betaFeedback

WorksPhysics and Astronomy

Thermally activated detection of dark particles in a weakly coupled quantum Ising ladder

Abstract

The Isingh2_h^2 integrable field theory emerges when two quantum critical Ising chains are weakly coupled. This theory possesses eight types of relativistic particles, among which the lightest one (B1B_1) has been predicted to be a dark particle, which cannot be excited from the ground state through (quasi-)local operations. The stability on one hand highlights its potential for applications, and on the other hand makes it challenging to be observed. Here, we point out that the mass of the B1B_1 dark particle mB1m_{B_1} appears as a thermally activated gap extracted from local spin dynamical structure factor at low frequency (ω≪mB1ω\ll m_{B_1}) and low temperatures (T≪mB1T \ll m_{B_1}). We then further propose that this gapped behavior can be directly detected via the NMR relaxation rate measurement in a proper experimental setup. Our results provide a practical criterion for verifying the existence of dark particles.

Introduction

Quantum criticality captures collective behaviors of many-body systems within the quantum critical region, regardless of their microscopic origins [1, 2]. At certain quantum critical point, conformal invariance may emerge associated with special scaling for dynamical and thermal behaviors [3, 4, 5]. Relevant perturbations at this point may result in new quantum integrability with exactly solvable spectrum and many-body wavefunctions [6, 7, 8, 9]. A paradigmatic model is the transverse field Ising chain (TFIC), whose quantum critical point (QCP) is governed by a central charge 1/21/2 conformal field theory [4]. Introducing Ising field perturbation into quantum critical TFIC further leads to the emergence of quantum E8E_{8} integrable field theory [6, 10]. Recently, these theoretical predictions have been confirmed in quasi-1D magnetic materials through combined efforts from experiments [11, 12, 13, 14, 15, 16].

Quantum integrable systems can also be categorized within the framework of coupled conformal field theory. Specifically, for the quantum Ising ladder composed of two weakly-coupled critical TFICs, its low-energy physics is described by the Ising22\mathrm{Ising}_{2}^{2} integrable field theory (IIFT) [7]. Among the 8 types of relativistic particles possessed by the IIFT, three of them (labeled as B1,3,5B_{1,3,5}) are “dark particles” [17], which are inherently prohibited from the ground state through any local or quasi-local spin operations, due to global selection rules. Remarkably, the lightest dark particle is theorized to be robust and long lifetime once prepared, thus may have potential applications in quantum information and quantum technology. The absence of single dark particle peaks in the zero-temperature dynamical structure factor (DSF) has been validated numerically [17, 18]. An additional advantage lies in the model construction, which can be simulated in Rydberg arrays or realized in quasi-1D magnets, e.g., CoNb2_{2}O6_{6}. However, although dark particles contribute to the spectra through multi-particle channels, their spectral signatures are expected to remain elusive in spectroscopy of materials predicted to exhibit Ising22\mathrm{Ising}_{2}^{2} physics [18]. Given the intriguing properties and significant potential applications of dark particles, it is desired to have a proper experimental setup capable of directly confirming the presence of single dark particles.

In this letter, after introducing the IIFT and emergent dark particles, we propose nuclear magnetic resonance (NMR) relaxation rate measurement at low temperatures, where the lightest dark particle’s mass manifests as the thermal activated gap extracted from the measurement. Quantitatively, we analyze asymptotic behaviors of the local spin DSF in low-frequency and low-temperature limit. To connect field theory result and lattice simulation, we analytically determine the relation between the gap and the interchain coupling. Our results show that the thermal activation gap extracted from the local DSF directly corresponds to the mass of the lightest dark particle mB1m_{B_{1}}, instead of the lightest visible particle mB2≈1.95mB1m_{B_{2}} \approx1.95m_{B_{1}} measured in other dynamical spectroscopies [19, 20, 21]. The characteristic feature can serve as a distinctive sign to confirm the existence of the lightest dark particle B1B_{1}.

The model.— The Hamiltonian of two weakly-coupled quantum critical TFICs [Figure 1(a)] follows

Figure 1. (a) Illustration of two weakly coupled Ising chains with transverse field JhxJh_x. (b) Single particle dispersion of the IIFT. The red line denotes A±1A_{\pm1}. The dotted black lines denote B1B_1, B3B_3, B5B_5 and the black lines denote B2B_2, B4B_4 and B6B_6 from below to above, respectively. The intersection of the dispersions with the P=0P=0 line reveals the particle mass.

H=−J(∑l=1,2∑n=1N−1(σnz(l)σn+1z(l)+hxσnx(l))+∑n=1NJiσnz(1)σnz(2))(1)H=-J\left(\sum_{l=1,2}\sum_{n=1}^{N-1}\left(\sigma_{n}^{z(l)}\sigma_{n+1}^{z(l)}+h_x\sigma_{n}^{x(l)}\right)+\sum_{n=1}^{N}J_i\sigma_{n}^{z(1)}\sigma_{n}^{z(2)}\right) \tag*{(1)}

where σnμ(1,2)\sigma_{n}^{\mu(1,2)} are Pauli matrices associated with spin operators Sμ=σμ/2S^\mu=\sigma^\mu/2 (ℏ=1\hbar=1, μ=x,y,z\mu=x,y,z) at site nn on chain (1) or (2). JJ and JiJ_i are intra- and inter-chain couplings, respectively. Here we focus on hx=1h_x=1 which is the critical field for each decoupled chain.

For Ji=0J_i=0, Jordan-Wigner transformation maps spins from each chain into a set of fermions, which can be further diagonalized by the Bogoliubov transformation [22, 23]. In the scaling limit, where the lattice spacing a→0a\to0 and J→∞J\to\infty with 2Ja=12Ja=1, the discrete fermions are recast as free fermionic field, leading to the central charge 1/21/2 conformal field theory Hc=1/2H_{c=1/2} [24, 25]. In parallel, scaling limit of σjz\sigma_j^z and σjx\sigma_j^x can be taken, conventionally referred to as order operator σ(x)\sigma(x) and energy operator ϵ(x)\epsilon(x). Historically, Bosonization is performed by combining the two sets of real fermions from each chain, leading to a free model for the bosonic field ϕ\phi [25]. Operator correspondences that will be useful are listed as σ(1)(x)σ(2)(x)→:cos⁡ϕ(x)/2:\sigma^{(1)}(x)\sigma^{(2)}(x)\to:\cos\phi(x)/2:, ϵ(1)(x)+ϵ(2)(x)→:cos⁡ϕ(x):\epsilon^{(1)}(x)+\epsilon^{(2)}(x)\to:\cos\phi(x): and ϵ(1)(x)−ϵ(2)(x)→:cos⁡Θ(x):\epsilon^{(1)}(x)-\epsilon^{(2)}(x)\to:\cos\Theta(x):, where the dual field Θ(x)\Theta(x) satisfies ∂Θ(x)/∂x=−∂ϕ(x)/∂t\partial\Theta(x)/\partial x=-\partial\phi(x)/\partial t.

