betaFeedback

WorksPhysics and Astronomy

Experimentally revealing anomalously large dipoles in a quantum-circuit dielectric

Abstract

Quantum two-level systems (TLSs) intrinsic to glasses induce decoherence in many modern quantum devices, such as superconducting qubits. Although the low-temperature physics of these TLSs is usually well-explained by a phenomenological standard tunneling model of independent TLSs, the nature of these TLSs, as well as their behavior out of equilibrium and at high energies above 1 K, remain inconclusive. Here we measure the non-equilibrium dielectric loss of TLSs in amorphous silicon using a superconducting resonator, where energies of TLSs are varied in time using a swept electric field. Our results show the existence of two distinct ensembles of TLSs, interacting weakly and strongly with phonons, where the latter also possesses anomalously large electric dipole moment. These results may shed new light on the low temperature characteristics of amorphous solids, and hold implications to experiments and applications in quantum devices using time-varying electric fields.

Two-level system (TLS) defects are known to decohere qubits[1, 2]. Furthermore, there is an effort to control their relaxation[3], and understand their dephasing[4] for future quantum information processors. TLS are also investigated as the focus of single-photon studies which extract their information, including their decoherence and couplings to other quantum elements[1, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. The science of TLSs has uncovered important phenomena including how noise is created on TLSs that are close to the frequency of qubits[14]. However, there is also the possibility that there are new opportunities[20, 21] or challenges ahead posed by TLSs because it is still unresolved how much qubit relaxation may result out of equilibrium. Specifically, the active use of a voltage gate in superconducting[22], semiconducting[23], and Majorana[24] qubits could cause unexpected changes in qubit coherence because time variance can reveal unexpected scientific results.

The properties of dielectrics in qubits, and amorphous materials at low energies T≲1 KT \lesssim1\,\mathrm{K}[25], are generally attributed to atoms or groups of atoms that can tunnel between two adjacent structural configurations. They can be described by a standard tunneling model (STM)[26, 27], which assumes the existence of independent TLSs with asymmetry energy Δ\Delta and tunneling amplitude Δ0\Delta_0 that are distributed universally, e.g., approximately describing all amorphous dielectrics, according to a distribution function P(Δ,Δ0)∝P0/Δ0P(\Delta,\Delta_0) \propto P_0/\Delta_0, where P0P_0 is a material constant. This leads to an energy independent density of states (DOS) as a function of the energy splitting E=Δ2+Δ02E = \sqrt{\Delta^2+\Delta_0^2}[26, 27, 28, 29]. This model accounts well for most of the low energy properties observed in a broad range of amorphous systems[28, 29, 30]. Within the STM the strength of the TLS-phonon coupling γ\gamma relates to the dimensionless tunneling strength, C0=P0γ2/(ρv2)C_0 = P_0\gamma^2/(\rho v^2), where ρ\rho is the mass density and vv the acoustic velocity. While the STM does not specify the value of the tunneling strength, experimentally it is found to be universally small, of order 10−310^{-3}, leading to weak and remarkably universal phonon attenuation properties in amorphous solids. Recent studies on material density[31, 32, 33], stress[34] and TLS-nucleus interactions[35] extend our knowledge of TLS origins. However, an understanding of the nature of TLSs and of the origin of low-temperature universality in the phonon attenuation of glasses is still lacking[30, 36, 37].

Individual TLSs have been studied via adiabatic adjustments of electric fields, and adjustments of strain field. The external field changes the TLS asymmetry energy Δ\Delta, which in turn modifies the TLS energy splitting E=Δ2+Δ02E = \sqrt{\Delta^2+\Delta_0^2}, while the tunneling energy Δ0\Delta_0 is fixed. Using static strain and electric fields to tune TLSs in qubits or resonators provides a spectroscopy method to extract energy splittings[38], couplings to phonons[38, 39], dipole moments[40, 41, 42], relaxation and decoherence rates[43] of individual TLSs, and recently even their locations[44, 45]. These experiments probe TLSs in resonance with the device in equilibrium conditions at ∼5\sim5 GHz, corresponding to energies of ∼kB(0.25 K)\sim k_{\mathrm{B}}(0.25\,\mathrm{K}), where kBk_{\mathrm{B}} is the Boltzmann constant. Consequently, good agreement with the STM predictions is found in the majority of the cases.

In a related study to this work[46], TLSs were swept through resonance with a resonator and emit the energy afterward as a phonon. By using fast-bias rates and large bias amplitudes, this technique allows the study of TLSs out of equilibrium, probing TLSs which are at high energies, but are temporarily lowered to the single-photon energy at the resonator. For a past measurement of amorphous silicon nitride[46], the bias rate dependence of the non-equilibrium dielectric loss is explained well by the STM and the Landau-Zener (LZ) effect[47] – it describes the scaling of non-equilibrium loss over a range of ac-field amplitudes and (swept) bias rates. The TLS dipole moment is then extracted from the data during scaling. It is well known that at equilibrium (zero bias field), the TLS microwave absorption is saturated at high driving powers (average photon number n≫1n\gg1) and the resonator loss decreases as a function of the driving power, as already shown five decades ago[48]. More recently, it was shown that a time-dependent bias electric field Eb(t)E_b(t) sweeps TLSs into resonance, with a rate proportional to E˙b\dot{E}_b[46, 47]. The dielectric loss increases with E˙b\dot{E}_b up to the intrinsic loss tangent tan⁡δ0\tan\delta_0, i.e. the loss in the equilibrium (E˙b=0\dot{E}_b=0) and low-power limit n≲1n\lesssim1, where TLSs

Steady-state and nonequilibrium loss

Figure 1. Steady-state and nonequilibrium loss. a Optical image and b schematic of the superconducting resonator device, where the total capacitance, CC, is made of a bridge of 4 equal capacitances (C1=C2=C3=C4=CC_1=C_2=C_3=C_4=C). Bias voltage VbV_b is applied across C1C_1 and C2C_2 (and also C3C_3 and C4C_4) in series. The inductor LL is coupled to a coplanar waveguide for transmission measurements. c Steady-state loss tangent tan⁡δs\tan\delta_s (red squares) plotted as a function of the average photon number nn at 20 mK. The loss fits to the standard model of TLS loss: tan⁡δs=tan⁡δ0/1+n/nc\tan\delta_s=\tan\delta_0/\sqrt{1+n/n_c} (blue), where tan⁡δ0=1.6×10−4\tan\delta_0=1.6\times10^{-4} is the intrinsic material loss measured at the single photon regime (n<1n<1) and nc=3.7n_c=3.7, is the quantum-classical crossover photon number. d Calculation of normalized non-equilibrium loss tan⁡δ/tan⁡δ0\tan\delta/\tan\delta_0 as a function of the dimensionless bias rate ξ=2v0/(πΩR02)=2εVb/(πω0p) E˙b\xi=2v_0/(\pi\Omega_{R0}^2)=2\varepsilon V_b/(\pi\omega_0 p)\,\dot{E}_b, based on the STM in the regime of strong saturation (Γ1,m/ΩR0≪1\Gamma_{1,m}/\Omega_{R0}\ll1) [47]. Inset: TLS-photon energy-level diagram as a function of the TLS energy bias rate vv. The states ∣g,nF⟩|g,n_F\rangle and ∣e,nF−1⟩|e,n_F-1\rangle (∣g⟩|g\rangle and ∣e⟩|e\rangle are the TLS ground and excited states, and ∣nF⟩|n_F\rangle is a photon number state) are connected by an adiabatic transition (with an avoided crossing gap equal to the TLS Rabi frequency ΩR\Omega_R) which may lead to single-photon loss; the probability for photon absorption is 1−e−πΩR2/(2v)1-e^{-\pi\Omega_R^2/(2v)}. e Loss measured at various microwave source powers (-60 dBm to 0 dBm with 5 dBm increment as indicated by the arrow) as a function of E˙b/n\dot{E}_b/n (proportional to ξ\xi). A periodic triangular bias voltage is applied, as indicated by the inset, with a fixed amplitude Eb,max⁡=0.44E_{b,\max}=0.44 V/μ\mum and varying modulation frequency fbf_b, resulting in the bias rate E˙b=4Eb,max⁡fb\dot{E}_b=4E_{b,\max}f_b. Data collapse at medium bias rates occurs below the intrinsic loss tan⁡δ0\tan\delta_0 (dashed-dotted black line), which scales according to LZ theory. The scaling deviates at the lowest rates, where TLS relaxation (T1T_1) processes dominate. Excess loss above tan⁡δ0\tan\delta_0 can be seen at E˙b/n≳10\dot{E}_b/n\gtrsim10 V/μ\mum s. f Electric field bias rate dependence of tan⁡δ\tan\delta at selected microwave source powers. Excess loss above tan⁡δ0\tan\delta_0 can be seen at fast bias rates, which reveals a data collapse at low powers Pac<−25P_{\mathrm{ac}}<-25 dBm.

are unsaturated. It should be noted that the observations were made previously in silicon nitride[46], but are expected to apply to all amorphous materials where the STM holds. A similar fast-bias technique has been used to demonstrate a defect maser based on TLS population inversion [20] and dynamical decoupling of TLSs from a resonator by multiple coherent resonant transitions [21].

An application of the STM to material data generally leads to the conclusion of small interactions between TLSs, related to an energy independent density of states. Early work on non-equilibrium measurements of dielectrics revealed small amplitude relaxations [49] which were understood from a small gap in the standard TLS density of states, formed by rare strong interactions [50]. Presently, modern measurements provide an opportunity for new discovery from the change of TLS energies at high rates. This allows fast lowering high-energy TLSs and they may undergo single-photon exchanges with a microwave resonator mode. This gives the opportunity to uncover phenomena on TLS density which is relevant to modern electric-biased quantum devices.

Here we employ the fast-bias technique of Ref. [46], but a different material, amorphous silicon is studied in this work. At low bias rates the dielectric loss follows the theory of Ref. [47] based on the STM, similar to previous measurements on silicon nitride [46]. However, at high bias rates we observe a striking excess loss, a loss much larger than the intrinsic loss, for all applied driving powers in the amorphous film under study. This result contradicts the common understanding of the STM and provides strong evidence for another loss mechanism from high-energy TLSs probed out of equilibrium. We analyze our data in terms of two types of TLSs, as proposed in Ref. [51], where the first type interacts weakly while the second interacts strongly with phonons, with their own contributions to the total loss. To our knowledge, our observations provide the first direct experimental evidence for the existence in amorphous solids of two types of TLSs with coupling to phonons that differ by an order of magnitude.

