Abstract
Geometric phase is an important tool to define the topology of the Hermitian and non-Hermitian systems. Besides, the range of coupling plays an important role in realizing higher topological indices and transition among them. With a motivation to understand the geometric phases for mixed states, we discuss finite temperature analysis of Hermitian and non-Hermitian topological models with extended range of couplings. To understand the geometric phases for the mixed states, we use Uhlmann phase and discuss the merit-limitation with respect extended range couplings. We extend the finite temperature analysis to non-Hermitian models and define topological invariant for different ranges of coupling. We include the non-Hermitian skin effect, and provide the derivation of topological invariant in the generalized Brillouin zone and their mixed state behavior also. We also adopt mixed geometric phases through interferometric approach, and discuss the geometric phases of extended-range (Hermitian and non-Hermitian) models at finite temperature.
Similar content being viewed by others
Introduction
Topological state of matter is the area of huge interest both from the theoretical and experimental aspects1,2,3,4. These states lack the conventional local order parameter to characterize the phase transition and can not be apprehended by the traditional Landau theory of symmetry breaking5,6. By emerging as a suitable platform for the realization of exotic particles other than the fermions and bossons, the area attracted a great attention7,8. A recent addition is the non-Hermitian topological phases. In non-Hermitian systems, the energy spectrum becomes complex and the band structure may change drastically based on the type of non-Hermitian factor9,10,11. Non-Hermitian systems are believed to be a better platform to understand the loss and gain in the areas of optical and open quantum systems12,13,14. However, majority of the Hermitian and non-Hermitian topological transitions have been recognized as the second order quantum transitions at zero temperature, and can be explained from the perspective of pure quantum states15,16,17.
In the pure states, Berry phase is an interesting phenomenon, which is a quantum phase and remains purely geometrical by excluding the dynamical contribution of the time evolution of states18,19,20. This can be called topological, if the phase is invariant under the path deformation of the evolving quantum state. For a quantum system, which is in contact with the environment, the evolution is affected and results in the production of noise in the form of thermal fluctuations. The quantum states are sensitive to these thermal noises and can become undefined at these instances. This creates a necessity to define topology at finite temperature and gives the idea of mixed quantum states20,21,22. The invariance of the geometrical phase during path deformation of mixed state evolution defines the topological invariant at finite temperature in the form of Uhlmann phase23,24,25,26,27,28,29,30,31. Recently there are also efforts to explain the finite temperature behavior based on ensemble geometric phases32,33 and Wilson loop27 approach for different topological models.
The topological states are protected by certain discrete symmetries and can be classified based on 10-fold symmetry classification. Different symmetry classes possess different physical properties and violation in symmetries reflect in the physical properties of the system2,3. One of the symmetry class called parity-time (\({\mathcal{P}\mathcal{T}}\)) symmetry has efficiently explained the physical properties of non-Hermitian systems34,35,36,37,38. However, the non-Hermitian systems possess a different periodic table for symmetry classifications which decides the physical property of the system39. These physical properties are maintained at zero temperature and the introduction of temperature destroys topological properties of the system. This is a temperature driven transition, where the zero temperature and high temperature represent the pure state and maximally mixed states respectively. The phase transition occurs at critical temperature \(T_c\), bellow which the initial topological properties of the systems are preserved31.
Topological phases are characterized based on the numbers called topological invariants, which are in correspondence with the number of localized edge modes40,41. It is possible to generate higher winding numbers (WNs) either by increasing the number of coupling sites (static method)42 or by periodic driving (dynamical method)43. However, this observation is true for BDI and AIII symmetry classes and holds good only for certain limit. The breaking of symmetry may violate the above property44. With the infinite neighbor coupling, one can obtain long-range models which are interesting from the perspective of massive edge modes and breaking of Lorentz invariance44,45,46,47,48,49,50. The long-range models have been experimentally realized in trapped ions51,52,53,54, atom coupled to multi- mode cavities55, magnetic impurities56,57 and simulated circuits58. They have a advantage of suppressing the finite sized effect over to short-range models59. The massive edge modes are found to be an effective qubits in topological computations48. On the other hand, non-Hermitian system exhibit sensitivity towards the boundary conditions, and there exists non-Hermitian skin effect which creates an extra localization of eigen states in the open boundary condition.
Motivation Dealing the topological state of matter at finite temperature as a condition of mixed state is an interesting area from the perspective of both theory and experimentation. Here our motivation is three-folded.
-
So far, there are many efforts to understand the behavior of geometric phase for mixed states through different approaches. In the work Ref.23, Viyuela et al., has shown that the Uhlmann phase can be an efficient tool to measure topology at finite temperature. The work of Zang et al. (Ref.27), gives the signature of analyzing the possible higher WNs through the mixed state approach (in a four band model). The authors of Refs.31,60 shows the limitations in Uhlmann phase, and explore interferometric phase as a better geometric phase at finite temperature. However, the question about measurement of higher WNs at finite temperature still remains unanswered. In a recent work, we have observed the staircase of topological transitions in 1D extended-range models42. The extended-range of coupling creates higher WNs, and the model reduces to short-range with the increase in the decay parameter. The staircase of transition occurs between even-even (odd-odd) WNs for even (odd) number of interacting neighbors. Here, we are motivated to understand the possibility to define higher WNs and transitions among them at finite temperature.
-
Mixed state approach can be an efficient tool to understand the geometry of non-Hermitian systems, where the system is connected to a thermal bath. Here we consider the non-Hermitian model (which exhibits local imbalance in the hopping amplitudes) and try to explain the topological invariant at finite temperature. Here we also extend our interest to understand the interplay of topology and long-range effects at finite temperature.
-
Some topological systems exhibit localization at criticality40,61,62,63,64,65,66,67,68,69,70,71, and hence there occurs a necessity of defining topology at gapless phases. In our work (Refs.72,73), we have observed a transition among critical regions across a multi-critical point. This is an interesting behavior which can be observed in models extended-range coupling. This finding signals that, bulk gap is not a necessary condition to perform bulk-boundary correspondence40,62. However, here we extend our studies towards topological invariant for mixed states at gapless condition and thereby to understand the relation between localized edge modes and topology of criticality at finite temperature.
Here we consider a topological model connected to a thermal bath at temperature T and can be effectively expressed through Gibb’s ensemble. In order to explain the topological invariant at finite temperature, we use two different approaches called Uhlmann and interferometric phases.
Model Hamiltonian and properties
We consider a generalized two band model for a one dimensional system, which can reflect both Hermitian and non-Hermitian properties based on the system parameter74.