In the presence of weak interchain coupling, the perturbed Hamiltonian follows

H=Hc=1/2(1)+Hc=1/2(2)+γ∫dxσ(1)(x)σ(2)(x),(2)H=H_{c=1/2}^{(1)}+H_{c=1/2}^{(2)}+\gamma\int dx\sigma^{(1)}(x)\sigma^{(2)}(x), \tag*{(2)}

which gives the Ising2_2 integrable field theory (IIFT), with (1, 2) referring to chain labels and γ\gamma is the rescaled interchain coupling. Its quantum integrability is more transparent by mapping to the bosonization form [7]

HB=∫dx[12(∂ϕ∂t)2+12(∂ϕ∂x)2+λcos⁡ϕ(x)2],(3)H_{\mathrm{B}}=\int dx\left[\frac{1}{2}\left(\frac{\partial\phi}{\partial t}\right)^2+\frac{1}{2}\left(\frac{\partial\phi}{\partial x}\right)^2+\lambda\cos\frac{\phi(x)}{2}\right], \tag*{(3)}

with λ=2γ\lambda=\sqrt{2\gamma} [26]. Note that Eq. (3) takes the same form as the sine-Gordon model [27], with ϕ\phi field in the former case defined on a Z2\mathbb{Z}_2 orbifold. More details can be found in Ref. [28].

The D8(1)D_8^{(1)} particles.— Spectrum and scattering matrices can be exactly solved for Eq. (3), characterized by the D8(1)D_8^{(1)} algebra [7]. The IIFT theory [7] accommodates 8 types of particles: 6 breathers (BnB_n, n=1,2,⋯ ,6n=1,2,\cdots,6), a soliton (A+1A_{+1}) and an anti-soliton (A−1A_{-1}). Their masses are related as mBn=mB1sin⁡(nπ/14)/sin⁡(π/14)m_{B_n}=m_{B_1}\sin(n\pi/14)/\sin(\pi/14) and mA±=mB1/(2sin⁡(π/14))m_{A_{\pm}}=m_{B_1}/(2\sin(\pi/14)). Each particle follows relativistic dispersion relation, i.e., E=m2+p2E=\sqrt{m^2+p^2}, where the “speed of light” is set as 1. In the context of quantum integrable field [8, 9], all excitations within the model can be described by single particles or their combinations, referred to as asymptotic states. An rr-particle state is denoted as ∣F1(θ1)…Fr(θr)⟩\lvert F_1(\theta_1)\ldots F_r(\theta_r)\rangle, with particle type FF and rapidity θ\theta, which carries eigenenergy E{r}=∑j=1rmPjcosh⁡θjE_{\{r\}}=\sum_{j=1}^{r}m_{P_j}\cosh\theta_j and momentum P{r}=∑j=1rmPjsinh⁡θjP_{\{r\}}=\sum_{j=1}^{r}m_{P_j}\sinh\theta_j. Scatterings between different states can be faced into two particle cases which are known in [7], setting up a systematical way to determine transition matrices through local field operators.

Global properties in the IIFT restrict the excitation channels. Taking advantage of the sine-Gordon model [29], topological charge QQ can be assigned to the soliton and antisoliton as +1+1 and −1-1, respectively, while Q=0Q=0 for the breathers. And for the parity conjugation C\mathcal{C} that operates as CϕC−1=−ϕ\mathcal{C}\phi\mathcal{C}^{-1}=-\phi, we have C∣Bn(θ)⟩=(−1)n∣Bn(θ)⟩\mathcal{C}\lvert B_n(\theta)\rangle=(-1)^n\lvert B_n(\theta)\rangle and C∣A±1(θ)⟩=∣A∓1(θ)⟩\mathcal{C}\lvert A_{\pm1}(\theta)\rangle=\lvert A_{\mp1}(\theta)\rangle. Studied in Ref. [17], excitation from the even-parity ground state to the odd-parity B1,3,5B_{1,3,5} via any local or quasi-local spin operations is forbidden. The B1,3,5B_{1,3,5} are referred to as dark particles. On the other hand, the sin⁡ϕ(x)\sin\phi(x) operator which connects odd-parity states with the ground state corresponds to ∏j=1lσjz(1)σjz(2)\prod_{j=1}^{l}\sigma_j^{z(1)}\sigma_j^{z(2)} (x=lax=la) in the spin representation. Since it involves macroscopic number of spins, such operation is strongly suppressed in the vacuum fluctuations since the coherent length of the fluctuation is exponentially cutoff because of the gap. As a result, the lightest particle B1B_1 cannot spontaneously decay once being prepared.

Thermal activated measurement.— Spin DSF measurements at low temperatures (T≪mB1T\ll m_{B_1}), such as inelastic neutron scattering (INS) [14, 15] and THz spectroscopy [20], are dominant by excitations directly from the ground state. Following discussion in the previous paragraph, energy levels of the dark particles cannot be resolved in these measurements. In contrast, we propose that the thermal activation behavior observed in NMR measurement can directly probe the lightest dark particle. Different from aforementioned detections that cover a rather larger energy range (e.g., 0<ω≲6mB10<\omega\lesssim6m_{B_1} in Ref. [20]), the NMR has low-energy detectability and measures locally. With input energy much smaller than mB2m_{B_2}, excitation from the ground state cannot happen, leading to dominant excitation from B1(0)B_1(0) to B1(p)B_1(p) [Figure 1(b)]. Intensity of such excitation is affected by the thermal distribution of B1B_1, reflecting in the thermal activation gap that can be determined experimentally in a relaxation process.

To capture the thermal activation gap, first we introduce the local spin DSF at finite temperature as following

CO(ω,T)=∫−∞∞dt eiωt⟨O(t,0)O†(0,0)⟩T,(4)C^{\mathcal{O}}(\omega,T)=\int_{-\infty}^{\infty}dt\,e^{i\omega t}\langle\mathcal{O}(t,0)\mathcal{O}^{\dagger}(0,0)\rangle_T, \tag*{(4)}

where O\mathcal{O} denotes the field theory counterpart of local spin operators. tt in the integral denotes real time, ω\omega and TT denote the frequency and temperature, respectively. In the NMR measurement, the spin-lattice relaxation rate is given by [30, 31]

1T1∼∣Ay∣2CY(ωn)+∣Az∣2CZ(ωn),(5)\frac{1}{T_1}\sim|A_y|^2 C^{\mathcal{Y}}(\omega_n)+|A_z|^2 C^{\mathcal{Z}}(\omega_n), \tag*{(5)}

with hyperfine coupling constant AjA_j (j=x,y,zj=x,y,z) and resonant frequency ωn\omega_n of NMR measurement. On the other hand, CxC^x is related to the spin-spin relaxation rate 1/T2=1/T1′+1/T2′1/T_2=1/T'_1+1/T'_2, where

1T1′=AT1,1T2′=∣Ax∣2Cx(ω→0),(6)\frac{1}{T'_1}=\frac{A}{T_1},\qquad\frac{1}{T'_2}=|A_x|^2 C^x(\omega\to0), \tag*{(6)}

with constant AA depending on microscopic details. In the next paragraph, we determine the thermal behavior of the local spin DSF.

