Competition of long-range interactions and noise at ramped quench dynamical quantum phase transition: The case of the long-range pairing Kitaev chain

Long-range interactions has been attracting great interest due to revealing surprising features [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14]. Long-range systems are often fascinating approach to analyze the validity of the hypotheses that are otherwise clearly perceived for prototypical short-range systems. The remarkable experimental advancement in ultracold atomic platforms has triggered a plethora of theoretical studies and opened up the possibility of engineering and fine-tuning long-range systems with great accuracy [15, 16, 17, 18, 19, 20].

Moreover, the unprecedented advancement in recent years is impeccable enough to study the none-equilibrium dynamics of long-range systems in a controled manner [21, 22, 23, 24]. Consequently, over the latest decades, theoretical and experimental research has raised a great deal of interest in non-equilibrium quantum phenomena [25], which has led to the discovery of some intriguing physics, including the observation of Kibble-Zurek phenomena [26, 27], discrete time-crystals [28], many-body localization [29] and the breaking of ergodicity [30, 31, 32].

In recent years, the concept of dynamical quantum phase transitions (DQPTs) have been introduced as nonequilibrium counterparts of thermal phase transitions [33, 34]. Within DQPTs real time plays the role of control parameter analogous to temperature in equilibrium phase transitions [35, 36, 37, 38, 39, 40, 41, 42, 43, 44]. While the conventional equilibrium phase transition is characterized by nonanalyticities in the thermal free energy, the DQPT is represented by the nonanalytical behavior of dynamical free energy [45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61]. DQPT displays a phase transition between dynamically emerging quantum phases, that takes place during the nonequilibrium coherent quantum time evolution under sudden/ramped quench [62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80] or time-periodic modulation of Hamiltonian [81, 82, 83, 84, 85, 86, 87, 88]. Furthermore, analogous to order parameters at equilibrium quantum phase transition, a dynamical topological order parameter is proposed to capture DQPTs [89, 90].

DQPT was observed experimentally in several studies [91, 92, 93, 94, 95, 96, 97] to confirm theoretical anticipation. Most of these researches associated with deterministic quantum evolution generated by ramping or a sudden quench of the Hamiltonian. However, relatively little attention has been devoted to the stochastic driving of thermally isolated systems with noisy Hamiltonian. In any real experiment, the simulation of the desired time dependent Hamiltonian is imperfect and noisy fluctuations are inevitable [98, 99, 100, 101]. Therefore, understanding the effects of noise in such systems is of utmost importance both in designing experiments and comprehend the results [102, 103, 104, 105, 106].

Despite numerous studies of DQPTs in a wide variety of long-range quantum systems [107, 108, 109, 110, 111, 112, 113], comparatively little attention has been paid to the noise effects [114] on long-range interaction properties. In the present work, we contribute to develop the systematic understanding of the competition between noise and long-range interaction at noisy ramped quench DQPT. For this purpose, we investigate the ramped quench DQPT of long-range pairing Kitaev model [13] in the presence of the white noise with Gaussian distribution [104]. We solve an exact master equation for the quench dynamics averaged over the noise distribution. This allows us to study the competition between the near-adiabatic quench dynamics of the gapped modes of the long-range pairing system and the accumulation of noise induced excitations.

We show that, for the quench across a single critical point, while the long-range pairing induces a region with three DQPTs time scales (three critical modes), this region shrinks in the presence of the noise. In addition, for a quench that crosses two critical points, the critical sweep velocity above which the DQPTs disappear, enhances by the long-range pairing while decreases in the presence of the noise. In other words, the noise has destructive effects on long-range pairing features.

The paper is organized as follows. In Sec. II, the dynamical free energy and DTOP of the two band Hamiltonians are discussed. In Sec. III, we present the model and review its exact solution and equilibrium phase transition. Section IV is dedicated to the numerical simulation of the noiseless case based on the analytical result. The effects of noise on the system is numerically studied in Section V. Section VI contains some concluding remarks.

Recently, an extension of the Kitaev model [120] which describes the algebraic decay of the tunneling and/or pairing terms has been intensively investigated [121, 13, 12, 122, 123]. This model describes experimental realizations of long-range topological superconductors [124, 125]. It has been shown that the phase diagram is modified in the presence of the long-range interactions [13, 122]. Moreover, this model exhibits algebraically localized edge states and an algebraic closing of the energy gap [13, 122]. However, when the pairing and tunneling terms are isotropic, exponential localization is recovered independent of the power-law exponent, as long as it is larger than unity [122, 12].

