betaFeedback

WorksPhysics and Astronomy

Electrostatic gate-controlled quantum interference in a high-mobility two-dimensional electron gas at the (La0.3_{0.3}Sr0.7_{0.7})(Al0.65_{0.65}Ta0.35_{0.35})O3_3/SrTiO3_3 interface

Abstract

We report quantum oscillations in magnetoresistance that are periodic in magnetic field (BB), observed at the interface between (La0.3_{0.3}Sr0.7_{0.7})(Al0.65_{0.65}Ta0.35_{0.35})O3_3 and SrTiO3_3. Unlike Shubnikov-de Haas oscillations, which appear at magnetic fields >7 > 7 T and diminish quickly as the temperature rises, these BB-periodic oscillations emerge at low fields and persist up to 10 K. Their amplitude decays exponentially with both temperature and field, specifying dephasing of quantum interference. Increasing the carrier density through electrostatic gating results in a systematic reduction in both the amplitude and frequency of the oscillations, with complete suppression beyond a certain gate voltage. We attribute these oscillations to the Altshuler-Aronov-Spivak effect, likely arising from naturally formed closed-loop paths due to the interconnected quasi-one-dimensional conduction channels along SrTiO3_3 domain walls. The relatively long phase coherence length (∼\sim 1.8 μμm at 0.1 K), estimated from the oscillation amplitude, highlights the potential of complex oxide interfaces as a promising platform for exploring quantum interference effects and advancing device concepts in quantum technologies, such as mesoscopic interferometers and quantum sensors.

Quantum interference – whether observed as the Aharonov–Bohm (AB) effect [1, 2], the Altshuler–Aronov–Spivak (AAS) effect [3, 4], or quantum conductance fluctuations [5] – offers valuable insights into electron dynamics within an electromagnetic environment. These phenomena allow direct measurement of the phase coherence length, which is critically important for developing quantum technologies such as quantum sensing, quantum computing, and quantum communication. The AB and AAS effects are manifested as conductance or resistance oscillations that are periodic in the magnetic flux, ϕ=BA\phi= BA, where BB is magnetic field and AA is area enclosed by electron trajectory typically within ring or cylindrical geometries. These oscillations differ from Shubnikov–de Haas (SdH) oscillations, which are periodic in the inverse magnetic field (1/B1/B) and arise from the Fermi energy crossing magnetically quantised electronic states. Notably, AAS oscillations persist even in the presence of disorder and survive ensemble averaging, making them robust in diffusive transport regimes. In contrast, AB oscillations are a hallmark of ballistic transport and tend to vanish when averaged over an ensemble of rings. Both AB and AAS oscillations have been observed in ring structures fabricated from III–V semiconductor heterostructure [6, 7], graphene [8, 9], and other two-dimensional (2D) materials.

The 2D electron gas (2DEG) at complex oxide interfaces offers a novel platform for exploring quantum interference in previously uncharted regimes. Unlike 2D electron systems in conventional III–V semiconductors, complex oxide interfaces such as LaAlO3/SrTiO3\mathrm{LaAlO}_{3}/\mathrm{SrTiO}_{3} (LAO/STO) host a strongly correlated 2DEG, leading to emergent phenomena like superconductivity [10, 11, 12] and magnetism [11, 12, 13, 14]. S. Goswami et al. fabricated superconducting quantum interference device at the superconducting LAO/STO interface and measured BB-periodic oscillations in voltage by applying an excitation current higher than its critical value [15]. Interestingly, AB-like oscillations were reported in a Hall bar device (channel width w=250w = 250 nm, length l=1.5 μml = 1.5\ \mu\mathrm{m}), but not in a nanoscale ring, both fabricated at the conducting LAO/STO interface [16].

To have a better understanding of the mechanism of BB-periodic oscillations in complex oxide interfaces and their tunability through external knobs such as electrostatic gating, we studied quantum transport in an ultrahigh mobility (35,000 cm2^{2}V−1^{-1}s−1^{-1}) 2DEG at the interface between (La0.3Sr0.7)(Al0.65Ta0.35)O3(\mathrm{La}_{0.3}\mathrm{Sr}_{0.7})(\mathrm{Al}_{0.65}\mathrm{Ta}_{0.35})\mathrm{O}_{3} (LSAT) and STO. Due to its smaller lattice mismatch (∼1%\sim1\%) compared to LAO/STO (∼3%\sim3\%), the LSAT/STO interface supports extremely high mobility 2DEG, reaching up to 50,000 cm2^{2}V−1^{-1}s−1^{-1} [17, 18]. Furthermore, the LSAT/STO interface hosts long-range ordered line dislocations that form a 2D square moiré pattern with a moiré lattice constant of ∼40\sim40 nm [19] and/or interconnected polar domain walls at low temperatures [20]. These naturally formed defect lines may serve as 1D conduction channels for electron transport, providing a promising platform for studying quantum interference phenomena.

