WorksPhysics and Astronomy
Tunable superconducting diode effect in higher-harmonic InSb nanosheet interferometers
Abstract
Superconducting diodes, characterized by the nonreciprocal supercurrent flow, have gained significant attention for their potential in dissipationless electronics. This study presents a superconducting quantum interference device (SQUID) composed of two Al-InSb nanosheet Josephson junctions. Utilizing prepatterned local backgates, we achieve a gate- and flux-tunable superconducting diode with controllable efficiency in both amplitude and sign. Numerical simulations attribute the diode effect to higher harmonics in the current-phase relation. Crucially, fractional Shapiro step experiments provide direct insights into the evolution of these higher harmonics with flux tuning, showcasing significant enhancements in the second-harmonic signatures of the SQUID near half-integer flux quanta. Furthermore, we investigate the microwave-assisted diode response and experimentally show that the polarity of the diode effect can be switched by the microwave power. These results demonstrate the potential of InSb nanosheet-based hybrid devices as highly tunable elements for use in dissipationless electronics.
The superconducting diode effect (SDE) is a nonreciprocal transport phenomenon, analogous to charge transport in semiconductor junctions, where dissipationless superconducting currents flow in one direction while being blocked or partially blocked in the opposite direction. This unique characteristic positions superconducting diodes as promising elements for applications in superconducting electronics, spintronics, and quantum technologies[1–4]. The SDE has recently attracted worldwide attention and has been the subject of study in various superconducting systems, including Josephson junctions (JJs) and junction-free superconductors[5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18]. The microscopic mechanisms underlying the SDE involve intrinsic material properties, including finite-momentum pairing[19, 20, 21], the magneto-chiral effect[22], and valley polarization[23], as well as extrinsic factors like trapped vortices and self-inductance[15, 24, 25].
For JJs, the current-phase relation (CPR) is typically expressed as a Fourier series, , where is the phase drop across the junction and represents the current component of the th harmonic term. A purely sinusoidal CPR results in equal amplitudes of positive and negative critical currents, thus precluding the SDE. However, high-transparency JJs generally possess a non-sinusoidal CPR, and in such higher-harmonic JJs (), a non-zero phase difference between harmonics can induce SDE, as illustrated by expressions like [23, 26]. Achieving such non-zero typically requires breaking time-reversal and inversion symmetries, which can arise from an interplay of external magnetic field with spin-orbit coupling or a nonuniform potential[6, 10, 27, 28].
Nonreciprocal critical currents in SQUIDs have been recognized since the 1970s[25, 29], but they have recently gained renewed attention due to the involvement of high-order harmonics in CPR[30, 31, 32, 33, 34, 35, 36, 37, 38, 39]. Theoretical proposals suggest that higher-harmonic CPR in SQUIDs with highly transmissive JJs can significantly affect supercurrent nonreciprocity, with predictions of a butterfly pattern diagram to illustrate diode efficiency[37, 38]. These theoretical frameworks set the stage for experimental investigations into SDE in various hybrid systems with SQUID configurations. Recent experiments have focused on quantum well-based hybrid superconductor-semiconductor interferometers, such as those made from Al-InAs and Al-Ge quantum wells, where gate-tunable elements are anticipated to enable parity-protected qubits[30, 31, 32, 33, 34, 35]. Tuning these diode responses with a microwave field is predicted to yield intriguing results, such as achieving an ideal SDE[32, 38] or even switching the polarity of the diode[40, 41], the latter of which has yet to be reported by experiments. These results offer valuable insights for creating a highly tunable superconducting diode.
InSb nanosheet is a two-dimensional (2D) semiconductor nanostructure developed in recent years[42, 43]. With its strong spin-orbit interaction, small effective mass, large Landé -factor[42, 44, 45], and 2D geometry, InSb nanosheets have emerged as a highly promising platform for the development of novel planar hybrid quantum devices[10, 46, 47, 48]. While prior studies have predominantly focused on devices incorporating single JJs, including the intrinsic diode effect in InSb nanosheets [10], research on nanosheet-based interferometers remains limited. The development of such interferometers holds the potential to introduce additional tuning knobs, facilitating the exploration of novel hybrid quantum devices on nanosheets.
In our work, we engineered the SDE in an InSb nanosheet interferometer with higher-harmonic junctions. As opposed to the intrinsic diode in single JJs[10], the SDE in a SQUID does not rely on spin-orbit coupling or in-plane Zeeman fields. Instead, tuning the two asymmetric SQUID arms enables a wider range of control over diode rectification efficiency. In the experiment, we demonstrate effective control of the supercurrent in each junction using two pre-patterned local backgates. This enables the realization of a gate- and flux-tunable superconducting diode with a controllable diode efficiency, both in amplitude (up to ) and direction. We analyze the mechanism and find that higher-order Josephson harmonics in CPR are indispensable for the SDE. We utilize fractional Shapiro steps to monitor the higher harmonic content in flux-tuning experiments. Notably, we observe significant enhancements in the second-harmonic signatures of the SQUID near half-integer flux quanta. Furthermore, we investigate the microwave response and experimentally demonstrate that the SDE can not only be tuned by microwave power, but also be switched. This may open up new possibilities for application with a microwave-controlled diode effect[40]. Numerical calculations qualitatively support all observations.