Dark particle detection.— We first study the transverse spin DSF CxC^x for finite but small energy (ω≪mB1\omega\ll m_{B_1}) at low temperature. According to previous paragraph, σx\sigma^x corresponds to cos⁡ϕ+cos⁡Θ\cos\phi+\cos\Theta in the bosonization. The operation of cos⁡ϕ\cos\phi preserves topological charge while cos⁡Θ\cos\Theta connects states with topological charge-1 difference. Terms with maximum Boltzmann weight resulted from ground state excitations only contribute to the DSF after reaching ωthreshold=mB2≈1.95mB1\omega_{\mathrm{threshold}}=m_{B_2}\approx1.95m_{B_1}, with channel ⟨0∣cos⁡ϕ∣B2(0)⟩\langle0|\cos\phi|B_2(0)\rangle, which vanishes as energy conservation can not hold for ω≪mB2\omega\ll m_{B_2}. Through the cluster expansion [32], the leading term appears at the order of e−mB1/Te^{-m_{B_1}/T} corresponding to ∣B1⟩→∣B1⟩|B_1\rangle\to|B_1\rangle channel [28],

C1,1x(ω,T)≈∑s=±1∫dθ2πe−mB1Tcosh⁡θ∣⟨B1(θ)∣cos⁡ϕ∣B1(sθ~)⟩∣2(ω+mB1cosh⁡θ)2−mB12(7)C^x_{1,1}(\omega,T)\approx\sum_{s=\pm1}\int\frac{d\theta}{2\pi}\frac{e^{-\frac{m_{B_1}}{T}\cosh\theta}|\langle B_1(\theta)|\cos\phi|B_1(s\widetilde{\theta})\rangle|^2}{\sqrt{(\omega+m_{B_1}\cosh\theta)^2-m_{B_1}^2}} \tag*{(7)}

with cosh⁡θ~=(ω+mB1cosh⁡θ)/mB1\cosh\widetilde{\theta}=(\omega+m_{B_1}\cosh\theta)/m_{B_1}. The leading behaviors of the integral in Eq. (7) can be obtained analytically [33]. For ω≪T≪mB1\omega\ll T\ll m_{B_1},

C1,1x(ω,T)≈eω/2T∣F11∣2πmB1e−mB1/T(−ln⁡ω4T−γE),(8)C^x_{1,1}(\omega,T)\approx\frac{e^{\omega/2T}|F^{11}|^2}{\pi m_{B_1}}e^{-m_{B_1}/T}\left(-\ln\frac{\omega}{4T}-\gamma_E\right), \tag*{(8)}

and for T≪ω≪mB1T\ll\omega\ll m_{B_1},

C1,1x(ω,T)≈∣F11∣2πmB1e−mB1/T[πTω−π4(Tω)3/2],(9)C^x_{1,1}(\omega,T)\approx\frac{|F^{11}|^2}{\pi m_{B_1}}e^{-m_{B_1}/T}\left[\sqrt{\frac{\pi T}{\omega}}-\frac{\sqrt{\pi}}{4}\left(\frac{T}{\omega}\right)^{3/2}\right], \tag*{(9)}

where F11=∣⟨0∣cos⁡ϕ∣B1(iπ)B1(0)⟩∣2=31.756F^{11}=|\langle0|\cos\phi|B_1(i\pi)B_1(0)\rangle|^2=31.756, originating in the transition element ⟨B1(0)∣cos⁡ϕ∣B1(0)⟩\langle B_1(0)|\cos\phi|B_1(0)\rangle. In both limits, e−mB1/Te^{-m_{B_1}/T} behavior is obtained for fixed ω\omega Figure 2 (a), resulting in the observable thermal activation gap mB1m_{B_1}. For the isothermal case, logarithmic divergence in ω\omega is found for ω≪T\omega\ll T Figure 2 (b). In contrast to the power law behavior for ω≫T\omega\gg T Figure 2 (c). Moreover, the subleading contribution comes from the transition ∣B2⟩→∣B2⟩|B_2\rangle\to|B_2\rangle, which is negligible since the corresponding Boltzmann weight e−mB2/T≈e−1.95mB1/T≪e−mB1/Te^{-m_{B_2}/T}\approx e^{-1.95m_{B_1}/T}\ll e^{-m_{B_1}/T} for T≪mB1T\ll m_{B_1}.

Asymptotic behaviors of the transverse spin DSF

Figure 2. Asymptotic behaviors of C1,1x(ω,T)C^x_{1,1}(\omega,T) Eq. (7) calculated from numerical integration with mB1=1m_{B_1}=1. (a) shows the result of fixed energy ω\omega and the fitted curve ln⁡C1,1x=−1.009/T+4.083\ln C^x_{1,1}=-1.009/T+4.083. (b) and (c) show the results with fixed temperature TT in the region ω≪T\omega\ll T and ω≫T\omega\gg T, respectively. The curve in (b) is fitted as C1,1x×108=−1.960ln⁡ω+3.647C^x_{1,1}\times10^8=-1.960\ln\omega+3.647, and in (c) C1,1x×108=0.736ω−0.504C^x_{1,1}\times10^8=0.736\omega^{-0.504}.

CzC^z and CyC^y Eq. (4) are beyond IIFT analytical form factor scheme, as σz\sigma^z and σy\sigma^y do not correspond to local fields in the bosonic theory. However, their thermal behaviors can be determined numerically. The first symmetry-allowed channel in CzC^z comes from ∣B1⟩→∣B1⟩|B_1\rangle\to|B_1\rangle, dominated by the zero momentum mode, which is indeed non-vanishing as elaborated in [28] following density-matrix renormalization group (DMRG) calculation. On the other hand, Cz(ω→0)C^z(\omega\to0) can be directly obtained by quantum Monte Carlo (QMC) simulation [34], which is calculated from 2(πτ)−1∑j⟨δSjz((2τ)−1)δSjz(0)⟩2(\pi\tau)^{-1}\sum_j\langle\delta S_j^z((2\tau)^{-1})\delta S_j^z(0)\rangle. Here τ\tau is the imaginary time and δSiz=Siz−⟨Siz⟩\delta S_i^z=S_i^z-\langle S_i^z\rangle [35, 36]. The thermal activation gap fitted from Figure 3 is about 0.58J0.58J at Jj=0.1J_j=0.1. However, this number can not compare directly with mB1m_{B_1} in the IIFT. For achieving the goal we need to analytically determine mB1m_{B_1} in the corresponding lattice model.

Temperature dependence of $C^z(\omega \to 0)$

Figure 3. Temperature dependence of Cz(ω→0)C^z(\omega\to0) [Eq. (4)] from QMC simulation with Ji=0.1J_i = 0.1, J=1J = 1 on a 2×40962 \times4096 lattice. Red line shows the fitted relation Cz∼e−0.58/TC^z \sim e^{-0.58/T}.