Here, we investigate magnetoresistance of LSAT/STO devices at temperatures down to 100 mK and in magnetic fields up to 16 T. Our LSAT/STO devices consist of 10 unit cells of LSAT film grown on TiO2_2-terminated STO (001) single crystals using pulse laser deposition (PLD) technique. The epitaxial growth of LSAT was precisely controlled by in-situ reflection high energy electron diffraction. During growth, the oxygen partial pressure (PO2P_{\mathrm{O}_2}) and temperature were maintained at 1×10−41 \times10^{-4} torr and 950°C, respectively. After growth, the samples were annealed in the oxygen atmosphere of pressure 25 mTorr and at the temperature of 600°C for filling the oxygen vacancies. As displayed in Figure 1a inset, the six-terminal Hall bar devices of width 50 μ\mum and length 160 μ\mum between the longitudinal voltage leads were fabricated by conventional photolithography technique using AlN films as hard mask. For transport measurements, the devices were electrically contacted with aluminum wires using wire bonder and then cooled from room temperature down to 100 mK in a dilution fridge. Simultaneous measurements of longitudinal resistance RxxR_{xx} and Hall resistance RyxR_{yx} were performed using an excitation current of 100 nA in static magnetic fields and 1 μ\muA in pulsed magnetic fields. We measured two devices, device-1 and device-2. The measurements on the as-grown device-1 were not feasible due to extreme inhomogeneity leading to high resistance. To enable resistance measurements and tune carrier density on this device, we employed electrostatic gating through the STO substrate of thickness 0.5 mm. For back-gate operation, the rear surface of the substrate was coated with silver paint, and a gold wire was attached at the center of this surface to serve as the gate electrode. We performed measurements at different tilt angles with respect to BB orientation using in-situ rotator probe. Measurements at temperatures above 1.2 K were performed in a helium-4 cryostat.

B-periodic oscillations in magnetoresistance

Figure 1. BB-periodic oscillations in magnetoresistance suggesting quantum interference. (a) Magnetic field dependence of longitudinal resistance Rxx(B)R_{xx}(B) at a few selected temperatures and at a back-gate voltage Vg=3V_g = 3 V. The curves are shifted vertically with spacing of 0.1 kΩ\Omega for better visuality of oscillations. The BB-periodic oscillations in Rxx(B)R_{xx}(B) in low fields (0 - 7 T), depicted by vertical dashed lines, survive above 1 K. Inset is a optical microscope image of the Hall-bar device with schematic of transport measurements. (b) A second order derivative of Rxx(B)R_{xx}(B) (blue solid line and left y axis) reveals BB-periodic oscillation along with higher harmonics. Rxx(B)R_{xx}(B) is displayed with red solid line and on right y axis. (c) Conductance (Inverse of Rxx(B)R_{xx}(B)) in the unit of conductance quantum e2/he^{2}/h (right y axis) and the oscillating conductance after subtracting the background of 1/Rxx1/R_{xx} (left y axis). The black solid lines are the 2nd2^{nd} order polynomial fit to the 1/Rxx1/R_{xx} data to subtract the monotonic background.

fig:1a shows the longitudinal resistance, Rxx(B)R_{xx}(B), measured at various fixed temperatures and at a gate voltage of Vg=3V_g = 3 V for device-1. The positive magnetoresistance (MR) exhibits an oscillatory behaviour across the entire magnetic field range (0-16 T). Most interestingly, the oscillations in low fields (0-7 T) are periodic in BB, as depicted by the vertical dashed lines, and persist up to temperatures of at least 9 K (see Figure 4 and Figure 5). However, the high-field oscillation amplitude damps with increasing temperature and vanishes entirely above 1.2 K (see Figure 6). While the high-field oscillations corresponds to the SdH effect, which is reported widely in STO-based 2DEGs [21, 22, 23, 24, 25, 26], the BB-periodic oscillations observed at an oxide interface without a mesoscopic ring and ballistic transport are particularly intriguing. The second-order derivative of Rxx(B)R_{xx}(B) (−d2Rxx/dB2-\mathrm{d}^{2}R_{xx}/\mathrm{d}B^{2}) in both field directions (up and down) reveals the existence of BB-periodic oscillations of period ∼1.7\sim1.7 T, along with higher harmonics (Figure 1b). Further, as displayed in Figure 1c, the amplitude of these oscillations is ∼40\sim40% of the quantum conductance (e2/he^{2}/h) and decays as the magnetic field increases (Figure 5).

Phase coherence length estimation from the temperature dependence of B-periodic oscillations

Figure 4. Phase coherence length LϕL_{\phi} estimation from the temperature dependence of BB-periodic oscillations. (a) Oscillating component of resistance at different temperatures. (b) Main panel: FFT of BB-periodic oscillations producing 3 additional harmonic peaks. Inset: Temperature dependence of FFT amplitude with linear fit. (c) Temperature and (d) gate dependence of estimated LϕL_{\phi}.

Damping of B-periodic oscillations with magnetic field and temperature

Figure 5. (a) Damping of BB-periodic oscillations with magnetic field and temperature. No oscillations are observed above 9 K. (b) Oscillations follow exponential decay with magnetic fields.

SdH oscillations after subtracting the monotonomous background of Rxx(B)

Figure 6. (a) SdH oscillations after subtracting the monotonous background of Rxx(B)R_{xx}(B). (b) Main panel: Temperature dependence of SdH oscillations amplitude ΔRxx\Delta R_{xx} at different magnetic fields fitted with L-K formula for cyclotron mass extraction. Inset: A linear fit to the inverse field dependence of normalized ΔRxx\Delta R_{xx} for the estimation of quantum mobility.

To understand the origin of BB-periodic oscillations and investigate their tunability through electrostatic gating, we measured the same device at different values of gate voltage (VgV_g) at 0.1 K. Figure 2a and Figure 2b show Rxx(B)R_{xx}(B) and Ryx(B)R_{yx}(B), respectively, for different VgV_g values. Measurements of Rxx(B)R_{xx}(B) for Vg<3V_g < 3 V and Ryx(B)R_{yx}(B) for Vg<3.8V_g < 3.8 V were not possible because this as-grown sample