Figure 1(a) shows a false-color scanning electron microscope image of an Al-InSb interferometer. Two planar JJs (JJ and JJ), each consisting of Al and InSb nanosheets with a gap of about 100 nm, are looped by the Al superconducting wires to form an SQUID (see Supplemental Material for a zoomed-in image of JJ and JJ[49]). The free-standing InSb nanosheets are grown by molecular-beam epitaxy, with a typical thickness of 10 to 100 nm. Details about the material and transport properties have been studied in previous reports[43, 47, 50]. A set of prepatterned metallic (Ti/Au) backgates are fabricated beneath the InSb nanosheets, isolated by a 15 nm HfO dielectric layer. They are used to independently tune the superconductivity of JJ and JJ by applying respective backgate voltages. InSb nanosheets are mechanically transferred by a tip, followed by a standardized aluminum evaporation process. Fabrication details are presented in Supplementary Material[49]. A standard DC measurement is performed in a dilution refrigerator with a base temperature of about 15 mK.
In Figure 1(c) and 1(d), we present the differential resistance of JJ (, 2) as a function of the respective backgate voltage and bias current . When measuring one JJ, a backgate voltage of 0 V is applied to pinch off the other JJ. The critical current of JJ can be tuned from a pinch-off status at V to approximately 15 nA at V, while the critical current of JJ can be tuned from pinch-off at V up to 40 nA at V. A slight decrease in the critical current is observed as continues to increase. This could be related to the distribution of carriers at high . At high values, carriers are pushed away from the superconductor-semiconductor interface due to electrostatic potential effects, thereby affecting the superconducting proximity effect and reducing . Additionally, excessive carrier accumulation at high may enhance electron-electron interactions and scattering, which could also contribute to the suppression of . Similar non-monotonic behavior of with has been reported in other JJ systems[51, 52]. Compared to JJ, the stochastic switching current in JJ exhibits a more pronounced response to thermal and charge fluctuations when tuning , as illustrated in Figure 1(d).

Figure 1. Device and characterization. (a) False-color scanning electron microscopy image of an Al-InSb SQUID and the cross-section of the Al-InSb nanosheet JJ that makes up the SQUID. The brown regular polygon regions are two free-standing InSb nanosheets. (b) Differential resistance of the SQUID as a function of bias current and perpendicular magnetic field at V and V. The cyan and orange arrows illustrate the extreme positions of the switching and retrapping valleys respectively. (c) and (d) Differential resistance as a function of bias current and backgate voltage for JJ and JJ. When one junction is measured, the other is completely pinched off by the corresponding backgate voltage.
Figure 1(b) shows the differential resistance of the SQUID as a function of perpendicular magnetic field and bias current at a gate configuration of V and V, where is swept from negative to positive in the ‘+’ direction. Three main observations are as follows: (1) The SQUID oscillation is periodic with a period of . This periodicity corresponds to an added external flux quantum , which gives the area of the inner loop . This value is slightly larger than the actual inner area of our SQUID (). This discrepancy could be due to the magnetic flux focusing effect or London penetration. (2) A set of fringes emerges in the oscillation valley. The discrete fringes observed may arise from Andreev bound states (ABSs) within the induced superconducting gap or from multiple Andreev reflections. As the critical current at the valley decreases, the finite voltage step at the superconducting transition drops below the bias voltage necessary to access the characteristic energy levels within the subgap, thereby facilitating the emergence of these fringes. (3) An asymmetry between the switching and retrapping branches is observed, representing an indirect signature of SDE. Overheating, however, can lead to an artificial SDE: once the junction switches to the resistive state, it fails to return to the superconducting state due to thermal effects until decreases below the smaller retrapping current, , resulting in . To exclude this artificial SDE caused by overheating, a standard approach in diode experiments involves comparing the switching current in the ‘+’ direction () with that in the opposite ‘−’ direction (), as illustrated in Figure 2(a).