where \(\chi _{x,y,z}\) are the pseudo-spin vectors as a function of quasi-momentum k and \(\sigma _{x,y,z}\) are the Pauli spin matrices. Under the Hermitian condition, the above Hamiltonian behaves \(H(k)=H(k)^{\dagger }\) and under non-Hermitian condition, it is \(H(k)\ne H(k)^{\dagger }\). Because of the Hermiticity, the left and right eigenstate are defined as,
with \(E_{\pm }(k)=\pm \sqrt{(\chi _x(k))^2+(\chi _y(k))^2+(\chi _z(k))^2}\). For the Hermitian condition, we work on the orthonormal basis, i.e.,\(|\psi _{k\pm }\rangle =|\varphi _{k\pm }\rangle\) and \(E_{\pm }(k)=E_{\pm }^*(k)\). For the non-Hermitian systems, we consider biorthonormal basis, i.e., \(\langle \psi _{k\nu }|\varphi _{k\mu }\rangle =\delta _{\nu ,\mu }\) and \(|\varphi _{k\mu }\rangle \langle \psi _{k\mu }|=1\). Here \(\nu\) and \(\mu\) represent the state ‘+ ’and ‘-’respectively. By normalizing with factor \(\sqrt{\langle \psi _{k,\pm }|\varphi _{k,\pm }\rangle }\), we get
with T as transpose operator. The geometric phase due to the adiabatic evolution is given by,
which yield the quantized values for the gapped topological phases19. For Hermitian systems, the azimuthal angle (\(\phi\)) is real and for non-Hermitian it is complex. This behavior is because of the pseudo-spin vectors, which are real for Hermitian systems and complex (at least one component) for non-Hermitian system respectively75, i.e., \(\phi =\phi _{real}(k)+\phi _{imaginary}(k)\). Here the real and imaginary parts contribute to the argument and amplitude respectively. The term \(\phi _{imaginary}(k)\) do not creates any effect on the topology of the system. The non-Hermiticity creates two exceptional points instead of single Dirac (gap closing) point. The curvature function around the exceptional points is given by \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\) respectively and the winding number is given by74,75,
where \(F1(k,\textbf{M})=\partial _k\tan ^{-1}\left( \frac{\chi _y^{re}(k)+\chi _x^{im}(k)}{\chi _x^{re}(k)+\chi _y^{im}(k)}\right)\) and \(F2(k,\textbf{M})=\partial _k\tan ^{-1}\left( \frac{\chi _y^{re}(k)-\chi _x^{im}(k)}{\chi _x^{re}(k)+\chi _y^{im}(k)}\right)\) respectively. (For the detailed derivation, refer the “Method” section). As our model contains imaginary component only in \(\chi _y\) term, the exceptional points are located at \((0, \chi _y^{im})\) and \((0, -\chi _y^{im})\) under the criticality condition \(\chi _x^2+\chi _y^2=0\). Due to the non-Hermiticity, each exceptional point induces its own origin of pseudo-spin space and corresponding winding vectors (WVs). Thus we work on extended winding vectors (EWVs) to understand the non-Hermitian effect in the parameter space. i.e.,
which correspond to parameter space of \(F_1(k,\textbf{M})\) and \(F_2(k,\textbf{M})\) respectively (For detailed study, please refer “Method” section).
Here we consider 1D Su-Schrieffer-Heeger (SSH) model in both Hermitian76 and non-Hermitian versions74,75,77. The model has its importance in understanding of different experimental realizations related to topology78,79,80. The model can be expressed with different range of coupling neighbors, without altering the symmetry of the system44,50. Based on the range of coupling, the model can be categorized as short-range, extended-range and long-range models.
Geometry at finite temperature
The advances in the experimental realizations of the topological matter has made a significant effort in defining the topology at finite temperature. In this regard, there are efforts in defining the topological invariant through Uhlmann and interferometric phases (mixed states) as equivalent tool to Berry phase (pure state)31 in the finite temperature limit (\(T\rightarrow 0\)). Here we study the possibility of defining extended-range systems at finite temperature.
When a system is in contact with the thermal bath at temperature T, it can be defined by the Gibb’s state as31,
with \(\Delta _k=2E(k,\textbf{M})\), which can be defined over the Bloch sphere using the polar coordinates \(r=\sqrt{\chi _y^2+\chi _z^2}\) and \(\phi =tan^{-1}\left( \frac{\chi _y(k)}{\chi _x(k)}\right)\). As the system is temperature dependent, the Bloch sphere is mapped to a sphere centered at the maximally mixed state. For the topological condition, this curve makes a closed loop and for non-topological the curve confines to a single side of the maximally mixed states respectively.
In case of Hermitian models, the method is straightforward, where the Gibbs’s state corresponds to a single parameter space \(F(k,\textbf{M})\). In case of non-Hermitian models, the Gibb’s state can be expressed in terms of parameter spaces \(F1(k,\textbf{M}),F2(k,\textbf{M})\) corresponding to individual exceptional points. Thus the effective geometry of the parameter space can be understood by the combined effect of \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\).
Uhlmann geometric phase Here, the pure states in the Hilbert space \(\mathscr {H}_A\bigotimes \mathscr {H}_B\) form a total space of fiber bundles over the mixed states of \(\mathscr {H}_A\). The geometric phase can be achieved from parallelism condition over the base manifold such that every infinitesimal change \(\delta k\) from the state \(\psi (k)^{\dagger }\) to \(\psi (k+\delta k)\) which equals the fidelity of density matrix20,21,22. i.e.,
where \(\rho (k)=\psi (k)\psi (k)^{\dagger }\) and has a gauge U(N) freedom such that \(\rho \rightarrow \psi (k)U(N)U^{\dagger }(k)\psi ^{\dagger }(k)\) remains unchanged. With the limit \(T\rightarrow 0\), the system drives towards pure state limit and the geometric phase becomes
Thus the Uhlmann geometric phase in pure state limit is technically equivalent to the Berry phase (For detailed study, please refer “Method” section). This methodology can be generalized to both Hermitian and non-Hermitian systems.
For the Hermitian systems the method is straightforward where we have single closing point and the parameter space corresponding to \(F(k,\textbf{M})\) gives the information of geometric phases. The method is slightly modified in the non-Hermitian systems, where we obtain two exceptional points. Thus, the Uhlmann phase can be calculated as the average of individual geometric phases corresponding to parameter spaces of \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\), i.e., \(W_U=\frac{W_{U1}+W_{U2}}{2}\).
Hermitian Su-Schrieffer-Heeger chain
Here we consider 1D Su-Schrieffer-Heeger chain with extended-range of intercell hopping, i.e.,
where t is the intracell hopping and \(t^{\prime }\) is the intercell hopping with power law decay. The term l represents the site index with \(\alpha\) as the decay parameter. With the limit \(l\rightarrow \infty\) (infinite neighbors) the model becomes long-range and as \(\alpha \rightarrow \infty\) the model reduces to short-range (where the system contains only \(W=1\) and \(W=0\) phases). This model resembles the physics of another famous model Kitaev chain in the isotropic limit, where the hopping parameter (J) and pairing parameter (\(\Delta\)) decays with same manner \(\frac{J}{l^{\alpha }}=\frac{\Delta }{l^{\alpha }}\)42,44. The schematic representation is given in Fig. 1a. After Fourier transformation, we write BdG Hamiltonian in the pseudo-spin basis as in Eq. (1) with coefficients
The model takes different phase diagrams and critical conditions based on the number of neighbors as shown in Fig. 4a–d. In this case, the quasi-energy dispersion and curvature function are real quantities given by,
At criticality, the energy dispersion vanishes and curvature function becomes non-analytic.
Non-Hermitian Su-Schrieffer-Heeger chain
Here we construct non-Hermitian SSH chain by introducing an imbalance through the intracell hopping parameter. The Hamiltonian can be written as75,77,
where \(\delta\) is the imbalance term. With the limit \(l\rightarrow \infty\) (infinite neighbors) the model becomes long-range and as \(\alpha \rightarrow \infty\) the model reduces to short-range (where the system contains only \(W=1\) and \(W=0\) phases). The term \(\frac{t^{\prime }}{l^{\alpha }}\) represents extended-range intercell coupling with \(\alpha\) as the decay parameter as schematically represented in Fig. 1b. After Fourier transformation, the Hamiltonian can be written in spin basis as in Eq. (1), with coefficients,
respectively. Due to imbalance in the intracell hopping, the energy dispersion becomes complex and thus the non-Hermiticity is introduced into the model. For the current model, energy dispersion is given by
The curvature function around exceptional points (0,\(\pm \chi _y^{im}(k\))) are given by
whose integral over the closed interval gives the WN. For the non-Hermitian case, WN is calculated as the average of individual WNs around the exceptional points75, i.e., \(W=\frac{W_1+W_2}{2}\). The model exhibits a interesting property called non-Hermitian skin effect, due to which the phase diagram, winding number and localized modes show the sensitivity towards the choice of boundary conditions. Initially we present the geometric properties in the periodic boundary conditions and skin effect in the later sections.
