev. Lett. 123, 150503 (2019).
125. C. Chen, X. Ding, J. Qin, Y. He, Y.-H. Luo, M.-C. Chen, C. Liu, X.-L. Wang,
W.-J. Zhang, H. Li, L.-X. You, Z. Wang, D.-W. Wang, B. C. Sanders, C.-Y. Lu,
and J.-W. Pan, “Observation of topologically protected edge states in a photonic
two-dimensional quantum walk,” Phys. Rev. Lett. 121, 100502 (2018).
126. J. Zak, “Berry’s phase for energy bands in solids,” Phys. Rev. Lett. 62,
2747–2750 (1989).
127. P. Deymier and K. Runge, Phase and Topology BT—Sound Topology, Duality,
Coherence and Wave-Mixing: an Introduction to the Emerging New Science of
Sound, P. Deymier and K. Runge, eds. (Springer, 2017), pp. 37–80.
128. A. P. Schnyder, S. Ryu, A. Furusaki, and A. W. W. Ludwig, “Classification of
topological insulators and superconductors in three spatial dimensions,” Phys.
Rev. B 78, 195125 (2008).
129. C.-K. Chiu, J. C. Y. Teo, A. P. Schnyder, and S. Ryu, “Classification of topologi-
cal quantum matter with symmetries,” Rev. Mod. Phys. 88, 035005 (2016).
130. C.-K. Chiu, H. Yao, and S. Ryu, “Classification of topological insulators and
superconductors in the presence of reflection symmetry,” Phys. Rev. B 88,
075142 (2013).
131. T. Morimoto and A. Furusaki, “Topological classification with additional sym-
metries from Clifford algebras,” Phys. Rev. B 88, 125129 (2013).
132. K. Shiozaki and M. Sato, “Topology of crystalline insulators and superconduc-
tors,” Phys. Rev. B 90, 165114 (2014).
133. M. S. Rudner and L. S. Levitov, “Topological transition in a non-Hermitian
quantum walk,” Phys. Rev. Lett. 102, 065703 (2009).
134. J. M. Zeuner, M. C. Rechtsman, Y. Plotnik, Y. Lumer, S. Nolte, M. S. Rudner,
M. Segev, and A. Szameit, “Observation of a topological transition in the bulk of
a non-Hermitian system,” Phys. Rev. Lett. 115, 040402 (2015).
135. C. Poli, M. Bellec, U. Kuhl, F. Mortessagne, and H. Schomerus, “Selective