The IIFT predicts the mass of B1B_1 as mB1=(λ/C1)4/7m_{B_1} = (\lambda/C_1)^{4/7}, with C1=0.33645⋯C_1 = 0.33645\cdots [26]. Following similar strategy as in [37], we fix the normalization condition for the lattice model [Eq. (1)] and its scaling limit [Eq. (2)}], i.e., ⟨γ∫dxσ(1)(x)σ(2)(x)⟩=⟨J∑n=1NJiσnz(1)σnz(2)⟩\langle\gamma\int dx \sigma^{(1)}(x)\sigma^{(2)}(x) \rangle= \langle J \sum_{n=1}^{N} J_i \sigma_n^{z(1)}\sigma_n^{z(2)} \rangle, which is recovered as γL⟨σz⟩2=JJiN⟨σz⟩2\gamma L\langle\sigma^z\rangle^2 = JJ_iN\langle\sigma^z\rangle^2 in the decoupled limit. Relation between γ\gamma and JiJ_i can be obtained by inserting the results for magnetizations (⟨σz⟩\langle\sigma^z\rangle and ⟨σ⟩\langle\sigma\rangle) [38]. Then we arrive at

mB1=(22C121/6e−1/4A3)4/7Ji4/7J,(10)m_{B_1} = \left(\frac{2\sqrt{2}}{C_1 2^{1/6}e^{-1/4}A^3}\right)^{4/7}J_i^{4/7}J, \tag*{(10)}

where the Glaisher’s constant A=1.282427⋯A = 1.282427\cdots. More details can be found in [28]. Coefficients determined here are consistent with numerical calculation [Fig. 4], where mB1m_{B_1} is obtained from the energy of the first excited state, following the DMRG method [39, 40].

Now we observe that the fitted gap from QMC is close to the predicted mB1=0.6384Jm_{B_1} = 0.6384J obtained from Eq. (10). The small deviation comes from finite temperature and finite size effect. Furthermore, the general relation for our

model Cy=(ω/J)2Cz/4C^y = (\omega/J)^{2}C^z/4 [10, 17] implies that σy\sigma^y channel is negligible when ω≪J\omega\ll J. Consequently, extracting the thermal activation gap from experiments allows for the verification of the existence of B1B_1.

Discussions.— Proper Ising-chain compounds may serve as to study dark particles. For instance, the quasi-1D magnet CoNb2_2O6_6, which was claimed to accommodate E8E_8 physics [20, 21]. The E8E_8 model emerges in a quantum critical Ising chain under the perturbation of a longitudinal field along the Ising spin direction [6]. In their setup, the transverse field is tuned to the putative 1D TFIC QCP in the 3D ordering dome at low temperature (T≪TNT \ll T_N), where the 3D order is considered to effectively provide the longitudinal field coupled to the transverse-field Ising chain in the material. Recent analysis [18] shows that the physics here is more comprehensively described by a quantum Ising ladder with D8(1)D_8^{(1)} structure. As the putative 1D QCP lives in the vicinity of the 3D QCP in the material, a 1D model is not sufficient for treating 3D fluctuations. Moreover, the aforementioned effective field from 3D ordering is suppressed by the strong magnetic frustration in the material.

For the model described in Eq. (1), the applied transverse field can serve as static magnetic field in an NMR setup [13]. A proper NMR measurement will not only reveal the existence of the lightest dark particle B1B_1 in the CoNb2_2O6_6, but also provide evidence to confirm the D8(1)D_8^{(1)} physics of the material, as the E8E_8 theory has no such hidden energy levels. Apart from the magnetic materials, it is also possible to directly simulate the Hamiltonian in Eq. (1) in Rydberg arrays.

As a starting point for the proposed detection, we assume that thermal equilibrium has been reached in this system. Considering that the phonon-spin coupling is typically weak, the perturbation would not change the “dark” properties. Though direct excitation from ground state to dark particles is forbidden, the transition from, e.g., ∣A±⟩|A_{\pm}\rangle to ∣B1⟩|B_1\rangle is permitted and ∣A±⟩|A_{\pm}\rangle can be excited from ground state. In contact with thermal reservoir, B1B_1 can be reached through a secondary process ∣0⟩→∣A±⟩→∣B1⟩|0\rangle\to|A_{\pm}\rangle\to|B_1\rangle. It is worth further investigating on the detailed thermalization procedure.

Conclusions.— To conclude, following the form factor approach and cluster expansion in the region ω,T≪mB1\omega,T \ll m_{B_1}, we analytically determine the thermal activation gap for σx\sigma^x channel, corresponding to the mass of the lightest dark particle. The same thermal activation gap is also obtained for σz\sigma^z following DMRG and QMC numerical calculations, which is in contrast to the first peak obtained from zero temperature spin DSF. Taking the advantage, we propose that a proper NMR experiment can detect the lightest dark particle through the relaxation rates 1/T11/T_1 and 1/T21/T_2 measurements where the mass of the lightest dark particle can be extracted as thermal activation gap. Potential material candidates, such as CoNb2_2O6_6 and compounds effectively described by the Ising ladder are suggested. Rydberg array and STM experiments could also directly simulate the required Ising ladder to probe the dark particles.

Acknowledgments

The work at Shanghai Jiao Tong University is supported by the National Natural Science Foundation of China Grant Nos. 12450004, 12274288 and the Innovation Program for Quantum Science and Technology Grant No. 2021ZD0301900. The work at Renmin University of China is supported by the National Key R&D Program of China (Grant No. 2023YFA1406500), and the National Natural Science Foundation of China (Grant Nos. 12334008 and 12174441). H. L. also acknowledges the supported by the Postdoctoral Fellowship Program of CPSF (No. GZC20241724).

Supplementary Material — Thermally activated detection of dark particles in a weakly coupled quantum Ising ladder

From the lattice model to the Isingh2^{2}_{h} integrable field theory (IIFT)

Consider the quantum critical transverse field Ising chain (TFIC) (Eq. (1) in the main text). Following the Jordan-Wigner transformation, spin operators can be mapped to fermionic operators cj†,cjc_j^\dagger,c_j as [?]

σjz=(cj†+cj)e±iπ∑l<jcl†cl,σjx=2cj†cj−1.(11)\sigma_j^z=(c_j^\dagger+c_j)e^{\pm i\pi\sum_{l<j}c_l^\dagger c_l},\qquad\sigma_j^x=2c_j^\dagger c_j-1. \tag*{(11)}

with {ci,cj†}=δij\{c_i,c_j^\dagger\}=\delta_{ij} and {ci,cj}={ci†,cj†}=0\{c_i,c_j\}=\{c_i^\dagger,c_j^\dagger\}=0. Consequently, the disorder operator is mapped as μj=∏k<jσkx=e±iπ∑k=1j−1ck†ck\mu_j=\prod_{k<j}\sigma_k^x=e^{\pm i\pi\sum_{k=1}^{j-1}c_k^\dagger c_k}. And we have σjz=(cj†+cj)μj\sigma_j^z=(c_j^\dagger+c_j)\mu_j.