Results

Nonequilibrium excess loss. As shown in Fig. 1a and Fig. 1b, the resonator consists of four equal bridge parallel-plate capacitors with a total capacitance, CC (C1=C2=C3=C4=CC_1=C_2=C_3=C_4=C). A dc bias field, EbE_b, is applied across the capacitors. The bridge layout is effective to isolate the ac resonance energy from the bias field input port. Standard transmission measurements of S21S_{21} are performed on the resonator as a function of the average photon number, nn, or microwave resonator field amplitude Eac=2nℏω0/(ϵV)E_{\mathrm{ac}}=\sqrt{2n\hbar\omega_0/(\epsilon V)}, where ℏ\hbar is the reduced Planck constant, ω0=2π×5.1 GHz\omega_0=2\pi\times5.1\,\mathrm{GHz} is the resonator resonance frequency, ϵ\epsilon is the permittivity, and V=2925 μm3V=2925\,\mu\mathrm{m}^{3} is the total capacitor volume. The material loss tangent tan⁡δ\tan\delta, equal to the inverse resonator quality 1/Qi1/Q_i, is then extracted. Figure 1c shows the steady state loss tangent, tan⁡δs\tan\delta_s, at zero bias field as a function of nn. The resonator photon power dependence arises from the saturation of TLSs, and the loss tangent tan⁡δs\tan\delta_s is expressed as tan⁡δs=tan⁡δ0/1+n/nc\tan\delta_s=\tan\delta_0/\sqrt{1+n/n_c} [48], where tan⁡δ0\tan\delta_0 is the intrinsic material loss measured in the single photon limit (n≪1n\ll1) and ncn_c is the quantum-classical crossover photon number. The fit yields nc=3.7n_c=3.7 and tan⁡δ0=1.6×10−4\tan\delta_0=1.6\times10^{-4}, corresponding to Qi=6200Q_i=6200.

When the bias field EbE_b is varied in time (see inset of Fig. 1e), the asymmetry energy of each TLS is modified as Δ(t)=Δ(0)−2pEb(t)cos⁡θ\Delta(t)=\Delta(0)-2pE_b(t)\cos\theta, where pp is the TLS dipole moment and θ\theta is the angle between p⃗\vec{p} and E⃗b\vec{E}_b. An ensemble of ground-state TLSs is swept through the resonance, described by the condition E(t)≡Δ02+Δ2(t)=ℏω0E(t)\equiv\sqrt{\Delta_0^2+\Delta^2(t)}=\hbar\omega_0 for each TLS, thereby leading to an enhanced loss. Close to resonance, the TLS energy changes at a rate of v=∣E˙∣/ℏ≈v01−(Δ0/ℏω0)2cos⁡2θv=|\dot{E}|/\hbar\approx v_0\sqrt{1-(\Delta_0/\hbar\omega_0)^2\cos^2\theta}, where v0=(2p/ℏ)E˙bv_0=(2p/\hbar)\dot{E}_b is the maximum bias rate. The dynamics of each resonant passage is of the LZ type, where an adiabatic transition corresponds to the excitation of a TLS by absorption of a single photon (see inset of Fig. 1d). The resulting non-equilibrium loss was analyzed in Ref. [47] and is shown to be a function of the two dimensionless parameters: the standard LZ parameter ξ=2v0/(πΩR02)\xi=2v_0/(\pi\Omega_{R0}^2) and the ratio Γ1,m/ΩR0\Gamma_{1,m}/\Omega_{R0}, where ΩR0=pEac/ℏ\Omega_{R0}=pE_{\mathrm{ac}}/\hbar is the maximum TLS Rabi frequency and Γ1,m\Gamma_{1,m} is the maximum TLS relaxation rate. As plotted in Fig. 1d, in the regime of strong saturation Γ1,m/ΩR0≪1\Gamma_{1,m}/\Omega_{R0}\ll1, the predicted normalized loss tan⁡δ/tan⁡δ0\tan\delta/\tan\delta_0 is a universal function of ξ\xi, approaching 1 in the non-adiabatic limit ξ≫1\xi\gg1. It should be noted that the intrinsic loss is recovered at high bias rates provided that the TLS DOS is energy-independent, as assumed by the STM. This is understood as a result of fast LZ passage time compared to 2π/ΩR02\pi/\Omega_{R0}, such that the swept TLSs remain unsaturated by the resonator field, as in the single-photon limit.

Fig. 1e shows the measured dielectric loss tangent tan⁡δ\tan\delta as a function of E˙b/n\dot{E}_b/n, which is proportional to the dimensionless bias rate ξ=2ϵVπω0p⋅(E˙b/n)\xi=\frac{2\epsilon V}{\pi\omega_0 p}\cdot(\dot{E}_b/n). The data demonstrate two distinct loss regimes separated by tan⁡δ0\tan\delta_0. Below tan⁡δ0\tan\delta_0 the curves show a single dependence on E˙b/n\dot{E}_b/n for all microwave driving powers, except for small deviations at small E˙b/n\dot{E}_b/n (or ξ<Γ1,m/ΩR0\xi<\Gamma_{1,m}/\Omega_{R0}) due to incoherent LZ transitions [46, 47]. The data collapse of the different curves is in accord with the theory of Ref. [47] and agrees with previous measurements in silicon nitride [46], suggesting that dielectric loss originates from standard TLSs. However, at higher values of E˙b/n\dot{E}_b/n, tan⁡δ\tan\delta strikingly exceeds the intrinsic loss tan⁡δ0=1.6×10−4\tan\delta_0=1.6\times10^{-4} and reaches a maximum of 5.2×10−45.2\times10^{-4} except for the high powers, Pac>−20 dBmP_{\mathrm{ac}>-20\,\mathrm{dBm}} (see Fig. 1f). The excess loss scales with E˙b/n\dot{E}_b/n at high driving powers, Pac>−20 dBmP_{\mathrm{ac}}>-20\,\mathrm{dBm} (see Fig. 1e), but scales with EbE_b at low driving powers, Pac<−30 dBmP_{\mathrm{ac}}<-30\,\mathrm{dBm} (see Fig. 1f. These distinguished saturation behaviors at small and large driving powers imply that the excess loss is due to a second type of TLSs. This saturation occurs at much higher driving powers (Pac∼−30 dBmP_{\mathrm{ac}} \sim-30\,\mathrm{dBm}) compared to ∼−60 dBm\sim-60\,\mathrm{dBm} for the saturation of standard TLSs responsible for the loss at low bias rates. It therefore suggests that the second type of TLSs has much higher relaxation rates. In addition, at small powers Pac<−25 dBmP_{\mathrm{ac}} < -25\,\mathrm{dBm} where the second type of TLSs are unsaturated, the loss shows a bias rate dependence (Fig. 1f), indicating a rate-(thus energy-) dependent DOS of the second type of TLSs. In comparison, the non-equilibrium loss of unsaturated standard TLSs is equal to the intrinsic value tan⁡δ0\tan\delta_0, irrespective of the bias rate.

Loss from two types of TLSs. We now analyze the data in a model which consists of two types of TLSs. First, we use the data at low to intermediate bias rates to calculate the loss resulting from the standard TLSs, tan⁡δ1\tan\delta_1. Figure 2a shows the loss of Fig. 1f normalized by tan⁡δ0\tan\delta_0, along with a numerical calculation of tan⁡δ/tan⁡δ0\tan\delta/\tan\delta_0 based on LZ theory (see Methods and Refs. [46, 47] for details). The calculated loss at a given bias rate E˙b\dot{E}_{\mathrm{b}} and photon number nn depends on the dipole moment pp and maximum relaxation rate Γ1,m\Gamma_{1,\mathrm{m}} which serve as fitting parameters. We conduct the fit in the low bias rate regime E˙b<300 V/μm s\dot{E}_{\mathrm{b}} < 300\,\mathrm{V}/\mu\mathrm{m}\,\mathrm{s} and obtain p1=11 Dp_1 = 11\,\mathrm{D} and Γ1,m=5.7 MHz\Gamma_{1,\mathrm{m}} = 5.7\,\mathrm{MHz} for the standard TLSs. In comparison to quasi-static tuned TLS measurements of amorphous silicon nitride[40] and thin-film crystalline alumina[42], the total average dipole moment pzp_z is 3-4 D, leading to a total representative moment of approximately p=6−7 Dp = 6 - 7\,\mathrm{D}. In two cases, the extracted moment seems larger. In dynamical tuned measurements of silicon nitride[46], a single representative dipole moment was extracted as p=7.9 Dp = 7.9\,\mathrm{D}, and this may partially be larger due to the different technique, rather than the material difference. Additionally, in amorphous alumina, some very large moments of pz∼10 Dp_z \sim10\,\mathrm{D} are observed[42] and a delocalized oxygen model might account for these unconventionally large moments[52]. We thus find that the first TLS moment type extracted from our amorphous silicon data, p1≈11 Dp_1 \approx11\,\mathrm{D}, to be within an expected range of standard TLSs, given large range of analyzed moments.

The calculations generally capture the loss well but show increasing variance with increasing E˙b\dot{E}_{\mathrm{b}}, since the contribution of the second type of TLSs becomes increasingly significant at higher bias rates. In Fig. 2b we plot the net excess loss tan⁡δ2\tan\delta_2 by subtracting the calculated standard TLS loss from the measured loss, i.e. tan⁡δ2=tan⁡δ−tan⁡δ1\tan\delta_2 = \tan\delta- \tan\delta_1. The single dependence on E˙b\dot{E}_{\mathrm{b}} for low driving powers Pac<−25 dBmP_{\mathrm{ac}} < -25\,\mathrm{dBm} then becomes apparent. The independence on driving power implies that the TLSs responsible for the excess loss are unsaturated in this power regime, such that the observed single curve

Loss analysis of the first and second types of TLSs

Figure 2. Loss analysis of the first and second types of TLSs. a Normalized loss tan⁡δ/tan⁡δ0\tan\delta/\tan\delta_0 (colored circles), where tan⁡δ0=1.6×10−4\tan\delta_0 = 1.6 \times10^{-4} is the intrinsic loss at low powers and zero bias rate, plotted at each driving power (in dBm). Colored solid curves are calculations based on LZ theory within the STM [47], performed at low bias rates E˙b<300 V/μm s\dot{E}_{\mathrm{b}} < 300\,\mathrm{V}/\mu\mathrm{m}\,\mathrm{s}. The calculation gives p1=11 Dp_1 = 11\,\mathrm{D} (1 D=0.21 eA˚1\,\mathrm{D} = 0.21\,\mathrm{e\AA}, where ee is the electron charge) and a maximum relaxation rate Γ1,m=5.7 MHz\Gamma_{1,\mathrm{m}} = 5.7\,\mathrm{MHz} for the standard (first) TLS type. b The excess loss tan⁡δ2=tan⁡δ−tan⁡δ1\tan\delta_2 = \tan\delta- \tan\delta_1 (colored circles) is attributed to the second type of TLSs. Colored solid curves are theoretical calculations of tan⁡δ2\tan\delta_2 based on an energy-dependent DOS of the TLS type. c Loss data (colored circles) corresponding to Fig. 1f are shown together with the theoretical calculations (colored solid lines), which combine both types of TLS loss.