enhancement of topologically induced interface states in a dielectric resonator
chain,” Nat. Commun. 6, 6710 (2015).
136. S. Weimann, M. Kremer, Y. Plotnik, Y. Lumer, S. Nolte, K. G. Makris, M.
Segev, M. C. Rechtsman, and A. Szameit, “Topologically protected bound states
in photonic parity–time-symmetric crystals,” Nat. Mater. 16, 433–438 (2017).
137. T. E. Lee, “Anomalous edge state in a non-Hermitian lattice,” Phys. Rev. Lett.
116, 133903 (2016).
138. D. Leykam, K. Y. Bliokh, C. Huang, Y. D. Chong, and F. Nori, “Edge modes,
degeneracies, and topological numbers in non-Hermitian systems,” Phys. Rev.
Lett. 118, 040401 (2017).
139. M. S. Rudner, M. Levin, and L. S. Levitov, “Survival, decay, and topological
protection in non-Hermitian quantum transport,” arXiv:1605.07652 (2016).
140. M. A. Bandres, S. Wittek, G. Harari, M. Parto, J. Ren, M. Segev, D.
N. Christodoulides, and M. Khajavikhan, “Topological insulator laser:
experiments,” Science 359, eaar4005 (2018).
141. G. Harari, M. A. Bandres, Y. Lumer, M. C. Rechtsman, Y. D. Chong, M.
Khajavikhan, D. N. Christodoulides, and M. Segev, “Topological insulator
laser: theory,” Science 359, eaar4003 (2018).
142. H. Shen, B. Zhen, and L. Fu, “Topological band theory for non-Hermitian
Hamiltonians,” Phys. Rev. Lett. 120, 146402 (2018).
143. X. Chen, Z.-C. Gu, Z.-X. Liu, and X.-G. Wen, “Symmetry-protected topological
orders in interacting bosonic systems,” Science 338, 1604–1606 (2012).
144. M. E. Tai, A. Lukin, M. Rispoli, R. Schittko, T. Menke, D. Borgnia, P. M. Preiss,
F. Grusdt, A. M. Kaufman, and M. Greiner, “Microscopy of the interacting
Harper–Hofstadter model in the two-body limit,” Nature 546, 519–523 (2017).
145. S. de Léséleuc, V. Lienhard, P. Scholl, D. Barredo, S. Weber, N. Lang, H. P.
Büchler, T. Lahaye, and A. Browaeys, “Observation of a symmetry-protected
topological phase of interacting bosons with Rydberg atoms,” Science 365,
775–780 (2019).
146. Y. Tenenbaum Katan, R. Bekenstein, M. Bandres, Y. Lumer, Y. Plotnik, and
M. Segev, “Induction of topological transport by long ranged nonlinearity,”
in Conference on Lasers and Electro-Optics, OSA Technical Digest (online)
(2016), paper FM3A.6.
147. L. J. Maczewsky, M. Heinrich, M. Kremer, S. K. Ivanov, M. Ehrhardt, F.
Martinez, Y. V. Kartashov, V. V. Konotop, L. Torner, D. Bauer, and A. Szameit,
“Nonlinearity-induced photonic topological insulator,” Science 370, 701–704
(2020).
148. N. P. Mitchell, L. M. Nash, D. Hexner, A. M. Turner, and W. T. M. Irvine,
“Amorphous topological insulators constructed from random point sets,” Nat.
Phys. 14, 380–385 (2018).
149. S. Aubry and G. André, “Analyticity breaking and Anderson localization in
incommensurate lattices,” Ann. Isr. Phys. Soc. 3, 133–140 (1980).
150. C. H. Lee, Y. Wang, Y. Chen, and X. Zhang, “Electromagnetic response of
quantum Hall systems in dimensions five and six and beyond,” Phys. Rev. B 98,
094434 (2018).
151. R. B. Laughlin, “Quantized Hall conductivity in two dimensions,” Phys. Rev. B
23, 5632–5633 (1981).
152. B. A. Bernevig and S.-C. Zhang, “Quantum spin Hall effect,” Phys. Rev. Lett.
96, 106802 (2006).
153. A. Cerjan, M. Wang, S. Huang, K. P. Chen, and M. C. Rechtsman, “Thouless
pumping in disordered photonic systems,” Light Sci. Appl. 9, 178 (2020).
154. F. Zangeneh-Nejad and R. Fleury, “Zero-index Weyl metamaterials,” Phys. Rev.