In this paper, we investigate how the noise affects features of the long-range interaction in the long-range pairing Kitaev model. Representing fermionic annihilation (creation) operators as $c_{n}(c_{n}^{\dagger})$, the Hamiltonian of the long-range pairing Kitaev model with linear time dependent chemical potential is given as

	$\displaystyle H$	$\displaystyle=$	$\displaystyle-w\sum_{n=1}^{N}\left(c_{n}^{\dagger}c_{n+1}+h.c.\right)-\mu(t)\sum_{n=1}^{N}\left(c_{n}^{\dagger}c_{n}-{{1}\over{2}}\right)$		(8)
			$\displaystyle+{{\Delta}\over{2}}\sum_{n,\ell}d_{\ell}^{-\alpha}\left(c_{n}c_{n+\ell}+c_{n+\ell}^{\dagger}c_{n}^{\dagger}\right)$

where $w$ denotes the hopping strength of the fermionic particles between adjacent lattice sites, $\Delta$ is the strength of the superconducting pairing term that decays with distance $l$ in a power law fashion characterized by exponent $\alpha$, and the onsite time dependent chemical potential $\mu(t)=\mu_{f}+vt$ changes from the initial value $\mu_{i}$ at time $t=t_{i}<0$ to the final values $\mu_{f}$ at $t=t_{f}=0$ with sweep velocity $v$. The effective distance $d_{\ell}$, between two sites denoted by $n$ and $n+\ell$ on the closed ring with $N$ sites, is given by the function $d_{\ell}=\min(\ell,N-\ell)$.

In the presence of the long-range pairing, the Hamiltonian Eq. (8) is exactly solvable in the momentum space [13]. Introducing the Nambu spinor $\Gamma^{\dagger}_{k_{m}}=(c_{k_{m}}^{\dagger},~{}c_{-k_{m}})$, the Fourier transformed Hamiltonian can be expressed as the sum of independent terms acting in the two-dimensional Hilbert space generated by $k$

$\displaystyle{\cal H}(t)=\frac{1}{2}\sum_{m=1}^{N/2}\Gamma_{k_{m}}^{\dagger}H^{(0)}_{k_{m}}(t)\Gamma_{k_{m}},$

(9)

where $H^{(0)}_{k_{m}}(t)$ (the superscript in $H^{(0)}_{k_{m}}(t)$ is introduced to denote noise-free driving) is given by

$\displaystyle H^{(0)}_{k_{m}}(t)$

$\displaystyle=\begin{pmatrix}-(2w\cos k_{m}+\mu(t))&i\Delta f_{\alpha}(k_{m})\\ -i\Delta f_{\alpha}(k_{m})&(2w\cos k_{m}+\mu(t))\end{pmatrix},$

(10)

where $f_{\alpha}(k_{m})=\sum_{\ell=1}^{N-1}{{\sin(k_{m}\ell)}/{d_{\ell}^{\alpha}}}$ is the Fourier transform of the superconducting gap term and $k_{m}=(2m-1)\pi/N,~{}~{}m=1,2,\cdots N/2$. In the thermodynamic limit $N\rightarrow\infty$, when $k_{m}$ gets continuous values, the function $f_{\alpha}(k_{m})$, is described as $f^{\infty}_{\alpha}(k)=-\frac{i}{2}\left(\mathbf{Li}_{\alpha}(e^{ik})-\mathbf{Li}_{\alpha}(e^{-ik})\right)$ with $\mathbf{Li}_{\alpha}(z)=\sum_{\ell=1}^{\infty}{{z^{\ell}}/{\ell^{\alpha}}}$ being the polylogarithmic function of $z$ that vanishes in the limit $k\to 0$ and $k\to\pi$ for $\alpha>1$. When $\alpha<1$ the polylogarithmic function only vanishes in the limit $k\to\pi$.

In the limit of $\alpha\rightarrow\infty$, the model reduces to that of the short-range Kitaev chain with only nearest-neighbor pairing which is exactly solvable and its topological properties were unravelled by Kitaev [120]. In this limit, for time-independent chemical potential $\mu(t)=\mu$ and $w=1$, the time-independent Hamiltonian undergoes topological quantum phase transitions at $\mu_{c}=\pm 2$, where the energy gap closes at $k=0,\pi$ [120]. For $\alpha>1$ the phase diagram and the topological properties of the long-range pairing Kitaev chain are identical to that of a short-range Kitaev chain (Fig. (2)). However, as $\alpha$ approaches $1$, the bulk gradually starts becoming gapped near $\mu=-2$ and for $\alpha<1$, $\mu=-2$ no longer remains a critical point [13].