is the equivalent of the intrinsic loss tan⁡δ0\tan\delta_0 of the standard TLSs, and will be denoted as tan⁡δ2,0\tan\delta_{2,0}. The fact that tan⁡δ2,0\tan\delta_{2,0} increases with E˙b\dot{E}_{\mathrm{b}} points to two features of the DOS of the contributing TLSs. First, the DOS is an increasing function of the TLS energy, because the loss is determined by the number of TLSs within the energy range ℏω0<E<ℏω0+pEb,max\hbar\omega_0 < E < \hbar\omega_0 + pE_{\mathrm{b,max}} that are swept through resonance. Second, Since E˙b\dot{E}_{\mathrm{b}} is varied by varying the modulation frequency fbf_{\mathrm{b}} with a fixed amplitude Eb,maxE_{\mathrm{b,max}}, the initial energies of the TLSs that are swept through resonance are independent of E˙b\dot{E}_b. This means that the large non-equilibrium DOS of these TLSs at the resonance energy ℏω0\hbar\omega_0 tends to restore its small equilibrium value by some mechanism. This mechanism becomes less effective as the sweep time reduces (E˙b\dot{E}_b increases).

Such a scenario arises in a previously proposed two-TLS model, which divides TLSs into two groups, distinguished by their interactions with phonons [51]. As a consequence of their distinct interactions, TLSs that are weakly coupled to phonons are abundant at low energies below ∼1\sim1 K and form the standard TLSs of the STM with an approximately energy-independent DOS ρ1\rho_1; TLSs that are strongly coupled to phonons are characterized by an energy-dependent DOS ρ2(E)\rho_2(E) exhibiting a soft (power-law) gap at low energies. This soft gap is a result of their mutual interactions with the standard TLSs [53, 54], as dictated by the Efros-Shklovskii mechanism for long-range interacting particles in glassy systems [55, 56]. As a result, strongly interacting (with phonons) TLSs are scarce at low energies, thus rarely observable in conventional measurements performed near equilibrium. However, in our measurement TLSs with maximum energy of ℏω0+p2Eb,max⁡\hbar\omega_0 + p_2 E_{b,\max}, where their densities are much larger, can be swept into resonance. Out of equilibrium, the interaction between the two types of TLSs acts to reform the equilibrium gap in the DOS of strongly interacting TLSs by rearrangement of the standard TLSs. This sets a typical time scale for the reformation of the gap, equal to a typical relaxation time of standard TLSs. One observes the excess loss when the bias field modulation frequency fbf_b exceeds the standard TLS relaxation rate, such that the reconstruction of the gap is incomplete. The non-equilibrium DOS ρ2(E,E˙b)\rho_2(E,\dot{E}_b) depends on energy due to the energy dependence of the equilibrium DOS ρ2,eq(E)\rho_{2,\mathrm{eq}}(E), and also depends on E˙b\dot{E}_b due to the time-dependent reformation of the gap. By extending the Efros-Shklovskii argument to our non-equilibrium situation we obtain an approximated expression for ρ2(E,E˙b)\rho_2(E,\dot{E}_b) (see Methods for details). The resulting non-saturated loss tangent tan⁡δ2,0(E˙b)=πp22ρ2(ℏω0,E˙b)/(3ϵ)\tan\delta_{2,0}(\dot{E}_b)=\pi p_2^2\rho_2(\hbar\omega_0,\dot{E}_b)/(3\epsilon), where p2p_2 is the dipole moment of the second type of TLSs, reads

tan⁡δ2,0(E˙b)=Aexp⁡[B∫0Cdx e−(D/E˙b)e2x],(1)\tan\delta_{2,0}(\dot{E}_b)=A\exp\left[B\int_0^C dx\,e^{-(D/\dot{E}_b)e^{2x}}\right], \tag*{(1)}

with the four fitting parameters AA, BB, CC, and DD. Here A=πp22ρ2,eq(ℏω0)/(3ϵ)A=\pi p_2^2\rho_{2,\mathrm{eq}}(\hbar\omega_0)/(3\epsilon) is the excess loss at equilibrium (E˙b=0\dot{E}_b=0), and is proportional to the equilibrium DOS ρ2,eq(ℏω0)\rho_{2,\mathrm{eq}}(\hbar\omega_0) at the resonance energy. B=(8π/3)ρ1uln⁡[1+E0/(ℏω0)]B=(8\pi/3)\rho_1 u\ln[1+E_0/(\hbar\omega_0)], where E0=min⁡{Emax⁡,ℏω0+p2Eb,max⁡}E_0=\min\{E_{\max},\hbar\omega_0+p_2E_{b,\max}\} with Emax⁡∼10 kBTE_{\max}\sim10\,k_\mathrm{B}T being the maximum energy of standard TLSs, is proportional to the phonon-mediated interaction strength uu between the two types of TLSs. C=ln⁡(Emax⁡/Δ0,min⁡)C=\ln(E_{\max}/\Delta_{0,\min}) is the upper limit of the integral over the normalized tunneling amplitudes of the standard TLSs x=ln⁡(Δ0/Δ0,min⁡)x=\ln(\Delta_0/\Delta_{0,\min}), where Δ0,min⁡\Delta_{0,\min} is a minimum cutoff for tunneling amplitudes [28, 29]. Finally, D=Γ1,m(Emax⁡)(Δ0,min⁡/Emax⁡)2Eb,max⁡D=\Gamma_{1,\mathrm{m}}(E_{\max})(\Delta_{0,\min}/E_{\max})^2E_{b,\max}, with De2C=Γ1,m(Emax⁡)Eb,max⁡D e^{2C}=\Gamma_{1,\mathrm{m}}(E_{\max})E_{b,\max} being the characteristic bias rate above which standard TLSs cannot change their state in order to equilibrate the DOS of the second type of TLSs. Fitting the collapsed data of Fig. 2(b) (for Pac<−25P_{\mathrm{ac}}<-25 dBm) yields A=1.8×10−8A=1.8\times10^{-8}, B=1.27B=1.27, C=7.8C=7.8 and D=7.2×10−3 V/μm sD=7.2\times10^{-3}\,\mathrm{V}/\mu\mathrm{m\,s}.

At Pac≥−15P_{\mathrm{ac}}\geq-15 dBm where a saturation effect is observed, we note that the normalized excess loss tan⁡δ2/tan⁡δ2,0\tan\delta_2/\tan\delta_{2,0}, which neutralizes the effect of energy-dependent DOS of the second TLS type, resumes a similar scaling with E˙b/n\dot{E}_b/n as the standard TLSs (Fig. 3). This provides a striking evidence for the TLS origin of the excess loss. We therefore repeat the numerical calculation based on LZ theory for the excess loss of Fig. 2b, similarly to the calculation performed for the standard TLSs in Fig. 2a. The colored solid curves shown in Fig. 2b are obtained for p2=110p_2=110 D and Γ1,m(2)=800\Gamma_{1,\mathrm{m}}^{(2)}=800 MHz for the second type of TLSs. Combining the theoretical calculations of tan⁡δ1\tan\delta_1 and tan⁡δ2\tan\delta_2 from the standard and second types of TLSs, respectively, the total loss tan⁡δ=tan⁡δ1+tan⁡δ2\tan\delta=\tan\delta_1+\tan\delta_2 agrees well with measured loss over the entire domain of the driving powers and bias rates explored in the experiment, as shown in Fig. 2c.

LZ scaling for the second TLS type

Figure 3. LZ scaling for the second TLS type. LZ analysis for the normalized excess loss tan⁡δ2/tan⁡δ2,0\tan\delta_2/\tan\delta_{2,0} (colored circles), where tan⁡δ2,0\tan\delta_{2,0} (inset) is the fit for the unsaturated excess loss at low powers Pac<−20P_{\mathrm{ac}}<-20 dBm. The normalized excess loss at high powers (0 to −15-15 dBm), where the second types of TLSs become saturated, obeys LZ scaling with the dimensionless bias rate ξ2=2eVℏω02⋅(E˙b/n)\xi_2=\frac{2eV}{\hbar\omega_0^2}\cdot(\dot{E}_b/n), similar to the loss due to standard TLSs [Fig. 2a]. The numerical calculation of tan⁡δ2/tan⁡δ2,0\tan\delta_2/\tan\delta_{2,0} using LZ theory (colored solid lines) yields p2=110p_2=110 D and Γ1,m(2)=800\Gamma_{1,\mathrm{m}}^{(2)}=800 MHz for the second TLS type. The LZ scaling verifies the TLS nature of the excess loss mechanism.

Comparison to the two-TLS model

From the calculation of tan⁡δ2\tan\delta_{2} we can extract information that allows us to examine the predictions of the two-TLS model. First, the ratio Γ1,m(2)/Γ1,m(1)\Gamma^{(2)}_{1,m}/\Gamma^{(1)}_{1,m} between the relaxation rates of the two types of TLSs is equal to (γ2/γ1)2(\gamma_{2}/\gamma_{1})^{2}, where γ1\gamma_{1} and γ2\gamma_{2} are the couplings of the two types of TLSs to phonons. Using the relaxation rates found above, we obtain γ2/γ1≈12\gamma_{2}/\gamma_{1}\approx12, which is slightly smaller than expected within the two TLS model[51, 57], but nevertheless describes distinct coupling strengths to phonons differing by an order of magnitude. Second, the ratio between the equilibrium DOS of the two types of TLSs at the resonance energy ℏω0≈0.25 kBK\hbar\omega_{0}\approx0.25\,\mathrm{k}_{\mathrm{B}}\mathrm{K} can be found from A/tan⁡δ0=p22ρ2,eq(ℏω0)/(p12ρ1)≈10−4A/\tan\delta_{0}=p_{2}^{2}\rho_{2,\mathrm{eq}}(\hbar\omega_{0})/(p_{1}^{2}\rho_{1})\approx10^{-4}. Together with the dipole moments p1=11 Dp_{1}=11\,\mathrm{D} and p2=110 Dp_{2}=110\,\mathrm{D}, we find ρ2,eq(ℏω0)/ρ1≈10−6\rho_{2,\mathrm{eq}}(\hbar\omega_{0})/\rho_{1}\approx10^{-6}. This is consistent with the general success of the STM in describing the low-temperature universality in some acoustic and thermodynamic properties of amorphous solids, since phonon attenuation is dominated by the standard (weakly interacting) TLSs at energies where γ22ρ2,eq(E)≪γ12ρ1\gamma_{2}^{2}\rho_{2,\mathrm{eq}}(E)\ll\gamma_{1}^{2}\rho_{1}. Lastly, by estimating the largest energy of standard TLSs as Emax⁡≈10 kBKE_{\max}\approx10\,\mathrm{k}_{\mathrm{B}}\mathrm{K}, from the parameter BB one obtains ρ1u≈4×10−2\rho_{1u}\approx4\times10^{-2}. According to the two-TLS model, the tunneling strength is given by C0=ρ1u⋅(γ1/γ2)C_{0}=\rho_{1u}\cdot(\gamma_{1}/\gamma_{2}), which for γ1/γ2≈1/12\gamma_{1}/\gamma_{2}\approx1/12 gives C0≈3×10−3C_{0}\approx3\times10^{-3}. This agrees with the universally small value of C0∼10−4−10−3C_{0}\sim10^{-4}-10^{-3} which is experimentally found to hold across a wide range of different amorphous solids [30].