Results
In the section we study the mixed state behavior of topological models with different range of couplings. Here we take the study of pseudo-spin vectors and geometric phase for this purpose. The pseudo-spin (winding) vectors play an important role in defining the topology of a two level system. The number of time WVs rotate or wrap the center of parameter space gives the WN42,81. Here we consider the normalized WVs, which wrap the center of the maximally mixed states. For topological condition, WVs wrap equal to that of WN and for non topological they keeps bouncing at one side of the parameter space. For the phase transition condition, the WVs show a discontinuity in their flow (3D representation) or curvature line touches the center of the parameter space (2D representation). It is interesting that, at criticality, the angle \(\tan \theta =\frac{\chi _y}{\chi _x}\) becomes indeterminate (0/0) as both \(\chi _x,\chi _y\rightarrow 0\) resulting in an ill-defined topological invariant. This situation is similar in case of non-Hermitian systems, where the azimuthal angle is complex. The modified angles around the exceptional points can be written as \(\tan (\theta _{1,2})=\frac{\chi _x^{(im)}+\chi _y^{(re)}}{\chi _x^{(re)}-\chi _y^{(im)}}\), which becomes indeterminate only if the terms \(\chi _y^{(re)},\chi _x^{(re)}\pm \chi _y^{(im)}\rightarrow 0\).
The situation is little different in long-range interaction, where the WVs take the form of polylogarithmic functions. Here \(\tan \theta = \frac{\chi _y}{\chi _x}\) becomes 0/0 at \(k=\pi (\forall \alpha )\) and \(k=0(\alpha >1)\) respectively. This is because, at \(k=0\), the function \(Li_{\alpha }(1)\propto \Gamma (\alpha -1)\) and remain divergent. Hence, for \(\alpha <1\) the WVs show a discontinuity around \(k=0\), resulting in the formation of removable singularities, whose integration gives the fractional WN.
Here we consider three different ranges to understand the interplay of neighboring coupling and mixed state behavior in topological (Hermitian/non-Hermitian) systems.
Short-range couplings
If the coupling parameter has a strength only up to first neighbor, then the model can be called as short-range model, i.e., the decay parameter \(\alpha \rightarrow \infty\). Here, we study the mixed state behavior of Hermitian and non-Hermitian SSH chain with short-range coupling.
Hermitian case
In this case, the Hamiltonian is given by Eq. (11) with \(\alpha \rightarrow \infty\), whose Fourier transform yield Eq. (12) in the form
The criticality occurs at \(t=t^{\prime }(k=0)\) and \(t=-t^{\prime }(k=\pi )\) by separating \(W=0\) and \(W=1\) phases as shown in Fig. 2a. The corresponding Gibb’s states are given by
At \(T=0\), the WVs wrap the origin of the parameter space only one time for \(W=1\) phases (Fig. 3a1), where as for \(W=0\), the WVs do not wrap the origin (Fig. 3a2). With the introduction of the arbitrary temperature (\(T=0.1\)), the WVs show some modifications but the nature remains same as previous. The further increase in temperature, the curves gradually move towards one corner of the parameter space which may finally result in breakdown of topological properties. On the other hand, the Uhlmann phase become equal to Berry phase in the pure state limit \((T\rightarrow 0)\) (Table 1). Our results are consistent with the previous work Ref.31.
Non-Hermitian case
In this case, the Hamiltonian is given by Eq. (14) with the limit \(\alpha \rightarrow \infty\), whose Fourier transform gives Eq. (15) in the form
The criticality condition is given by \(t=-t^{\prime }\pm \delta (k=0)\) and \(t=t^{\prime }\pm \delta (k=\pi )\) which separate phases \(W=0,1/2\) and \(W=1\) as shown in Fig. 2b. Instead of single gap closing point, we get two exceptional points which also produce two parameter spaces for the WVs. The combined effect in the parameter spaces decides the topological behavior of the system. The EWVs are given by,
and corresponding Gibb’s states are given by
At \(T=0\), the EWVs encircle the origin once in both parameter space for \(W=1\) case (Fig. 3b1,b2). For \(W=1/2\), the EWVs encircle the origin of only one parameter space as shown in Fig. 3b3,b4. With the introduction of arbitrary temperature, the curve starts shifting towards one end of the parameter space. The higher temperature may destroy the topological properties of the system. The Uhlmann phase is given by Eq. (10), which becomes equal to Berry phase in the pure state limit \(T\rightarrow 0\) (Table 1).
Extended-range couplings
When the coupling strength is more than one nearest neighbor, the model is called extended model44 and the phase diagram varies based on the number of coupling neighbors as shown in Figs. 4a–d and 6a–d. With the increase of neighbors, the possible WN also increases up to certain level and the model reduces to short-range limit with increasing value of decay parameter. This creates a staircase of topological transitions and we can observe a pattern of transitions among even-even and odd-odd WNs, based on the number of interacting neighbors42.
To understand the mixed state behavior, we consider the simple case \(r=2\), where there are only two neighbor couplings for Hermitian and non-Hermitian situations.
Hermitian case
In this case, model Hamiltonian is given by Eq. (11) with \(r=2\), whose Fourier transform is gives Eq. (12) in the form
The criticality condition is given by
which separate \(W=0,1\) and \(W=2\) phases. The phase diagram is given by Fig. 4c. The Gibb’s states corresponding to above parameter space are given by,
At \(T=0\), the WVs encircle the origin two (one) times for \(W=2 (W=1)\) phases as shown in Fig. 5a,b. With the introduction of an arbitrary temperature, the enclosed area shrinks but the nature remains same. The further increase of the temperature results in the localization of curves towards a corner of the parameter space which results in the breakdown of the topological properties. The Uhlmann phase is given by Eq. (10), which shows a different behavior for extended-range coupling. The Uhlmann phase recognizes \(W=0\) and \(W=1\), but fails to recognize \(W=2\). This result is also holds same to other extended-range models, with finite number of neighbor couplings. This is because, the Uhlmann phase fails to recognize multiple winding of curves around the center of parameter space. The results are presented in Table 2.
Non-Hermitian case
In this case, model Hamiltonian is given by Eq. (14) with \(r=2\), whose Fourier transform is gives Eq. (15) in the form
The criticality occurs at
which separates \(W=0,1/2,1,3,2,2,1\) and 0 topological phases as shown in Fig. 6c.
The EWVs are given by
The Gibb’s states corresponding to above parameter space are given by,
In non-Hermitian extended models, the topological properties are determined by the average behavior of parameter space corresponding to \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\). If the EWVs encircle the origin equal (unequal) number of times, results in integer (fractional) WNs. Sometimes the EWVs encircle one of the origin even (odd) number times and other origin zero times, resulting in integer (fractional) WNs. For \(W=2\) case, the EWVs encircle the center of each parameter space two times (Fig. 7a3,b3), while they encircle each centers only once for \(W=1\). For \(W=1/2\) case, the EWVs encircle one of the center once and do not encircle the other (Fig. 7a1,b1). For \(W=3/2\) case, the EWVs encircle one of the origin twice and the other only once (Fig. 7a2,b2). In some special cases, EWVs encircle the origin of one parameter space twice and do not encircle the other, which also result in \(W=1\) (Fig. 7a4,b4). In non-Hermitian cases, the Uhlmann phase is given by the average of geometric phase with respect to parameter spaces \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\), given by \(W_U=\frac{W_{U1+W_{U2}}}{2}\) in the pure state limit (\(T\rightarrow 0\)). The Uhlmann phase recognizes \(W=0,1/2\) and \(W=1\) phases, but fails to recognize \(W=3/2\) and \(W=2\). This is due to the limitation of the geometry, that Uhlmann, approach do not recognizes the multiple number of windings around the origin. For the same reason, Uhlmann phase fails to recognize the above mentioned special cases, even though the resulting phase is \(W=1\). A detailed study is given in Table 2. These results also holds for the extended-range models with higher number of neighboring couplings.