Gating effect on electrical properties and quantum oscillations

Figure 2. Gating effect on electrical properties and quantum oscillations. (a) Rxx(B)R_{xx}(B) and (b) Ryx(B)R_{yx}(B) at different values of VgV_g. Dashed line in (b) is the linear fit to Ryx(B)R_{yx}(B) at Vg=3.8V_g = 3.8 V. (c) Zero-field offset of RyxR_{yx} as a function RxxR_{xx}. Non-linear dependence of RyxR_{yx} on RxxR_{xx} in zero field indicates the existence of charge inhomogeneity at the interface. (d) Carrier density, nn (left y axis), and mobility, μ\mu (right y axis) as a function of VgV_g. (e) An evolution of oscillations with varying VgV_g. BB-periodic oscillations completely suppress for Vg>20V_g > 20 V. (f) FFT of BB-periodic oscillations reveals two peaks, the main peak at 0.580.58 T−1^{-1} and the harmonic peak at 1.161.16 T−1^{-1}. (g) Variation in the oscillations frequency with VgV_g. For the best estimation of frequency, we assign the minima to integer and linear fit the plot of minima positions vs their number.

was highly resistive due to extreme charge inhomogeneity. The inhomogeneity is evident from the RyxR_{yx} offset at zero magnetic field, which decreases progressively with increasing VgV_g and approaches zero for Vg≥20V_g \geq20 V. To identify if a misalignment in Hall voltage contacts is not a cause of RyxR_{yx} offset, we plot zero-field RyxR_{yx} as a function of RxxR_{xx} in Figure 2c. A non-linear dependence of RyxR_{yx} on RxxR_{xx} confirms that the RyxR_{yx} offset is due to the inhomogeneous distribution of carriers at the interface, not the misalignment of contacts.

From Figure 2b, the slope of Ryx(B)R_{yx}(B) decreases progressively with increasing VgV_g, indicating an increase in the electron population within the quantum well formed at the interface. This trend is further corroborated by a reduction in both the zero-field RxxR_{xx} and the positive MR (Figure 2a). We estimate the carrier density, nn, through a linear fit to Ryx(B)R_{yx}(B) as shown by a dashed red line for Vg=3.8V_g = 3.8 V in Figure 2b, and plot nn and mobility μ\mu as a function of VgV_g in Figure 2d. At Vg=3.8V_g = 3.8 V, nn is 2.3×10122.3 \times10^{12} cm−2^{-2}, which increases with VgV_g up to 50 V and saturates at a maximum value of ∼7.5×1012\sim7.5 \times10^{12} cm−2^{-2} beyond 50 V. This behaviour of n(Vg)n(V_g) closely resembles previously reported trends at LAO/STO interfaces [25, 27]. The saturation of nn with increasing VgV_g can be attributed to the bending of the STO conduction band, which allows electrons to escape the quantum well and become trapped by defects [25, 28]. The mobility μ(Vg)\mu(V_g) follows the same trend as n(Vg)n(V_g), except that the μ\mu starts decreasing with VgV_g above

60 V. For Vg=3.8V_g = 3.8 V, μ\mu is ∼21,700\sim21{,}700 cm2^{2}V−1^{-1}s−1^{-1} and reaches a maximum value of ∼35,000\sim35{,}000 cm2^{2}V−1^{-1}s−1^{-1} at Vg=60V_g = 60 V. A linear proportionality of μ\mu with nn indicates the high quality of the electron gas, however, the decrease in μ\mu above 60 V indicates a slight growth in ionised defects, which enhances the scattering rate while the density remains almost unchanged.

As shown in Figure 2e and as a colour contour plot in Figure 7, the oscillations exhibit a systematic shift in position as VgV_g increases. Specifically, as indicated by the vertical dashed lines, the SdH oscillations at high fields (B>7B > 7 T) shift toward higher BB, consistent with the chemical potential moving to higher energies with increasing positive VgV_g. In contrast, BB-periodic oscillations, also marked by dashed lines, shifts toward lower BB. Notably, the BB-periodic oscillations amplitude diminishes with increasing VgV_g and the oscillations nearly vanish at Vg=20V_g = 20 V, where the density exceeds ∼5×1012\sim5 \times10^{12} cm−2^{-2} and zero-field RyxR_{yx} approaches zero. It is worth noting that the samples reported in Ref. [18], with densities on the order of 101310^{13} cm−2^{-2}, did not exhibit BB-periodic oscillations, consistent with the emergence of BB-periodic oscillations only at lower carrier densities.

Colour contour plot of oscillations evolution with $V_g$

Figure 7. Colour contour plot of oscillations evolution with VgV_g at T=0.1T = 0.1 K. Blue colour represents oscillations minima and red maxima. The vertical dashed line at B∼7B \sim7 T divide the region of BB-periodic AAS oscillations and SdH oscillations. The decay of BB-periodic oscillations and their shift to lower BB with increasing VgV_g is clearly visible. However, we don’t see any amplitude change with VgV_g for SdH oscillations.

From the fast Fourier transform (FFT) analysis of BB-periodic oscillations (Figure 2f), we extract a primary frequency of 0.580.58 T−1^{-1} (corresponding to a period B0=1.7B_0 = 1.7 T) and a second harmonic at 1.11.1 T−1^{-1}. As displayed in Figure 2g, the frequency of the BB-periodic oscillations decreases linearly with increasing VgV_g, a trend confirmed by both FFT analysis and a plot of the oscillation minima positions versus their index (inset of Figure 2g).