Further comparison with the two-TLS model is achieved by studying the bias rate dependence of the loss at various bias amplitudes. This is shown in Figure 4a for driving power of −35 dBm-35\,\mathrm{dBm}. It reveals a clear bias amplitude dependence at high bias rates, indicating an energy-dependent DOS for the second TLS type. At the highest bias rates E˙b≫De22C\dot{E}_{b}\gg D_{e2}^{2\mathrm{C}}, where De22C=Γ1,m(Emax⁡)Eb,max⁡∼5×104 V/μm sD_{e2}^{2\mathrm{C}}=\Gamma_{1,m}(E_{\max})E_{b,\max}\sim5\times10^{4}\,\mathrm{V}/\mu\mathrm{m}\,\mathrm{s}, the gap reconstruction in the non-equilibrium DOS ρ2(E,E˙b)\rho_{2}(E,\dot{E}_{b}) is negligible. Thus ρ2(E,E˙b≫Γ1,m(Emax⁡)Eb,max⁡)\rho_{2}(E,\dot{E}_{b}\gg\Gamma_{1,m}(E_{\max})E_{b,\max}) approaches the equilibrium DOS ρ2,eq(E=ℏω0+p2Eb,max⁡)\rho_{2,\mathrm{eq}}(E=\hbar\omega_{0}+p_{2}E_{b,\max}) at the highest energy from which TLSs are brought into resonance. The excess loss at the highest bias rate is therefore tan⁡δ2,max⁡≈πp22ρ2,eq(ℏω0+p2Eb,max⁡)/(3ε)\tan\delta_{2,\max}\approx\pi p_{2}^{2}\rho_{2,\mathrm{eq}}(\hbar\omega_{0}+p_{2}E_{b,\max})/(3\varepsilon) and thus proportional to the equilibrium DOS at this energy. Within the two-TLS model, ρ2,eq(E)\rho_{2,\mathrm{eq}}(E) is predicted to have a power-law energy dependence below the maximum energy of the standard TLSs Emax⁡E_{\max}, followed by a logarithmic behavior at higher energies. The former results from interactions between the two types of TLSs, whereas the latter is a result of interactions among the strongly interacting TLSs [50, 54]. With a dipole moment p2=110 Dp_{2}=110\,\mathrm{D}, and bias amplitudes Eb,max⁡=0.13 V/μmE_{b,\max}=0.13\,\mathrm{V}/\mu\mathrm{m}, 0.32 V/μm0.32\,\mathrm{V}/\mu\mathrm{m} and 0.44 V/μm0.44\,\mathrm{V}/\mu\mathrm{m}, the corresponding energies ℏω0+p2Eb,max⁡≈p2Eb,max⁡\hbar\omega_{0}+p_{2}E_{b,\max}\approx p_{2}E_{b,\max} are estimated as 3 kBK3\,\mathrm{k}_{\mathrm{B}}\mathrm{K}, 8 kBK8\,\mathrm{k}_{\mathrm{B}}\mathrm{K} and 12 kBK12\,\mathrm{k}_{\mathrm{B}}\mathrm{K}. Figure 4b shows the maximum excess loss tan⁡δ2,max⁡\tan\delta_{2,\max} as a function of p2Eb,max⁡p_{2}E_{b,\max}, which is a manifestation of the equilibrium DOS of the second TLS type at this energy. Our measurements are most likely taken in the regime where p2Eb,max⁡p_{2}E_{b,\max} is the vicinity of or outside the gap edge, namely in the logarithmic dependence domain.

Bias amplitude dependence of the maximum loss

Figure 4. a tan δ\delta measured as a function of E˙b\dot{E}_{b} for different bias field amplitudes Eb,max⁡E_{b,\max}. b The measured maximum excess loss tan⁡δ2,max⁡\tan\delta_{2,\max} as a function of p2Eb,max⁡p_{2}E_{b,\max}, proportional to the equilibrium DOS of the second type of TLSs at this energy, ρ2,eq(p2Eb,max⁡)\rho_{2,\mathrm{eq}}(p_{2}E_{b,\max}). The blue line shows a (weak) logarithmic dependence of the measured tan⁡δ2,max⁡\tan\delta_{2,\max}, which indicates that we most likely measured in the regime where p2Eb,max⁡p_{2}E_{b,\max} is in the vicinity of or outside the gap edge.

Discussion

Using a fast-swept bias field in addition to microwave fields in a resonator, we have studied the non-equilibrium loss in amorphous silicon. The data suggests the existence of two types of TLSs. At low bias rates the dielectric loss is determined by standard TLSs, which are weakly coupled to phonons. At high bias rates the dielectric loss deviates from the STM predictions, but agree well with the two-TLS model used. Analysis of the bias rate dependent loss indicates a gap in the DOS and the fitting yields a large electric dipole moment for the second TLS type. Recent works [58, 59] have examined the broad distribution of the coupling of the TLSs to phonons due to polaron effect [60], and the TLS-TLS interactions could lead to the reduction of TLS DOS. However, the bias application cannot release TLSs suppressed by the polaron effect because the polaron bath is formed by phonons almost instantaneously. Furthermore, a single continuously distributed TLS DOS cannot account for our experimental observations explained by two moments and relaxation parameters. Technically, the fittings of the data for the entire domain of driving powers and bias rates contain eight parameters (see Fig. 2c). These include the dipole moments p1≈11p_1 \approx11 D, p2≈110p_2 \approx110 D and relaxation rates Γ1,m(1)≈5.7\Gamma^{(1)}_{1,m} \approx5.7 MHz, Γ1,m(2)≈800\Gamma^{(2)}_{1,m} \approx800 MHz for the two types of TLSs, and the four parameters AA, BB, CC and DD used to fit the unsaturated excess loss at low and intermediate powers. The accuracy of these parameters depends on the details of the LZ analysis and the fitting precision is better than 10%. Also, our main results do not depend on the preciseness of these parameters, since the distinct dipole moments and relaxation rates of the two kinds of TLSs result from the distinct driving power dependence of the two contributions to the loss, below and above the intrinsic loss tangent tan⁡δ0\tan\delta_0. Experimental data consistent with the existence of two types of TLSs, characterized by distinct couplings to phonons, was previously reported [63, 64]. Our results here go beyond previous experiments as data attests directly to the presence of two types of TLSs, the bimodality of their couplings to phonons, and their electric dipole interaction strengths.

The location of the second TLS type (corresponding to p2p_2) must be in the film, similar to the standard type. The parallel plate capacitor geometry allows for measuring loss from TLSs from the dielectric within the capacitor alone, effectively eliminating TLS contributions to the loss elsewhere because electric fields are negligible outside of the capacitor. The second TLS type, similar to the standard TLSs, is believed to be uniformly distributed throughout the film similar to the first type, because TLSs in the bulk film have a 100% participation ratio, whereas those residing on the metal-dielectric interface in the capacitor would have a participation ratio on the order of 1% (using a thickness of a few nanometers for the interface, as is standard and expected).

The dipole moment of 110 D extracted for the second type of TLSs is anomalously large and it only appears out of equilibrium due to a gap in its density of states. One likely candidate for this TLS is tunneling nanoclusters of atoms, see e.g., Ref [62]. Insights of the two types of TLSs gleaned from the experiment are of general interest to understand TLS properties in amorphous solids, and of increasing importance due to their impact in quantum information science. The fast bias technique has been previously used to study the non-equilibrium loss for a couple different materials: silicon nitride[46] and alumina[21]. However, we have not yet seen excess loss in these materials. The two-TLS model is silent with respect to the electric dipole moments of the TLSs, which can vary between materials. A recent study suggests that mechanical and dielectric loss in amorphous silicon originates from two different types of TLSs [65]. Our measurement here is sensitive to the size of the electric dipole moment of the TLSs, which is plausibly material dependent. To check the generality of the second TLS type in other amorphous solids, one possibility is to repeat our non-equilibrium protocol, but measure both the dielectric and acoustic responses, as large acoustic response out of equilibrium is expected irrespective of the value of the electric dipole moment. Alternatively, one could also measure the thermal conductivity, which is proportional to square of the TLS-phonon coupling, and compare the results with and without a rapidly varying bias field. Such studies would allow detailed characterization of TLSs at low and high energies. This could clearly establish if high-energy TLSs will appear from high sweep rates in qubits and reduce their coherence, as well as help to uncover an appropriate microscopic model.