Long-range couplings
With infinite number of coupling neighbors,the model becomes long-range and the pseudo-spin vectors are expressed in terms of polylogarithmic function. Expansions of polylogarithmic functions gives82,
where \(\Gamma\) function becomes ill-defined for the region \(\alpha <1\) around \(k=0\). This creates a removable singularity in the parameter space, where the winding vectors covers only a half rotation around the axis and bounce back to the original configuration.Thus, they fail to give a integer winding number but produces fractional winding numbers.
Hermitian case
In this case, Hamiltonian is given by Eq. (11) with \(l\rightarrow \infty\), whose Fourier transform gives Eq. (12) in the form
where the term \(Li_{\alpha }\) gives the polylogarithmic function82, which is the consequence of long-range effect. Here the criticality occurs at
as shown in Fig. 4d. The Gibb’s states corresponding to above parameter space are given by,
Due to long-range effect, the WVs behave as \(\sin (k)\) instead of \(\sin (nk)\). Thus we obtain only two topological regions with \(W=0\) and \(W=1\). For \(W=1\) the WVs encircle the axis once and for \(W=0\) they do not encircle the axis. Due to polylogarithmic nature, \(Li_{\alpha }(1)\) show a discontinuous region for \(\alpha <1\), which results in an ill-defined topological region for \(\alpha <1\)42,44,50. Here we observe a discontinuity at \(k=0\), which acts as a removable singularity and yields fractional WN (\(W=1/2\)) (Fig. 8a1). For \(1<\alpha <2\), we observe a less populated WVs around \(k=0\), which do not alters the resulting geometric phase (Fig. 8a2). For the range \(\alpha >2\), we observe a homogeneous distribution of WVs, which is equivalent to the short-range limit, i.e., \(W=1\) (Fig. 8a3). With the introduction of an arbitrary temperature, we observe a variation in density of WVs as shown in Fig. 8b1–b3. In the pure state limit (\(T\rightarrow 0\)), Uhlmann phase fails to recognize the fractional WN but it recognizes the integer WN as shown in Table 3.
Non-Hermitian case
In this case, Hamiltonian is given by Eq. (14) with \(l\rightarrow \infty\), whose Fourier transform gives Eq. (15) in the form
The criticality occurs at
as shown in phase diagram Fig. 6d. The EWVs are given by
The Gibb’s states corresponding to above parameter space are given by,
Here we can observe two kinds of fractional WNs. For \(\alpha <1\) we find fractional WN as a result of polylogarithmic nature, (which are topologically ill-defined) and for \(\alpha >1\) as a result of non-Hermiticity. At \(T=0\), the EWVs corresponding to the parameter space corresponding to \(F1(k,\textbf{M})\) and \(F2(k,\textbf{M})\) show a discontinuity for \(\alpha <1\) region, resulting in an ill defined topological invariant (Fig. 9a1,b1). But this this acts as a removable singularity and yield \(W=1/4\) for \(\alpha <1\) region. For \(1<\alpha <2\), the EWVs encircle the axis once with a less populated arrows around \(k=0\) (Fig. 9a2,b2). However, this does not influences the geometric phase of the system, and we obtain \(W=1\) for \(\delta -\text {Li}_{\alpha }(1)<t<-\delta -\text {Li}_{\alpha }(-1)\) region. For \(\alpha >2\), we get the short-range limit with \(W=1\) where the EWVs encircle the axis of both the parameter space (Fig. 9a3,b3). For the region \({\mp }\delta -\text {Li}_{\alpha }({\mp }1)<t<\pm \delta -\text {Li}_{\alpha }(\pm 1)\), we get \(W=1/2\) where the EWVs encircle the axis of any one of the parameter space (Fig. 9 a4,b4). With the introduction of arbitrary temperature, the EWVs show the modifications as shown in Fig. 9 (Fig. 9c1–c4,d1–d4). In the pure state limit \(T\rightarrow 0\), the Uhlmann phase recognizes the integer and fractional phases for \(\alpha >1\) but fails to recognize the phases in the region \(\alpha <1\). A detailed study is given in Table 3.
From the above study, we understand the limitation of Uhlmann phase in explaining the topological invariants at finite temperature, especially for extended-range models. In extended-range (Hermitian and non-Hermitian) models, the Uhlmann phase fails to reproduce the phase diagram at pure state limit. Here we use another important approach, interferometric geometric phase to explain the finite temperature behavior of extended-range topological models.
Interferometric geometric phase
Here a different geometric phase for mixed states is introduced through the concept of interferometer83. The purification of normalized of state \(|\omega \rangle \in \mathscr {H}_{\omega }\) is
with \(\mathscr {H}_{\omega }=\mathscr {H}_S \otimes \mathscr {H}_A\) and \(|\psi _i\rangle \in \mathscr {H}_A\). Here \(\mathscr {H}_S\) is the Hilbert space of the system and \(\mathscr {H}_A\) is the Hilbert space spanned by the ancillary states with i dimensions. By tracing over the ancillary states, we write the density matrix as, \(\rho =Tr_A(|\omega \rangle \langle \omega |)\). By parameterizing the density operators with a continuous parameter k, such that the eigenvalues of \(\rho (k)\) for each k yield non-degenerate values,
The eigenstates evolve in parallel manner such that two infinitesimally separated eigenstates in \(\mathscr {H}_s\) show parallel transport in order to fix the phase ambiguity of \(|\omega \rangle\) through gauge fixing. The parallelism is given by
which yield the interferometric phase of \(\rho (k)\) across the Brillouin zone. i.e.,
where \(V_i(2\pi )=exp[-\oint dk^{\prime }\langle \psi _i(k^{\prime })|\partial _{k^{\prime }}\psi _i(k^{\prime })\rangle ]\). It is to be noted that, the state \(|\psi _k\rangle\) is affected by parallel transport, while the ancillary states \(|\psi _{k^{\prime }}\rangle\) are not. Hence, the interferometric phase reduces to Berry phase only if \(\rho (k)\) is a parameterized density operator of pure states31,60.
Thus, we can observe the higher WN as well as fractional WNs in the pure state limit. This geometric phase overcomes the limitations of Uhlmann phase and can be effectively used as topological measure in Hermitian and non-Hermitian systems. For the Hermitian system, the method remains straightforward. For the non-Hermitian systems, the interferometric phase can be calculated for the individual parameter spaces corresponding to \(F_{1,2}(k,\textbf{M})\) and combined effect can be observed through the average \(\frac{W_{I1}+W_{I2}}{2}\). Thus, we understand that interferometric geometric phase is a better tool to study the mixed state behavior of extended-range (Hermitian and non-Hermitian) topological models. A comparison is given in Fig. 10.
Topology at gapless condition
Localization is an important property which has direct relation with the topological invariant. We can observe a one to correspondence between the number of localized modes and neighbor coupling up to some extent44. However, increase in the number of neighbor coupling result in the generation of higher winding numbers. With the increase of decay parameter, we observe a staircase of topological transitions among corresponding even-even and odd-odd winding numbers42. The bulk phases contain localized modes, protected by certain discrete symmetries. This naturally exhibits bulk-boundary correspondence and it is believed that, bulk gap is necessary to exhibit bulk-boundary correspondence. Recently, there have been observations of localization even at criticality exhibiting bulk-boundary correspondence, signaling the bulk-boundary correspondence even in the absence of bulk gap72,73,84. By definition, topological invariant is ill-defined at criticality, but there are efforts to define topology at criticality through different means. Here we consider a case of localization at criticality and define the topological invariant by excluding the infinitesimal neighborhood of singular points in the Brillouin zone64.
where \(\{k_i\}\) is the set of critical points in the momentum space. For the Hermitian case, we find a unique behavior at \(k=\pi\) criticality. The critical line \(\mu =-t^{\prime }(-1+\frac{1}{2^{\alpha }})\) line contains a multi-critical point at \(\alpha =1\), where the critical line \(\alpha <1\) witness the transition \(W:2\rightarrow 1\) and \(\alpha >1\) witnesses \(W:0\rightarrow 1\) respectively (Fig. 4c). During the transition \(W:2\rightarrow 1\), out of two edge modes, one localizes at criticality and the other transfers to the \(W=1\) gapped phase. Thus, on the same critical line, we observe a localized mode for \(\alpha <1\) and non localization for \(\alpha >1\), which creates a topological transition along criticality across a multi-critical point72,73,84.