As we know, our LSAT/STO device is not fabricated in a ring geometry; instead, it is a Hall bar with a length of 160 μ\mum and a width of 50 μ\mum. For a micro-scale spatially inhomogeneous sample, the transport mobility extracted using the Drude formula can be a rather crude estimate. To obtain a more reliable measure of electron scattering, we therefore estimate the quantum mobility by analyzing the SdH oscillations using the Lifshitz–Kosevich formalism [29]. The transport mobility is insensitive to small angle scattering, whereas quantum mobility is sensitive to all scattering events and is typically one order of magnitude smaller than transport mobility in 2D systems in which the carriers and ionized impurities are spatially separated. From the field-dependent of the SdH oscillations amplitude (see Figure 6b), we extract a quantum mobility of μq∼2530\mu_q \sim2530 cm2^2V−1^{-1}s−1^{-1}, one order of magnitude smaller than the transport mobility. Now, we calculate the Fermi wavelength as λF=2π/n=16\lambda_F = \sqrt{2\pi/n} = 16 nm, and the corresponding mean free path as l=hμq/eλF≈65l = h\mu_q/e\lambda_F \approx65 nm (at Vg=3.0V_g = 3.0 V). Comparing these values with the estimated circumference of the effective closed loop (LL) ∼123\sim123 nm (as L=2πℏ/eB0L = \sqrt{2\pi\hbar/eB_0} and B0=1.7B_0 = 1.7 T), it is evident that the transport is not ballistic but diffusive. Therefore, the observed BB-periodic quantum oscillations cannot be attributed solely to the AB effect. Instead, these oscillations are likely a result of AAS interference,

which typically arises from coherent backscattering along time-reversed paths in disordered systems or in networks of multiply connected loops [30, 31]. Moreover, the presence of resistance maxima at B=0B = 0 across all temperatures (Figure 1a) and exponential decay with magnetic field (Figure 5b) further support the interpretation of AAS oscillations.

To verify the influence of magnetic flux on the BB-periodic oscillations, we measured the same sample by rotating it with respect to magnetic field orientation at T=0.1T = 0.1 K, as illustrated in the inset of Figure 3a. In the low-field regime (B<12B < 12 T), the MR decreases with increasing tilt angle from 0° (out-of-plane field orientation) to 90° (in-plane field orientation); however, the MR shows a non-monotonic change in fields higher than 12 T. More interesting is the evolution of oscillations with tilt angle as displayed in Figure 3b. The amplitude of all oscillations decays with increasing angle. The period of BB-periodic oscillations reduces by ∼24%\sim24\% with increasing angle from 0° to 20°, and these oscillations almost vanishes at 40°. For large angles (60°–90°), new oscillations, whose amplitude and position do not dramatically change with field orientation, are resolved. These oscillations systematically follow thermal damping and almost vanish for T>1.2T > 1.2 K (Figure 3c). FFT analysis (Figure 3d) of 1/B1/B dependence of oscillations reveals a single frequency of 10.4 T. The cyclotron mass is estimated from the fitting of temperature dependence of oscillations amplitude with the Lifshitz–Kosevich (L-K) formula (Figure 3e) given below [29].

Impact of B orientation on quantum oscillations

Figure 3. Impact of BB orientation on quantum oscillations (a) RxxR_{xx} and (b) −d2Rxx/dB2-\mathrm{d}^{2}R_{xx}/\mathrm{d}B^{2} at different tilt angles measured at 0.1 K. The inset of (a) is a schematic of the magnetic field orientation, where θ=0∘\theta= 0^\circ and 90∘90^\circ are the angles when the field is perpendicular and parallel to the (001) plane, respectively. (c) Inverse field dependence of quantum oscillations measured at θ=90∘\theta= 90^\circ. (d) FFT spectra of oscillations at different temperatures. (e) Thermal damping of oscillations amplitude fitted with Lifshitz-Kosevich formula for cyclotron mass estimation.

R(T)=R02π2kBmcT/ℏeBsinh⁡(2π2kBmcT/ℏeB)(1)R(T) = R_0 \frac{2\pi^2 k_B m_c T/\hbar eB}{\sinh(2\pi^2 k_B m_c T/\hbar eB)} \tag*{(1)}

The extracted mass mc≈1.9 mem_c \approx1.9\,m_e indicates that these single-frequency oscillations originate from the quantization of 3D Fermi surface of STO [32]. The existence of 3D SdH oscillations suggests that a portion of the electrons reside deeper into the STO, consistent with previous observations at similar interfaces [26, 33]. However, the influence of 3D carriers on the AAS oscillations is not yet well understood. Based on our current understanding, we expect the transport contributions from 2D confined channels at the interface and from 3D bulk-like channels in STO to be largely decoupled. Both in our previous work [26] and in the present study, we find that SdH oscillations associated with 3D carriers are significantly different than those originating from 2D carriers, indicating that when the magnetic field is applied perpendicular to the interface, electrical conduction and the resulting quantum interference effects are dominated by 2D carriers.

Next, to understand the origin of the observed AAS effect in our Hall bar-patterned LSAT/STO, we consider two possible scenarios for the existence of naturally formed closed-loop paths: (1) a 2D network of line dislocations forming a square moiré pattern, resulting from complete lattice relaxation at the LSAT/STO interface [19]; and (2) interconnected domain walls originated below cubic-to-tetragonal phase transition of STO [20, 34, 35, 36]. If the 2DEG is spatially inhomogeneous—as indicated by the gate-tunable zero-field offset observed in RxyR_{xy} (Figure 2b and Figure 2c)—then conduction through interconnected quasi-1D channels can occur in either scenario. In the naturally created network of such closed-loop paths, phase-coherent electrons can traverse time-reversed paths around loops, leading to constructive interference. However, the formation of a 2D moiré pattern is an intrinsic property of the LSAT/STO interface and should lead to a consistent oscillation period across samples grown under similar conditions. In contrast, STO domain walls form randomly, and their configurations are not expected to be identical between samples, leading to variations in oscillation periodicity. To test this, we examined another LSAT/STO device (device-2) grown in same conditions as device-1 and observed oscillations with a period of approximately 3 T (Figure 8), nearly twice that of the device-1. This significant difference indicates that the BB-periodic oscillations in LSAT/STO originate from conduction through randomly arranged and sufficiently interconnected STO domain walls, driven by spatial inhomogeneity. Further evidence for domain walls formation is provided by the temperature dependence of the resistance. Rxx(T)R_{xx}(T) of device-1 measured while cooling from room temperature to 1.5 K (Figure 9a), exhibits two distinct anomalies at ∼100\sim100 K and ∼40\sim40 K. The anomaly near 100 K is widely attributed to the cubic-to-tetragonal structural transition in STO [20, 37, 38]. And, the anomaly around 40 K, which shows pronounced hysteresis between cooling and warming cycles (Figure 9b) and strong gate dependence [20], is attributed to the formation of polar domain walls.