Methods

Experimental setup

The resonators are fabricated with Al/a-Si/Al trilayer films on highly resistive silicon substrates. The bottom aluminum layer (100 nm) and the top aluminum layer (250 nm) are sputtered via DC sputtering. The low-stress amorphous silicon layer (250 nm) is grown by PECVD at 100°C. The film can have oxygen- and hydrogen-based impurities related to the low growth temperature and SiH4SiH_4 used in the growth, either of which might be responsible for the moments in p1p_1 (see discussion above). The resonator (Fig. 1a) is defined by photolithography and subsequent etching. It consists of four equal bridge parallel-plate capacitors (Figs. 1a and 1b), which is modified from that in ref.31. Assuming the dielectric constant for the amorphous silicon is 11.5, the capacitor has a total capacitance C=1.2C = 1.2 pF (C1−C4=CC_1 - C_4 = C). The vias are etched by SF6SF_6 to connect the capacitor to the meander inductor. The resonator is inductively coupled to the coplanar transmission line for standard transmission measurements S21S_{21}. The measurements were performed at the base temperature of the dilution refrigerator (Tbase≈20T_{\mathrm{base}} \approx20 mK). A dc bias field, Eb=Vb/2dE_b = V_b/2d, where d=250d = 250 nm is the thickness of the amorphous silicon dielectric, is applied across the capacitors. The bridge layout is effective to isolate the microwave resonance energy from the bias field input port. A triangular bias voltage is applied. The time dependence of the loss for a resonator when a periodic bias field is applied is studied in Ref.[46]. The maximum bias is Eb,max⁡=0.44E_{b,\max} = 0.44 V/μ\mum, and the fastest bias frequency is fb=4.5f_b = 4.5 MHz. The resonance frequency f0=5.1f_0 = 5.1 GHz. S21S_{21}’s are measured as a function of photon number, nn and bias rate, E˙b\dot{E}_b. The corresponding loss tan⁡δ\tan\delta, equal to the inverse resonator quality 1/Qi1/Q_i, is found from each S21S_{21} measurement. See Supplementary Note 1 for more wiring and attenuation details in the measurement setup.

Loss due to LZ transitions within the STM

The Hamiltonian of a single TLS driven by the resonator electric field E⃗res(t)=E⃗accos⁡(ω0t)\vec{E}_{\mathrm{res}}(t)=\vec{E}_{\mathrm{ac}}\cos(\omega_{0}t) and by a time-dependent bias electric field E⃗b(t)\vec{E}_{b}(t) is H=(1/2)(Δ(t)σz+Δ0σx)−p⃗⋅E⃗accos⁡(ωt)σzH=(1/2)(\Delta(t)\sigma_{z}+\Delta_{0}\sigma_{x})-\vec{p}\cdot\vec{E}_{\mathrm{ac}}\cos(\omega t)\sigma_{z}, where σi\sigma_{i} (i=x,y,zi=x,y,z) are the Pauli matrices, p⃗\vec{p} is the TLS dipole moment, Δ(t)=Δ(0)−2p⃗⋅E⃗b(t)\Delta(t)=\Delta(0)-2\vec{p}\cdot\vec{E}_{b}(t) is the asymmetry energy tuned by the bias field and Δ0\Delta_{0} is the tunneling energy. In the instantaneous eigenbasis of the TLS, the corresponding Hamiltonian is

H=12E(t)σz−p⃗⋅E⃗accos⁡(ω0t)(Δ(t)E(t)σz−Δ0E(t)σx),(2)H=\frac{1}{2}E(t)\sigma_{z}-\vec{p}\cdot\vec{E}_{\mathrm{ac}}\cos(\omega_{0}t)\left(\frac{\Delta(t)}{E(t)}\sigma_{z}-\frac{\Delta_{0}}{E(t)}\sigma_{x}\right), \tag*{(2)}

where E(t)=Δ2(t)+Δ02E(t)=\sqrt{\Delta^{2}(t)+\Delta_{0}^{2}} is the TLS energy splitting, assumed to be slowly varying on the time scale 2π/ω02\pi/\omega_{0}. A TLS with Δ0<ℏω0\Delta_{0}<\hbar\omega_{0} can be swept through resonance at time t0t_{0} for which E(t0)=ℏω0E(t_{0})=\hbar\omega_{0}. Near this resonance the energy splitting can be expanded as E(t)≈ℏω0+ℏv(t−t0)E(t)\approx\hbar\omega_{0}+\hbar v(t-t_{0}) [46, 47], where v=E˙(t0)/ℏ=v01−(Δ0/ℏω0)2cos⁡θv=\dot{E}(t_{0})/\hbar=v_{0}\sqrt{1-(\Delta_{0}/\hbar\omega_{0})^{2}}\cos\theta with θ\theta the angle between p⃗\vec{p} and E⃗b\vec{E}_{b}, and v0=(2p/ℏ)E˙b(t0)v_{0}=(2p/\hbar)\dot{E}_{b}(t_{0}) the maximum bias rate.

The TLS absorption in the absence of the bias field can be separated into the so-called relaxation absorption, arising from the longitudinal term (∝σz\propto\sigma_{z}) in the brackets of Eq. (2), and the resonant absorption resulting from the transverse coupling (∝σx\propto\sigma_{x}) [28, 29, 66, 67]. In the regime ω0≫kBT/ℏ≫Γ1,Γ2\omega_{0}\gg k_{B}T/\hbar\gg\Gamma_{1},\Gamma_{2} considered in this paper (Γ1\Gamma_{1} and Γ2\Gamma_{2} are the TLS relaxation and decoherence rates, respectively), the resonant mechanism dominates and the longitudinal term in Eq. (2) can be neglected. The TLS dipole moment induced by the resonator field can therefore be written as ⟨p(t)⟩=−pcos⁡θ(Δ0/ℏω0)⟨σx(t)⟩=ℜ[χEace−iω0t]\langle p(t)\rangle=-p\cos\theta(\Delta_{0}/\hbar\omega_{0})\langle\sigma_{x}(t)\rangle=\Re\left[\chi E_{\mathrm{ac}}e^{-i\omega_{0}t}\right], where χ=χ′+iχ′′\chi=\chi^{\prime}+i\chi^{\prime\prime} is the TLS electric susceptibility. The imaginary part of the susceptibility, χ′′\chi^{\prime\prime}, yields the imaginary part ε′′\varepsilon^{\prime\prime} of the dielectric constant due to TLSs upon averaging over the ensemble of TLSs (χ′\chi^{\prime} gives the shift in the real part of the dielectric constant due to TLSs, which is small compared to the material dielectric constant ε\varepsilon). The loss tangent tan⁡δ=ε′′/ε\tan\delta=\varepsilon^{\prime\prime}/\varepsilon is thus proportional to the out of phase (∝sin⁡(ω0t)\propto\sin(\omega_{0}t)) component of ⟨σx(t)⟩\langle\sigma_{x}(t)\rangle. To calculate this component, we transform to the frame of reference rotating around the zz axis with frequency ω0\omega_{0}, using the unitary transformation UR=eiω0tσz/2U_{R}=e^{i\omega_{0}t\sigma_{z}/2}. Since URσxUR†=cos⁡(ω0t)σx−sin⁡(ω0t)σyU_{R}\sigma_{x}U_{R}^{\dagger}=\cos(\omega_{0}t)\sigma_{x}-\sin(\omega_{0}t)\sigma_{y}, in the rotating frame of reference the relation χ′′=(ℏΩR/Eac2)⟨σy⟩\chi^{\prime\prime}=(\hbar\Omega_{R}/E_{\mathrm{ac}}^{2})\langle\sigma_{y}\rangle holds [67], where ΩR=ΩR0cos⁡θ(Δ0/ℏω0)\Omega_{R}=\Omega_{R0}\cos\theta(\Delta_{0}/\hbar\omega_{0}) is the TLS Rabi frequency, with its maximum value ΩR0=pEac/ℏ\Omega_{R0}=pE_{\mathrm{ac}}/\hbar. Moreover, application of the rotating wave approximation yields the LZ Hamiltonian

HLZ=ℏ2[v(t−t0)σz+ΩRσx],(3)H_{\mathrm{LZ}}=\frac{\hbar}{2}\left[v(t-t_{0})\sigma_{z}+\Omega_{R}\sigma_{x}\right], \tag*{(3)}

governing the dynamics of a TLS resonant passage [47]. The probability for photon absorption in a transition is the famous LZ probability for an adiabatic transition from the initial state ∣g,n⟩|g,n\rangle to the final state ∣e,n−1⟩|e,n-1\rangle Fig. 1(d), Pad=1−e−πΩR2/(2v)P_{\mathrm{ad}}=1-e^{-\pi\Omega_{R}^{2}/(2v)}. Finite TLS relaxation and decoherence rates Γ1\Gamma_{1} and Γ2\Gamma_{2} can be taken into account by means of the Bloch equations within the rotating frame of reference [67]. The Hamiltonian (3) can be written as the Hamiltonian of a spin-1/2 particle in a magnetic field, HLZ=−S⃗⋅B⃗H_{\mathrm{LZ}}=-\vec{S}\cdot\vec{B}, where S⃗=ℏσ⃗/2\vec{S}=\hbar\vec{\sigma}/2 and B⃗=−(ΩR,0,v(t−t0)/ℏ)\vec{B}=-(\Omega_{R},0,v(t-t_{0})/\hbar). The corresponding Bloch equations read d⟨S⃗⟩/dt=⟨S⃗⟩×B⃗−(Γ2⟨Sx⟩,Γ2⟨Sy⟩,Γ1(⟨Sz⟩−Sz0))d\langle\vec{S}\rangle/dt=\langle\vec{S}\rangle\times\vec{B}-(\Gamma_{2}\langle S_{x}\rangle,\Gamma_{2}\langle S_{y}\rangle,\Gamma_{1}(\langle S_{z}\rangle-S_{z0})), where Sz0=(ℏ/2)tanh⁡(ℏω0/2kBT)S_{z0}=(\hbar/2)\tanh(\hbar\omega_{0}/2k_{B}T). In our calculations we neglect pure dephasing of the resonant TLSs, such that Γ2=Γ1/2\Gamma_{2}=\Gamma_{1}/2. Assuming TLS relaxation into the phonon bath, one has Γ1(Δ0)=Γ1,m(Δ0/ℏω0)2\Gamma_{1}(\Delta_{0})=\Gamma_{1,m}(\Delta_{0}/\hbar\omega_{0})^{2}, where Γ1,m\Gamma_{1,m} is the maximum relaxation rate for resonant TLSs with energy splitting E=ℏω0E=\hbar\omega_{0}. If the bias duration is longer than the TLS relaxation time T1=1/Γ1T_{1}=1/\Gamma_{1}, the steady state solution depends on time via the detuning from resonance δE=v(t−t0)\delta E=v(t-t_{0}), and ε′′\varepsilon^{\prime\prime} can be obtained by averaging χ′′\chi^{\prime\prime} over the ensemble of TLSs with the distribution function P(E,Δ0)=ρ(E)/E2−Δ02P(E,\Delta_{0})=\rho(E)/\sqrt{E^{2}-\Delta_{0}^{2}}, where ρ\rho is the TLS DOS, assumed to be energy-independent within the STM. The integration over energies can then be replaced by integration over time, leading to the expression

tan⁡δtan⁡δ0=3∫01dy y2∫01x dx1−x22vπΩR2∫−∞∞⟨σy(t)⟩ dt,(4)\frac{\tan\delta}{\tan\delta_{0}}=3\int_{0}^{1}dy\,y^{2}\int_{0}^{1}\frac{x\,dx}{\sqrt{1-x^{2}}}\frac{2v}{\pi\Omega_{R}^{2}}\int_{-\infty}^{\infty}\langle\sigma_{y}(t)\rangle\,dt, \tag*{(4)}

where tan⁡δ0=πρp2/(3ε)\tan\delta_{0}=\pi\rho p^{2}/(3\varepsilon) is the intrinsic loss tangent and the integration variables are x=Δ0/(ℏω0)x=\Delta_{0}/(\hbar\omega_{0}) and y=cos⁡θy=\cos\theta (in terms of these variables v=v01−x2yv=v_{0}\sqrt{1-x^{2}}y and ΩR=ΩR0xy\Omega_{R}=\Omega_{R0}xy). This loss is a function of the parameters ξ=2v0/(πΩR02)\xi=2v_{0}/(\pi\Omega_{R0}^{2}) and Γ1,m/ΩR0\Gamma_{1,m}/\Omega_{R0}, and can be calculated by a numerical integration of the Bloch equations.