Lett. 125, 054301 (2020).
155. X. Zhang, K. Ding, X. Zhou, J. Xu, and D. Jin, “Experimental observation of
an exceptional surface in synthetic dimensions with Magnon polaritons,” Phys.
Rev. Lett. 123, 237202 (2019).
156. Q. Wang, K. Ding, H. Liu, S. Zhu, and C. T. Chan, “Exceptional cones in 4D
parameter space,” Opt. Express 28, 1758–1770 (2020).
157. L. J. Maczewsky, K. Wang, A. A. Dovgiy, A. E. Miroshnichenko, A. Moroz,
M. Ehrhardt, M. Heinrich, D. N. Christodoulides, A. Szameit, and A. A.
Sukhorukov, “Synthesizing multi-dimensional excitation dynamics and
localization transition in one-dimensional lattices,” Nat. Photonics 14, 76–81
(2020).
158. C. Qin, F. Zhou, Y. Peng, D. Sounas, X. Zhu, B. Wang, J. Dong, X. Zhang, A.
Alù, and P. Lu, “Spectrum control through discrete frequency diffraction in the
presence of photonic gauge potentials,” Phys. Rev. Lett. 120, 133901 (2018).
159. Z.-D. Cheng, Q. Li, Z.-H. Liu, F.-F. Yan, S. Yu, J.-S. Tang, Z.-W. Zhou, J.-S.
Xu, C.-F. Li, and G.-C. Guo, “Experimental implementation of a degenerate
optical resonator supporting more than 46 Laguerre-Gaussian modes,” Appl.
Phys. Lett. 112, 201104 (2018).
160. A. Perez-Leija, R. Keil, A. Kay, H. Moya-Cessa, S. Nolte, L.-C. Kwek, B. M.
Rodríguez-Lara, A. Szameit, and D. N. Christodoulides, “Coherent quantum
transport in photonic lattices,” Phys. Rev. A 87, 012309 (2013).
161. E. Lustig, L. Maczewsky, T. Biesenthal, Z. Yang, Y. Plotnik, A. Szameit, and
M. Segev, “Experimentally realizing photonic topological edge states in 3D,”
Society of America, 2020), paper FW3A.2.
162. And, therefore, it has the same topological bandgap and bands in both repre-
sentations. Since the transformation between synthetic space and real space is
a unitary transformation, the dispersion of the system is preserved between the
synthetic space and real space.
163. G. Harari, M. A. Bandres, Y. Lumer, Y. Plotnik, D. N. Christodoulides, and
M. Segev, “Topological lasers,” in Conference on Lasers and Electro-Optics
(Optical Society of America, 2016), paper FM3A.3.
164. B. Bahari, A. Ndao, F. Vallini, A. El Amili, Y. Fainman, and B. Kanté,
“Nonreciprocal lasing in topological cavities of arbitrary geometries,” Science
358, 636–640 (2017).
165. Y. Zeng, U. Chattopadhyay, B. Zhu, B. Qiang, J. Li, Y. Jin, L. Li, A. G. Davies,
E. H. Linfield, B. Zhang, Y. Chong, and Q. J. Wang, “Electrically pumped topo-
logical laser with valley edge modes,” Nature 578, 246–250 (2020).
166. I. Amelio and I. Carusotto, “Theory of the coherence of topological lasers,” Phys.
Rev. X 10, 041060 (2019).
167. Y. G. N. Liu, P. Jung, M. Parto, W. E. Hayenga, D. N. Christodoulides, and
M. Khajavikhan, “Topological Haldane lattice,” in Conference on Lasers and
Electro-Optics (Optical Society of America, 2020), paper FW3A.1.
168. K. Wang, A. Dutt, K. Y. Yang, C. C. Wojcik, J. Vučković, and S. Fan,
“Generating arbitrary topological windings of a non-Hermitian band,” Science
371, 1240–1245 (2021)
169. M. Segev, Y. Ophir, B. Fischer, and G. Eisenstein, “Mode locking and frequency
tuning of a laser diode array in an extended cavity with a photorefractive phase
conjugate mirror,” Appl. Phys. Lett. 57, 2523–2525 (1990).
170. L. Yuan, A. Dutt, M. Qin, S. Fan, and X. Chen, “Creating locally interacting
Hamiltonians in the synthetic frequency dimension for photons,” Photon. Res. 8,
B8–B14 (2020).
171. C. Joshi, A. Farsi, A. Dutt, B. Y. Kim, X. Ji, Y. Zhao, A. M. Bishop, M. Lipson,