B-periodic oscillations in the longitudinal resistance and Hall resistance

Figure 8. BB-periodic oscillations in the longitudinal resistance (top) and Hall resistance (b) of device-2. The periodicity of oscillations is ≈3.0\approx3.0 T. These measurements were conducted without any electrostatic gating.

Temperature dependence of resistance $R(T)$

Figure 9. Temperature dependence of resistance R(T)R(T) of device-1. (a) R(T)R(T) in the temperature range of 300 - 1.5 K measured while cooling. (b) Comparison of R(T)R(T) measured while cooling and warming in the low temperature regime. The dashed vertical lines depict the anomalies observed in R(T)R(T).

To estimate the phase coherence length (LϕL_{\phi}), we further analyzed the temperature dependence of the BB-periodic oscillations of device-1. After subtracting a monotonic background from the data in Figure 1a, we plot the oscillatory component of the resistance at different temperatures in Figure 4a. As expected, the oscillation amplitude—and correspondingly, the FFT amplitude (Figure 4b)—decreases with increasing temperature, consistent with the suppression of phase coherence with temperature. In a closed loop of circumference LL, the oscillations amplitude decays exponentially with the ratio L/LϕL/L_{\phi} [6]:

A(T)=A0exp⁡(−NLLϕ(T))(2)A(T) = A_0 \exp\left(-N\frac{L}{L_{\phi}(T)}\right) \tag*{(2)}

where Lϕ(T)∝T−pL_{\phi}(T) \propto T^{-p} and NN (= 1, 2....) is the number of harmonics. The temperature dependence of the FFT amplitude for the first two harmonics (inset of Figure 4b) fits well with Eq. 2 for p=1.14p = 1.14, indicating the suppression of quantum interference due to charge fluctuations [39]. Using the fitting parameters and estimated circumference of the effective closed loop (LL) ∼\sim123 nm, we calculate Lϕ=1.84L_{\phi} = 1.84 μ\mum at 0.1 K for the first harmonic, which is twice as large as that for a LAO/STO device with channel lengths nearly 100 times shorter [16]. We show the temperature dependence of LϕL_{\phi} from the first two harmonics in Figure 4c. The extracted exponent p∼1.1p \sim1.1 suggests that decoherence in LSAT/STO closely resembles that in ballistic mesoscopic systems, where Lϕ(T)∝T−1L_{\phi}(T) \propto T^{-1} and dephasing is primarily driven by electron-electron interactions [6, 9].

Assuming a similar temperature dependence of the oscillation amplitude at other VgV_g, and using the oscillation frequencies shown in Figure 2g, we estimate the LϕL_{\phi} for various VgV_g values. LϕL_{\phi} decreases almost linearly with increasing VgV_g (Figure 4d), following the trend in oscillation frequency. It is worth noting that the amplitude of the BB-periodic oscillations also diminishes with increasing VgV_g, and these oscillations vanish completely for Vg>20V_g > 20 V. This behavior can be attributed to the increase in carrier density with VgV_g, which reduces the Fermi wavelength λF\lambda_F. A shorter λF\lambda_F allows more wave modes to contribute with different phases, resulting in partial dephasing and a reduction in oscillation amplitude due to mode averaging. Similar suppression of oscillations amplitude with increasing VgV_g has been reported in GaAs/AlGaAs heterostructures [40] and graphene [9]. The simultaneous reduction in both oscillation frequency and amplitude with increasing VgV_g can also be attributed to a decrease in spatial inhomogeneity. As carriers accumulate at the interface, the charge distribution becomes more uniform, reducing the size of naturally-formed closed-loop paths. Eventually, this evolution will lead to the disappearance of interconnected closed-loop paths at some point and completely vanish BB-periodic oscillations.

The characteristic mesoscopic length scale extracted from the observed oscillations is on the order of 0.1 μ\mum, raising the question of whether it corresponds to the STO domain size and whether AAS oscillations can persist in the presence of spatially varying domains. Although we do not have a direct measurement of ferroelectric domain sizes in LSAT/STO, studies on LAO/STO interfaces report domain widths ranging from 0.1 μ\mum - 10 μ\mum, depending on strain, thermal history, polarization, and carrier density. A direct one-to-one correspondence between the oscillation-derived loop size and the macroscopic domain spacing is therefore not expected. Importantly, AAS oscillations can survive spatial variations in domain size or disorder because they originate from constructive interference between time-reversed electron trajectories, whose phases remain identical under ensemble averaging. Consistent with this picture, BB-periodic oscillations have been observed in multiwalled carbon nanotubes despite a wide distribution of tube diameters ranging from 1 to 20 nm [41, 42, 43]. Moreover, theoretical studies of AAS oscillations in disordered networks and granular conductors show that although the underlying loop-area distribution may be broad, the effective contribution is narrowed by phase-coherence constraints and magnetic-field averaging [44]. Using the empirical relation ΔA/A=B0/Bmax⁡\Delta A/A = B_0/B_{\max}, where Bmax⁡B_{\max} (= 9 T) is the field where the oscillations completely vanish and B0B_0 (= 1.7 T) is the period of oscillations for device-1, we estimate an effective relative loop-area dispersion of ∼0.2\sim0.2, consistent with theoretical expectations for random network systems. Future studies such as direct imaging of domain walls at LSAT/STO interfaces with scanning probe techniques could further test the proposed connection between mesoscopic transport and naturally-formed domain structures and provide a better understanding of the microscopic origin of the robust BB-periodic oscillations.