In the regime Γ1,m/ΩR0≪1\Gamma_{1,m}/\Omega_{R0}\ll1, dissipation can be neglected for ξ≫Γ1,m/ΩR0\xi\gg\Gamma_{1,m}/\Omega_{R0}, and one obtains the expression for the dielectric loss tangent [47]

tan⁡δtan⁡δ0=3∫01dy y2∫01x dx1−x22vπΩR2[1−e−πΩR2/(2v)].(5)\frac{\tan\delta}{\tan\delta_{0}}=3\int_{0}^{1}dy\,y^{2}\int_{0}^{1}\frac{x\,dx}{\sqrt{1-x^{2}}}\frac{2v}{\pi\Omega_{R}^{2}}\left[1-e^{-\pi\Omega_{R}^{2}/(2v)}\right]. \tag*{(5)}

By performing numerical integration over xx and yy, one obtains the normalized loss tan⁡δ/tan⁡δ0\tan\delta/\tan\delta_{0} as a function of the dimensionless bias rate ξ\xi, approaching unity at ξ≫1\xi\gg1, as shown in Fig. 1(d).

Loss due to the second type of TLSs

Let us derive the expression for the non-equilibrium DOS of the second type of TLSs, ρ2(E,E⃗b)\rho_{2}(E,\vec{E}_{b}), under the assumption that their energy-dependent equilibrium DOS results from interactions between the two types of TLSs, as motivated by the Efros-Shklovskii Coulomb gap [55, 56] and discussed in Refs. [51, 54]. In equilibrium, the DOS is reduced by interaction with standard TLSs that break the Efros-Shklovskii stability criterion E+E′−2u/R3>0E+E^{\prime}-2u/R^{3}>0, where E′E' is the energy of the standard TLS, u/R3u/R^3 is the mutual TLS-TLS interaction and RR is the distance between the TLSs. We note that interactions exist also among the second type of TLSs, but since they are scarce at low energies, the dominating interactions are with the standard TLSs. As discussed in Refs. [51, 54], at higher energies interactions among the second type of TLSs lead to weaker logarithmic energy dependence of the DOS [see Figure 4]; here we neglect these interactions and consider the gap in the DOS at low energies due to interactions between the two types of TLSs. In the non-equilibrium situation of our experiment, only those standard TLSs that are capable of changing their state during the sweep time 1/(2fb)1/(2f_b) will contribute to the reconstruction of the gap in the DOS. For fb≪Γ1f_b \ll\Gamma_1, where Γ1\Gamma_1 is a typical relaxation rate of standard TLSs, the DOS ρ2(E,fb≪Γ1)\rho_2(E,f_b \ll\Gamma_1) should approach the DOS at equilibrium, ρ2,eq(E)\rho_{2,\mathrm{eq}}(E), discussed in Ref. [51]. On the other hand, in the limit of instantaneous sweep, fb≫Γ1f_b \gg\Gamma_1, the non-equilibrium DOS ρ2(E,fb≫Γ1)\rho_2(E,f_b \gg\Gamma_1) should satisfy the condition ρ2(E,fb≫Γ1)≈ρ2,eq(E+p2Eb,max⁡)\rho_2(E,f_b \gg\Gamma_1) \approx\rho_{2,\mathrm{eq}}(E+p_2 E_{b,\max}). Taking into account the fraction 1−e−Γ1/(2fb)1-e^{-\Gamma_1/(2f_b)} of standard TLSs that can flip during time 1/(2fb)1/(2f_b), where Γ1=Γ1(E′,Δ0)\Gamma_1=\Gamma_1(E',\Delta_0) is the relaxation time of standard TLSs with energy E′E' and tunneling energy Δ0\Delta_0, we obtain the expression

ρ2(E,fb)≈∏d3R[1−d3R∫0E0ρ1dE′∫Δ0,min⁡E′dΔ0Δ0Θ(2uR3−E−E′)(1−e−Γ12fb)ρ2,eq(E+p2Eb,max⁡)]≈exp⁡[−ρ1∫0E0dE′∫Δ0,min⁡E′dΔ0Δ0∫d3R Θ(2uR3−E−E′)(1−e−Γ12fb)ρ2,eq(E+p2Eb,max⁡)]=exp⁡[ρ1∫0E0dE′∫Δ0,min⁡E′dΔ0Δ0∫d3R Θ(2uR3−E−E′)e−Γ12fb]ρ2,eq(E),(6)\begin{aligned} \rho_2(E,f_b) &\approx\prod_{d^3R}\left[1-d^3R\int_0^{E_0}\rho_1dE'\int_{\Delta_{0,\min}}^{E'}\frac{d\Delta_0}{\Delta_0}\Theta\left(\frac{2u}{R^3}-E-E'\right)\left(1-e^{-\frac{\Gamma_1}{2f_b}}\right)\rho_{2,\mathrm{eq}}(E+p_2E_{b,\max})\right]\\ &\approx\exp\left[-\rho_1\int_0^{E_0}dE'\int_{\Delta_{0,\min}}^{E'}\frac{d\Delta_0}{\Delta_0}\int d^3R\,\Theta\left(\frac{2u}{R^3}-E-E'\right)\left(1-e^{-\frac{\Gamma_1}{2f_b}}\right)\rho_{2,\mathrm{eq}}(E+p_2E_{b,\max})\right]\\ &=\exp\left[\rho_1\int_0^{E_0}dE'\int_{\Delta_{0,\min}}^{E'}\frac{d\Delta_0}{\Delta_0}\int d^3R\,\Theta\left(\frac{2u}{R^3}-E-E'\right)e^{-\frac{\Gamma_1}{2f_b}}\right]\rho_{2,\mathrm{eq}}(E), \tag*{(6)} \end{aligned}

where Θ(x)\Theta(x) is the step function and E0=min⁡{Emax⁡,ℏω0+p2Eb,max⁡}E_0=\min\{E_{\max},\hbar\omega_0+p_2E_{b,\max}\} with Emax⁡∼10 KE_{\max}\sim10\,\mathrm{K} being the maximum energy of standard TLSs. This expression fulfills both limiting conditions discussed above. The exponent in the last line of Eq. (6) gives the enhancement of the equilibrium DOS ρ2,eq(E)\rho_{2,\mathrm{eq}}(E) due to standard TLSs with relaxation rates Γ1≲fb\Gamma_1\lesssim f_b. Using E˙b=2fbEb,max⁡\dot{E}_b=2f_bE_{b,\max} and assuming relaxation of standard TLSs to be dominated by interaction with phonons, corresponding to the rate Γ1(E′,Δ0)=Γ1,m(E′)(Δ0/E′)2\Gamma_1(E',\Delta_0)=\Gamma_{1,\mathrm{m}}(E')(\Delta_0/E')^2 with Γ1,m(E′)∝E′3\Gamma_{1,\mathrm{m}}(E')\propto E'^3 [28], we obtain

ρ2(E,E˙b)=ρ2,eq(E)exp⁡[8π3ρ1u∫0E0dE′E+E′∫0ln⁡(E′/Δ0,min⁡)dx e−Γ1,m(E′)(Δ0,min⁡/E′)2(Eb,max⁡/E˙b)e2x].(7)\rho_2(E,\dot{E}_b)=\rho_{2,\mathrm{eq}}(E)\exp\left[\frac{8\pi}{3}\rho_1u\int_0^{E_0}\frac{dE'}{E+E'}\int_0^{\ln(E'/\Delta_{0,\min})}dx\,e^{-\Gamma_{1,\mathrm{m}}(E')(\Delta_{0,\min}/E')^2(E_{b,\max}/\dot{E}_b)e^{2x}}\right]. \tag*{(7)}

where x=ln⁡(Δ0/Δ0,min⁡)x=\ln(\Delta_0/\Delta_{0,\min}). Since the major contribution to the integral over the tunneling amplitudes Δ0\Delta_0 comes from slow TLSs with small Δ0\Delta_0, we can further simplify the last expression by replacing E′E' by Emax⁡E_{\max} in the last integral and in the upper limit. This results in the expression

ρ2(E,E˙b)≈ρ2,eq(E)exp⁡[8π3ρ1uln⁡(1+E0E)∫0ln⁡(Emax⁡/Δ0,min⁡)dx e−Γ1,max⁡(Emax⁡)(Δ0,min⁡/Emax⁡)2(Eb,max⁡/E˙b)e2x].(8)\rho_2(E,\dot{E}_b)\approx\rho_{2,\mathrm{eq}}(E)\exp\left[\frac{8\pi}{3}\rho_1u\ln\left(1+\frac{E_0}{E}\right)\int_0^{\ln(E_{\max}/\Delta_{0,\min})}dx\,e^{-\Gamma_{1,\max}(E_{\max})(\Delta_{0,\min}/E_{\max})^2(E_{b,\max}/\dot{E}_b)e^{2x}}\right]. \tag*{(8)}

At low power driving, where the second type of TLSs is non-saturated, the excess loss is tan⁡δ2,0(E˙b)=πp22ρ2(ℏω0,E˙b)/(3ϵ)\tan\delta_{2,0}(\dot{E}_b)=\pi p_2^2\rho_2(\hbar\omega_0,\dot{E}_b)/(3\epsilon) and thus takes the form of Eq. (1). This loss serves as the intrinsic loss for the second TLS type, and is used for scaling the excess loss in Fig. (3). The LZ analysis described above is then carried out for the normalized excess loss tan⁡δ2/tan⁡δ2,0\tan\delta_2 / \tan\delta_{2,0} at all driving powers.