Furthermore, the Majorana spinor ψ=(ψL,ψR)T\psi=(\psi_L,\psi_R)^T is introduced with left (ψL\psi_L)- and right (ψR\psi_R)- components [25]

ψL(j)=(−1)ja(cj†e−iπ/4+cjeiπ/4),ψR(j)=(−1)ja(cj†eiπ/4+cje−iπ/4),(12)\psi_L(j)=\frac{(-1)^j}{\sqrt{a}}(c_j^\dagger e^{-i\pi/4}+c_j e^{i\pi/4}),\qquad\psi_R(j)=\frac{(-1)^j}{\sqrt{a}}(c_j^\dagger e^{i\pi/4}+c_j e^{-i\pi/4}), \tag*{(12)}

where aa is the lattice spacing. Commutation relations follow {ψR(i),ψR(j)}={ψL(i),ψL(j)}=δij/a\{\psi_R(i),\psi_R(j)\}=\{\psi_L(i),\psi_L(j)\}=\delta_{ij}/a, {ψR(i),ψL(i)}=0\{\psi_R(i),\psi_L(i)\}=0. Then the relations between spin operators and Majorana spinor are directly obtained as

σjx=−aiψR(j)ψL(j),σjz=(−1)ja2(ψR(j)+ψL(j))e±iπ∑l<j(−aiψR(l)ψL(l)+1)/2.(13)\sigma_j^x=-ai\psi_R(j)\psi_L(j),\qquad\sigma_j^z=(-1)^j\sqrt{\frac{a}{2}}(\psi_R(j)+\psi_L(j))e^{\pm i\pi\sum_{l<j}(-ai\psi_R(l)\psi_L(l)+1)/2}. \tag*{(13)}

By taking the scaling limit that a→0a\to0 and JaJa keeps finite, the critical TFIC can be described by a free Hamiltonian in continuous space in terms of the Majorana spinor

Hms=ψ†(x)(−iγ5∂∂xψ(x))(14)\mathcal{H}_{ms}=\psi^\dagger(x)\left(-i\gamma^5\frac{\partial}{\partial x}\psi(x)\right) \tag*{(14)}

with x=jax=ja, γ5=σz\gamma^5=\sigma^z and 2Ja=12Ja=1. In the field theory language, σjz\sigma_j^z, μj\mu_j and σjx\sigma_j^x are referred to as order operator σ(x)\sigma(x), disorder operator μ(x)\mu(x) and energy operator ϵ(x)\epsilon(x), respectively.

Then we consider two copies of the quantum critical TFIC. Since the Majorana spinor is real, two sets of Majorana spinors ψ(1,2)\psi^{(1,2)} obtained from the two chains can be further gathered into complex Dirac spinors χ\chi as [25]

χ=12(ψ(1)+iψ(2)),χ†=12(ψ(1)−iψ(2)).(15)\chi=\frac{1}{\sqrt{2}}\left(\psi^{(1)}+i\psi^{(2)}\right),\qquad\chi^\dagger=\frac{1}{\sqrt{2}}\left(\psi^{(1)}-i\psi^{(2)}\right). \tag*{(15)}

Then the Dirac spinors can be bosonized into free massless bosonic field, following the bosonization rule

χR=αRN:eiϕR:,χL=αLN:eiϕL:,(16)\chi_R=\frac{\alpha_R}{\sqrt{N}}:e^{i\phi_R}:,\qquad\chi_L=\frac{\alpha_L}{\sqrt{N}}:e^{i\phi_L}:, \tag*{(16)}

where the normal ordering :::: enforces annihilation operators to the right and αL,R\alpha_{L,R} ensure the anti-commutation relation of χR,L\chi_{R,L} [25]. Right- and left- going components of the bosonic field are introduced via the mode expansion ϕ(x,t)=ϕR(x−t)+ϕL(x+t)\phi(x,t)=\phi_R(x-t)+\phi_L(x+t), i.e.,

ϕR(t−x)=ϕ0R2π−QR2N(t−x)−i2π∑n≠0aˉnne−2πin[(t−x)/N]\phi_R(t-x)=\frac{\phi_{0R}}{2\pi}-\frac{Q_R}{2N}(t-x)-\frac{i}{2\pi}\sum_{n\ne0}\frac{\bar a_n}{n}e^{-2\pi in[(t-x)/N]}
ϕL(t+x)=ϕ0L2π+QL2N(t+x)−i2π∑n≠0anne−2πin[(t+x)/N],(17)\phi_L(t+x)=\frac{\phi_{0L}}{2\pi}+\frac{Q_L}{2N}(t+x)-\frac{i}{2\pi}\sum_{n\ne0}\frac{a_n}{n}e^{-2\pi in[(t+x)/N]}, \tag*{(17)}

where NN is the system size, and QR,LQ_{R,L} are conjugation of the zero modes ϕ0R,L\phi_{0R,L}, satisfying [QR,ϕ0R]=−[QL,ϕ0L]=i/2[Q_R,\phi_{0R}]=-[Q_L,\phi_{0L}]=i/2.

ana_n, aˉn\bar{a}_n are related to boson creation and annihilation operators (a~n\tilde{a}_n and a~n†\tilde{a}^{\dagger}_n) by

an={−ina~n(n>0),i−na~−n†(n<0),aˉn={−ina~−n(n>0),i−na~n†(n<0).(18)a_n = \begin{cases} -i\sqrt{n}\tilde{a}_n & (n>0),\\ i\sqrt{-n}\tilde{a}^{\dagger}_{-n} & (n<0), \end{cases} \qquad \bar{a}_n = \begin{cases} -i\sqrt{n}\tilde{a}_{-n} & (n>0),\\ i\sqrt{-n}\tilde{a}^{\dagger}_{n} & (n<0). \end{cases} \tag*{(18)}

with non-vanishing commutators [aˉn,aˉm]=[an,am]=nδn+m,0[\bar{a}_n,\bar{a}_m]=[a_n,a_m]=n\delta_{n+m,0}. The dual field of ϕ\phi is introduced as Θ(x,t)=ϕR(x−t)−ϕL(x+t)\Theta(x,t)=\phi_R(x-t)-\phi_L(x+t) which satisfies

∂Θ∂x=−∂ϕ∂t.(19)\frac{\partial\Theta}{\partial x}=-\frac{\partial\phi}{\partial t}. \tag*{(19)}

Thus we obtain the effective free bosonic field theory with the Hamiltonian density

Hfb=∂2ϕ(x)∂x2,(20)\mathcal{H}_{fb}=\frac{\partial^2\phi(x)}{\partial x^2}, \tag*{(20)}

corresponding to a conformal field theory with central charge c=1c=1 [41].

Taking the scaling limit of Eq. (13), we have ϵ(1,2)(x)=iψR(1,2)(x)ψL(1,2)(x)\epsilon^{(1,2)}(x)=i\psi_R^{(1,2)}(x)\psi_L^{(1,2)}(x). Several operator correspondences in the bosonization representation can be derived by inserting the bosonization rules Eq. (16) and using the Baker-Hausdorff formula for normal ordering operators [42]

:eA::eB:=e[A+,B−]:eA+B:(21):e^A::e^B:=e^{[A^+,B^-]}:e^{A+B}: \tag*{(21)}

if [A+,B−][A^+,B^-] is c-number, where O^=O^++O^−\hat{O}=\hat{O}_++\hat{O}_- and +/−+/- denotes the creation/annihilation piece of the operator, which are summarized in TABLE. I [25, 43].

spin fieldbosonized
σ(1)(x)σ(2)(x)\sigma^{(1)}(x)\sigma^{(2)}(x):cos⁡[ϕ(x)/2]::\cos[\phi(x)/2]:
ϵ(1)(x)+ϵ(2)(x)\epsilon^{(1)}(x)+\epsilon^{(2)}(x):cos⁡[ϕ(x)]::\cos[\phi(x)]:
ϵ(1)(x)−ϵ(2)(x)\epsilon^{(1)}(x)-\epsilon^{(2)}(x):cos⁡[Θ(x)]::\cos[\Theta(x)]:
ϵ(1)(x)ϵ(2)(x)\epsilon^{(1)}(x)\epsilon^{(2)}(x)∂γϕ∂γϕ (γ=x,t)\partial_\gamma\phi\partial_\gamma\phi\ (\gamma=x,t)
σ(1)(x)μ(2)(x)\sigma^{(1)}(x)\mu^{(2)}(x):cos⁡[Θ(x)/2]::\cos[\Theta(x)/2]:
μ(1)(x)μ(2)(x)\mu^{(1)}(x)\mu^{(2)}(x):sin⁡[ϕ(x)/2]::\sin[\phi(x)/2]:

Table I. Bosonization correspondences for spin operators in the scaling limit.