In the time dependent case $\mu(t)=\mu_{f}+vt$, the instantaneous eigenvalues and eigenvectors of time dependent Hamiltonian Eq.(8), are given by

	$\displaystyle\varepsilon^{\pm}_{k_{m}}$	$\displaystyle=$	$\displaystyle\pm\varepsilon_{k_{m}}=\pm\sqrt{h^{2}_{z}(k_{m},t)+h^{2}_{xy}(k_{m})},$		(11)
	$\displaystyle|\chi^{-}_{k_{m}}(t)\rangle$	$\displaystyle=$	$\displaystyle\cos(\frac{\theta_{k_{m}}(t)}{2})|\uparrow\rangle-{\it i}\sin(\frac{\theta_{k_{m}}(t)}{2})|\downarrow\rangle,$
	$\displaystyle|\chi^{+}_{k_{m}}(t)\rangle$	$\displaystyle=$	$\displaystyle-{\it i}\sin(\frac{\theta_{k_{m}}(t)}{2})|\uparrow\rangle+\cos(\frac{\theta_{k_{m}}(t)}{2})|\downarrow\rangle,$

where,

$$\cos(\frac{\theta_{k_{m}}(t)}{2})=\frac{\varepsilon_{k_{m}}-h_{z}({k_{m}},t)}{\sqrt{2\varepsilon_{k_{m}}(\varepsilon_{{k_{m}}}-h_{z}({k_{m}},t))}},$$

$$\sin(\frac{\theta_{k_{m}}(t)}{2})=\frac{h_{xy}(k_{m})}{\sqrt{2\varepsilon_{{k_{m}}}(\varepsilon_{{k_{m}}}-h_{z}({k_{m}},t))}},$$

with $h_{xy}({k_{m}})=\Delta f_{\alpha}(k_{m})$, and $h_{z}({k_{m}},t)=2w\cos({k_{m}})+\mu(t)$, and $|\chi^{\pm}_{k_{m},t}\rangle$ are the adiabatic basis of the system.

In such a case, if the system is prepared in its ground state at $t_{i}\rightarrow-\infty$ ($\mu_{i}\ll\mu_{c}=-2$), the probability that the $k$:th mode is found in the upper level at $t$ is given as (see Appendix A)

	$\displaystyle p_{k}=e^{-\pi\gamma^{2}/4}|U_{22}\cos(\frac{\theta_{k_{m}}(t)}{2})-\frac{\gamma e^{-{\it i}\pi/4}}{\sqrt{2}}U_{12}\sin(\frac{\theta_{k_{m}}(t)}{2})|^{2},$
			(12)

with $U_{22}=D_{\nu}(x),~{}~{}U_{12}=D_{\nu-1}(x),$ where, $D_{\nu}(x)$ is the parabolic cylinder function [126, 127], $\gamma=\Delta f_{\alpha}(k_{m})/\sqrt{2v}$, $\nu={\it i}\gamma^{2}/2$, $x=2e^{{\it i}3\pi/4}\sqrt{v}\tau_{k}$, and $\tau_{k}=((\mu_{f}+vt)/2+w\cos(k))/v$.

In this section, we report the results of our numerical simulations, based on an analytical approach, to investigate the dynamics of the model using the notion of DQPTs. To this end, we consider the linear quenching of the chemical potential $\mu(t)=\mu_{f}+vt$, changes from initial value $\mu_{i}\rightarrow-\infty$, where the system is prepared in its ground state, to various final values $\mu_{f}=1,1.95,4$ at $t_{f}=0$. In addition, to better understand the effect of long-range pairing on the dynamics of system after the ramp quench, we will focus only on the case $\alpha>1$ where the location of the critical points in the parameter space is not altered by varying $\alpha$.

As mentioned the noises are ubiquitous and indispensable in any physical system. Specifically, when energy is transferred into or out of an otherwise isolated system via a quench in the laboratory, there will inevitably be time dependent fluctuations (”noise”) in this transfer. In this section we investigate the effects of noise on the dynamical phase diagram of the long-range pairing Kitaev model. For this purpose, we add a noise to the time dependent chemical potential $\mu(t)=\mu_{f}+vt+R(t)$, where $R(t)$ is a random fluctuation confined to the ramp interval $[t_{i},t_{f}=0[$, with vanishing mean, $\langle R({t})\rangle=0$. We use white noise with Gaussian two-point correlations $\langle R(t)R(t^{\prime})\rangle=\xi^{2}\delta(t-t^{\prime})$ where $\xi$ characterizes the strength of the noise ($\xi^{2}$ has units of time). White noise is approximately equivalent to fast colored noise with exponentially decaying two-point correlations (Ornstein-Uhlenbeck process) [114]. In the presence of noise the transition probability is obtained by numerically solving the exact master equation [128, 129, 130, 131] for the averaged density matrix $\rho_{k_{m}}(t)$ of the noisy system

	$\displaystyle{d\over dt}\rho_{k_{m}}(t)=-i[H^{(0)}_{k_{m}}(t),\rho_{k_{m}}(t)]-\frac{\xi^{2}}{2}[H_{1},[H_{1},\rho_{k_{m}}(t)]],$
			(13)

where $H^{(0)}_{k_{m}}(t)$ is the noise-free Hamiltonian while $R(t)H_{1}=-R(t)\sigma^{z}$ expresses the “noisy” part for the full Hamiltonian $H^{(\xi)}_{k_{m}}(t)=H^{(0)}_{k_{m}}(t)+R(t)H_{1}$. The master equation, Eq.(13), is solved within the quench interval $t\in\,[t_{i},0[$.