Data Availability

The data in the main text and Supplementary Materials are available from the corresponding authors upon request.

Acknowledgments

L.Y. would like to thank F. C. Wellstood and S. Dutta for their discussions and support. S.M. acknowledges support by the A. von Humboldt foundation. Y.J.R. acknowledges Lawrence Livermore National Laboratory operated by Lawrence Livermore National Security, LLC, for the U.S. Department of Energy, National Nuclear Security Administration under Contract DE-AC52-07NA2. A.L.B. and A.M. acknowledge the support by Carrol Lavin Bernick Foundation Research Grant (2020-2021), NSF CHE-1900568 grant and LINK Program of the NSF and Louisiana Board of Regents. M.S. acknowledges support from the Israel Science Foundation (Grant No. 2300/19).

Author Contributions

L.Y., Y.J.R., C.H. and K.D.O. designed and performed the experiments, L.Y., K.D.O. and S.M. analysed the data and wrote the paper. L.Y. and Y.J.R. processed the samples. S.M., M.S. A. M. and A.L.B. developed and carried out the theoretical work and numerical modeling. All authors discussed the results and commented on the manuscript.

Competing Interests

The authors declare no competing interests.

Experimentally revealing anomalously large dipoles in a quantum-circuit dielectric,

Yu et al.

Additional Details of the Experimental Setup

The measurements are performed in a dilution refrigerator at a temperature of less than 20 mK. Supplementary Figure 1 shows the schematic of the detailed wirings of the measurement setup. A vector network analyzer (VNA) is used to carry out the S21S_{21} measurements. The input line is heavily filtered with multiple attenuators at different stages. An additional 12 GHz K&L low-pass filter is added. The output signal is routed through two low-noise circulators (4--12 GHz) which are thermally anchored at mixing chamber plate. The output signal is amplified by a high electron mobility transistor (HEMT) at 4 K. For the bias line, a −10-10 dB attenuator is added at room temperature. A 12 GHz K&L low-pass filter and a RC filter are placed at mixing chamber stage, which gives a cut-off frequency of 10 MHz. A time-varying bias voltage as a triangular waveform is applied. The maximum bias is Eb=0.44E_{\mathrm{b}} = 0.44 V/μ\mum, and the fastest applied bias frequency is fb=4.5f_{\mathrm{b}} = 4.5 MHz. In previous related measurements, the time-dependent loss, particularly when the bias rate changes sign, was studied[1]. It is known that the change to the measured average loss is negligible. The base temperature of the dilution refrigerator remains below 20 mK throughout all measurements, such that no significant heating from bias leakage current is observed.

Schematic of the experimental wiring for measuring the resonator

Figure 1 (PDF p. 15). Schematic of the experimental wiring for measuring the resonator.

Input Microwave Power Calibration

To determine the input power on chip due to attenuation in the input line, throughput S21S_{21}’s (bypassing the mounted device) at various source powers have been measured at room temperature. This is particularly important to accurately determine the average photon number in the resonator. The attenuation depends on the signal frequency and the source power from the VNA. Four relatively high source power (10 dBm, 0 dBm, −10-10 dBm and −20-20 dBm) are used to obtain sufficient signal to noise ratio. As shown in Supplementary Fig. 2a, linear fits are conducted on all four semilog S21S_{21} plots. The frequency dependence is obtained by averaging the four slopes at different applied source powers. As shown in Supplementary Fig. 2b, the source power dependence of the input power at zero frequency is determined by plotting the intercepts as a function of the source powers. The obtained slope is one, which suggests any nonlinearities in the system associated with the power is negligible. Therefore, the eventual input power at mixing chamber can be expressed as

Throughput calibration measurements and linear fit of intercepts

Figure 2 (PDF p. 16). a Throughput calibration measurements of microwave source power at room temperature. b Plot and linear fit of the intercepts from the fits in a.

P=−1.87×10−9f+Pac−70.2(9)P = -1.87 \times10^{-9} f + P_{\mathrm{ac}} - 70.2 \tag*{(9)}

This expression is used to calibrate the photon number on the measured resonator throughout the main manuscript.

Extraction of loss from S21S_{21}

Magnitude of $|S_{21}|$ versus frequency and raw $S_{21}$ data in the IQ plane

Figure 3 (PDF p. 17). a An example of magnitude of ∣S21∣|S_{21}| vs frequency measured at Pac=−29 dBmP_{\mathrm{ac}}=-29\,\mathrm{dBm}. b Raw S21S_{21} data (blue dots) are plotted in IQ plane in the lower left. The red curve is a fitting of Supplementary Eq. 2. The black line represents the scaling of amplitude and the rotation from (1,0)(1,0). The yellow line represents a ϕ\phi rotation from a small impedance mismatch. We plot the data with adjusted amplitude and electric delays on the right.

The quality factor extraction is following the fitting function[2]

S21=Amp×eiθ×1−Q∣Q^e−1∣eiϕ1+2iQω−ω0ω0.(10)S_{21} = Amp \times e^{i\theta} \times\frac{1-Q\lvert\widehat{Q}_{e}^{-1}\rvert e^{i\phi}}{1+2iQ\frac{\omega-\omega_{0}}{\omega_{0}}}. \tag*{(10)}

The AmpAmp and θ\theta are the adjusted parameters for the raw data due to the attenuations, amplifications, and electric delays in both input and output line, and QQ is the total quality factor. Because of the impedance mismatch at two ports, Q^e\widehat{Q}_{e} is the external quality factor and also a complex number, and ϕ\phi is the angle of Q^e−1\widehat{Q}_{e}^{-1}. Fig. 3a shows an exemplary S21S_{21} data (magnitude vs frequency) measured at Pac=−29 dBmP_{\mathrm{ac}}=-29\,\mathrm{dBm}. The corresponding IQ plot of the data are shown in Fig. 3b: The lower left curve (blue dots) and the right curve (blue dots) are data before and after the adjusted AmpAmp and θ\theta. From the fitting, we extract Qe=1/Re⁡{Q^e−1}≈7300Q_{e}=1/\operatorname{Re}\{\widehat{Q}_{e}^{-1}\}\approx7300, Qi≈9000Q_{i}\approx9000, and ϕ=5.7∘\phi=5.7^\circ when the input photon number n=1422n=1422.

Supplementary References

  1. Khalil, M. S. et al. Landau-Zener population control and dipole measurement of a two-level-system bath. Phys. Rev. B 90, 100201 (2014).

  1. Khalil, M. S., Stoutimore, M. J. A., Wellstood, F. C. & Osborn, K. D. An analysis method for asymmetric resonator transmission applied to superconducting devices. J. Appl. Phys. 111, 054510 (2012).