Now we turn on weak interchain coupling, the continuum theory of Eq. (1) is given by the action

AIsingh2=Ac=1/2(1)+Ac=1/2(2)+λ′∫dxdt σ(1)σ(2),(22)\mathcal{A}_{\mathrm{Ising}_h^2}=\mathcal{A}^{(1)}_{c=1/2}+\mathcal{A}^{(2)}_{c=1/2}+\lambda'\int dxdt\,\sigma^{(1)}\sigma^{(2)}, \tag*{(22)}

where each chain is described by a conformal field theory Ac=1/2(1,2)\mathcal{A}^{(1,2)}_{c=1/2}, and the rescaled coupling strength λ′∝λ7/4\lambda'\propto\lambda^{7/4}. The last term in Eq. (22) implies that additional interaction cos⁡(ϕ/2)\cos(\phi/2) should be added to the free boson theory Eq. (20), which formally gives the Isingh2\mathrm{Ising}_h^2 action Eq. (3). Eq. (3) appears in the same from as a sine-Gordon model, while the difference lies in that the ϕ\phi field is defined on a Z2\mathbb{Z}_2 orbifold [7], rather than on a circle.

Finite temperature correlation function

Local spin DSF at finite temperature follows

CO(ω,T)=∫−∞∞dt eiωt⟨O(t,0)O†(0,0)⟩T,(23)C^{\mathcal{O}}(\omega,T)=\int_{-\infty}^{\infty}dt\,e^{i\omega t}\langle O(t,0)O^\dagger(0,0)\rangle_T, \tag*{(23)}

where OO denotes the field theory counterpart of local spin operators. Using field theory language, the DSF can be expressed in the Lehmann spectral representation as,

CO(ω,T)=1Z∑i,fCi,fO(ω,T),(24)C^{\mathcal{O}}(\omega,T)=\frac{1}{Z}\sum_{i,f}C^{\mathcal{O}}_{i,f}(\omega,T), \tag*{(24)}

with Ci,fO\mathcal{C}^{\mathcal{O}}_{i,f} labelling the contribution of excitation from #ii-particle states to #ff-particle states and the partition function Z=Tr⁡e−H/T=∑n=0∞Zn\mathcal{Z}=\operatorname{Tr}e^{-H/T}=\sum_{n=0}^{\infty}Z_n. Explicitly,

Ci,fO(ω,T)=∑i,f∫dθ1′⋯dθi′(2π)iAi∫dθ1⋯dθf(2π)fAfe−Ei/T∣⟨P1′(θ1′)⋯Pi′(θi′)∣O∣P1(θ1)⋯Pf(θf)⟩∣2δ(ω+Ei−Ef),(25)\mathcal{C}^{\mathcal{O}}_{i,f}(\omega,T)=\sum_{i,f}\int\frac{d\theta'_1\cdots d\theta'_i}{(2\pi)^i A_i}\int\frac{d\theta_1\cdots d\theta_f}{(2\pi)^f A_f}e^{-E_i/T}\left|\langle P'_1(\theta'_1)\cdots P'_i(\theta'_i)|\mathcal{O}|P_1(\theta_1)\cdots P_f(\theta_f)\rangle\right|^2\delta(\omega+E_i-E_f), \tag*{(25)}

and

Zn=∫dθ1⋯dθn(2π)nAne−En/T⟨P1(θ1)⋯Pn(θn)∣P1(θ1)⋯Pn(θn)⟩,(26)Z_n=\int\frac{d\theta_1\cdots d\theta_n}{(2\pi)^n A_n}e^{-E_n/T}\langle P_1(\theta_1)\cdots P_n(\theta_n)|P_1(\theta_1)\cdots P_n(\theta_n)\rangle, \tag*{(26)}

where the Boltzmann constant kB=1k_B=1, i\mathbf{i} labels asymptotic state containing ii particles, and Ai=∏l∈inl!A_i=\prod_{l\in i}n_l! with nln_l counting the particle number of type ll. At low temperature T≪mB1T\ll m_{B_1} with the Boltzmann factor exp⁡(−Ei/T)\exp(-E_i/T) serving as a controlled parameter, a regularized linked cluster expansion can be obtained [32]

CO(ω,T)=∑i,f=0∞Di,fO(ω,T),(27)\mathcal{C}^{\mathcal{O}}(\omega,T)=\sum_{i,f=0}^{\infty}D^{\mathcal{O}}_{i,f}(\omega,T), \tag*{(27)}

where D0,fO=C0,fOD^{\mathcal{O}}_{0,f}=C^{\mathcal{O}}_{0,f}, D1,fO=C1,fO−Z1C0,f−1OD^{\mathcal{O}}_{1,f}=C^{\mathcal{O}}_{1,f}-Z_1C^{\mathcal{O}}_{0,f-1}, D2,fO=C2,fO−Z1C1,f−1O+(Z12−Z2)C0,f−2OD^{\mathcal{O}}_{2,f}=C^{\mathcal{O}}_{2,f}-Z_1C^{\mathcal{O}}_{1,f-1}+(Z_1^2-Z_2)C^{\mathcal{O}}_{0,f-2}, ⋯\cdots.

Non-vanishing B1→B1B_1\to B_1 channel

Because the local spin along yy or zz direction is highly non-local in the IIFT, CzC^z and CyC^y [Eq. (S14)] are beyond IIFT analytical form factor scheme. However, their thermal behaviors can be determined numerically. The first symmetry allowed channel in CzC^z comes from ∣B1⟩→∣B1⟩|B_1\rangle\to|B_1\rangle, dominant by zero momentum mode. Its non-vanishing spectral weight can be confirmed by calculating ∣⟨B1(0)∣σz(1,2)(0)∣B1(0)⟩∣2|\langle B_1(0)|\sigma^{(1,2)}_z(0)|B_1(0)\rangle|^2 in parallel to Eq. (7), effectively captured by WzW^z defined through

Wα≡∑n[L⟨B1(0)∣σnα(1,2)∣B1(0)⟩L]2/N,(28)W^\alpha\equiv\sum_n\left[\mathcal{L}\langle B_1(0)|\sigma^{\alpha(1,2)}_n|B_1(0)\rangle_{\mathcal{L}}\right]^2/N, \tag*{(28)}

where α=x,z\alpha=x,z, NN is the system size and L\mathcal{L} indicates the states for the lattice model. ∣B1(0)⟩L|B_1(0)\rangle_{\mathcal{L}} stands for the first excited state of the lattice Hamiltonian [Eq. (1)] Through DMRG calculation, Figure S1 shows that both WxW^x and WzW^z converge to a finite value as the lattice size N→∞N\to\infty, suggesting the leading behavior Cz∼e−mB1/TC^z\sim e^{-m_{B_1}/T}.