Figure 2. Gate- and flux-tunable SDE. (a) The measured voltage as a function of and at a gate configuration of . is swept from zero to positive values (indicated by the red arrow) in the red branch, and subsequently from zero to negative values (indicated by the blue arrow) in the blue branch, allowing the extraction of switching currents and for two opposite sweep directions. (b) and , extracted from panel (a), plotted as a function of . The blue dotted line corresponds to the diode efficiency. (c) and , simulated using two different parameters, as a function of . Parameters in the left panel consider only a realistic pH, while parameters in the right include pH and the first two harmonics (. Parameters , obtained at a gate configuration of , are used for both plots. (d) and (e) Efficiency as a function of and , respectively, obtained from the experiment and theoretical model. Here, is fixed at V. In panel (d), the pixel distribution along the axis is non-uniform due to unequal spacing in sampling.
To quantify SDE, we study and obtained in two opposite sweep directions. The red branch in Figure 2(a) corresponds to the current sweeping from zero to a positive value in the ‘+’ direction, while the blue branch corresponds to sweeping from zero to a negative value in the ‘−’ direction. An asymmetric configuration of the SQUID is established in the measurement using and , with nA and nA (see Supplemental Material for the determination method of and in the SQUID[49]). and as functions of the external magnetic flux are extracted and shown as green and orange curves, respectively, in Figure 2(b). A clear asymmetry in the bias current direction is observed, indicating the presence of SDE. The polarity of the SDE changes around , as reflected in the diode efficiency , defined as
Next, we attempt to understand the origin of the SDE in this experiment. Recognizing that loop inductance can result in an SDE in SQUIDs, we model an inductive SQUID following Ref. 53. The total current passing through the SQUID is the sum of the currents through JJ and JJ:
Here and represent the phase drops across JJ and JJ, respectively. The loop inductance introduces additional flux into the system, with its magnitude directly proportional to the circulating current in the loop:
where indicates a flux induced by the external magnetic field and represents the flux quantum. We disregard inductance asymmetry between branches as they are negligible in our SQUID setup. Given a specific , or can be determined by identifying the maximum or minimum concerning .
To analyze the impact of inductance on the SDE, we first assume a standard sinusoidal CPR for JJ and JJ. i.e., , . A realistic in SQUID is the sum of the geometric and kinetic inductances, here estimated to be 100 pH (see Supplemental Material for details[49]). The left panel in Figure 2(c) shows a numerical fit with pH. We find that, simply including a realistic does not adequately capture the asymmetry in the switching current observed in the experiment. To address this asymmetry, should be at least an order of magnitude larger than the realistic value. Thus, the contribution of higher-order harmonics to the SDE is significant in our SQUID, even if the higher harmonic content is relatively small. For simplicity, we assume that JJ and JJ exhibit the same harmonic content, leading to the CPR:
Here, represents the th harmonic content, and is a normalized coefficient ensuring and . To simplify the analysis, we focus on the first two harmonics in this study. The right panel in Figure 2(c) illustrates a numerical fit with , and pH. Incorporating contributions from the first two harmonics, with a small , effectively captures crucial experimental characteristics.
We now return to quantify the SDE and discuss its dependence on gate voltage and external flux. Our measurements demonstrate that the flux-dependent SDE can be tuned with varying gate voltages (see Supplementary Material for details[49]). To systematically quantify the SDE, we extract as a function of and external flux at V, as depicted in Figure 2(d). The key observation is that both the magnitude and the sign of can be tuned by and . When V, for and for . Conversely, when V, the scenario is reversed. The experiment yields a maximum of around 10%. Numerical simulation results in Figure 2(e) show as a function of and . In the simulation, we assume the CPR in JJ () remains in the form of , with , , and as a gate-dependent switching current in JJ.
The experiment and numerical simulation highlight a critical value of V, where changes sign, corresponding to a symmetric SQUID case (). The absence of asymmetry between and leads to a decrease in . However, in practical experiments, especially when and are nearly equal, fluctuations in diode polarity are more likely to be triggered by the stochastic nature of , as seen at V in Figure 2(d).
The agreement between the experiment results and the theoretical model highlights the critical role of the high-order harmonic term in CPR. While the CPR of single junctions is not measured due to the absence of a reference junction with a significantly larger critical current, the nonsinusoidal CPR in single JJs inevitably contributes to the emergence of high-order harmonics in the SQUID. These harmonics can be effectively investigated through the AC Josephson effect. In an SQUID without high-order harmonics, quantized voltage plateaus (Shapiro steps) appear at , with being an integer, the Planck constant and the microwave frequency. When the SQUID includes a high-order harmonic term, such as with , fractional quantized voltage plateaus emerge at . Thus, by monitoring the evolution of these fractional Shapiro steps in flux-tuning experiments, one can gain direct insights into the behavior of the relevant high-order harmonic term in the SQUID.
In the following, we study the microwave response of our system. The microwave is coupled to the sample via an antenna positioned approximately 1 mm away. Due to frequency-dependent coupling and the lack of impedance matching between the antenna and the sample, accurately quantifying the microwave power reaching the sample is challenging. Therefore, as is common in similar experiments, we report only the nominal microwave power output from the generator at room temperature.