In conclusion, we have fabricated high-mobility Hall-bar-patterned LSAT/STO devices and performed magnetotransport measurements down to 0.1 K. At low magnetic fields (0–7 T), the devices exhibit a positive MR superimposed with pronounced quantum oscillations that are periodic in magnetic field BB, indicating fully phase-coherent transport. The amplitude of these BB-periodic oscillations decays exponentially with increasing temperature and magnetic field strength, consistent with dephasing of quantum interference. We attribute the BB-periodic oscillations to AAS interference, most likely arising from naturally formed closed-loop paths due to the spatial inhomogeneity, which supports high-mobility conduction along the interconnected SrTiO3_3 domain walls. The observed gate-tunability, combined with the relatively long phase coherence length (∼1.8\sim1.8 μ\mum at 0.1 K) underscores the potential of complex oxide interfaces as a platform for studying quantum interference phenomena and for developing quantum technologies such as mesoscopic interferometers and quantum sensors.

Acknowledgements: K. R. acknowledge support from the National High Magnetic Field Laboratory, supported by the National Science Foundation through NSF/DMR-2128556 and the state of Florida and the US Department of Energy (DoE). This work was performed as part of the DoE BES project "Science of 100 Tesla". This study has been partially supported through the EUR grant NanoX no. ANR-17-EURE-0009 in the framework of the "Programme des Investissements dAvenir". This work at NUS is supported by the Ministry of Education (MOE), SIngapore, under its Tier-2 Academic Research Fund (AcRF) and by the MOE Tier-3 Grant (MOE-MOET32023-0003) ’Quantum Geometric Advantage’.

APPENDICES

Appendix 1 To investigate the temperature range over which the BB-periodic oscillations persist, we measured the device-1 in a helium-4 cryostat at temperatures ranging from 1.5 K to 9.0 K at Vg=3.0V_g = 3.0 V. These measurements were carried out in pulsed magnetic fields of pulse length ∼300\sim300 ms. Figure 5a display Rxx(B)R_{xx}(B) measured at different temperatures. The zero-field RxxR_{xx} from these measurements differs slightly, consistent with changes in electrical conduction commonly observed in STO-based interfaces after thermal cycling. Such changes are often attributed to oxygen vacancy redistribution [37, 45], leading to slight variation in carrier density and mobil-

ity. BB-periodic oscillations remain visible at least up to 9 K (Figure 5a); and the amplitude of these oscillations exhibits an exponential decay with increasing magnetic field strength (Figure 5b).

Appendix 2 To analyse the SdH oscillations observed for θ=0∘\theta= 0^\circ, we subtracted the monotonous background of Rxx(B)R_{xx}(B) data displayed in Figure 1(a). At the lowest temperature 0.1 K, we observed large amplitude low-frequency oscillations superimposed with higher frequency oscillations (Figure 6a). As the temperature increases, the high-frequency oscillations start damping quickly and the low-frequency oscillations survive at least up to ∼1.1\sim1.1 K. By fitting the temperature dependence of oscillations amplitude with L-K formula as shown in the main panel of Figure 6b, we estimated the cyclotron mass of ∼2.1me\sim2.1m_e. Further, analyzing the exponential growth of oscillations with BB (inset of Figure 6b), we extracted the quantum mobility of ∼2530\sim2530 cm2^2V−1^{-1}s−1^{-1}, one order of magnitude smaller than the transport mobility estimated from Drude formula.

Appendix 3 To further illustrate the evolution of quantum oscillations with VgV_g, we replot the data from Figure 2e as a color contour map (Figure 7). In this representation, the blue regions correspond to oscillation minima and the red regions to maxima. This format clearly reveals a shift of the BB-periodic oscillations (AAS oscillations) toward lower magnetic fields and SdH oscillations toward higher magnetic fields with increasing VgV_g.