Relation between $W^{z/x}$ and $N$

Figure S1. Relation between Wz/xW^{z/x} and NN obtained from DMRG calculation with Ji=0.1J_i=0.1 and J=1J=1. Dashed lines show the fitting function Wz=−1.5172/N+0.0969W^z=-1.5172/N+0.0969 and Wx=−0.6331/N+0.0732W^x=-0.6331/N+0.0732.

Relation between the gap and interchain coupling

In the IIFT [Eq. (3)], relation between the rescaled interchain coupling λ\lambda and the gap Δ\Delta is known as [26]

λ=−2Γ(1/8)πΓ(7/8)[Δπ4sin⁡π14Γ(4/7)Γ(1/14)]4/7≡C1Δ4/7.(29)\lambda=-\frac{2\Gamma(1/8)}{\pi\Gamma(7/8)}\left[\frac{\Delta\sqrt{\pi}}{4\sin\frac{\pi}{14}}\frac{\Gamma(4/7)}{\Gamma(1/14)}\right]^{4/7}\equiv C_1\Delta^{4/7}. \tag*{(29)}

To obtain the the gap in terms of lattice model parameters, a scaling relation is needed for the interchain coupling of discrete and continuum cases. Since the model is obtained from two quantum critical transverse field Ising chains (TFICs) with weak interchain Ising coupling, TFIC provide us with a starting point. Following similar strategy in Ref. [37], to fix the normalization, we have

⟨γ∫dxσ(1)(x)σ(2)(x)⟩=⟨JJi∑j=1Nσjz(1)σjz(2)⟩.(30)\left\langle\gamma\int dx \sigma^{(1)}(x)\sigma^{(2)}(x) \right\rangle= \left\langle J J_i \sum_{j=1}^{N}\sigma_j^{z(1)}\sigma_j^{z(2)} \right\rangle. \tag*{(30)}

Relation between JiJ_i and γ\gamma should be recovered from the decoupled limit of Eq. (30), s.t. γL⟨σ⟩2=JJiN⟨σz⟩2\gamma L\langle\sigma\rangle^2 = J J_i N\langle\sigma^z\rangle^2. Using the results in TFIC [38]

⟨σˉ⟩2=(21/12e−1/8A3/2M1/8)2,A=1.2824271291…\langle\bar{\sigma}\rangle^2 = \left(2^{1/12}e^{-1/8}A^{3/2}M^{1/8}\right)^2,\qquad A=1.2824271291\ldots
M=2J(1−hx),⟨σjz(1,2)⟩=(1−hx2)1/8,(31)M=2J(1-h_x),\qquad\left\langle\sigma_j^{z(1,2)}\right\rangle=(1-h_x^2)^{1/8}, \tag*{(31)}

we substitute the expressions in Eq. (30) and take the hx→1h_x\to1 limit, obtaining

γ=2J7/4Ji21/6e−1/4A3.(32)\gamma=\frac{2J^{7/4}J_i}{2^{1/6}e^{-1/4}A^3}. \tag*{(32)}

Finally, consider the normalization before and after bosonization that

⟨γ∫dxσ(1)(x)σ(2)(x)⟩=⟨λ∫dxcos⁡ϕ(x)/2⟩,(33)\left\langle\gamma\int dx\sigma^{(1)}(x)\sigma^{(2)}(x)\right\rangle=\left\langle\lambda\int dx\cos\phi(x)/2\right\rangle, \tag*{(33)}

the coupling constants are related by γ=2λ\gamma=\sqrt{2}\lambda as shown in [26]. As a result,

Δ=(22C121/6e−1/4A3)4/7Ji4/7J.(34)\Delta=\left(\frac{2\sqrt{2}}{C_1 2^{1/6}e^{-1/4}A^3}\right)^{4/7}J_i^{4/7}J. \tag*{(34)}

By setting J=1J=1, Eq. (34) implies that Δ=2.3797Ji4/7\Delta=2.3797J_i^{4/7}. This is in good agreement with the numerical results shown in Figure 4.

Relation between $m_{B_1}$ and $J_i$

Figure 4. Relation between mB1m_{B_1} and JiJ_i. The black dots are calculated from DMRG method with N=400N = 400 for each chain and J=1J = 1 with fitted result mB1=2.3443Ji0.5674m_{B_1} = 2.3443J_i^{0.5674} (red line). It agrees well with analytical result mB1=2.3797Ji4/7m_{B_1} = 2.3797J_i^{4/7} from Eq. (10).