Figure 3(a) shows differential resistance as a function of microwave power and bias current at different . The microwave driving frequency is 6 GHz. The minima in the curves correspond to the Shapiro steps in the - characteristic. The low-order integer Shapiro step regions are marked in Figure 3(a). Three main observations are as follows: (1) In addition to the integer Shapiro steps, half-integer steps are also observed at zero , appearing as dips between adjacent integer Shapiro step lobes. This indicates the presence of higher-order harmonics in our SQUID. (2) As is tuned from 0 to , the first half-integer Shapiro step region appears between the first Shapiro step lobes of index 0 and . At the same time, the zero-voltage region shrinks significantly. These behaviors can also be directly identified by comparing the - curves at and 0.5 under the same microwave power. As shown in Figure 3(b), when fixing dBm, a weakly developed quantized voltage step at emerges at , while a sharp voltage jump occurs instead at .
Figure 3. Enhanced second-harmonic signatures near half-integer flux quanta. (a) Differential resistance as a function of and when is tuned from 0 to . The microwave frequency is 6 GHz, and the gate configuration is V and V. The black arrow on the axis indicates the sweeping direction of the dc current bias. The top brown and darkgreen arrows point to the power values of and -8 dBm. (b) Voltage-current curves at and , as extracted from panel (a) at dBm. Voltage is normalized to . Half-integer Shapiro steps appear at , highlighted by the black arrows. (c-d) Differential resistance versus and at dBm and -8 dBm. The integer and half-integer Shapiro step regions have been labeled, with half-integer steps primarily observed near half-integer flux quanta. (e) Calculated differential resistance map as a function of the dc and ac current bias at . Here, a RSJ model is used, disregarding Joule overheating and nonlinear resistance effects. At (cyan arrow) and 1.53 (orange arrow), the differential resistance versus and is plotted in panel (f) and (g), respectively. Parameters used for all simulation plots are , , , GHz.
(3) Higher-order half-integer Shapiro step regions are expanded by tuning to . For instance, in the area around (-10 dBm, 55 nA) in Figure 3(a), which corresponds to the Shapiro step of index , there is a notable vertical expansion of the Shapiro step region as is tuned from 0 to .
Studying the -dependent Shapiro steps at fixed provides a comprehensive understanding of the effect of external flux on higher-order harmonics. Figure 3(c) and 3(d) show the cases of and -8 dBm, respectively. We observe, first, that in Figure 3(c) and 3(d), the Shapiro step islands, characterized by an index of 0.5, become apparent at half-integer values of . These islands gradually disappear once deviates from half-integer values. Second, we observe a periodic oscillation of the Shapiro step regions with indices and as a function of , with a period of . These half-integer Shapiro step regions reach maxima at half-integer values of and minima at integer values of .
To analyze the dynamics of the SQUID in the microwave field, we consider a resistively shunted junction (RSJ) model with a current bias, disregarding Joule overheating and nonlinear resistance effects. The time-dependent phase difference and the voltage drop are given by
Where and are the amplitudes of the dc and ac amplitudes, is the microwave frequency, is a shunt resistance in parallel with the interferometer and is the total CPR of the interferometer, which, as mentioned above, considers only the first two harmonics. By solving this equation numerically, we find that, despite the simplicity of the model, the results qualitatively capture the main features of this work. Figure 3(e) presents the simulated Shapiro pattern at and at a drive frequency of 6 GHz. At and 1.53, indicated by the cyan and orange arrows, the calculated differential resistances versus and are depicted in Figure 3(f) and 3(g). These results agree well with the main features observed in Figure 3(c) and 3(d).
Next, we discuss the reasons behind the evolution of the half-integer Shapiro steps. Given that the CPR in JJ () follows , the CPR of the SQUID at and thus can be
given by
By tuning , the ratio between the second and first harmonics increases from at to at . The enhanced second-harmonic content influences the behavior of the half-integer Shapiro steps near . In the symmetric limit where , the first harmonic can be completely suppressed at , with the second harmonic () contributes to the supercurrent. This regime corresponds to the transport of pairs of Cooper pairs. However, the relatively small magnitude of the second harmonics makes the direct observation of the second harmonic-induced supercurrent component particularly challenging. In addition, we have noticed that in Figure 3, under the influence of DC bias, magnetic flux, and microwave power, we did not observe the pronounced asymmetric Shapiro step formation or the displaced Shapiro steps phenomenon that is typically seen in JJ array structures[54] or asymmetric SQUID ratchets[38, 55]. The absence of such clear asymmetry is related to the relatively low content of higher-order harmonic terms in the CPR of the SQUID (see Supplemental Material for a simulation with a higher ).