For the region \(\alpha <1\), the WVs show two loops, with the inner one touching the center of the parameter space (Fig. 11 a), while for \(\alpha >1\), WVs show a single loop touching the center (Fig. 11b). The change in configuration occurs at \(\alpha =1\) as shown in Fig. 11c. Even though, all the configurations are occurring on the line \(t=-t^{\prime }(-1+\frac{1}{2^{\alpha }})\), the topological behavior of them are different. Due to the localization property, we observe different WNs even at criticality. But at a finite temperature, Uhlmann phase does not recognizes the WNs at criticality. This limitation can be overcome by using interferometric phase, where we can recognize the WNs at criticality. The comparison of geometric phases are given in Table 4.
Non-Hermitian skin effect
So far, we have constructed the phase diagram and calculated the topological invariant by assuming the periodic boundary condition to Hermitian and non-Hermitian SSH models. Under general conditions, topological invariant is calculated using Bloch band structure and is in good agreement with the number of localized edge modes. This scenario is true in Hermitian systems, while non-Hermitian systems behave differently. Non-Hermitian systems are sensitive to the boundary conditions, which produce different phase diagram for periodic and open boundary conditions. This behavior is due to the phenomenon called non-Hermitian skin effect, which signals the necessity of the construction of ‘non-Bloch topological invariant’. Due to non-Hermitian skin effect, the topologically protected edge modes depends on the spacial dimension and the symmetry class rather than the topological invariant34. The bulk-edge correspondence holds good for the open boundary condition and this creates a necessity to define topological invariant in open chain. In 2018, Yao et al, formulated this framework77 which works on the generalize Brillouin zone and topological invariant is given by
(The construction of generalized Brillouin zone is given in “Method” section). The current form of the modified winding number is due to the presence of chiral symmetry. Under the periodic boundary conditions, bulk Hamiltonian is given by Eq. (1) with \(k\in \left[ 0,2\pi \right]\), whereas under open boundary condition Hamiltonian is given with \(k\in \left[ 0,2\pi \right)\). The presence of sub lattice symmetry creates \(\mathbb {Z}\) topological phase in the presence of line gap complex energy spectrum. The topological invariant has the similar form of that of Hermitian conditions. For the region with point gap complex energy spectrum, there exists \(\mathbb {Z}\oplus \mathbb {Z}\) topological phases, and invariant is given by,
The \(W_P\) becomes zero for an integer \(W_L\), whereas \(W_P\) becomes integer value for fractional \(W_L\).
Here we present the geometric phases of non-Hermitian SSH chain with different range of couplings (Fig. 12), which is different than its counterpart in periodic boundary condition (Fig. 12a1,a2). It clearly shows the sensitivity of phase diagram, towards the choice of boundary conditions. In generalized Brillouin zone, there can occur a significant calculation errors in defining the topological invariant for extended-range due to various reasons89,90. In addition, the Uhlmann phase shows a limitation in defining the geometric phase at finite temperature. The higher winding numbers are not defined by the Uhlmann phase and the phase boundary does not shows an exact point of transition (Fig. 12b1,b2). Instead, the phase is not exactly integer quantized (at least in extended-range) add the phase boundary is spreaded. On the other hand the interferometric phase shows the similar behavior that of Berry phase in the finite temperature (\(T\rightarrow 0\)) limit (Fig. 12c1,c2). However, so far the physical interpretation of Uhlmann phase has not been revealed in an extensive way, which creates a difficulty in interpreting the physical meaning of geometric phases at finite temperature. The Uhlmann phase also has a limitation in establishing the relation between the geometry at finite temperature and the bulk-edge correspondence23,31. The Uhlmann phase works on the principle of the parallel transport of density matrices at finite temperature, thus it recognized the critical temperature and the topological region, while it fails to determine the number of edge modes due to the memory effect31,60. On the other hand, the interferometric phase works straightforwardly based on the principle of ancillary states unlike Uhlmann phase (which works on operator method), and behaves similar to the Berry phase at \(T\rightarrow 0\) limit. However, one can predict the criticality, critical temperature and geometric phases of mixed state topological systems through different means (both for periodic and open boundary conditions), but the understanding towards the bulk-edge correspondence needs some more study.
Discussion
Mixed state behavior of quantum system is an efficient way to understand the thermal fluctuations, especially to define the topology at finite temperature (\(T\rightarrow 0\)). There are a few studies in literature to understand the finite temperature limit of topological system with short-range and long-range interactions. Here we made an attempt to understand the behavior of extended topological models in finite temperature limit. We have extended our observations to the non-Hermitian topological system to understand the geometry and geometric phases at finite temperature.
Introduction of extended range interaction in a topological system is an efficient way to generate the higher WNs, which is an static way. The similar effect can be done by the quenching/periodic driving, which is a dynamical method. In an equilibrium topological system, generally the WN is in correspondence with the localized edge modes in the gapped phases (at least up to some range \(r<L/2\). Sometimes this may not be true if certain symmetries like time reversal is broken). Here we made an effort to study the extended topological chains in the finite temperature limit. We have used the mixed geometric phases like Uhlmann and interferometric phases to understand the topology at finite temperature. We have found that Uhlmann approach has a limitation to define the higher WNs and this can be verified by defining the winding vectors in the form of Gibb’s state. We use interferometric geometric phase to find the geometric phase at finite temperature as shown in Fig. 10. Long-range topological models are the platform to realize the massive edge modes (along with Majorana zero modes), which can also effectively used as topological qubit. Here we obtain fractional WNs for the region where massive edge modes dominate. We carry out finite temperature and observe that the Uhlmann phase has a limitation in defining the fractional WNs. The authors of Ref.60 have worked on similar model and mentioned a different undefined region for Uhlmann phase, due to the nature of superconducting pairing term.
Non-Hermiticity is a prominent area of quantum mechanics and here we have adopted biorthonormal basis vectors to define the topology of the models. The concept can be extended to the mixed states and hence it is possible to understand the geometric phases of non-Hermitian models at finite temperature. However, the non-Hermitian topological models are quite different than the Hermitian systems and have a separate set of periodic table of symmetry protected classes. Moreover, they are sensitive to the boundary condition, and phase diagram can differ according to that. Here we have analyzed chiral non-Hermitian SSH chain and their topology at finite temperature. We observe that the Uhlmann phase recognizes the fractional and integer phases in the short-range limit, but fails to recognize the same in extended-range. It is also to be noted that the similar situation has been observed in the open boundary condition (in the presence of non-Hermitian skin effect), where the Uhlmann phase shows a limitation in defining the higher winding numbers.
As per the traditional definition of topology, WN can be well defined in the gapped phases and ill-defined at the gapless regions. But the experimental and theoretical study at criticality strengthens the possibility of localized states even at criticality. This creates a need of defining topology at criticality and we adopted modified definition to address this issue. The study of Uhlmann phase shows a limitation in defining topological invariant at criticality and fails to recognize the transition among gapless regions across a multi-critical point.
