# Fileset

[sciadv.adp3725.pdf](https://mdr.nims.go.jp/filesets/f31d647c-f293-4b0a-accf-8bbcddc1a3ac/download)

## Creator

Giorgio Di Battista, Kin Chung Fong, Andrés Díez-Carlón, [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), Dmitri K. Efetov

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Infrared single-photon detection with superconducting magic-angle twisted bilayer graphene](https://mdr.nims.go.jp/datasets/da212f75-9806-4af9-aee7-7b5373485e87)

## Fulltext

Infrared single-photon detection with superconducting magic-angle twisted bilayer grapheneDi Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e1 of 7C O N D E N S E D  M AT T E R  P H Y S I C SInfrared single-photon detection with superconducting magic-angle twisted bilayer grapheneGiorgio Di Battista1, Kin Chung Fong2,3, Andrés Díez-Carlón1, Kenji Watanabe4,Takashi Taniguchi5, Dmitri K. Efetov1,6*The moiré superconductor magic-angle twisted bilayer graphene (MATBG) shows exceptional properties, with an electron (hole) ensemble of only ~1011 carriers per square centimeter, which is five orders of magnitude lower than traditional superconductors (SCs). This results in an ultralow electronic heat capacity and a large kinetic inductance of this truly two-dimensional SC, providing record-breaking parameters for quantum sens-ing applications, specifically thermal sensing and single-photon detection. To fully exploit these unique super-conducting properties for quantum sensing, here, we demonstrate a proof-of-principle experiment to detect single near-infrared photons by voltage biasing an MATBG device near its superconducting phase transition. We observe complete destruction of the SC state upon absorption of a single infrared photon even in a 16–square micrometer device, showcasing exceptional sensitivity. Our work offers insights into the MATBG-photon interaction and demonstrates pathways to use moiré superconductors as an exciting platform for revolutionary quantum devices and sensors.INTRODUCTIONSuperconducting materials are at the heart of advanced technolo-gies, as they are central active elements for modern quantum com-puting, quantum sensing, and quantum metrology applications. In particular, nanopatterned superconducting thin films have gained attention for ultrasensitive photodetection (1–3), as these combine a low heat capacity and a sharp superconducting transition. When a photon is absorbed in such a device, it breaks Cooper pairs and generates quasiparticles above the superconducting gap, thereby introducing a change in impedance. Harnessing this mechanism, superconductor-based detectors, such as transition-edge sensors (4, 5), superconducting nanowires (6–8), hot electron bolometers (9), kinetic inductance detectors (10), and Josephson junctions (11–13), are among the best photodetectors for applications de-manding high sensitivity, e.g., communication, radio astronomy (14), quantum network (15), and spectroscopy (16).Two-dimensional superconductors offer a unique approach to single-photon detection (SPD), due to their reduced electronic heat capacity and electron-phonon coupling, leading to a large temperature rise of the electron ensemble upon absorption of single photons (17–21). It has been recently found that the flat electronic band structure produced by stacking two layers of graphene twisted at a “magic” angle leads to a unique superconducting phase (22, 23). The discovery has now expanded into an entire family of graphene-based superconductors (24–27), not only sparking intense inves-tigations to understand the fundamental physics of their alleged unconventional superconducting states but also prompting explo-ration of their potential applications (28–30). Specifically, moiré superconductor magic-angle twisted bilayer graphene (MATBG) can be a promising material for kinetic inductance detectors (10). When a single photon is absorbed in an MATBG that is embed-ded in a resonator, its kinetic inductance will increase, causing a shift of the resonance frequency, δf. This shift is proportional to the ratio between the density of the generated quasiparticles (δnqp) to Cooper pairs (ns): δf ~ δnqp/ns (10) and would be large for MATBG because we expect a low ns as the intrinsic moiré superlattice of MATBG has a record-low carrier density of n ~ 1011 cm−2, which is ~5 orders of magnitude lower than conventional superconduc-tors (Fig. 1B). In this material, even a minute amount of quasi-particles generated by a single low-energy photon can induce a substantial change in the kinetic inductance, opening a promis-ing avenue to extend SPD across a broader spectral range.In this study, we take the first step to develop an SPD based on superconducting MATBG and perform a proof-of-principle exper-iment to demonstrate the capability of detecting single photons. We illuminate the device at millikelvin temperatures with a highly at-tenuated 1550-nm laser source and monitor the induced photovolt-age (Vph), as shown in the schematic drawing of Fig. 1A.RESULTSThe optical image (inset of Fig. 1F) shows a typical device. The van der Waals stack consists of two graphene sheets rotated at a global twist angle of ~1.1° encapsulated into insulating hexagonal boron nitride (hBN) layers. The metallic graphite gate underneath the het-erostructure is used to electrostatically tune the carrier concentra-tion in the MATBG by applying an external gate voltage. Figure 1D shows the four-terminal longitudinal resistance Rxx of device A (θ = 1.04° ± 0.02) as a function of the moiré filling factor ν (filling of electrons per moiré unit cell) for temperatures ranging from T = 50 mK up to T = 6 K. At electrostatic doping levels corresponding to the half-filling of the moiré unit cell (ν = −2), we observe an insulat-ing state flanked by a pronounced superconducting dome (22). In Fig. 1E, the measurement of Rxx versus T at the optimal doping of 1Fakultät für Physik, Ludwig-Maximilians-Universität, Schellingstrasse 4, München 80799, Germany. 