and A. L. Gaeta, “Frequency-domain quantum interference with correlated pho-
tons from an integrated microresonator,” Phys. Rev. Lett. 124, 143601 (2020).
172. A. K. Tusnin, A. M. Tikan, and T. J. Kippenberg, “Nonlinear states and dynam-
ics in a synthetic frequency dimension,” Phys. Rev. A 102, 023518 (2020).
173. K. Wang, B. A. Bell, A. S. Solntsev, D. N. Neshev, B. J. Eggleton, and A. A.
Sukhorukov, “Multidimensional synthetic chiral-tube lattices via nonlinear
frequency conversion,” Light Sci. Appl. 9, 132 (2020).
174. I. Martin, G. Refael, and B. Halperin, “Topological frequency conversion in
strongly driven quantum systems,” Phys. Rev. X 7, 041008 (2017).
175. L. Yuan, M. Xiao, Q. Lin, and S. Fan, “Synthetic space with arbitrary dimen-
sions in a few rings undergoing dynamic modulation,” Phys. Rev. B 97, 104105
(2018).
176. E. Lustig, Y. Sharabi, and M. Segev, “Topological aspects of photonic time crys-
tals,” Optica 5, 1390–1395 (2018).
177. N. H. Lindner, G. Refael, and V. Galitski, “Floquet topological insulator in semi-
conductor quantum wells,” Nat. Phys. 7, 490–495 (2011).
178. G. Puentes and O. Santillán, “Zak phase in discrete-time quantum walks,”
arXiv:1506.08100v2 (2015).
179. J. Boutari, A. Feizpour, S. Barz, C. D. Franco, M. Kim, W. S. Kolthammer, and
I. A. Walmsley, “Large scale quantum walks by means of optical fiber cavities,”
J. Opt. 18, 94007 (2016).
180. M. S. Rudner, N. H. Lindner, E. Berg, and M. Levin, “Anomalous edge states
and the bulk-edge correspondence for periodically driven two-dimensional
systems,” Phys. Rev. X 3, 031005 (2013).
Eran Lustig is a Ph.D. student at the Technion—Israel Institute
of Technology. He received his B.Sc. in physics and a B.Sc. in
electrical engineering at the Technion. He is the recipient of the
Leonard and Diane Sherman Interdisciplinary Graduate School
Fellowship (2016) and the recipient of the prestigious Adams
fellowship for doctoral students for outstanding Israeli doc-
toral students in the exact sciences. In his research, he studies
theoretically and experimentally the propagation of light in non-
spatial dimensions, and he takes interest in using machine learning approaches in
experimental physics.
Moti Segev is the Robert J. Shillman Distinguished Professor of

Physics and Electrical Engineering at the Technion, Israel. He
received his B.Sc. and Ph.D. from the Technion in 1985 and 1990.
After his postdoc at Caltech, he joined Princeton as Assistant
Professor (1994), becoming Associate Professor in 1997 and
Professor in 1999. Subsequently, Moti went back to Israel, and
in 2009 he was appointed as Distinguished Professor (highest
academic rank at the Technion). Moti’s interests are mainly in
nonlinear optics, photonics, solitons, sub-wavelength imaging, lasers, quantum simu-
lators, and quantum electronics, although he finds entertainment in more demanding
fields such as basketball and hiking. He has won numerous international awards,
among them the 2007 Quantum Electronics Prize of the European Physics Society,
the 2009 Max Born Award of the The Optical Society, and the 2014 Arthur Schawlow
Prize of the American Physical Society. In 2011, he was elected to the Israel Academy
of Sciences and Humanities, and in 2015 he was elected to the National Academy of
Science (NAS) of the United States of America. In 2014, Moti Segev won the Israel
Prize in Physics and Chemistry (highest honor in Israel), and in 2019 he won the
EMET Prize (Israel). However, above all his personal achievements, he takes pride
in the success of his graduate students and postdocs; among them are currently 23
professors in the USA, Germany, Taiwan, Croatia, Italy, India, China, and Israel, and
many hold senior R&D positions in the industry.