To conclude, we have analyzed the mixed state behavior of Hermitian and non-Hermitian topological models and tried to calculate the geometric phase at \(T\rightarrow 0\) limit. Among the study of geometric phase for mixed states, Uhlmann phase has showed a limitation in defining topological invariant for extended-range at finite temperature. This limitation can be overcome by the interferometric phase in the pure state limit. We also extend our study to non-Hermitian models and analyze the limitation of Uhlmann phase in defining the topological invariants for extended-range models. We have analyzed the geometric phases at open boundary condition and tried to understand the behavior of geometric phases in the presence of non-Hermitian skin effect. We understand that the physics of Uhlmann phase is complicated and there needs further study in this regard. We agree with the statement of Viyuela et al.29, and Anderson et al.31 that the Uhlmann phase does not determine the fate of edge modes at finite temperature’ and we extend validation of this statement to the extended range interaction of Hermitian and non-Hermitian models. We also find that interferometric phase can serve as an efficient tool calculate the topology of (Hermitian and non-Hermitian) extended-range models.
Method
Derivation of effective winding number for non-Hermitian systems
The generalized expression for WN is given by the Eq. (4), with periodic boundary condition the expression is
Due non-Hermiticity, the Eigen vectors of the Hamiltonian can be written as75
where
The non-Hermiticity induces the complexity in winding vectors(at in least one component) with the criticality condition \(\chi _x^2+h\chi _y^2=0\). Here we observe exceptional points (at least two), instead of a single Dirac cone (band gap closing point), whose position can be understood by the complex analysis as,
As our model contains imaginary component only in \(\chi _y\) term, the exceptional points are located at \((0, \chi _y^{im})\) and \((0, -\chi _y^{im})\) under the criticality condition \(\chi _x^2+\chi _y^2=0\). Due to the non-Hermiticity, each exceptional point induces its own origin of pseudo-spin space and corresponding winding vectors. Thus we work on extended winding vectors to understand the non-Hermitian effect in the parameter space. i.e.,
which correspond to parameter space \(F_1(k,\textbf{M})\) and \(F_2(k,\textbf{M})\) respectively (Fig. 13).
As a result of non-Hermiticity the azimuthal angle becomes complex, i.e., \(\phi =\phi ^{re}+i\phi ^{im}\), where the real and imaginary parts contribute to the argument and amplitude respectively. Thus the complex angle can be calculated as,
where \(\tan (\phi _1)=\frac{\chi _y^{re}(k)+\chi _x^{im}(k)}{\chi _x^{re}(k)+\chi _y^{im}(k)}\) and \(\tan (\phi _2)=\frac{\chi _y^{re}(k)+\chi _x^{im}(k)}{\chi _x^{re}(k)-\chi _y^{im}(k)}\). Thus azimuthal angle \(\phi _1(\phi _2)\) corresponding to first (second) exceptional point can be expressed in real angles. The integration of curvature functions \(F1(k,\textbf{M}),F2(k,\textbf{M})\) corresponding these azimuthal angles give the WN as i.e.,
Encircling of these extended winding vectors around the centers of both the parameter space (equal time) yields integer WN. If the encircling is only around one of them yields fractional and neither of them yields \(W=0\) respectively.
Generalized Brillouin zone and topological invariant
Due to the existence of non-Hermitian skin effect, the topological phase diagram becomes sensitive to boundary conditions. Here we adopt the non-Block band theory to treat the topological models under open boundary conditions85. For a short-range model (Eq. (14) with \(\alpha \rightarrow \infty\)), the real space Eigen equation leads
Considering the ansatz \(|\psi \rangle =\sum _{j}|\phi ^{(j)}\rangle\), where each \(|\phi ^{(j)}\rangle\) takes the form (omitting the j index temporarily) \((\phi _{n,A},\phi _{n,B})=\beta ^n(\phi _{A},\phi _{B})\). We obtain the relation
the energy equation is given by
Under the limit \(E\rightarrow 0\), we obtain the roots
Restoring the j index, we obtain
Thus the general solution for \(\psi\) can be written as linear superposition of \(\phi\) as \(\psi =\sum _{n}\beta ^{(n)}\phi\). For the spectrum in thermodynamic limit, the bulk-Eigenstates follow the rule \(|\beta _1|=|\beta _2|\), which leads to
For the case \(r=1\), the non-Hermitian model behaves similar to that of the Hermitian counterpart88.
Topological invariant Due to non-Bloch band structure, the components get modified as \(e^{ik}\rightarrow \beta ,e^{-ik}\rightarrow \beta ^{-1}\) and the Hamiltonian is given by
where \(\sigma _{\pm }=\frac{(\sigma _x+i\sigma _y)}{2}\). Here the k term takes the complex value \(k\rightarrow k-i\ln (r_0)\). The energy relations are given by
where \(|\tilde{u_R}\rangle =\sigma _z|u_R\rangle ,|\tilde{u_L}\rangle =\sigma _z|u_L\rangle\) which follow the condition \(\langle u_L|u_R\rangle =\langle \tilde{u_L}|\tilde{u_R}\rangle =1,\langle u_L|\tilde{u_R}\rangle =\langle \tilde{u_L}|u_R\rangle =0\) leading to the Q-matrix \(Q(\beta )=|\tilde{u_R}(\beta )\rangle \langle \tilde{u_L}(\beta )|-|u_R(\beta )\rangle \langle u_L(\beta )|\). The topological invariant in generalized Brillouin zone is given by
where H(k) can be written in the corrected form as
which defines the topological invariant in the generalized Brillouin zone10,85,86,87,88.
Extended-range couplings For the extended-range models such as \(r=2\), the real space wavefunctions can be written as,
with energy relation
Since, the energy equation takes quartic form, there occurs four roots with
The topological invariant can be calculated by using Eq. (60), where
where \(\beta\) is given by Eq. (64). The construction of generalized Brillouin zone is highly sensitive towards the system parameters and there can occur the problems of calculation accuracy10,86,87,88,89,90. This construction can be unreliable due to numerical diagonalization errors and sensitivity towards matrix dimensions89. Thus it can be really difficult and time consuming to construct the generalize Brillouin zone for extended models to verify the edge mode behavior. However, we observe some efforts to define topological invariant in generalized Brillouin zone for various extended models10,77,86, which is similar to our model. Here we limit our study to the second interacting neighbor where the Uhlmann phase at finite temperature shows its limitation in recognizing the higher topological invariant.
Derivation of Uhlmann geometric phase
The parallelism condition is explained by the H\(\ddot{u}\)bner relation,
where \(\psi =\sqrt{\rho },p_i\) and \(|u_i\rangle\) are the eigenvector and eigenvalues of \(\rho\) respectively. For a Bloch sphere (of two level system), the above relation becomes
with
For the Abelian condition, Eq. (67) becomes
Over the closed loop, the group element becomes
with \(B=\frac{1}{2}\oint dk(\partial _k\psi )(\sqrt{p_1}-\sqrt{p_2})^2\). It is to be noted that, the term B need not be periodic, even though \(\theta _k\) is periodic. Uhlmann phase is given by the phase factor of Eq. (69).
where \(\psi _{||}(k)=\psi (k)U(k)\). For \(\phi (0)=0\), we get
With the limit \(T\rightarrow 0\), the system drives towards pure state limit and Eq. (71) becomes
Similar methodology can be adopted to define the geometric phase for open boundary condition. Here the term B (Eq. 69) behaves independent of periodicity, which signals the generalization of Uhlmann phase to incorporate non-Hermitian skin effect. By replacing the Block wave function \(u_k\) with non-Block function \(\beta\), this can be achieved. The parallelism of the density matrix allows the wavefunctions to define Uhlmann phase at finite temperature. The ‘Q’ matrix defines the parallel condition and the trace of the matrix defined the geometric phase in the generalized Brillouin zone.
Data availability
The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.
References
Stanescu, T. D. Introduction to Topological Quantum Matter & Quantum Computation (CRC Press, 2016).
Haldane, F. D. M. Nobel lecture: Topological quantum matter. Rev. Modern Phys. 89, 040502 (2017).
Ortmann, F., Roche, S. & Valenzuela, S. O. Topological Insulators: Fundamentals and Perspectives (Wiley, 2015).