In the final section of this study, we delve into the diode response by tuning microwave powers. The theoretical investigations have revealed that as power levels increase, the efficiency can potentially reach 100%[38, 41, 56]. This ideal diode has been experimentally reported[18, 32]. A further theoretical study recently revealed that tuning may also switch the diode polarity[40, 41], but this has not yet been experimentally reported.
In Figure 4(a), we measure the voltage drop across the SQUID under microwave irradiation with a drive frequency of GHz. As the microwave power increases, the switching current is gradually suppressed. At a constant , we measure the curves under different by sweeping the current in two opposite directions, thus obtaining and as functions of . Consequently, we determine by utilizing and and plot them against in Figure 4(b). Despite the presence of noise in the signal, it is clear that the diode polarity, in addition to being periodically influenced by (or magnetic flux ), is also governed by . A switch in diode polarity occurs when is around -14 dBm. The reproducibility of this result can be partially validated by observing similar behavior over two consecutive magnetic flux quantum periods. At the same time, we notice that the fluctuations in become more pronounced at elevated . This can be partly attributed to the increased stochasticity of the switching current resulting from thermal fluctuations associated with higher microwave power. Indeed, it is more likely due to the suppression of and at elevated power . When the switching current is smaller, identical fluctuation levels in and lead to more pronounced fluctuations in , given that
Figure 4. Tuning diode nonreciprocity with the microwave power at 8 GHz. (a) The voltage drop across the SQUID as a function of and at zero . (b) The experimental diode efficiency as a function of at different . is recalibrated for each line by subtracting offsets. (c) The numerically simulated efficiency as a function of with varying . (d) Efficiency as a function of at a fixed , with in the top panel and in the bottom panel. The red curves in panel (b) and (d), fitted to , are obtained using a Savitzky-Golay filter.
To further rule out the possibility of experimental artifacts, we directly sweep while setting to and to measure the diode polarity switching, as illustrated in Figure 4(d). The diode polarity switching can be observed around -15 dBm. Figure. Figure 4(c) presents a numerically simulated result using the RSJ model with previously obtained parameters. are selected at values before reaching the first node of the zeroth Shapiro step lobe. The simulated result qualitatively agrees with the experimental observations. As is tuned from to , the sign of changes accordingly. The sign-reversal of the diode efficiency can be intuitively understood as follows: In the fast-driving limit and considering the zero-voltage step, the amplitudes of the Josephson harmonics in the CPR are modulated by the zeroth Bessel function of the first kind[41]. Specifically, , . As increases, both and can change signs, leading to a sign change of the diode efficiency. The microwave-controlled switching of diode polarity offers new possibilities for distinguishing -periodic topological Josephson junctions in asymmetric SQUIDs[40].
In conclusion, we have developed a gate- and flux-tunable superconducting diode utilizing a SQUID composed of aluminum and InSb nanosheets. Theoretical analyses indicate that higher harmonics dominate the SDE rather than . Fractional Shapiro step experiments with flux tuning, showing significant enhancements in the second-harmonic signatures of the SQUID near half-integer flux quanta, as confirmed by numerical simulations. Furthermore, by integrating numerical calculations with microwave experiments, we show that the diode polarity can be switched by tuning the microwave power in SQUIDs. These findings present an exciting perspective for exploring highly tunable superconducting diodes on nanosheet-based interferometers and realizing a parity-protected qubit. They may also have potential applications in discerning -periodic Josephson junctions.
Author contributions
H.Q.X. supervised the project. X.W., H.S., and S.Y. fabricated devices. X.W. and J.-Y.W. performed measurements. X.W. did the simulation. D.P. and J.Z. developed the nanosheet material. X.W., P.Z., and H.Q.X. analyzed the data and wrote the manuscript with inputs from all the authors.
Acknowledgements
This work is supported by the NSFC (Grant Nos. 92165208, 12004039, 11874071, 92365103, 12374480, 12374459, 61974138 and 92065106). D.P. acknowledges the support from Youth Innovation Promotion Association, Chinese Academy of Sciences (Nos. 2017156 and Y20210143).
Data analysis and data availability
The raw data and the analysis files are available at https://doi.org/10.5281/zenodo.13268561
References
- [5]E. Ando, Y. Miyasaka, T. Li, J. Ishizuka, T. Arakawa, Y. Shiota, T. Moriyama, Y. Yanase, and T. Ono, Observation of superconducting diode effect, Nature 584, 373 (2020).DOI
- [6]G. Mazur, N. van Loo, D. van Driel, J.-Y. Wang, G. Badawy, S. Gazibegovic, E. Bakkers, and L. Kouwenhoven, The gate-tunable josephson diode, arXiv preprint arXiv:2211.14283 (2022).arxiv.org/abs/2211.14283
- [7]Y. M. Itahashi, T. Ideue, Y. Saito, S. Shimizu, T. Ouchi, T. Nojima, and Y. Iwasa, Nonreciprocal transport in gate-induced polar superconductors rtiо3, Sci. Adv. 6, eaay9120 (2020).