References

  1. [1]Y. Aharonov and D. Bohm, Significance of electromagnetic potentials in the quantum theory, Physical review 115, 485 (1959).DOI
  2. [2]R. A. Webb, S. Washburn, C. Umbach, and R. Lai- bowitz, Observation of h/e aharonov-bohm oscillations in normal-metal rings, Physical Review Letters 54, 2696 (1985).
  3. [3]B. Al’Tshuler, A. Aronov, and B. Spivak, The aaronov-bohm effect in disordered conductors, Jetp Lett 33, 94 (1981).
  4. [4]D. Y. Sharvin and Y. V. Sharvin, Magnetic flux quantization in a cylindrical film, J. Exp. Theor. Phys. Lett 34 (1981).
  5. [5]P. A. Lee and A. D. Stone, Universal conductance fluctuations in metals, Physical review letters 55, 1622 (1985).
  6. [6]A. Hansen, A. Kristensen, S. Pedersen, C. Sørensen, and P. Lindelof, Mesoscopic decoherence in aharonov-bohm rings, Physical Review B 64, 045327 (2001).DOI
  7. [7]Ö. Gül, N. Demarina, C. Blöamers, T. Rieger, H. Lüth, M. Lepsa, D. Grützmacher, and T. Schäpers, Flux periodic magnetoconductance oscillations in GaAs/InAs core/shell nanowires, Physical Review B 89, 045417 (2014).DOI
  8. [8]S. Russo, J. B. Oostinga, D. Wehenkel, H. B. Heersche, S. S. Sobhani, L. M. Vandersypen, and A. F. Morpurgo, Observation of aharonov-bohm conductance oscillations in a graphene ring, Physical Review B—Condensed Matter and Materials Physics 77, 085413 (2008).DOI
  9. [9]J. Dauber, M. Oellers, F. Venn, A. Epping, K. Watanabe, T. Taniguchi, F. Hassler, and C. Stampfer, Aharonov-bohm oscillations and magnetic focusing in ballistic graphene rings, Physical Review B 96, 205407 (2017).DOI
  10. [10]A. Caviglia, S. Gariglio, N. Reyren, D. Jaccard, T. Schneider, M. Gabay, S. Thiel, G. Hammerl, J. Mannhart, and J.-M. Triscone, Electric field control of the LaAlO3/SrTiO3 interface ground state, Nature 456, 624 (2008).
  11. [11]L. Li, C. Richter, J. Mannhart, and R. C. Ashoori, Coexistence of magnetic order and two-dimensional superconductivity at LaAlO3/SrTiO3 interfaces, Nature physics 7, 762 (2011).DOI
  12. [12]J. A. Bert, B. Kalisky, C. Bell, M. Kim, Y. Nikita, H. Y. Hwang, and K. A. Moler, Direct imaging of the coexistence of ferromagnetism and superconductivity at the LaAlO3/SrTiO3 interface, Nature physics 7, 767 (2011).DOI
  13. [13]A. Brinkman, M. Huijben, M. Van Zalk, J. Huijben, U. Zeitler, J. Maan, W. G. van der Wiel, G. Rijnders, D. H. Blank, and H. Hilgenkamp, Magnetic effects at the interface between non-magnetic oxides, Nature materials 6, 493 (2007).pubmed.ncbi.nlm.nih.gov/17546035
  14. [14]B. Kalisky, J. A. Bert, B. B. Klopfer, C. Bell, H. K. Sato, M. Hosoda, Y. Nikita, H. Y. Hwang, and K. A. Moler, Critical thickness for ferromagnetism in LaAlO3/SrTiO3 heterostructures, Nature communications 3, 922 (2012).DOI
  15. [15]S. Goswami, E. Mulazimoglu, A. M. Monteiro, R. Wölbing, D. Koelle, R. Kleiner, Y. M. Blanter, L. M. Vander-sypen, and A. D. Caviglia, Quantum interference in an interfacial superconductor, Nature nanotechnology 11, 861 (2016).DOI
  16. [16]P. Irvin, H. Lee, J.-W. Lee, M. Briggeman, S. Lu, A. Annadi, G. Cheng, M. Tomczyk, J. Li, M. Huang, et al., Strong aharonov-bohm quantum interference in simply connected LaAlO3/SrTiO3 structures, Physical Review B 100, 161103 (2019).DOI
  17. [17]Z. Huang, K. Han, S. Zeng, M. Motapothula, A. Y. Borisevich, S. Ghosh, W. Lü, C. Li, W. Zhou, Z. Liu, M. Coey, T. Venkatesan, and A. Ariando, The effect of polar fluctuation and lattice mismatch on carrier mobility at oxide interfaces, Nano Letters 16, 2307 (2016).DOI
  18. [18]K. Han, Z. Huang, S. Zeng, M. Yang, C. Li, W. Zhou, X. R. Wang, T. Venkatesan, J. Coey, M. Goiran, et al., Electrical properties and subband occupancy at the (La,Sr)(Al, Ta)O3/SrTiO3 interface, Physical Review Materials 1, 011601 (2017).DOI
  19. [19]M. Burian, B. F. Pedrini, N. Ortiz Hernandez, H. Ueda, C. Vaz, M. Caputo, M. Radovic, and U. Staub, Buried moiré supercells through SrTiO3 nanolayer relaxation, Physical Review Research 3, 013225 (2021).DOI
  20. [20]K. Han, W. Li, M. Xie, S. Zeng, Y. Bai, J. Jiang, L. Xu, P. Yin, P. Chen, Z. Huang, and A. Ariando, Effect of polar domain walls on macroscopic electrical properties at the (La,Sr)(Al,Ta)O3/SrTiO3 interfaces, Applied Physics Letters 127, 191602 (2025).DOI
  21. [21]M. Ben Shalom, A. Ron, A. Palevski, and Y. Dagan, Shubnikov-de haas oscillations in SrTiO3/LaAlO3 interface, Physical review letters 105, 206401 (2010).
  22. [22]A. Fête, S. Gariglio, C. Berthod, D. Li, D. Stornaiuolo, M. Gabay, and J.-M. Triscone, Large modulation of the shubnikov-de haas oscillations by the rashba interaction at the LaAlO3/SrTiO3 interface, New Journal of Physics 16, 112002 (2014).DOI
  23. [23]A. McCollam, S. Wenderich, M. Kruize, V. Guduru, H. Molegraaf, M. Huijben, G. Koster, D. H. Blank, G. Rijnders, A. Brinkman, et al., Quantum oscillations and subband properties of the two-dimensional electron gas at the LaAlO3/SrTiO3 interface, APL materials 2 (2014).arxiv.org/abs/1207.7003