Bernevig, B. A. & Hughes, T. L. Topological Insulators and Topological Superconductors (Princeton University Press, 2013).
Hasan, M. Z. & Kane, C. L. Colloquium: Topological insulators. Rev. Modern Phys. 82, 3045 (2010).
Moore, J. E. The birth of topological insulators. Nature 464, 194–198 (2010).
Kitaev, A. Y. Unpaired majorana fermions in quantum wires. Physics-Uspekhi 44, 131 (2001).
Sarma, S. D., Freedman, M. & Nayak, C. Majorana zero modes and topological quantum computation. NPJ Quant. Informat. 1, 1–13 (2015).
Gong, Z. et al. Topological phases of non-Hermitian systems. Phys. Rev. X 8, 031079 (2018).
Yokomizo, K. & Murakami, S. Non-bloch band theory of non-Hermitian systems. Phys. Rev. Lett. 123, 066404 (2019).
Ge, Z.-Y. et al. Topological band theory for non-Hermitian systems from the dirac equation. Phys. Rev. B 100, 054105 (2019).
Ashida, Y., Gong, Z. & Ueda, M. Non-Hermitian physics. Adv. Phys. 69, 249–435 (2020).
Longhi, S. Parity-time symmetry meets photonics: A new twist in non-Hermitian optics. EPL (Europhys. Lett.) 120, 64001 (2018).
Moiseyev, N. Non-Hermitian Quantum Mechanics (Cambridge University Press, 2011).
Arouca, R., Lee, C. & Smith, C. M. Unconventional scaling at non-Hermitian critical points. Phys. Rev. B 102, 245145 (2020).
Cats, P., Quelle, A., Viyuela, O., Martin-Delgado, M. & Smith, C. M. Staircase to higher-order topological phase transitions. Phys. Rev. B 97, 121106 (2018).
Chen, S., Wang, L., Hao, Y. & Wang, Y. Intrinsic relation between ground-state fidelity and the characterization of a quantum phase transition. Phys. Rev. A 77, 032111 (2008).
Berry, M. Classical adiabatic angles and quantal adiabatic phase. J. Phys. A Math. General. 18, 15 (1985).
Wilczek, F. & Shapere, A. Geometric Phases in Physics Vol. 5 (World Scientific, 1989).
Buchleitner, A., Viviescas, C. & Tiersch, M. Entanglement and Decoherence: Foundations and Modern Trends Vol. 768 (Springer Science & Business Media, 2008).
Uhlmann, A. Parallel transport and “quantum holonomy’’ along density operators. Rep. Math. Phys. 24, 229–240 (1986).
Uhlmann, A. On berry phases along mixtures of states. Annalen der Physik 501, 63–69 (1989).
Viyuela, O. et al. Observation of topological uhlmann phases with superconducting qubits. NPJ Quant. Inform. 4, 1–6 (2018).
Quan, H. & Cucchietti, F. Quantum fidelity and thermal phase transitions. Phys. Rev. E 79, 031101 (2009).
Mera, B., Vlachou, C., Paunković, N. & Vieira, V. R. Uhlmann connection in fermionic systems undergoing phase transitions. Phys. Rev. Lett. 119, 015702 (2017).
Tong, D., Sjöqvist, E., Kwek, L. C. & Oh, C. H. Kinematic approach to the mixed state geometric phase in nonunitary evolution. Phys. Rev. Lett. 93, 080405 (2004).
Zhang, Y., Pi, A., He, Y. & Chien, C.-C. Comparison of finite-temperature topological indicators based on uhlmann connection. Phys. Rev. B 104, 165417 (2021).
Amin, S. T., Mera, B., Vlachou, C., Paunković, N. & Vieira, V. R. Fidelity and uhlmann connection analysis of topological phase transitions in two dimensions. Phys. Rev. B 98, 245141 (2018).
Viyuela, O., Rivas, A. & Martin-Delgado, M. Symmetry-protected topological phases at finite temperature. 2D Mater. 2, 034006 (2015).
Viyuela, O., Rivas, A. & Martin-Delgado, M. Two-dimensional density-matrix topological fermionic phases: Topological uhlmann numbers. Phys. Rev. Lett. 113, 076408 (2014).
Andersson, O., Bengtsson, I., Ericsson, M. & Sjöqvist, E. Geometric phases for mixed states of the kitaev chain. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 374, 20150231 (2016).
Pi, A., Zhang, Y., He, Y. & Chien, C.-C. Proxy ensemble geometric phase and proxy index of time-reversal invariant topological insulators at finite temperatures. Phys. Rev. B 105, 085418 (2022).
Molignini, P. & Cooper, N. Topological phase transitions at finite temperature. arXiv preprint arXiv:2208.08994 (2022).
Kawabata, K., Shiozaki, K., Ueda, M. & Sato, M. Symmetry and topology in non-Hermitian physics. Phys. Rev. X 9, 041015 (2019).
Kawabata, K., Ashida, Y., Katsura, H. & Ueda, M. Parity-time-symmetric topological superconductor. Phys. Rev. B 98, 085116 (2018).
Wang, X., Liu, T., Xiong, Y. & Tong, P. Spontaneous pt-symmetry breaking in non-Hermitian kitaev and extended kitaev models. Phys. Rev. A 92, 012116 (2015).
Shiozaki, K. & Ono, S. Symmetry indicator in non-Hermitian systems. Phys. Rev. B 104, 035424 (2021).
Agarwal, K. S. & Joglekar, Y. N. Pt-symmetry breaking in a kitaev chain with one pair of gain-loss potentials. Phys. Rev. A 104, 022218 (2021).
Zhou, H. & Lee, J. Y. Periodic table for topological bands with non-Hermitian symmetries. Phys. Rev. B 99, 235112 (2019).
Verresen, R., Jones, N. G. & Pollmann, F. Topology and edge modes in quantum critical chains. Phys. Rev. Lett. 120, 057001 (2018).
Niu, Y. et al. Majorana zero modes in a quantum ising chain with longer-ranged interactions. Phys. Rev. B 85, 035110 (2012).
Kartik, Y., Kumar, R. R., Rahul, S., Roy, N. & Sarkar, S. Topological quantum phase transitions and criticality in a longer-range kitaev chain. Phys. Rev. B 104, 075113 (2021).
Tong, Q.-J., An, J.-H., Gong, J., Luo, H.-G. & Oh, C. Generating many majorana modes via periodic driving: A superconductor model. Phys. Rev. B 87, 201109 (2013).
Alecce, A. & Dell’Anna, L. Extended kitaev chain with longer-range hopping and pairing. Phys. Rev. B 95, 195160 (2017).
Lepori, L., Vodola, D., Pupillo, G., Gori, G. & Trombettoni, A. Effective theory and breakdown of conformal symmetry in a long-range quantum chain. Ann. Phys. 374, 35–66 (2016).
Jäger, S. B., Dell’Anna, L. & Morigi, G. Edge states of the long-range kitaev chain: an analytical study. Phys. Rev. B 102, 035152 (2020).
Ares, F., Esteve, J. G., Falceto, F. & de Queiroz, A. R. Entanglement entropy in the long-range kitaev chain. Phys. Rev. A 97, 062301 (2018).
Viyuela, O., Vodola, D., Pupillo, G. & Martin-Delgado, M. A. Topological massive dirac edge modes and long-range superconducting hamiltonians. Phys. Rev. B 94, 125121 (2016).
Vodola, D., Lepori, L., Ercolessi, E. & Pupillo, G. Long-range ising and kitaev models: Phases, correlations and edge modes. N. J. Phys. 18, 015001 (2015).
Vodola, D., Lepori, L., Ercolessi, E., Gorshkov, A. V. & Pupillo, G. Kitaev chains with long-range pairing. Phys. Rev. Lett. 113, 156402 (2014).