- [8]J.-X. Lin, P. Siriviboon, H. D. Scammell, S. Liu, D. Rhodes, K. Watanabe, T. Taniguchi, J. Hone, M. S. Scheurer, and J. Li, Zero-field superconducting diode effect in small-twist-angle trilayer graphene, Nat. Phys. 18, 1221 (2022).
- [9]C. Baumgartner, L. Fuchs, A. Costa, S. Reinhardt, S. Gronin, G. C. Gardner, T. Lindemann, M. J. Manfra, P. E. Faria Junior, D. Kochan, J. Fabian, N. Paradiso, and C. Strunk, Supercurrent rectification and magnetochiral effects in symmetric josephson junctions, Nat. Nanotechnol. 17, 39 (2022).
- [10]B. Turini, S. Zaman, M. Carrega, A. Iorio, E. Strambini, F. Giazotto, V. Zannier, L. Sorba, and S. Heun, Josephson diode effect in high-mobility insb nanoflags, Nano Lett. 22, 8502 (2022).
- [11]H. Wu, Y. Wang, Y. Xu, P. K. Sivakumar, C. Pasco, U. Filippozzi, S. S. Parkin, Y.-J. Zeng, T. McQueen, and M. N. Ali, The field-free josephson diode in a van der waals heterostructure, Nature 604, 653 (2022).DOI
- [12]F. K. de Vries, E. Portolés, G. Zheng, T. Taniguchi, K. Watanabe, T. Ihn, K. Ensslin, and P. Rickhaus, Gate-defined josephson junctions in magic-angle twisted bilayer graphene, Nat. Nanotechnol. 16, 760 (2021).
- [13]J. Díez-Mérida, A. Díez-Carlón, S. Yang, Y.-M. Xie, X.-J. Gao, J. Senior, K. Watanabe, T. Taniguchi, X. Lu, A. P. Higginbotham, K. T. Law, and D. K. Efetov, Symmetry-broken josephson junctions and superconducting diodes in magic-angle twisted bilayer graphene, Nat. Commun. 14, 2396 (2023).
- [14]M. Trahms, L. Melischek, J. F. Steiner, B. Mahendru, I. Tamir, N. Bogdanoff, O. Peters, G. Reecht, C. B. Winkelmann, F. von Oppen, and K. J. Franke, Diode effect in josephson junctions with a single magnetic atom, Nature 615, 628 (2023).
- [15]A. Sundaresh, J. I. Väyrynen, Y. Lyanda-Geller, and L. P. Rokhinson, Diamagnetic mechanism of critical current non-reciprocity in multilayered superconductors, Nat. Commun. 14, 1628 (2023).
- [16]S. Matsuo, T. Imoto, T. Yokoyama, Y. Sato, T. Lindemann, S. Gronin, G. C. Gardner, M. J. Manfra, and S. Tarucha, Josephson diode effect derived from short-range coherent coupling, Nat. Phys. 19, 1636 (2023).
- [17]N. Loftizadeh, W. F. Schiela, B. Pekerten, P. Yu, B. H. Elfeky, W. M. Strickland, A. Matos-Abiague, and J. Shabani, Superconducting diode effect sign change in epitaxial al-inas josephson junctions, Commun. Phys. 7, 120 (2024).
- [18]H. Su, J.-Y. Wang, H. Gao, Y. Luo, S. Yan, X. Wu, G. Li, J. Shen, L. Lu, D. Pan, J. Zhao, P. Zhang, and H. Q. Xu, Microwave-assisted unidirectional superconductivity in al-inas nanowire-al junctions under magnetic fields, Phys. Rev. Lett. 133, 087001 (2024).
- [19]B. Pal, A. Chakraborty, P. K. Sivakumar, M. Davydova, A. K. Gopi, A. K. Pandeya, J. A. Krieger, Y. Zhang, M. Date, S. Ju, N. Yuan, N. B. M. Schröter, L. Fu, and S. S. P. Parkin, Josephson diode effect from cooper pair momentum in a topological semimetal, Nat. Phys. 18, 1228 (2022).
- [20]N. F. Yuan and L. Fu, Supercurrent diode effect and finite-momentum superconductors, Proc. Natl. Acad. Sci. 119, e2119548119 (2022).arxiv.org/abs/2311.11087
- [21]M. Davydova, S. Prembabu, and L. Fu, Universal josephson diode effect, Sci. Adv. 8, eabq0309 (2022).DOI
- [22]H. F. Legg, D. Loss, and J. Klinovaja, Superconducting diode effect due to magnetochiral anisotropy in topological insulators and rashba nanowires, Phys. Rev. B 106, 104501 (2022).