  24. [24]F. Trier, G. E. Prawiroatmodjo, Z. Zhong, D. V. Christensen, M. von Soosten, A. Bhowmik, J. M. G. Lastra, Y. Chen, T. S. Jespersen, and N. Pryds, Quantization of hall resistance at the metallic interface between an oxide insulator and SrTiO3, Physical Review Letters 117, 096804 (2016).DOI
  25. [25]K. Rubi, J. Gosteaue, R. Serra, K. Han, S. Zeng, Z. Huang, B. Warot-Fonrose, R. Arras, E. Snoeck, M. Goiran, and W. Escoffier, Aperiodic quantum oscillations in the two-dimensional electron gas at the LaAlO3/SrTiO3 interface, npj Quantum Materials 5, 1 (2020).DOI
  26. [26]K. Rubi, D. R. Candido, M. Dumen, S. Zeng, A. Ammerlaan, F. Bangma, M. K. Chan, M. Goiran, A. Ariando, S. Chakraverty, et al., Unconventional quantum oscillations and evidence of nonparabolic electronic states in quasi-two-dimensional electron system at complex oxide interfaces, Physical Review Research 6, 043231 (2024).DOI
  27. [27]A. Joshua, S. Pecker, J. Ruhman, E. Altman, and S. Ilani, A universal critical density underlying the physics of electrons at the LaAlO3/SrTiO3 interface, Nature communications 3, 1129 (2012).DOI
  28. [28]J. Biscaras, S. Hurand, C. Feuillet-Palma, A. Rastogi, R. Budhani, N. Reyren, E. Lesne, J. Lesueur, and N. Bergeal, Limit of the electrostatic doping in two-dimensional electron gases of LaXO3 (X= Al, Ti)/SrTiO3, Scientific reports 4, 6788 (2014).
  29. [29]D. Shoenberg, Magnetic oscillations in metals (Cambridge university press, 2009).
  30. [30]C. Umbach, C. Van Haesendonck, R. Laibowitz, S. Washburn, and R. A. Webb, Direct observation of ensemble averaging of the aharonov-bohm effect in normal-metal loops, Physical review letters 56, 386 (1986).pubmed.ncbi.nlm.nih.gov/10033174
  31. [31]C. Naud, G. Faini, and D. Mailly, Aharonov-bohm cages in 2d normal metal networks, Physical Review Letters 86, 5104 (2001).
  32. [32]A. Bhattacharya, B. Skinner, G. Khalsa, and A. V. Suslov, Spatially inhomogeneous electron state deep in the extreme quantum limit of strontium titanate, Nature communications 7, 12974 (2016).
  33. [33]H. Yan, S. Zeng, K. Rubi, G. J. Omar, Z. Zhang, M. Goiran, W. Escoffier, and A. Ariando, Ionic modulation at the laalo3/ktao3 interface for extreme high-mobility two-dimensional electron gas, Advanced Materials Interfaces , 2201633 (2022).DOI
  34. [34]B. Kalisky, E. M. Spanton, H. Noad, J. R. Kirtley, K. C. Nowack, C. Bell, H. K. Sato, M. Hosoda, Y. Xie, Y. Hikita, et al., Locally enhanced conductivity due to the tetragonal domain structure in LaAlO₃/SrTiO₃ heterointerfaces, Nature materials 12, 1091 (2013).DOI
  35. [35]M. Honig, J. A. Sulpizio, J. D. Doria, A. Joshua, E. Zeldov, and S. Ilani, Local electrostatic imaging of striped domain order in LaAlO₃/SrTiO₃, Nature materials 12, 1112 (2013).DOI
  36. [36]H. H. Ma, S. Scharinger, S. Zeng, D. Kohlberger, M. Lange, A. Stöhr, X. R. Wang, T. Venkatesan, R. Kleiner, J. Scott, et al., Local electrical imaging of tetragonal domains and field-induced ferroelectric twin walls in conducting SrTiO₃, Physical review letters 116, 257601 (2016).DOI
  37. [37]M. Z. Minhas, A. Muller, F. Heyroth, H. H. Blaschek, and G. Schmidt, Temperature dependent giant resistance anomaly in LaAlO₃/SrTiO₃ nanostructures, Scientific Reports 7, 5215 (2017).DOI
  38. [38]Y. Kwak, W. Han, J. S. Lee, J. Song, and J. Kim, Hysteretic temperature dependence of resistance controlled by gate voltage in LaAlO₃/SrTiO₃ heterointerface electron system, Scientific Reports 12, 6458 (2022).DOI
  39. [39]G. Seelig and M. Büttiker, Charge-fluctuation-induced dephasing in a gated mesoscopic interferometer, Physical Review B 64, 245313 (2001).DOI
  40. [40]C. Ford, T. Thornton, R. Newbury, M. Pepper, H. Ahmed, D. Peacock, D. Ritchie, J. Frost, and G. Jones, Electrostatically defined heterojunction rings and the aharonov–bohm effect, Applied physics letters 54, 21 (1989).DOI
  41. [41]A. Bachtold, C. Strunk, J.-P. Salvetat, J.-M. Bonard, L. Forró, T. Nussbaumer, and C. Schönenberger, Aharonov–bohm oscillations in carbon nanotubes, Nature 397, 673 (1999).DOI
  42. [42]A. Fujiwara, K. Tomiyama, H. Suematsu, M. Yumura, and K. Uchida, Quantum interference of electrons in multiwall carbon nanotubes, Physical Review B 60, 13492 (1999).DOI
  43. [43]M.-G. Kang, T. Morimoto, N. Aoki, J.-U. Bae, Y. Ochiai, and J. P. Bird, Aharonov–bohm effect in the magnetoresistance of a multiwalled carbon nanotube with tunneling contacts, Physical Review B 77, 113408 (2008).DOI
  44. [44]Y. V. Sharvin, Weak localization and oscillatory magnetoresistance of cylindrical metal films, Physica B: Condensed Matter 126, 288 (1984).DOI
  45. [45]N. J. Goble, R. Akrobetu, H. Zaid, S. Sucharitakul, M.-H. Berger, A. Sehirlioglu, and X. P. A. Gao, Anisotropic electrical resistance in mesoscopic LaAlO₃/SrTiO₃ devices with individual domain walls, Scientific Reports 7, 44361 (2017).DOI

Paper details

vlnc:W_3vt5jt37
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