Deng, X.-L., Porras, D. & Cirac, J. I. Effective spin quantum phases in systems of trapped ions. Phys. Rev. A 72, 063407 (2005).
Britton, J. W. et al. Engineered two-dimensional ising interactions in a trapped-ion quantum simulator with hundreds of spins. Nature 484, 489–492 (2012).
Hauke, P. et al. Complete devil’s staircase and crystal-superfluid transitions in a dipolar xxz spin chain: A trapped ion quantum simulation. N. J. Phys. 12, 113037 (2010).
Roy, N., Sharma, A. & Mukherjee, R. Quantum simulation of long-range x y quantum spin glass with strong area-law violation using trapped ions. Phys. Rev. A 99, 052342 (2019).
Douglas, J. S. et al. Quantum many-body models with cold atoms coupled to photonic crystals. Nat. Photon. 9, 326–331 (2015).
Zhang, K., Wang, P. & Song, Z. Majorana flat band edge modes of topological gapless phase in 2d kitaev square lattice. Sci. Rep. 9, 1–9 (2019).
Ménard, G. C. et al. Long range coherent magnetic bound states in superconductors. arXiv preprint arXiv:1506.06666 (2015).
Amin, S. T., Mera, B., Paunković, N. & Vieira, V. R. Information geometric analysis of long range topological superconductors. J. Phys. Condensed Matter 31, 485402 (2019).
Gong, Z.-X. et al. Topological phases with long-range interactions. Phys. Rev. B 93, 041102 (2016).
Bhattacharya, U. & Dutta, A. Topological footprints of the kitaev chain with long-range superconducting pairings at a finite temperature. Phys. Rev. B 97, 214505 (2018).
Jones, N. G. & Verresen, R. Asymptotic correlations in gapped and critical topological phases of 1d quantum systems. J. Stat. Phys. 175, 1164–1213 (2019).
Verresen, R., Thorngren, R., Jones, N. G. & Pollmann, F. Gapless topological phases and symmetry-enriched quantum criticality. arXiv preprint arXiv:1905.06969 (2019).
Thorngren, R., Vishwanath, A. & Verresen, R. Intrinsically gapless topological phases. arXiv preprint arXiv:2008.06638 (2020).
Verresen, R. Topology and edge states survive quantum criticality between topological insulators. arXiv preprint arXiv:2003.05453 (2020).
Kestner, J., Wang, B., Sau, J. D. & Sarma, S. D. Prediction of a gapless topological haldane liquid phase in a one-dimensional cold polar molecular lattice. Phys. Rev. B 83, 174409 (2011).
Cheng, M. & Tu, H.-H. Majorana edge states in interacting two-chain ladders of fermions. Phys. Rev. B 84, 094503 (2011).
Fidkowski, L., Lutchyn, R., Nayak, C. & Fisher, M. Majorana zero modes in 1d quantum wires without long-ranged superconducting order. APS 2012, Q29-002 (2012).
Sau, J. D., Halperin, B., Flensberg, K. & Sarma, S. D. Number conserving theory for topologically protected degeneracy in one-dimensional fermions. Phys. Rev. B 84, 144509 (2011).
Kraus, C. V., Dalmonte, M., Baranov, M. A., Läuchli, A. M. & Zoller, P. Majorana edge states in two atomic wires coupled by pair-hopping. arXiv preprint arXiv:1302.0701 (2013).
Scaffidi, T., Parker, D. E. & Vasseur, R. Gapless symmetry-protected topological order. Phys. Rev. X 7, 041048 (2017).
Jiang, H.-C., Li, Z.-X., Seidel, A. & Lee, D.-H. Symmetry protected topological luttinger liquids and the phase transition between them. Sci. Bull. 63, 753–758 (2018).
Kumar, R. R., Kartik, Y. R., Rahul, S. & Sarkar, S. Multi-critical topological transition at quantum criticality. Sci. Rep. 11, 1–20 (2021).
Rahul, S., Kumar, R. R., Kartik, Y. & Sarkar, S. Majorana zero modes and bulk-boundary correspondence at quantum criticality. J. Phys. Soc. Jpn. 90, 094706 (2021).
Zhu, B., Ke, Y., Zhong, H. & Lee, C. Dynamic winding number for exploring band topology. Phys. Rev. Res. 2, 023043 (2020).
Yin, C., Jiang, H., Li, L., Lü, R. & Chen, S. Geometrical meaning of winding number and its characterization of topological phases in one-dimensional chiral non-Hermitian systems. Phys. Rev. A 97, 052115 (2018).
Su, W.-P., Schrieffer, J. & Heeger, A. Soliton excitations in polyacetylene. Phys. Rev. B 22, 2099 (1980).
Yao, S. & Wang, Z. Edge states and topological invariants of non-Hermitian systems. Phys. Rev. Lett. 121, 086803 (2018).
Zeuner, J. M. et al. Observation of a topological transition in the bulk of a non-Hermitian system. Phys. Rev. Lett. 115, 040402 (2015).
Weimann, S. et al. Topologically protected bound states in photonic parity-time-symmetric crystals. Nat. Mater. 16, 433–438 (2017).
Poli, C., Bellec, M., Kuhl, U., Mortessagne, F. & Schomerus, H. Selective enhancement of topologically induced interface states in a dielectric resonator chain. Nat. Commun. 6, 1–5 (2015).
Sarkar, S. Quantization of geometric phase with integer and fractional topological characterization in a quantum ising chain with long-range interaction. Sci. Rep. 8, 1–20 (2018).
Olver, F. W., Lozier, D. W., Boisvert, R. F. & Clark, C. W. NIST Handbook of Mathematical Functions Hardback and CD-ROM (Cambridge University Press, 2010).
Sjöqvist, E. et al. Geometric phases for mixed states in interferometry. Phys. Rev. Lett. 85, 2845 (2000).
Kumar, R. R., Roy, N., Kartik, Y., Rahul, S. & Sarkar, S. Topological phase transition at quantum criticality. arXiv preprint arXiv:2112.02485 (2021).
Okuma, N., Kawabata, K., Shiozaki, K. & Sato, M. Topological origin of non-Hermitian skin effects. Phys. Rev. Lett. 124, 086801 (2020).
Fu, Y. & Zhang, Y. Anatomy of open-boundary bulk in multiband non-Hermitian systems. Phys. Rev. B 107, 115412 (2023).
Zhang, K., Yang, Z. & Fang, C. Correspondence between winding numbers and skin modes in non-Hermitian systems. Phys. Rev. Lett. 125, 126402 (2020).
Kawabata, K., Okuma, N. & Sato, M. Non-bloch band theory of non-Hermitian hamiltonians in the symplectic class. Phys. Rev. B 101, 195147 (2020).
Yang, Z., Zhang, K., Fang, C. & Hu, J. Non-Hermitian bulk-boundary correspondence and auxiliary generalized Brillouin zone theory. Phys. Rev. Lett. 125, 226402 (2020).
Lin, R., Tai, T., Li, L. & Lee, C.H. Topological Non-Hermitian skin effect. arXiv preprint arXiv:2302.03057 (2021).
Acknowledgements
SS would like to acknowledge DST (CRG/2021/000996) for the support and RRI, SN Bose library for the books and journals. YRK is grateful to AMEF for providing the PhD fellowship. The authors would like to acknowledge Utso Bhattacharya, BS Ramachandra, C Sivaram, Rahul S and Ranjith Kumar R for useful discussions.
Author information
Authors and Affiliations
Contributions
Y.R.K. identified the problem and also wrote the manuscript. Both Y.R.K. and S.S. have analyzed the results and reviewed the manuscript.
Corresponding author
Ethics declarations
Competing Interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Kartik, Y.R., Sarkar, S. Mixed state behavior of Hermitian and non-Hermitian topological models with extended couplings. Sci Rep 13, 6431 (2023). https://doi.org/10.1038/s41598-023-33449-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-33449-9
- Springer Nature Limited