References

  1. [1]Martinis, J. M., Cooper, K. B., McDermott, R., Steffen, M., Ansmann, M., Osborn, K. D., Cicak, K., Oh, S., Pappas, D. P., Simmonds, R. W., & Yu, C. C. Decoherence in Josephson Qubits from Dielectric Loss. Phys. Rev. Lett. 95, 210503 (2005).
  2. [2]Wang, C., Axline, C., Gao, Y. Y., , Brecht, T., Chu, Y., Frunzio, L., Devoret, M. H. & Schoelkopf, R. J. Surface participation and dielectric loss in superconducting qubits. Appl. Phys. Lett. 96, 072505 (2015).
  3. [3]Klimov, P. V. et al. Fluctuations of Energy-Relaxation Times in Superconducting Qubits. Phys. Rev. Lett. 121, 090502 (2018).DOI
  4. [4]Burnett, J. et al. Decoherence benchmarking of superconducting qubits. npj Quantum Inf. 5, 54 (2019).
  5. [5]Müller, C., Cole, J. H. & Lisenfeld, J. Towards understanding two-level-systems in amorphous solids: insights from quantum circuits. Rep. Prog. Phys 82, 124501 (2019).DOI
  6. [6]Oliver, W. D. & Welander P. B. Materials in superconducting quantum bits. MRS Bull. 38, 816 (2013).
  7. [7]Shalibo, Y. et al. Lifetime and Coherence of Two-Level Defects in a Josephson Junction. Phys. Rev. Lett. 105, 177001 (2010).
  8. [8]Zaretskey, V., Suri, B., Novikov, S., Wellstood, F. C. & Palmer, B. S. Spectroscopy of a Cooper-pair box coupled to a two-level system via charge and critical current. Phys. Rev. B 87, 174522 (2013).DOI
  9. [9]Christensen, B. G. et al. Anomalous charge noise in superconducting qubits. Phys. Rev. B 100, 140503 (2019).
  10. [10]Pappas, D. P., Vissers, M. R., Wisbey, D. S., Kline, J. S. & Gao, J. Two Level System Loss in Superconducting Microwave Resonators. IEEE Trans. Appl. Supercond. 21, 871 (2011).
  11. [11]Paik, H. & Osborn, K. D. Reducing quantum-regime dielectric loss of silicon nitride for superconducting quantum circuits. Appl. Phys. Lett. 96, 072505 (2010).DOI
  12. [12]Müller, C., Lisenfeld, J., Shnirman, A. & Poletto, S., Interacting two-level defects as sources of fluctuating high-frequency noise in superconducting circuits. Phys. Rev. B 92, 035442 (2015).DOI
  13. [13]Schlör, S. et al. Correlating Decoherence in Transmon Qubits: Low Frequency Noise by Single Fluctuators. Phys. Rev. Lett. 123, 190502 (2019).DOI
  14. [14]de Graaf, S. E. et al. Suppression of low-frequency charge noise in superconducting resonators by surface spin desorption. Nat. Commun. 9, 1143 (2018).DOI
  15. [15]Gao, J. et al. Experimental evidence for a surface distribution of two-level systems in superconducting lithographed microwave resonators. Appl. Phys. Lett. 92, 152505 (2008).DOI
  16. [16]Burnett, J. et al. Evidence for interacting two-level systems from the 1/f noise of a superconducting resonator. Nat. Commun. 5, 4119 (2014).DOI
  17. [17]Faoro, L. & Ioffe, L. B. Interacting tunneling model for two-level systems in amorphous materials and its predictions for their dephasing and noise in superconducting microresonators. Phys. Rev. B 91, 014201 (2015).DOI
  18. [18]Burin, A. L., Matityahu, S. & Schechter, M. Low-temperature 1/f noise in microwave dielectric constant of amorphous dielectrics in Josephson qubits. Phys. Rev. B 92, 174201 (2015).DOI
  19. [19]Connors, E. J., Nelson, J., Qiao, H., Edge, L. F. & Nichol, J. M. Low-frequency charge noise in Si/SiGe quantum dots. Phys. Rev. B 100, 165305 (2019).
  20. [20]Rosen, Y. J., Khalil, M. S., Burin, A. L. & Osborn, K. D. Random-Defect Laser: Manipulating Lossy Two-Level Systems to Produce a Circuit with Coherent Gain. Phys. Rev. Lett. 116, 163601 (2016).DOI
  21. [21]Matityahu, S. et al. Dynamical decoupling of quantum two-level systems by coherent multiple Landau–Zener transitions. npj Quantum Inf. 5, 114 (2019).DOI
  22. [22]Casparis, L., Connolly, M. R., Kjaergaard, M., Pearson, N. J., Kringhøj, A., Larsen, T. W., Kuemmeth, F., Wang, T., Thomas, C., Gronin, S., Gardner, G. C., Manfra, M., J. Marcus. C. M. & Pettersson, K. D. Superconducting gatemon qubit based on a proximitized two-dimensional electron gas. Nat. Nanotech. 13, 915 (2018).DOI
  23. [23]Kawakami, E., Scarlino, P., Ward, D. R., Braakman, F. R., Savage, D. E., Lagally, M. G., Friesen, M., Coppersmith, S. N., Eriksson, M. A. & Vandersypen, L. M. K. Electrical control of a long-lived spin qubit in a Si/SiGe quantum dot. Nat. Nanotech. 9, 666 (2014).DOI
  24. [24]Plugge, S., Rasmussen, A., Reinthold, E. & Flensberg, K. Majorana box qubits. New J. Phys. 19, 012001 (2017).
  25. [25]Zeller, C. R. & Pohl, R. O. Thermal Conductivity and Specific Heat of Noncrystalline Solids. Phys. Rev. B 4, 2029 (1971).
  26. [26]Phillips, W. A. Tunneling states in amorphous solids. J. Low Temp. Phys. 7, 351 (1972).
  27. [27]Anderson, P. W., Halperin, B. I. & Varma, C. M. Anomalous low-temperature thermal properties of glasses and spin glasses. Philos. Mag. J. Theor. Exp. Appl. Phys. 25, 1 (1972).
  28. [28]Phillips, A. W. Two-level states in glasses. Rep. Prog. Phys. 50, 1657 (1987).
  29. [29]Hunklinger, S. & Raychaudhuri, A. K. Chapter 3: Thermal and Elastic Anomalies in Glasses at Low Temperatures. Prog. Low Temp. Phys. 9, 265 (1986).DOI
  30. [30]Pohl, R. O., Liu, X. & Thompson, E. J. Low-temperature thermal conductivity and acoustic attenuation in amorphous solids. Rev. Mod. Phys. 74, 991 (2002).DOI
  31. [31]Queen, D. R. et al. Two-level systems in evaporated amorphous silicon. J. Non-Cryst. Solids 426, 19 (2015).
  32. [32]Molina-Ruiz, M. et al. Two-level systems and growth-induced metastability in hydrogenated amorphous silicon. Mater. Res. Express 7, 095201 (2020).DOI
  33. [33]Liu, X. et al. Comparing amorphous silicon prepared by electron-beam evaporation and sputtering toward eliminating atomic tunneling states. J. Alloys Compd. 855, 157431 (2021).DOI
  34. [34]Southworth, D. R. et al. Stress and Silicon Nitride: A Crack in the Universal Dissipation of Glasses. Phys. Rev. Lett. 102, 225503 (2009).
  35. [35]Bartkowiak, M. et al. Nuclear Quadrupole Moments as a Microscopic Probe to Study the Motion of Atomic Tunneling Systems in Amorphous Solids. Phys. Rev. Lett. 110, 205502 (2013).
  36. [36]Yu, C. C. & Leggett, A. J. Low temperature properties of amorphous materials: Through a glass darkly. Comments Condens. Matter Phys. 14, 231 (1988).
  37. [37]Liu, X. et al. Amorphous Solid without Low Energy Excitations. Phys. Rev. Lett. 78, 4418 (1997).DOI
  38. [38]Grabovskij, G. J., Peichl, T., Lisenfeld, J., Weiss, G. & Ustinov, A. V. Strain Tuning of Individual Atomic Tunneling Systems Detected by a Superconducting Qubit. Science 338, 232 (2012).DOI
  39. [39]Brehm, J. D., Bilmes, A., Weiss, G., Ustinov, A. V. & Lisenfeld J. Transmission-line resonators for the study of individual two-level tunneling systems. Appl. Phys. Lett. 111, 112601 (2017).DOI
  40. [40]Sarabi, B., Ramanayaka, A. N., Burin, A. L., Wellstood, F. C. & Osborn, K. D. Projected Dipole Moments of Individual Two-Level Defects Extracted Using Circuit Quantum Electrodynamics. Phys. Rev. Lett. 116, 167002 (2016).DOI
  41. [41]Lisenfeld, J. et al. Electric field spectroscopy of material defects in transmon qubits. npj Quantum Inf. 5, 1 (2019).DOI
  42. [42]Hung, C. C., Yu, L., Foroozani, N., Fritz, S., Gerthsen, D. & Osborn, K. Probing Hundreds of Individual Quantum Defects in Polycrystalline and Amorphous Alumina. Phys. Rev. Applied 17, 034025 (2022).DOI
  43. [43]Lisenfeld, J., Bilmes, A., Matityahu, S., Zanker, S., Marthaler, M., Schechter, M., Schön, G., Shnirman, A., Weiss, G. & Ustinov, A. V., Decoherence spectroscopy with individual two-level tunneling defects. Sci. Rep. 6, 23786 (2016).DOI
  44. [44]Bilmes, A., Megrant, A., Klimov, P., Weiss, G., Martinis, J. M., Ustinov, A. V. and Lisenfeld J. Resolving the positions of defects in superconducting quantum bits. Sci. Rep. 10, 3090 (2020).
  45. [45]Bilmes, A., Voloshenuik, S., Brehm, J. D., Ustinov, A. V. & Lisenfeld, J. Quantum sensors for microscopic tunneling systems, npj Quantum Inf. 7, 27 (2021).DOI
  46. [46]Khalil, M. S. et al. Landau-Zener population control and dipole measurement of a two-level-system bath. Phys. Rev. B 90, 100201 (2014).DOI
  47. [47]Burin, A. L., Khalil, M. S. & Osborn, K. D. Universal Dielectric Loss in Glass from Simultaneous Bias and Microwave Fields. Phys. Rev. Lett. 110, 157002 (2013).
  48. [48]Von Schickfus, M. & Hunklinger, S. Saturation of the dielectric absorption of vitreous silica at low temperatures. Phys. Lett. A 64, 144 (1977).DOI
  49. [49]Natelson, D., Rosenberg, D. & Osheroff, D. D. Evidence for Growth of Collective Excitations in Glasses at Low Temperatures. Phys. Rev. Lett. 80, 4689 (1998).DOI
  50. [50]Burin, A. L. Dipole Gap Effects in Low Energy Excitation Spectrum of Amorphous Solids. Theory for Dielectric Relaxation. J. Low Temp. Phys. 100, 309 (1995).
  51. [51]Schechter, M. & Stamp, P. C. E. Inversion symmetric two-level systems and the low-temperature universality in disordered solids. Phys. Rev. B 88, 174202 (2013).
  52. [52]DuBois, T. C., Per, M. C., Russo, S. P. & Cole, J. H. Delocalized Oxygen as the Origin of Two-Level Defects in Josephson Junctions. Phys. Rev. Lett. 110, 077002 (2013).
  53. [53]Churkin, A., Barash, D. & Schechter, M. Nonhomogeneity of the density of states of tunneling two-level systems at low energies. Phys. Rev. B 89, 104202 (2014).DOI
  54. [54]Churkin, A., Gabdank, I., Burin, A. & Schechter, M. The strain gap in a system of weakly and strongly interacting two-level systems. Preprint at http://arxiv.org/abs/1307.0868 (2014).
  55. [55]A. L. Efros, A. L. and Shklovskii, B. I. Coulomb gap and low temperature conductivity of disordered systems. J. Phys. C 8, L49 (1975).
  56. [56]Baranovskii, S. D., Efros, A. L., Gelmont, B. L. & Shklovskii B. I. Coulomb gap in disordered systems: computer simulation. J. Phys. C 12, 1023 (1979).
  57. [57]Gaita-Ariño, A. & Schechter, M. Identification of Strong and Weak Interacting Two-Level Systems in KBr:CN. Phys. Rev. Lett. 107, 105504 (2011).
  58. [58]Carruzzo, H. M. & Yu, C. C. Why phonon scattering in glasses is universally small at low temperatures. Phys. Rev. Lett. 124, 075902 (2020).DOI
  59. [59]Carruzzo, H. M., Bilmes, A., Lisenfeld, J., Yu, Z., Wang, B., Wan, Z., Schmidt, J. R. & Yu, C. C. Distribution of two-level system couplings to strain and electric fields in glasses at low temperatures. Phys. Rev. B 104, 134203 (2021).
  60. [60]Agarwal, K., Martin, I., Lukin, M. D. & Demler, E. Polaronic model of two-level systems in amorphous solids. Phys. Rev. B 87, 144201 (2013).
  61. [62]Lubchenko, V. & Wolynes, P. G. Intrinsic Quantum Excitations of Low Temperature Glasses. Phys. Rev. Lett. 87, 195901 (2001).
  62. [63]Matityahu, S., Shnirman, A., Schön, G. & Schechter, M. Decoherence of a quantum two-level system by spectral diffusion. Phys. Rev. B 93, 134208 (2016).
  63. [64]Kirsh, N., Svetitsky, E., Burin, A. L., Schechter, M. & Katz, N. Revealing the nonlinear response of a tunneling two-level system ensemble using coupled modes. Phys. Rev. Materials 1, 012601(R) (2017).
  64. [65]M. Molina-Ruiz, M., Rosen, Y. J., Jacks, H. C., Abernathy, M. R., Metcalf, T. H., Liu, X., DuBois, J. L. & Hellman, F. Origin of mechanical and dielectric losses from two-level systems in amorphous silicon. Phys. Rev. Materials 5, 035601 (2021).
  65. [66]Jäckle, J., Piché, L., Arnold, W. & Hunklinger, S. Elastic effects of structural relaxation in glasses at low temperatures. J. Non-Cryst. Solids 20, 365 (1976).
  66. [67]Carruzzo, H. M., Grannan, E. R. & Yu, C. C. Nonequilibrium dielectric behavior in glasses at low temperatures: Evidence for interacting defects. Phys. Rev. B 50, 6685 (1994).

Paper details

vlnc:W_vf7rnrdp
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