- [23]J.-X. Hu, Z.-T. Sun, Y.-M. Xie, and K. T. Law, Josephson diode effect induced by valley polarization in twisted bilayer graphene, Phys. Rev. Lett. 130, 266003 (2023).
- [24]D. Suri, A. Kamra, T. N. Meier, M. Kronseder, W. Belzig, C. H. Back, and C. Strunk, Non-reciprocity of vortex-limited critical current in conventional superconducting micro-bridges, Appl. Phys. Lett. 121, 102601 (2022).
- [25]T. A. Fulton, L. N. Dunkleberger, and R. C. Dynes, Quantum interference properties of double josephson junctions, Phys. Rev. B 6, 855 (1972).
- [26]B. Zhang, Z. Li, V. Aguilar, P. Zhang, M. Pendharkar, C. Dempsey, J. S. Lee, S. D. Harrington, S. Tan, J. S. Meyer, M. Houzet, C. J. Palmstrom, and S. M. Frolov, Evidence of φ0-josephson junction from skewed diffraction patterns in sn-insb nanowires (2024), arXiv:2212.00199 [cond-mat.mes-hall].
- [27]A. Rasmussen, J. Danon, H. Suominen, F. Nichele, M. Kjaergaard, and K. Flensberg, Effects of spin-orbit coupling and spatial symmetries on the josephson current in sns junctions, Phys. Rev. B 93, 155406 (2016).
- [28]T. Yokoyama, M. Eto, and Y. V. Nazarov, Anomalous josephson effect induced by spin-orbit interaction and zeeman effect in semiconductor nanowires, Phys. Rev. B 89, 195407 (2014).
- [29]T. Fulton and R. Dynes, Current-phase relations in superconducting bridges, Phys. Rev. Lett. 25, 794 (1970).
- [30]C. Ciaccia, R. Haller, A. C. Drachmann, T. Lindemann, M. J. Manfra, C. Schrade, and C. Schönenberger, Gate-tunable josephson diode in proximitized inas supercurrent interferometers, Phys. Rev. Res. 5, 033131 (2023).DOI
- [31]C. Ciaccia, R. Haller, A. C. Drachmann, T. Lindemann, M. J. Manfra, C. Schrade, and C. Schönenberger, Charge-4e supercurrent in a two-dimensional inas-al superconductor-semiconductor heterostructure, Commun. Phys. 7, 1 (2024).
- [32]M. Valentini, O. Sagi, L. Baghmanyan, T. de Gijsel, J. Jung, S. Calcaterra, A. Ballabio, J. Aguilera Servin, K. Aggarwal, M. Janik, and et al., Parity-conserving cooper-pair transport and ideal superconducting diode in planar germanium, Nat. Commun. 15, 169 (2024).
- [33]M. Gupta, G. V. Graziano, M. Pendharkar, J. T. Dong, C. P. Dempsey, C. Palmström, and V. S. Pribiag, Gate-tunable superconducting diode effect in a three-terminal josephson device, Nat. commun. 14, 3078 (2023).DOI
- [34]T. W. Larsen, M. E. Gershenson, L. Casparis, A. Kringhøj, N. J. Pearson, R. P. McNeil, F. Kuemmeth, P. Krogstrup, K. D. Petersson, and C. M. Marcus, Parity-protected superconductor-semiconductor qubit, Phys. Rev. Lett. 125, 056801 (2020).
- [35]A. Leblanc, T. Tangchinchai, Z. S. Momtaz, E. Kiyooka, J.-M. Hartmann, G. T. Fernandez-Bada, Z. Scherübl, B. Brun, V. Schmitt, S. Zilbmann, R. Maurand, E. Dumeur, S. De Franceschi, and F. m. c. Lefloch, From nonreciprocal to charge-4e supercurrent in ge-based josephson devices with tunable harmonic content, Phys. Rev. Res. 6, 033281 (2024).DOI
- [36]P. Zhang, A. Zarassi, L. Jarjat, V. V. de Sande, M. Pendharkar, J. S. Lee, C. P. Dempsey, A. P. McFadden, S. D. Harrington, J. T. Dong, H. Wu, A. H. Chen, M. Hocevar, C. J. Palmstrøm, and S. M. Frolov, Large second-order Josephson effect in planar superconductor-semiconductor junctions, SciPost Phys. 16, 030 (2024).DOI
- [37]Y. V. Fominov and D. S. Mikhailov, Asymmetric higher-harmonic squid as a josephson diode, Phys. Rev. B 106, 134514 (2022).DOI
- [38]R. S. Souto, M. Leijnse, and C. Schrade, Josephson diode effect in supercurrent interferometers, Phys. Rev. Lett. 129, 267702 (2022).DOI
- [39]S. Reinhardt, T. Ascherl, A. Costa, J. Berger, S. Gronin, G. C. Gardner, T. Lindemann, M. J. Manfra, J. Fabian, D. Kochan, et al., Link between supercurrent diode and anomalous josephson effect revealed by gate-controlled interferometry, Nat. Commun. 15, 4413 (2024).DOI
- [40]J. J. Cuozzo, W. Pan, J. Shabani, and E. Rossi, Microwave-tunable diode effect in asymmetric squids with topological josephson junctions, Phys. Rev. Res. 6, 023011 (2024).DOI
- [41]R. Seon, R. Souto, M. Leijnse, C. Schrade, M. Valentini, G. Katsaros, and J. Danon, Tuning the josephson diode response with an ac current, Phys. Rev. Res. 6, L022002 (2024).DOI
- [42]M. De La Mata, R. Leturcq, S. R. Plissard, C. Rolland, C. Magén, J. Arbiol, and P. Caroff, Twin-induced insb nanowires: A convenient high mobility quantum system, Nano Lett. 16, 825 (2016).DOI
- [43]D. Pan, D. X. Fan, N. Kang, J. H. Zhi, X. Z. Yu, H. Q. Xu, and J. H. Zhao, Free-standing two-dimensional single-crystalline insb nanosheets, Nano Lett. 16, 834 (2016).DOI
- [44]H. A. Nilsson, P. Caroff, C. Thelander, M. Larsson, J. B. Wagner, L.-E. Wernersson, L. Samuelson, and H. X. Griant, level-dependent g factors in insb nanowire quantum dots, Nano Lett. 9, 3151 (2009).DOI
- [45]Z. Lei, C. A. Lehner, E. Cheah, C. Mittag, M. Karalic, W. Wegscheider, K. Ensslin, and T. Ihn, Gate-defined quantum point contact in an insb two-dimensional electron gas, Phys. Rev. Res. 3, 023042 (2021).DOI
- [46]F. K. de Vries, M. L. Sol, S. Gazibegovic, R. L. M. o. h. Veld, S. C. Balk, D. Car, E. P. A. M. Bakkers, L. P. Kouwenhoven, and J. Shen, Crossed andreev reflection in insb flake josephson junctions, Phys. Rev. Res. 1, 032031 (2019).DOI
- [47]N. Kang, D. Fan, J. Zhi, D. Pan, S. Li, C. Wang, J. Guo, J. Zhao, and H. Xu, Two-dimensional quantum transport in free-standing insb nanosheets, Nano Lett. 19, 561 (2019).DOI
- [48]A. Iorio, A. Crippa, B. Turini, S. Salimian, M. Carrega, L. Chirolli, V. Zannier, L. Sorba, E. Strambini, F. Giazotto, et al., Half-integer shapiro steps in highly transmissive insb nanoflag josephson junctions, Phys. Rev. Res. 5, 033015 (2023).DOI
- [49]See Supplemental Material at URL–will-be-inserted-by-publisher for the loop inductance estimation, device fabrication, numerical simulation, and additional data for the device.
- [50]X. Wu, H. Su, C. Zeng, J.-Y. Wang, S. Yan, D. Pan, J. Zhao, P. Zhang, and H. Q. Xu, Superconducting switching jump induced missing first shapiro step in al-insb nanosheet josephson junctions (2024), arXiv:2403.07370 [cond-mat.supr-con].DOI
- [51]S. Yan, H. Su, D. Pan, W. Li, Z. Lyu, M. Chen, X. Wu, L. Lu, J. Zhao, J.-Y. Wang, et al., Supercurrent, multiple andreev reflections and shapiro steps in inas nanosheet josephson junctions, Nano Lett. 23, 6497 (2023).DOI
- [52]C.-Z. Li, A.-Q. Wang, C. Li, W.-Z. Zheng, A. Brinkman, D.-P. Yu, and Z.-M. Liao, Fermi-arc supercurrent oscillations in dirac semimetal josephson junctions, Nat. Commun. 11, 1150 (2020).DOI
- [54]B. Chesca, M. Gaifullin, D. John, J. Cox, S. Savel’ev, and C. Mellor, The influence of magnetic vortices motion on the inverse ac josephson effect in asymmetric arrays, Appl. Phys. Lett. 121, 162601 (2022).DOI
- [55]I. Zapata, R. Bartussek, F. Sols, and P. Hänggi, Voltage rectification by a squid ratchet, Phys. Rev. Lett. 77, 2292 (1996).DOI
- [56]J. Clarke and A. I. Braginski, The SQUID handbook: Applications of SQUIDs and SQUID systems (John Wiley & Sons, 2006).DOI