文献1

Uploaded by

Copyright:

Available Formats

文献1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

文献1

Uploaded by

Copyright:

Available Formats

426 Vol. 13, No.

2 / June 2021 / Advances in Optics and Photonics Review

insulators. We will do it on an intuitive basis and attempt to view these principles

Topological insulators in a nut shell. (a) Topological insulator in two dimensions: a

of the topological bandgap, which determines the degree of topological protection of

1.2 Photonic Topological Insulators

1.3 Role of Dimensionality and Connectivity in Topological Physics

1.4 Historic Development of Synthetic Dimensions in Photonics

phases [71]. In photonics, however, modulating effectively the dielectric properties

2. BASIC PHYSICS OF TOPOLOGY IN SYNTHETIC DIMENSIONS

2.2 Richness of Topological Realm

fully understand high-dimensional topological phases or complex topological phases

3. DESCRIPTION AND DEFINITION OF NON-SPATIAL DIMENSIONS

where L̃ is a loop on the extended Brillouin zone of k̃ = (k x , kα ). Since the implica-

3.1 Topological Photonics in Parameter Space

where c n (c n† ) is the annihilation (creation) operator related to site n, h.c. is the

where ti = t˜i + λi cos(2π bi i + φi ), with i = {x , y }. In Eq. (9), each AA model is

(a)–(c) 1D Andre–Aubry lattice exhibiting the physics of 2D integer quantum Hall

of the topological phenomena as much as it does in other approaches. When a param-

3.2 Synthetic Space Relying on Modal Ladders

where p̂ x is the momentum operator in x̂ , M is a mass term, and ω is a fixed param-

Ĥtot = Ĥ0 + V̂ (t) = Ĥ0 + κ x̂ cos(ω D t + φ), (12)

where ω D = ω − 1 with 1  ω and φ is a constant phase of the modulation. Now, the

Mode ladder of a harmonic potential. The coupling between modes is achieved by

+ J s ,λ (|λ, s − 1ihλ, s | + h.c), (14)

mode-lock is 10 [169]. Here is where the concept of topology in synthetic dimensions

3.3 Using Time-Bins as Synthetic Dimension

synthetic dimension. In the third approach, the synthetic dimension is a coordinate—

U1D = T↑ R(θ2 )T↓ R(θ1 ), (17)

U = Ty R(θ2 )Tx R(θ1 ). (18)

(a)–(e) 2D topology in DTQW [125]. (a) Topological phase diagram of the 2D

As in previous approaches discussed, beyond extending the dimensionality of the

4. FUTURE DIRECTIONS AND VISION

REFERENCES AND NOTES

4. D. Hsieh, D. Qian, L. Wray, Y. Xia, Y. S. Hor, R. J. Cava, and M. Z. Hasan, “A

24. A. B. Khanikaev, S. H. Mousavi, W.-K. Tse, M. Kargarian, A. H. MacDonald,

42. A. B. Khanikaev and G. Shvets, “Two-dimensional topological photonics,” Nat.

60. O. Boada, A. Celi, J. I. Latorre, and M. Lewenstein, “Quantum simulation of an

on Lasers and Electro-Optics, OSA Technical Digest (Optical Society of

117. S. Barkhofen, T. Nitsche, F. Elster, L. Lorz, A. Gábris, I. Jex, and C. Silberhorn,

135. C. Poli, M. Bellec, U. Kuhl, F. Mortessagne, and H. Schomerus, “Selective

154. F. Zangeneh-Nejad and R. Fleury, “Zero-index Weyl metamaterials,” Phys. Rev.

171. C. Joshi, A. Farsi, A. Dutt, B. Y. Kim, X. Ji, Y. Zhao, A. M. Bishop, M. Lipson,

Moti Segev is the Robert J. Shillman Distinguished Professor of

You might also like

where ω D = ω − 1 with 1 ω and φ is a constant phase of the modulation. Now, the