2Quantum Engineering and Computing Group, Raytheon BBN Technologies, Cambridge, MA 02138, USA. 3Department of Physics, Harvard Uni-versity, Cambridge, MA 02138, USA. 4Research Center for Functional Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan. 5International Center for Materials Nanoarchitectonics, National Institute for Mate-rials Science, 1-1 Namiki, Tsukuba 305-0044, Japan. 6Munich Center for Quantum Science and Technology (MCQST), München, Germany.*Corresponding author. Email: dmitri.​efetov@​lmu.​deCopyright © 2024 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY). Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024mailto:dmitri.​efetov@​lmu.​dehttp://crossmark.crossref.org/dialog/?doi=10.1126%2Fsciadv.adp3725&domain=pdf&date_stamp=2024-09-18Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e2 of 7ν = −2.45 reveals a superconducting transition with a normal state resistance of ~10 kilohm and a critical temperature of Tc ~ 2.8 K, calculated as 50% of its normal state resistance (see the Supplemen-tary Materials for a complete transport characterization).Figure 1F plots the I-V characteristic of the superconducting state of device A measured in a four-terminal current-biased scheme at T = 35 mK. Here, we observe a clear hysteretic behavior with re-spect to the sweeping direction of the bias current Ibias, character-ized by ΔI = Ic − Ir ~ 15 nA, where Ic and Ir are the switching and retrapping current, respectively (see the Supplementary Materials). The hysteresis loops are ubiquitous in MATBG (31), potentially due to a current-induced self-heating hotspot (32, 33) when the MATBG is in the normal state.Photoresponse measurementsWe can bias our device near the normal-superconductor transition to enable SPD. When the photon is absorbed, it breaks Cooper pairs and produces a voltage output. To prevent the detector to “latch” permanently into a stable resistive state where it no longer detects photons, we implement a self-reset circuitry (Fig. 1C). The circuit is constituted by a voltage divider with load resistor R2 < < Rres + RMATBG. Here, Rres is a residual resistance (arising from the contact resistance and the metallic leads), and RMATBG is the four-terminal resistance of the device active region, which is sketched as a variable resistor. The voltage bias scheme maintains a constant voltage across the source and drain contacts of the device (Vbias). In this way, the increase in resistance induced by the switching of the MATBG detector in the normal state diverts part of the cur-rent into the load resistor R2, reducing the current flowing in the detector and, consequently, the Joule heating, analogously to an electrothermal feedback (34, 35). Once the current is reduced, the detector returns to the superconducting state. As illustrated in Fig. 1C, the voltage probes in the four-terminal scheme are con-nected to a room-temperature low-noise amplifier, and the output of which is fed to an oscilloscope or an analog-to-digital converter to measure the voltage over time induced by the photons. When the MATBG transitions into the resistive state upon photon ab-sorption, we register a spike in Vph, the detector resets itself, and we can measure the statistics of counts as a function of bias volt-age, laser power, and temperature.To perform the photoresponse measurements, we mounted the MATBG device in a dilution refrigerator and provided optical exci-tation with a 1550-nm laser diode. The beam was collimated in the sample space (~4-mm spot diameter) allowing illumination of the entire device area. The incident laser power was then controlled using a programmable optical attenuator (see Materials and Meth-ods). In our experimental setup, both bias and readout leads were heavily filtered to ensure millikelvin electron temperature at the phMATBGVbiasλ = 1550 nm LNAR1R2RresMATBGCryoV0VbiasVphRxx(kilohm)Rxx(kilohm)T (K)Tc 2.8 KT = 50 mKT = 6 K5500 1-2 -1-4 -3 2 3 401010-110010110210310151520 25 500003 05300.51204-2-40-500IbiasdirectionV)Vm(Ibias (nA)Ibias (nA)V (mV)ACBEFDd (nm)n (cm-2)10111011001012 1013 1014 1015 1016 1017Deposited films Exfoliated Non-moiré MoiréInterfaceCompoundsThin-film TwTT o-ddiimensionalGraphene-basedMATBGMATTGBLGTLGMAT4GMAT5G LAO/STOFeSe/STOLSCO/LCO*NbSe2ZrNClWTe2BSCCO*BSCCO1LMoS2WSi* NbN*MoSi*NbTiN*MgB2*TiNNb*Al**YBCOFig. 1. Superconducting MATBG as an ultrasensitive material for SPD. (A) The incident near-infrared photon breaks Cooper pairs generating a photovoltage output, Vph. (B) Logarithmic plot of film thickness d versus carrier density n for various superconductors (19, 24–28, 30, 46–50). MATTG, magic-angle twisted trilayer graphene; MAT4G, magic-angle twisted four-layer graphene; MAT5G, magic-angle twisted five-layer graphene; BLG, Bernal bilayer graphene; TLG, rhombohedral trilayer graphene. The materials previously used for photodetection applications are marked with an asterisk (3, 6, 18–20). (C) Simplified circuit diagram used to measure MATBG’s photore-sponse. The near-infrared photons, incident on the voltage-biased MATBG (R1 = 1 megohm, R2 = 1 kilohm, and Rres = 53 kilohm for device A), induce voltage spikes in the detector (sketched as a variable resistor) that are recorded using an oscilloscope or an analog-to-digital converter. LNA, low-noise amplifier. (D) Longitudinal resistance Rxx of device A (θ = 1.04°) as a function of the filling factor ν for successive temperatures T ranging from 50 mK to 6 K. A pronounced superconducting state is observed for −3 < ν < −2. (E) Rxx versus T at the optimal doping of ν = −2.45. (F) I-V curve measured at the optimal doping, displaying a hysteretic behavior with respect to the sweep-ing direction of the bias current, highlighted in the top inset. In the bottom inset, the optical image of the MATBG device. In white dashed box, the measured area is A ~ 16 μm2. Scale bar, 3 μm.Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e3 of 7sample stage. The constrained electrical bandwidth available in the experiment imposes limitations on the maximum detector count rate and the speed of the reset circuitry but still allows to properly studying the statistics of the photoinduced counts. We extensively describe the optoelectronic setup used in our experiment and the method used to register the counts in the Supplementary Materials.Figure 2A illustrates examples of the photovoltage traces Vph(t), measured over time across the MATBG detector when exposed to the laser beam radiation in the configuration described in Fig. 1C. We observe voltage spikes, emerging as we increase the incident laser power, which we attribute to photoinduced switching events from the superconducting to the normal state. To confirm the origin of the voltage spikes, which are the “clicks” of our detector, we investi-gate their stochastic nature by producing histograms of counts with 1-s bins and extracting the mean (μhist) and variance (σ2hist) of the sampling distribution (36). As reported in the inset of Fig. 2B, for all the histograms, the mean equals the variance, as prescribed by a Poisson process. We further demonstrate the agreement with this statistic by plotting the Poisson distribution on top of the histo-grams, with the extracted μhist and σ2hist (solid lines). The excellent agreement between the experimentally registered counts and the statistical model confirms that our observation is compatible with the photon shot noise generated by the highly attenuated continu-ous wave (CW) laser source.In addition, we examine the average click height as a function of the bias voltage, Vbias. These results are overlaid on the I-V curve (top inset in Fig. 2A), which was measured in the configuration de-scribed in Fig. 1C. We find that the generated photovoltage closely matches the voltage in the normal state across all explored bias volt-ages: Vph(Vbias) ≈ V(Vbias). This observation indicates that the inci-dent photons induce a complete transition of the MATBG detector from its superconducting state to the normal state.Single-photon sensitivity by superconducting MATBGTo investigate the observed photoresponse, in Fig. 3A, we compare the photon count rate (PCR) (counts recorded per second) for different bias voltages (Vbias) without light (empty dots) and with an excitation wavelength of λ = 1550 nm for different laser powers (filled dots). When the detector operates at a bias voltage far from the transition (Vc), the PCR is orders of magnitude higher upon illumination than in the dark, while as Vbias approaches Vc, a sudden increase in false-positive (dark) counts occurs, ultimately domi-nating the detector’s response. We fit the PCR versus Vbias curve under illumination with a sigmoid function (solid line in Fig. 3A, bottom) and observe that the experimental data exhibit a tendency to saturation at Vbias ~ 0.997 Vc. These saturations are intrinsic to the process of SPD, rather than extrinsic, e.g., speed of the mea-surement circuitry (see the Supplementary Materials), and resem-ble the photon counts in superconducting nanowire SPDs (SNSPDs) (7). In SNSPDs, the saturation of the PCR as a function of current bias indicates that the internal detector efficiency (37), without coupling, reaches unity. Conversely, in our experiment, the PCR curve does not entirely saturate, implying that the intrinsic effi-ciency attained is not 100%.We can demonstrate that the registered counts are triggered by single near-infrared photons. For this purpose, we explore how the count rate evolves as a function of the CW laser power over sev-eral orders of magnitude at two different bias points. To provide a quantitative description of the light-induced count rate, we esti-mate the power density incident on the MATBG (PL) in the ap-proximation of a Gaussian beam (38). From PL, we can calibrate the number of incident photons per square micrometer in a time window τ as 〈Nphoton〉 = τ · PL/hν, where hν = 1.28 × 10−19 J is the energy of a single photon at λ = 1550 nm. Choosing τ = 5 ms, which is close to the typical detector recovery time, a laser power density of PL = 10 aW/μm2 corresponds to 〈Nphoton〉 = 0.4 photons incident/μm2 in a time window of 5 ms (see the Supplementary Materials). Under illumination with a weak coherent light source (36), the probability of detecting m photons in a detection time window reduces to ~〈Nphoton〉m /m!. Figure 3B shows the PCR as a function of 〈Nphoton〉. The measured detection probability increas-es linearly with 〈Nphoton〉 over >3 orders of magnitude with an offset 0.37 15 58 146 µm2aWµm2aWµm2aWµm2aWCounts per 1-s binytili baborP50 10 15 20 2500.100.300.10.200.51µhist (s-1)2 hist (s-1)10-110-1101101Vph (mV)t (s)0.98 1 1.02V)Vm( 0.50.1Vbias /VcClick height 200.10.50.10.50.10.50.10.54 6 8 10A BFig. 2. Statistics of the light-induced clicks. (A) Raw photovoltage time traces, Vph(t), measured at increasing laser powers for λ = 1550 nm. Top right inset: Average click height measured as a function of Vbias. The click heights are overlaid on the I-V curve, measured in the configuration described in Fig. 1C. (B) Histograms of counts in 1-s bins for the same laser powers in (A) measured over ~103-s time window. The inset shows that the extracted variance of counts, σ2hist, equals to its mean μhist. The agree-ment with the Poisson statistic is confirmed by plotting the Poisson distribution on top of the histograms, with the extracted μhist and σ2hist (solid lines).Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e4 of 7due to the dark counts, demonstrating single-photon sensitivity of the MATBG superconducting detector (1). As reported for other SPDs (39), we observe that the count rate deviates from linearity at low photon fluxes, when it enters the noise level defined by the dark counts and at high photon fluxes when it saturates because of the limited bandwidth of measurement circuitry.The trace measured at 0.995 Vc shows the same overall behavior as the one measured at 0.989 Vc, with a higher detection probability and dark count rate due to the increase in the intrinsic quantum efficiency as we approach Vc. In the Supplementary Materials, we show several raw photovoltage time traces at different Vbias and PL values from which we extract the PCR reported in Fig. 3 and detail the method used to register the counts in the MATBG detector. For completeness, we also demonstrate single-photon sensitivity under pulsed light excitation (see the Supplementary Materials).Detector performance at higher temperaturesTo provide further insights on the photodetection mechanism in MATBG, in Fig. 4A, we present the PCR versus Vbias at six different temperatures ranging from 35 to 800 mK, both with and without laser excitation (filled and empty dots, respectively). The PCRs with illumination are consistent with the sigmoid function and exhibit a tendency to saturation for Vbias ~ 0.997 Vc. Using the linear scaling of the PCR with laser power (fig. S15), we confirm the SPD from our MATBG device up to ~0.7 K. The single-photon PCR eventually vanishes when temperature rises up to 0.8 K, at which the dark count dominates the PCR. Figure 4B plots the SPD efficiency as a function of Vbias and the dark count rate (PCR without illumination) at various temperatures. Here, the efficiency is defined as the ratio of counts detected per second to photons incident per second in the area (A ~ 16 μm2) between the two voltage probes (white dashed box in the optical image of Fig. 1F). The dark count rate (right-hand side of the y axis) exhibits two distinct Vbias dependence above and below Vbias = 0.998 Vc. When Vbias > 0.998 Vc, a sharp increase in dark counts occurs. This justifies the abrupt rise of PCR under illumi-nation when Vbias ~ Vc. When Vbias < 0.998 Vc, the dark counts rise gradually because of background photons coupling through the optical fiber connected at the room-temperature optical port (40). The detection efficiency on the left-hand side of the y axis is at maximum and gradually decreases as the temperature rises, akin to observations in other SPDs (37, 40). To further investigate this trend, we extract the efficiency at three different Vbias values from the sigmoid fit of Fig. 4A and plot them against temperature in Fig. 4C. The efficiency decreases as temperature rises. We attribute this to the increase in the thermal conductance. Elevated temperatures enhance heat transfer out of the electrons (41), reducing the probability of latching into the resistive state by a self-heating effect. This argument is supported by thermal transport measurements in the superconducting state on the same system, which report a rapid increase in thermal conductivity within the range 35 mK < T < 800 mK (30). Figure 4D plots the trade-off between SPD efficiency and dark count rate at various temperatures to determine the optimal operating condition of our MATBG detector. We observe that the most favorable SPD performance is achieved in the plateau region, where the efficiency approaches its maximum value while maintaining a low dark count rate (11).DISCUSSIONIn addition to the demonstration of SPD, our experiment offers in-sight to the MATBG superconductivity via its interaction with pho-tons. Since the incident photon energy (~0.8 eV) greatly exceeds the flat bands’ width (~10 meV) (42) and the superconducting gap’s size (~1 meV) (30, 43), we can approximate MATBG’s absorption to be the same as bilayer graphene. Using the transfer matrix formalism (see the Supplementary Materials), we estimate ~5.3%. With the measured PCR at the saturation plateau in Fig. 4B, we estimate the internal efficiency of our SPD as ~10−3/0.053 to 0.019. Two factors can limit the internal efficiency in our setup. First, the effective area 10-310-210-11001010.985 0.99 0.995 1051015 Vc 0.989 aW/µm2aW/µm2 aW/µm2aW/µm2Vc 0.995 Dark1229 18373Vbias Vc/PCR (Hz)PCR (Hz)10-2 100 102 10410-410-310-210-110010110210-1 101 103A BPL ( )PCR (Hz)in 1 s/ µm2µm20.995 0.989 Vc Vc aWFig. 3. Single-photon sensitivity by superconducting MATBG. (A) Top: PCR as a function of voltage bias Vbias for four different laser powers (filled dots) and in the dark (empty dots). Bottom: PCR versus Vbias for PL = 73 aW/μm2 on a linear scale. The orange line is a fit with a sigmoid function. The PCR shows a tendency to saturation at ~0.997 Vc. The vertical dashed lines are the bias points at which we performed the PCR versus PL measurements reported in (B). (B) PCR versus average incident photon number 〈Nphoton〉 in a 1-s time window per square micrometer for two different bias points (Vbias = 0.995 Vc and Vbias = 0.989 Vc). On the top x axis, the corresponding in-cident CW power density PL = 〈Nphoton〉 ∙ hν/τ. The solid lines are linear fits (with an offset due to dark counts), showing that the detection probability evolves linearly with 〈Nphoton〉. The gray dashed line depicts a quadratic power dependence.Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e5 of 7of the MATBG contributing to the photoresponse could be much smaller than the entire area of the device. This argument would agree with the measurements of twist angle inhomogeneity by local probes techniques on MATBG (44, 45). Twisted moiré materials are charac-terized by an intrinsic disorder due to the local variation of twist an-gle within the same sample. Because of the relaxation of the lattice structure, the atoms between the two twisted layers of graphene may not align to the same global twist angle, resulting in a narrower su-perconducting area in MATBG. If we assume an internal efficiency of ~1 at the saturation plateau, then we can estimate a lower limit for the effective superconducting area contributing to the photoresponse as Aeff ~ 0.019 and A ≈ 0.3 μm2. If the superconducting channel fully percolates between the two voltage probes, which are spaced about 3 μm apart, then we estimate the channel’s width to be approximately 100 nm. This scenario would be similar to the SNSPDs in which the absorbed photon generates a resistive domain of quasiparticles to produce a readout signal (6). Unlike SNSPDs, however, the hotspot in MATBG would expand with self-sustained Joule heating, leading to a complete breakdown of superconductivity across the whole su-perconducting path, as observed in the inset of Fig. 2A. Hence, heat dissipation, which reduces the latching probability, could be the sec-ond factor limiting the internal efficiency, as supported by the temperature-dependent data in Fig. 4C. Future applications may explore SPD readout mechanisms that do not involve a complete switching of the entire MATBG device into the normal state and could use more sensitive probes to monitor the light-induced chang-es in voltage or kinetic inductance (10).In conclusion, our experimental work has successfully demon-strated that the superconducting state of MATBG can be used to detect single near-infrared photons. This result strongly motivates further investigation to extend single-photon capability to lower en-ergies using MATBG and other low-carrier density graphene-based superconductors (24–27). Pursuing this route necessitates further research effort to understand the intricate interplay between inci-dent photons and these alleged unconventional superconducting phases. Our investigation has contributed important insights into the physical process underlying the observed MATBG’s photores-ponse. These insights will play a pivotal role in the development of theoretical models and in the design of innovative quantum devices that exploit the unique characteristics of these materials, ultimately advancing the field of quantum technology.MATERIALS AND METHODSDevice fabricationThe MATBG devices were fabricated using the “cut-and-stack” tech-nique. A stamp made of propylene carbonate and polydimethylsi-loxane was prepared and mounted on a glass slide. The stamp was used to pick up the top layer of hBN. The hBN layer was then used to pick up the two halves of graphene, which had been precut using an atomic force microscopy tip. The graphene halves were carefully rotated to a target twist angle of 1.1°. To complete the heterostruc-ture, the entire stack was fully encapsulated with a bottom hBN layer. A graphite layer was added at the bottom of the stack, serving 1010-710-610-510-410-310-310-210-1100101102203000.9850.8 K0.7 K0.6 K0.3 K0.035 K0.45 K0.99 0.995 10.985 0.99 0.995 11020010501050101550101550Darkλ = 1550 nmPCR (Hz)Vbias/VcVbias/VcBC DA0 0.05 0.10.993 Vc0.995 Vc0.997 Vc000.20.20.40.40.60.60.810.8Detection efficiency 10-3T (K)00.20.40.60.81Detection efficiency 10-3Detection efficiencyDark count rate (Hz)Dark count rate (Hz)0.035 K0.3 K0.45 K 0.8 K0.7 K0.6 K0.035 K0.3 K0.45 K 0.8 K0.7 K0.6 KFig. 4. Detector performance at higher temperatures. (A) PCR as a function of voltage bias Vbias and temperature T upon illumination (filled dots) and in the dark (empty dots). The continuous lines are fit with the sigmoid function. (B) Filled markers indicate detection efficiency versus Vbias at different temperatures. Empty markers indicate dark count rate versus Vbias at different temperatures. The detection efficiency is defined as the ratio of counts detected per second to photons incident per sec-ond in the area (A ~ 16 μm2) between the two voltage probes. (C) Detection efficiency versus T for three different bias points extracted from the sigmoidal fit of (A). (D) Trade-off between detection efficiency and dark count rate for different temperatures.Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e6 of 7as a local backgate for the device. The full stack was then deposited onto a Si/SiO2 chip and etched into a Hall bar geometry. Last, edge contacts made of Cr/Au (5/50 nm) were evaporated onto the device to establish electrical connections.Photoresponse measurementsTo perform the photoresponse measurements, we placed the device on the cold finger of a dilution refrigerator (BlueFors-SD250), housed in a gold-coated oxygen-free copper box. The dilution refrigerator (base temperature of 35 mK) was optimized for low-frequency trans-port of low-carrier density two-dimensional superconducting mate-rials. A two-stage RC low-pass filter was mounted at the 1-K still plate of the dilution refrigerator combined with an additional radio frequency filter mounted at the mixing chamber stage to ensure millikelvin electron temperature and reject high-frequency noise (fig. S6). The overall bandwidth of the readout was <1 kHz (fig. S9). In the SPD experiment described in Fig. 1C, the bias voltage was applied at the source contact with a voltage generator (Keithley 2400) in series with a 1/1000 voltage divider (R1 = 1 megohm and R2 = 1 kilohm for device A). The voltage probes were connected to a room-temperature 1-MHz-bandwidth low-noise amplifier (SR-560). A room-temperature low-pass filter with a sharp cutoff at ~1 to 10 kHz was used for rejecting white noise outside of the readout bandwidth. The amplified signal was fed to a sampling oscilloscope with variable bandwidth up to 600 MHz (UHF-Scope Zurich Instrument) or a 100-kHz-bandwidth analog-to-digital converter (UHF-Aux In Zurich Instrument). The optical excitation was provided by a 1550-nm laser diode (Taiko PDL M1). The light was fed into the dilution refrigerator to the MATBG detector, through a single-mode optical fiber coupled with a collimator mounted few centimeters on top of the sample space, allowing illumination of the entire device area of ~4-mm spot diameter (fig. S7). To control the incident laser power, a programmable optical attenuator (JGR OA5 l) was used, enabling precise adjustment over several orders of magnitude. Extensive de-scription and schematics of the optoelectronic setup used in the ex-periment are available in the Supplementary Materials.Transport measurementsThe longitudinal resistance Rxx was measured using standard low-frequency lock-in techniques (Stanford Research SR860). To con-trol the carrier density, a voltage was applied to the graphite metallic gate using a Keithley 2400 voltage source connected in series with a 100-megohm resistor. For measuring the current-biased I-V curves in the four-terminal configuration, the bias current was supplied at the source contact using a voltage generator (Keithley 2400) con-nected in series with a bias resistor of 10 megohm. The voltage probes were connected to a room-temperature low-noise preampli-fier (SR-560), and the output signal was measured using a digital multimeter (Keithley 2700).Supplementary MaterialsThis PDF file includes:Supplementary TextFigs. S1 to S20ReferencesREFERENCES AND NOTES  1.  G. N. Gol’tsman, O. Okunev, G. Chulkova, A. Lipatov, A. Semenov, K. Smirnov, B. Voronov, A. Dzardanov, C. Williams, R. Sobolewski, Picosecond superconducting single-photon optical detector. Appl. Phys. Lett. 79, 705–707 (2001).  2.  A. D. Semenov, G. N. Gol’tsman, A. A. Korneev, Quantum detection by current carrying superconducting film. Physica C 351, 349–356 (2001).  3.  D. V. Morozov, A. Casaburi, R. H. Hadfield, Superconducting photon detectors. Contemp. Phys. 62, 69–91 (2021).  4.  B. Cabrera, R. M. Clarke, P. Colling, A. J. Miller, S. Nam, R. W. Romani, Detection of single infrared, optical, and ultraviolet photons using superconducting transition edge sensors. Appl. Phys. Lett. 73, 735–737 (1998).  5.  K. D. Irwin, G. C. Hilton, Transition-edge sensors. Topics Appl. Phys. 99, 63–149 (2005).  6. I . Holzman, Y. Ivry, Superconducting nanowires for single-photon detection: Progress, challenges, and opportunities. Adv. Quant. Technol. 2, 1800058 (2019).  7.  F. Marsili, V. B. Verma, J. A. Stern, S. Harrington, A. E. Lita, T. Gerrits, I. Vayshenker, B. Baek, M. D. Shaw, R. P. Mirin, S. W. Nam, Detecting single infrared photons with 93% system efficiency. Nat. Photonics 7, 210–214 (2013).  8. V . B. Verma, B. Korzh, A. B. Walter, A. E. Lita, R. M. Briggs, M. Colangelo, Y. Zhai,  E. E. Wollman, A. D. Beyer, J. P. Allmaras, H. Vora, D. Zhu, E. Schmidt, A. G. Kozorezov,  K. K. Berggren, R. P. Mirin, S. W. Nam, M. D. Shaw, Single-photon detection in the mid-infrared up to 10 μm wavelength using tungsten silicide superconducting nanowire detectors. APL Photonics 6, 056101 (2021).  9.  B. S. Karasik, S. V. Pereverzev, A. Soibel, D. F. Santavicca, D. E. Prober, D. Olaya,  M. E. Gershenson, Energy-resolved detection of single infrared photons with λ = 8 μm using a superconducting microbolometer. Appl. Phys. Lett. 101, 052601 (2012).  10.  P. K. Day, H. G. LeDuc, B. A. Mazin, A. Vayonakis, J. Zmuidzinas, A broadband superconducting detector suitable for use in large arrays. Nature 425, 817–821 (2003).  11. E . D. Walsh, D. K. Efetov, G.-H. Lee, M. Heuck, J. Crossno, T. A. Ohki, P. Kim, D. Englund,  K. C. Fong, Graphene-based Josephson-junction single-photon detector. Phys. Rev. Appl. 8, 024022 (2017).  12.  G. H. Lee, D. K. Efetov, W. Jung, L. Ranzani, E. D. Walsh, T. A. Ohki, T. Taniguchi, K. Watanabe, P. Kim, D. Englund, K. C. Fong, Graphene-based Josephson junction microwave bolometer. Nature 586, 42–46 (2020).  13. E . D. Walsh, W. Jung, G. H. Lee, D. K. Efetov, B.-I. Wu, K.-F. Huang, T. A. Ohki, T. Taniguchi,  K. Watanabe, P. Kim, D. Englund, K. C. Fong, Josephson junction infrared single-photon detector. Science 372, 409–412 (2021).  14.  D. J. Benford, S. H. Mosley, Cryogenic detectors for infrared astronomy: The single aperture far-infrared (SAFIR) observatory. Nucl. Instrum. Methods Phys. Res. Sect. A 520, 379–383 (2004).  15. E . Knill, R. Laflamme, G. J. Milburn, A scheme for efficient quantum computation with linear optics. Nature 409, 46–52 (2001).  16.  J. A. Lau, V. B. Verma, D. Schwarzer, A. M. Wodtke, Superconducting single-photon detectors in the mid-infrared for physical chemistry and spectroscopy. Chem. Soc. Rev. 921–941 (2023).  17.  P. Seifert, J. R. D. Retamal, R. L. Merino, H. H. Sheinfux, J. N. Moore, M. A. Aamir,  T. Taniguchi, K. Watanabe, K. Kadowaki, M. Artiglia, M. Romagnoli, D. K. Efetov, A high-Tc van der Waals superconductor based photodetector with ultra-high responsivity and nanosecond relaxation time. 2D Mater. 8, 035053 (2021).  18.  R. L. Merino, P. Seifert, J. Duran Retamal, R. Mech, T. Taniguchi, K. Watanabe, K. Kadowaki, R. H. Hadfield, D. K. Efetov, Two-dimensional cuprate nanodetector with single telecom photon sensitivity at T = 20 K. 2D Mater. 10, 021001 (2023).  19. I . Charaev, D. A. Bandurin, A. T. Bollinger, I. Y. Phinney, I. Drozdov, M. Colangelo,  B. A. Butters, T. Taniguchi, K. Watanabe, X. He, O. Medeiros, I. Božović, P. Jarillo-Herrero,  K. K. Berggren, Single-photon detection using high-temperature superconductors.  Nat. Nanotechnol. 18, 343–349 (2023).  20.  G. J. Orchin, D. De Fazio, A. Di Bernardo, M. Hamer, D. Yoon, A. R. Cadore, I. Goykhman,  K. Watanabe, T. Taniguchi, J. W. A. Robinson, R. V. Gorbachev, A. C. Ferrari, R. H. Hadfield, Niobium diselenide superconducting photodetectors. Appl. Phys. Lett. 114, 251103 (2019).  21.  R. Kokkoniemi, J.-P. Girard, D. Hazra, A. Laitinen, J. Govenius, R. E. Lake, I. Sallinen,  V. Vesterinen, M. Partanen, K. Y. Tan, P. Hakonen, M. Möttönen, Bolometer operating at the threshold for circuit quantum electrodynamics. Nature 586, 47–51 (2020).  22.  Y. Cao, V. Fatemi, S. Fang, K. Watanabe, T. Taniguchi, E. Kaxiras, P. Jarillo-Herrero, Unconventional superconductivity in magic-angle graphene superlattices. Nature 556, 43–50 (2018).  23.  R. Bistritzer, A. H. MacDonald, Moiré bands in twisted double-layer graphene. Proc. Natl. Acad. Sci. U.S.A. 108, 12233–12237 (2011).  24.  J. M. Park, Y. Cao, K. Watanabe, T. Taniguchi, P. Jarillo-Herrero, Tunable strongly coupled superconductivity in magic-angle twisted trilayer graphene. Nature 590, 249–255 (2021).  25.  J. M. Park, Y. Cao, L.-Q. Xia, S. Sun, K. Watanabe, T. Taniguchi, P. Jarillo-Herrero, Robust superconductivity in magic-angle multilayer graphene family. Nat. Mater. 21, 877–883 (2022).  26. H . Zhou, T. Xie, T. Taniguchi, K. Watanabe, A. F. Young, Superconductivity in rhombohedral trilayer graphene. Nature 598, 434–438 (2021).  27. H . Zhou, L. Holleis, Y. Saito, L. Cohen, W. Huynh, C. L. Patterson, F. Yang, T. Taniguchi,  K. Watanabe, A. F. Young, Isospin magnetism and spin-polarized superconductivity in Bernal bilayer graphene. Science 375, 774–778 (2022).Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024Di Battista et al., Sci. Adv. 10, eadp3725 (2024)     18 September 2024S c i e n c e  A d v a n c e s  |  R e s e ar  c h  A r t i c l e7 of 7  28.  P. Seifert, X. Lu, P. Stepanov, J. R. D. Retamal, J. N. Moore, K. C. Fong, A. Principi, D. K. Efetov, Magic-angle bilayer graphene nano-calorimeters: Towards broadband, energy-resolving single photon detection. Nano Lett. 20, 3459–3464 (2020).  29.  F. J. Gonzalez, M. S. Lodge, M. Ishigami, R. A. Klemm, A. Rathod, K. L. Lina, A. C. Bowman,  F. Hernandez, C. J. Fredricksen, C. Cariker, R. E. Peale, Antenna-coupled graphene Josephson-junction terahertz detector. MRS Adv. 8, 148–151 (2023).  30.  G. Di Battista, P. Seifert, K. Watanabe, T. Taniguchi, K. C. Fong, A. Principi, D. K. Efetov, Revealing the thermal properties of superconducting magic-angle twisted bilayer graphene. Nano Lett. 22, 6465–6470 (2022).  31.  Y. Cao, D. Rodan-Legrain, J. M. Park, N. F. Q. Yuan, K. Watanabe, T. Taniguchi,  R. M. Fernandes, L. Fu, P. Jarillo-Herrero, Nematicity and competing orders in superconducting magic-angle graphene. Science 372, 264–271 (2021).  32.  W. J. Skocpol, M. R. Beasley, M. Tinkham, Self-heating hotspots in superconducting thin-film microbridges. J. Appl. Phys. 45, 4054–4066 (1974).  33.  M. Tinkham, J. U. Free, C. N. Lau, N. Markovic, Hysteretic I − V curves of superconducting nanowires. Phys. Rev. B 68, 134515 (2003).  34.  A. J. Kerman, J. K. W. Yang, R. J. Molnar, E. A. Dauler, K. K. Berggren, Electrothermal feedback in superconducting nanowire single-photon detectors. Phys. Rev. B 79, 100509 (2009).  35.  K. D. Irwin, An application of electrothermal feedback for high resolution cryogenic particle detection. Appl. Phys. Lett. 66, 1998 (1995).  36.  R. Loudon, Quantum Theory of Light (Oxford University Press, ed. 3, 2000).  37.  B. Baek, A. E. Lita, V. B. Verma, S. W. Nam, Superconducting a-WxSi1−x nanowire single-photon detector with saturated internal quantum efficiency from visible to  1850 nm. Appl. Phys. Lett. 98, 251105 (2011).  38.  O. Svelto, Principles of Lasers (Springer, ed. 5, 2012).  39.  R. H. Hadfield, Single-photon detectors for optical quantum information applications. Nat. Photonics 3, 696–705 (2009).  40. T . Yamashita, S. Miki, W. Qiu, M. Fujiwara, M. Sasaki, Z. Wang, Temperature dependent performances of superconducting nanowire single-photon detectors in an ultralow-temperature region. Appl. Phys. Express 3, 102502 (2010).  41.  F. C. Wellstood, C. Urbina, J. Clarke, Hot-electron effects in metals. Phys. Rev. B 49, 5942–5955 (1994).  42.  Y. Cao, V. Fatemi, A. Demir, S. Fang, S. L. Tomarken, J. Y. Luo, J. D. Sanchez-Yamagishi,  K. Watanabe, T. Taniguchi, E. Kaxiras, R. C. Ashoori, P. Jarillo-Herrero, Correlated insulator behaviour at half-filling in magic-angle graphene superlattices. Nature 556, 80–84 (2018).  43.  M. Oh, K. P. Nuckolls, D. Wong, R. L. Lee, X. Liu, K. Watanabe, T. Taniguchi, A. Yazdani, Evidence for unconventional superconductivity in twisted bilayer graphene. Nature 600, 240–245 (2021).  44.  A. Uri, S. Grover, Y. Cao, J. A. Crosse, K. Bagani, D. Rodan-Legrain, Y. Myasoedov,  K. Watanabe, T. Taniguchi, P. Moon, M. Koshino, P. Jarillo-Herrero, E. Zeldov, Mapping the twist-angle disorder and Landau levels in magic-angle graphene. Nature 581, 47–52 (2020).  45. T . Benschop, T. A. de Jong, P. Stepanov, X. Lu, V. Stalman, S. J. van der Molen, D. K. Efetov, M. P. Allan, Measuring local moiré lattice heterogeneity of twisted bilayer graphene. Phys. Rev. Res. 3, 013153 (2021).  46.  W. N. Kang, C. U. Jung, K. H. P. Kim, M.-S. Park, S. Y. Lee, H.-J. Kim, E.-M. Choi, K. H. Kim,  M.-S. Kim, S.-I. Lee, Hole carrier in MgB2 characterized by Hall measurements. Appl. Phys. Lett. 79, 982–984 (2001).  47.  A. Gozar, G. Logvenov, L. F. Kourkoutis, A. T. Bollinger, L. A. Giannuzzi, D. A. Muller,  I. Bozovic, High-temperature interface superconductivity between metallic and insulating copper oxides. Nature 455, 782–785 (2008).  48. N . A. Saveskul, N. A. Titova, E. M. Baeva, A. V. Semenov, A. V. Lubenchenko, S. Saha,  H. Reddy, S. I. Bogdanov, E. E. Marinero, V. M. Shalaev, A. Boltasseva, V. S. Khrapai,  A. I. Kardakova, G. N. Goltsman, Superconductivity behavior in epitaxial TiN films points to surface magnetic disorder. Phys. Rev. Appl. 12, 054001 (2019).  49.  A. Stangl, A. Palau, G. Deutscher, X. Obradors, T. Puig, Ultra-high critical current densities of superconducting YBa2Cu3O7-δ thin films in the overdoped state. Sci. Rep. 11, 8176 (2021).  50. H . Ge, Y. R. Jin, X.-H. Song, High quality NbTiN films fabrication and rapid thermal annealing investigation. Chinese Phys. B 28, 077402 (2019).  51.  A. J. Annunziata, O. Quaranta, D. F. Santavicca, A. Casaburi, L. Frunzio, M. Ejrnaes,  M. J. Rooks, R. Cristiano, S. Pagano, A. Frydman, D. E. Prober, Reset dynamics and latching in niobium superconducting nanowire single-photon detectors. J. Appl. Phys. 108, 084507 (2010).  52.  X. Du, D. E. Prober, H. Vora, C. B. McKitterick, Graphene-based bolometers. Graphene 2D Mater. 1, 1–22 (2014).  53.  G. F. Burkhard, E. T. Hoke, M. D. McGehee, Accounting for interference, scattering, and electrode absorption to make accurate internal quantum efficiency measurements in organic and other thin solar cells. Adv. Mater. 22, 3293–3297 (2010).  54.  M. Furchi, A. Urich, A. Pospischil, G. Lilley, K. Unterrainer, H. Detz, P. Klang, A. M. Andrews, W. Schrenk, G. Strasser, T. Mueller, Microcavity-integrated graphene photo-detector. Nano Lett. 12, 2773–2777 (2012).  55.  D. K. Efetov, R.-J. Shiue, Y. Gao, B. Skinner, E. D. Walsh, H. Choi, J. Zheng, C. Tan, G. Grosso, C. Peng, J. Hone, K. C. Fong, D. Englund, Fast thermal relaxation in cavity-coupled graphene bolometers with a Johnson noise read-out. Nat. Nanotechnol. 13, 797–801 (2018).Acknowledgments: We acknowledge discussion with P. Seifert and R. H. Hadfield. G.D.B., A.D.-C., and D.K.E. acknowledge the support of the Center for NanoScience (CeNS). Funding: D.K.E. acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 852927), the funding from the EU EIC Pathfinder program, project FLATS (grant agreement no. 101099139), and the German Research Foundation (DFG) under the priority program SPP2244 (project no. 535146365). K.W. and T.T. acknowledge support from the Elemental Strategy Initiative conducted by the MEXT, Japan (grant number JPMXP0112101001) and JSPS KAKENHI (grant numbers 19H05790, 20H00354, and 21H05233). Author contributions: D.K.E. and G.D.B. conceived and designed the experiments. G.D.B. and A.D.-C. fabricated the devices. G.D.B. performed the measurements. G.D.B., D.K.E., and K.C.F. analyzed the data. T.T. and K.W. contributed materials. D.K.E. and K.C.F. supported the experiments. G.D.B., D.K.E., and K.C.F. wrote the paper. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.Submitted 21 March 2024 Accepted 13 August 2024 Published 18 September 2024 10.1126/sciadv.adp3725Downloaded from https://www.science.org at National Institute for Materials Science on September 24, 2024 Infrared single-photon detection with superconducting magic-angle twisted bilayer graphene INTRODUCTION RESULTS Photoresponse measurements Single-photon sensitivity by superconducting MATBG Detector performance at higher temperatures DISCUSSION MATERIALS AND METHODS Device fabrication Photoresponse measurements Transport measurements Supplementary Materials This PDF file includes: REFERENCES AND NOTES Acknowledgments