References

References

  1. [1]S. Sachdev, Quantum Phase Transitions, 2nd ed. (Cambridge University Press, 2011).
  2. [2]A. Altland and B. D. Simons, Condensed Matter Field Theory, 2nd ed. (Cambridge University Press, 2010).
  3. [3]A. M. Polyakov, Conformal symmetry of critical fluctuations, Pisma Zh. Eksp. Teor. Fiz. 12, 538 (1970).
  4. [4]A. Belavin, A. Polyakov, and A. Zamolodchikov, Infinite conformal symmetry in two-dimensional quantum field theory, Nucl. Phys. B 241, 333 (1984).
  5. [5]D. Friedan, Z. Qiu, and S. Shenker, Conformal invariance, unitarity, and critical exponents in two dimensions, Phys. Rev. Lett. 52, 1575 (1984).DOI
  6. [6]A. B. Zamolodchikov, Integrals of motion and S- matrix of the (scaled) T = T_c Ising model with magnetic field, Int. J. Mod. Phys. A 4, 4235 (1989).DOI
  7. [7]A. LeClair, A. Ludwig, and G. Mussardo, Integrability of coupled conformal field theories, Nucl. Phys. B 512, 523 (1998).DOI
  8. [8]F. A. Smirnov, Form Factors in Completely Integrable Models of Quantum Field Theory (WORLD SCIENTIFIC, 1992).
  9. [9]G. Mussardo, Statistical Field Theory: An Introduction to Exactly Solved Models in Statistical Physics, 1st ed., Oxford graduate texts (Oxford Univ. Press, New York, NY, 2010).
  10. [10]J. Wu, M. Kormos, and Q. Si, Finite-temperature spin dynamics in a perturbed quantum critical Ising chain with an E₈ symmetry, Phys. Rev. Lett. 113, 247201 (2014).pubmed.ncbi.nlm.nih.gov/25541800
  11. [11]Z. Wang, T. Lorenz, D. I. Gorbunov, P. T. Cong, Y. Kohama, S. Niesen, O. Breunig, J. Engelmayer, A. Herman, J. Wu, K. Kindo, J. Wosnitza, S. Zherlitsyn, and A. Loidl, Quantum criticality of an Ising-like spin-1/2 antiferromagnetic chain in a transverse magnetic field, Phys. Rev. Lett. 120, 207205 (2018).DOI
  12. [12]Z. Zhang, K. Amelin, X. Wang, H. Zou, J. Yang, U. Nagel, T. Rõ om, T. Dey, A. A. Nugroho, T. Lorenz, J. Wu, and Z. Wang, Observation of E₈ particles in an Ising chain antiferromagnet, Phys. Rev. B 101, 220411 (2020).DOI
  13. [13]A. W. Kinross, M. Fu, T. J. Munsie, H. A. Dabkowska, G. M. Luke, S. Sachdev, and T. Imai, Evolution of quantum fluctuations near the quantum critical point of the transverse field Ising chain system CoNb₂O₆, Phys. Rev. X 4, 031008 (2014).DOI
  14. [14]H. Zou, Y. Cui, X. Wang, Z. Zhang, J. Yang, G. Xu, A. Okutani, M. Hagiwara, M. Matsuda, G. Wang, G. Mussardo, K. Hódsági, M. Kormos, Z. He, S. Kimura, R. Yu, W. Yu, J. Ma, and J. Wu, E₈ spectra of quasi-one-dimensional antiferromagnet BaCo₂V₂O₈ under transverse field, Phys. Rev. Lett. 127, 077201 (2021).
  15. [15]X. Wang, K. Puzniak, K. Schmalzl, C. Balz, M. Matsuda, A. Okutani, M. Hagiwara, J. Ma, J. Wu, and B. Lake, Spin dynamics of the E₈ particles, Sci. Bull. 69, 2974 (2024).DOI
  16. [16]J. Yang, X. Wang, and J. Wu, Magnetic excitations in the one-dimensional Heisenberg-Ising model with external fields and their experimental realizations, J. Phys. A: Math. Theor. 56, 013001 (2023).DOI
  17. [17]Y. Gao, X. Wang, N. Xi, Y. Jiang, R. Yu, and J. Wu, Spin dynamics and dark particle in a weak-coupled quantum Ising ladder with D₈⁽¹⁾ spectrum, Phys. Rev. B 111, L201117 (2025).DOI
  18. [18]N. Xi, X. Wang, Y. Gao, Y. Jiang, R. Yu, and J. Wu, Emergent D₈⁽¹⁾ spectrum and topological soliton excitation in CoNb₂O₆, arXiv:2403.10785.
  19. [19]C. M. Morris, N. Desai, J. Viirok, D. Hüvonen, U. Nagel, T. Rõõm, J. W. Krizan, R. J. Cava, T. M. McQueen, S. M. Koohpayeh, R. K. Kaul, and N. P. Armitage, Duality and domain wall dynamics in a twisted Kitaev chain, Nature Phys. 17, 832 (2021).DOI
  20. [20]K. Amelin, J. J. Engelmayer, J. Viirok, U. Nagel, T. Rõom, T. Lorenz, and Z. Wang, Experimental observation of quantum many-body excitations of E₈ symmetry in the Ising chain ferromagnet CoNb₂O₆, Phys. Rev. B 102, 104431 (2020).DOI
  21. [21]R. Coldea, D. A. Tennant, E. M. Wheeler, E. Wawrzynska, D. Prabhakaran, M. Telling, K. Habicht, P. Smeibidl, and K. Kiefer, Quantum criticality in an Ising chain: Experimental evidence for emergent E₈ symmetry, Science 327, 177 (2010).DOI
  22. [22]P. Pfeuty, The one-dimensional Ising model with a transverse field, Ann. Phys. 57, 79 (1970).DOI
  23. [23]E. Lieb, T. Schultz, and D. Mattis, Two soluble models of an antiferromagnetic chain, Ann. Phys. 16, 407 (1961).
  24. [24]J. B. Zuber and C. Itzykson, Quantum field theory and the two-dimensional Ising model, Phys. Rev. D 15, 2875 (1977).DOI
  25. [25]D. Boyanovsky, Field theory of the two-dimensional Ising model: Conformal invariance, order and disorder, and bosonization, Phys. Rev. B 39, 6744 (1989).pubmed.ncbi.nlm.nih.gov/9947319
  26. [26]P. Baseilhac, One-point functions in integrable coupled minimal models, Nucl. Phys. B 594, 607 (2001).
  27. [27]S. Coleman, Quantum sine-Gordon equation as the massive Thirring model, Phys. Rev. D 11, 2088 (1975).
  28. [28]See Supplemental Material at [url] which includes Refs. [7, 22, 25, 26, 32, 38, 41? -43], for details of the derivation of the integrable field theory, cluster expansion of the correlation at finite temperature, non-vanishing B₁ → B₁ contribution in the DSF and derivation of mB₁ for the lattice model.
  29. [29]A. O. Gogolin, A. A. Nersesian, and A. M. Tsvelik, Bosonization and Strongly Correlated Systems (Cambridge University Press, Cambridge, 2004).
  30. [30]T. Moriya, Nuclear magnetic relaxation near the Curie temperature, Prog. Theor. Phys. 28, 371 (1962).DOI
  31. [31]J. Yang, W. Yuan, T. Imai, Q. Si, J. Wu, and M. Kormos, Local dynamics and thermal activation in the transverse-field Ising chain, Phys. Rev. B 106, 125149 (2022).DOI
  32. [32]B. Pozsgay and G. Takács, Form factor expansion for thermal correlators, J. Stat. Mech.: Theory Exp. 2010 (11), P11012.DOI
  33. [33]J. Wu, M. Kormos, and Q. Si, Finite-temperature spin dynamics in a perturbed quantum critical Ising chain with an E₈ symmetry, Phys. Rev. Lett. 113, 247201 (2014).pubmed.ncbi.nlm.nih.gov/25541800
  34. [34]M. E. J. Newman and G. T. Barkema, Monte Carlo Methods in Statistical Physics (Clarendon Press, Oxford, 1999).DOI
  35. [35]Y. Fan, J. Yang, W. Yu, J. Wu, and R. Yu, Phase diagram and quantum criticality of Heisenberg spin chains with Ising anisotropic interchain couplings, Phys. Rev. Res. 2, 013345 (2020).DOI
  36. [36]M. Randiera, N. Trivedi, A. Moreo, and R. T. Scalettar, Pairing and spin gap in the normal state of short coherence length superconductors, Phys. Rev. Lett. 69, 2001 (1992).
  37. [37]X. Wang, M. Oshikawa, M. Kormos, and J. Wu, Magnetization oscillations in a periodically driven transverse field Ising chain, Phys. Rev. B 110, 195101 (2024).DOI
  38. [38]B. M. McCoy and T. T. Wu, The Two-Dimensional Ising Model (Harvard University Press, Cambridge, MA and London, England, 1973).
  39. [39]U. Schollwöck, The density-matrix renormalization group in the age of matrix product states, Ann. Phys. 326, 96 (2011).DOI
  40. [40]S. R. White, Density matrix formulation for quantum renormalization groups, Phys. Rev. Lett. 69, 2863 (1992).pubmed.ncbi.nlm.nih.gov/10046608
  41. [41]M. Oshikawa and I. Affleck, Defect lines in the Ising model and boundary states on orbifolds, Phys. Rev. Lett. 77, 2604 (1996).
  42. [42]E. Fradkin, Field Theories of Condensed Matter Physics, 2nd ed. (Cambridge University Press, 2013).DOI
  43. [43]L. P. Kadanoff and A. C. Brown, Correlation functions on the critical lines of the baxter and Ashkin-Teller models, Ann. Phys. 121, 318 (1979).DOI

Paper details

vlnc:W_u36z8tt8
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents