# Fileset

[GrPDt8_Clean_JJAP2.pdf](https://mdr.nims.go.jp/filesets/99292dfa-f02b-430d-a07a-b0ff3e112d4b/download)

## Creator

[Takuya Iwasaki](https://orcid.org/0000-0002-1103-2433), Yodai Sato, Makoto Ogo, Byunghun Oh, [Daichi Kozawa](https://orcid.org/0000-0002-0629-5589), [Ryo Kitaura](https://orcid.org/0000-0001-8108-109X), [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), [Satoshi Moriyama](https://orcid.org/0000-0001-6599-6433), Junichi Fujikata

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Photo-thermoelectric effect-driven detection of optical communication light in graphene/hBN heterostructures](https://mdr.nims.go.jp/datasets/c3bd862a-79bc-452e-97f0-f0732ae4b33f)

## Fulltext

Japanese Journal of Applied Physics REGULAR PAPERPhoto-thermoelectric effect-driven detection of optical communicationlight in graphene/hBN heterostructuresTakuya Iwasaki1 ∗, Yodai Sato2, Makoto Ogo2, Byunghun Oh2, Daichi Kozawa1, Ryo Kitaura1,Kenji Watanabe3, Takashi Taniguchi1, Satoshi Moriyama2, and Junichi Fujikata41Research Center for Materials Nanoarchitectonics, National Institute for Materials Science, Tsukuba, Ibaraki305-0044, Japan2Department of Electrical and Electronic Engineering, Tokyo Denki University, Adachi, Tokyo 120-8551, Japan3Research Center for Electronic and Optical Materials, National Institute for Materials Science, Tsukuba, Ibaraki305-0044, Japan4Institute of Post-LED Photonics, Tokushima University, 2-1, Minami-Josanjima, Tokushima 770-8506, JapanWe report on the photodetection properties of high-quality graphene encapsulated by hexagonal boron nitrideunder illumination with optical communication light. We demonstrate a gate-tunable photocurrent and zero-biasswitching cycle operation at room temperature. Through gate- and temperature-dependent photocurrent mea-surements, we determine that the dominant photoresponse mechanism is the photo-thermoelectric effect. At lowtemperatures, the photocurrent in finite doping regions correlates with the Seebeck coefficient, while sharp peaksemerge near the charge neutrality point due to an edge-excited photocurrent. Our study provides guidelines forhigh-performance graphene-based optoelectronic devices.Graphene has garnered attention due to its excellent electronic and optical properties,such as a broad light absorption range, high conductivity, and high carrier mobility.1) Owingto its ultrafast photo-excitation dynamics, graphene is particularly anticipated as a buildingblock for high-speed optoelectronic devices in fiber-optic communication bands from 1260to 1625 nm. Studies have explored photodetection at the low-loss wavelength of 1550 nmusing graphene on SiO2,2–5) carbon nanotubes,6) Si waveguides (WGs),7–14) SiN WGs,15,16)plasmonic cavities,17) and metamaterials.18) A high responsivity of ∼50 mA/W was demon-strated in graphene/Si WG-integrated devices,9) and graphene-metamaterial hybrid devicesdemonstrated an ultrafast operating speed of >500 GHz.18)Despite such high performance, the quality of graphene in previous studies was limiteddue to the use of chemical vapor deposition-grown graphene6,10, 14–18) and/or gate dielectrics,which could degrade the quality of the product.14–16) To enhance graphene photodetector∗E-mail: IWASAKI.Takuya@nims.ac.jp1/12Jpn. J. Appl. Phys. REGULAR PAPERperformance, improving its electronic quality (e.g., carrier mobility) is crucial. Encapsulationof exfoliated graphene with hexagonal boron nitride (hBN), forming a hBN/graphene/hBNheterostructure, is a promising approach.19) A previous study using hBN/graphene/hBN het-erostructures on Si WGs demonstrated high-speed operation and high responsivity; however,the low carrier mobility of 1000 cm2/V·s, below that of typical graphene devices, mighthave limited performance.11) Recent time-resolved measurements have demonstrated that theoperating speed of high-quality graphene devices based on hBN/graphene/hBN heterostruc-tures reaches 220 GHz for visible light excitation.20) Accordingly, a hBN/graphene/hBNheterostructure is key to achieving high-performance photodetectors. Investigating photode-tection characteristics for optical communication bands in high-quality hBN/graphene/hBNdevices is imperative; however, this area has received limited exploration. In particular, adevice consisting of hBN/graphene/hBN heterostructures on SiO2/Si substrates, a typicalhigh-quality graphene device structure, has not been adopted so far.Furthermore, understanding the photoresponse mechanism is crucial for improving pho-todetector performance. Although the photoresponse mechanism varies with device designand materials, an intrinsic photodetection mechanism in graphene is hypothesized to be aphoto-thermoelectric (PTE) effect,21,22) generating a photoresponse through a temperaturegradient and the thermoelectric (Seebeck) effect. This gradient is caused by the ultrafastcooling of photo-excited-hot carriers in graphene.23–25) As the PTE-induced photoresponseemerges without an external bias voltage,14–16,18, 21, 22) it enables zero-bias and zero-darkcurrent operation, leading to low noise and low power consumption in device applications.In this study, we investigate the photodetection properties of the hBN/graphene/hBNdevice on a typical SiO2/Si substrate under simple top illumination at a wavelength of 1550nm. We examine the gate- and temperature-dependence of the photoresponse in our device toelucidate the photoresponse mechanism in hBN/graphene/hBN heterostructures. Our resultssuggest that the dominant mechanism of the photocurrent generation is the PTE effect, allowingthe gate-tunable zero-bias photoresponse.For device fabrication, single-layer graphene (SLG) and hBN flakes were prepared on a 90-nm SiO2/Si substrate through mechanical exfoliation from bulk crystals. The hBN/SLG/hBNheterostructure was assembled via a modified dry transfer method as described in 26) . Electron-beam (EB) lithography and reactive ion etching (SF6 plasma) were employed to selectively etchthe top hBN layer to expose the contact region, ensuring SLG was not etched due to fluorinationof its surface.27) Subsequently, contact electrodes (Cr/Pd/Au) were deposited directly over theEB resist, followed by liftoff. The schematic cross-section and optical image of the fabricated2/12Jpn. J. Appl. Phys. REGULAR PAPERFig. 1. (a) Schematic cross-section of our device. (b) Drain current as a function of 𝑉g at 𝑉sd = 1 mV, 𝑇 =300 K in the dark. The inset exhibits the optical image of our device. The represented scale bar is 10 𝜇m. Thedotted lines represent the edge of graphene.device are displayed in Fig. 1(a) and the inset of Fig. 1(b), respectively. Measurements wereconducted in a vacuum using a helium optical cryostat, allowing temperature (𝑇) controldown to 8 K. A two-terminal DC drain current (𝐼d) of the device was measured by applyinga source-drain bias voltage (𝑉sd) using a source-measure unit. A highly doped Si substrateserved as a back gate to adjust the Fermi energy of graphene. The device was illuminatedusing a semiconductor laser at a wavelength of 1550 nm and a spot diameter of ∼4.5 𝜇m.Given that the laser spot size exceeds that of the graphene channel, we referred to the laserpower as the internal power (𝑃in) at the device.Initially, we examine the standard transport property of our device at room temperature.Figure 1(b) illustrates the drain current as a function of the back-gate voltage (𝑉g) at 𝑇 =300 K, indicating typical ambipolar behavior with the charge neutrality point (CNP) locatedat 𝑉g ∼ 0 V. This behavior suggests minimal external doping in the graphene due to hBNencapsulation. The field-effect mobility is estimated by 𝜇FE = (𝑑𝐼d/𝑑𝑉g)𝐿/(𝑊𝑉sd𝐶g), where𝐿 is the channel length, 𝑊 is the channel width, 𝐶g is the gate capacitance per unit area. Weobtain 𝜇FE ∼ 9600 cm2/V·s for electrons and 𝜇FE ∼ 19000 cm2/V·s for holes, higher than thatin typical graphene on SiO2 and WG-integrated graphene devices.11–16)Next, we explore the photoresponse in our device. Figure 2(a) displays the transientphotoresponse characteristics at zero-bias voltage (𝑉sd ∼ 0). We monitor the change in 𝐼dwhen turning on and off the light illumination. At𝑉g = 1 V, a finite photoresponse is observed,as presented in the top panel of Fig. 2(a). Conversely, a negative photoresponse at 𝑉g = −1 Vis also observed, as depicted in the bottom panel of Fig. 2(a). These results indicate that thesign and intensity of the photoresponse depend on 𝑉g, and our device operates at zero bias.The on/off tracing speed is instantaneous on a 10-ms timescale (the inset of Fig. 2(a)), limited3/12Jpn. J. Appl. Phys. REGULAR PAPERFig. 2. Photodetection at 𝑇 = 300 K. (a) Zero-bias switching cycles at 𝑉g = 1 V (top), −1 V (bottom) with𝑃in = 345 𝜇W. The inset shows the photoresponse at illumination switching. The green-shaded regionrepresents the data from the device under illumination. (b) Seebeck coefficient (top) and zero-bias photocurrent(bottom) as a function of 𝑉g and 𝑃in. (c) Bias voltage dependence of the photocurrent with 𝑃in = 345 𝜇W. Theinset exhibits the 𝐼d-𝑉sd curves in the dark and under illumination at 𝑉g = 1 V. (d) Responsivity andphotocurrent (inset) as a function of 𝑃in at zero bias.by our measurement system.Investigating the gate-dependence of the photocurrent provides insight into the photore-sponse mechanism. Here, the photocurrent (𝐼ph) is defined as the difference between the draincurrent under illumination (𝐼light) and in the dark (𝐼dark), i.e., 𝐼ph = 𝐼light − 𝐼dark. The bottompanel of Fig. 2(b) exhibits the dependence of 𝐼ph on 𝑉g at zero bias (𝐼dark ∼ 0). The signof 𝐼ph switches upon crossing the CNP, whereas the photocurrent vanishes at the CNP. Inother words, the photocurrent is negative (positive) when the Fermi energy lies in the valence(conduction) band. This feature can be attributed to the PTE effect,22) and is discussed below.The sign reversal of 𝐼ph across the CNP could reflect the dependence of the Seebeck4/12Jpn. J. Appl. Phys. REGULAR PAPERcoefficient (𝑆) on the carrier density.21,22, 28, 29) The Seebeck coefficient, which converts atemperature gradient into a built-in voltage, can be derived from the transport characteristicsby the modified Mott formula:21,22, 28, 30)𝑆 = −𝜋2𝑘2B𝑇3𝑒𝐺𝑑𝐺𝑑𝑉g𝑑𝑉g𝑑𝐸����𝐸=𝐸F(1)where 𝑘B is the Boltzmann constant, 𝑒 is the elementary charge, 𝐺 (= 𝐼d/𝑉sd) is the electricalconductance, and 𝐸F is the Fermi energy. Using the Fermi energy of SLG, 𝐸F = ℏ𝑣F√𝜋𝑛 (ℏ isthe Planck constant divided by 2𝜋, 𝑣F ∼ 106 m/s is the Fermi velocity of graphene31)) and thecarrier density, 𝑛 = 𝐶g(𝑉g − 𝑉CNP)/𝑒 (𝑉CNP is the gate voltage at the CNP), we calculate theSeebeck coefficient as shown in the top panel of Fig. 2(b). The absolute value of 𝑆 increasesas doping deceases, reaching a maximum value of ∼200 𝜇V/K, in good agreement withvalues from the literature for high-quality graphene/hBN devices.29) Upon further decreasingthe doping, the 𝑆 value decreases and vanishes at the CNP. This is due to the oppositecontribution of electrons and holes near the CNP to 𝑆. Consequently, the sign of 𝑆 switchesat the CNP, analogous to the dependence of 𝐼ph on 𝑉g in our device.In our configuration, although the graphene channel is smaller than the laser spot, itis possible that one of the two graphene-electrode interfaces heats up significantly due todeviation in the center location of the laser spot. Additionally, the graphene channel away fromthe electrodes is n-doped (p-doped) for 𝑉g > 𝑉CNP(< 𝑉CNP), while the chemical potentialnear the graphene-electrode interface is pinned by the doping from the metal electrode.32)Consequently, an n-i (p-i) interface forms in the graphene channel, where regions with different𝑆 connect. This creates a temperature gradient and induces the PTE effect in our device.To analyze the PTE effect, we calculate the temperature difference resulting in the pho-toresponse using Δ𝑇PTE = 𝑉ph/Δ𝑆, where 𝑉ph(= 𝐼ph𝑅) is the PTE-induced photovoltageand 𝑅(= 𝐺−1) is the resistance. Considering 𝑅 ∼ 780 Ω and Δ𝑆 ∼ 117 𝜇V/K at 𝑉g = 1V, and 𝑃in = 345 𝜇W, we obtain Δ𝑇PTE ∼ 0.1 K. Furthermore, we estimate the tempera-ture increase induced by illumination (Δ𝑇) by considering heat flow as a plane-wavefront;Δ𝑇 = (𝐿/2𝑡gr𝑊) (𝑃in𝛼/𝜅),21,33) where 𝑡gr is the thickness, 𝛼 = 2.3% is the absorption coeffi-cient, and 𝜅 ∼5300 W/m·K is the thermal conductivity of SLG.33) This calculation yieldsΔ𝑇 ∼0.6 K, in the same order of magnitude and consistent with Δ𝑇PTE. It is noted that illuminationof both source and drain electrodes could reduce net photocurrent due to the cancellationof symmetric photocurrents from opposite-directional graphene-electrode interfaces,22,34, 35)potentially leading to an underestimation of Δ𝑇PTE. From this analysis, we deduce that thePTE effect predominates in the photoresponse mechanism of our device.5/12Jpn. J. Appl. Phys. REGULAR PAPERIt should be considered that there are other potential mechanisms for photocurrent gener-ation, such as photovoltaic, bolometric, and photogating effects. However, if the photovoltaiceffect was dominant, a sign switch of 𝐼ph would not occur upon crossing the CNP.3,18) Thebolometric effect requires a bias voltage to observe a photocurrent,36) contrary to the zero-biasphotoresponse in our device. The photogating effect is unlikely in our device because grapheneis isolated from potential charge traps by the hBN encapsulation.6,37) Moreover, photogatingeffects from a doped Si substrate should not occur for wavelengths >1100 nm, as the excitationenergy is lower than the bandgap of Si.37)We reaffirm the zero-bias photoresponse in the 𝐼ph-𝑉sd and 𝐼d-𝑉sd characteristics depicted inFig. 2(c) and its inset, respectively. For𝑉g = 1 V (−1 V), the photocurrent increases (decreases)with increasing 𝑉sd, becoming zero at 𝑉sd ∼ −0.45 mV (−0.2 mV). This reflects the polarityof the photovoltage, where the bias voltage enhances (cancels out) the photovoltage when itspolarity is the same (opposite). At finite 𝑉sd, the photoconductive effect might also contributeto enhancing 𝐼ph.The figure-of-merit values are instrumental in evaluating the photodetector performanceof our device. As shown in the bottom panel of Fig. 2(b) and the inset of Fig. 2(d), the absolutevalues of 𝐼ph increase with 𝑃in. The maximum responsivity 𝑅I(= 𝐼ph/𝑃in) is ∼0.55 mA/W atzero bias (Fig. 2(d)). Assuming the noise current originates from the thermal Johnson-Nyquistcontribution,38) we estimate the minimum noise equivalent power (NEP =√4𝑘B𝑇/𝑅/𝑅I) to be∼9.5 nW/Hz−1/2. These values are comparable to conventional graphene photodetectors (zerobias, 1550-nm top illumination).2,3, 11) It should be noted that the PTE-induced photocurrentis dependent on the laser spot position and is enhanced when the laser is focused on agraphene-electrode interface. Thus, the device performance could be improved by enlargingthe graphene channel size, introducing the asymmetric geometry, and focusing on one side ofthe graphene-electrode interfaces.We then consider the low-temperature characteristics and temperature dependence of thephotoresponse. Figure 3(a) and its inset display the 𝐼d-𝑉g and 𝐼d-𝑉sd characteristics at 𝑇 = 20K, respectively. At low 𝑇 , the CNP shifts slightly to 𝑉g ∼ −0.4 V. We confirm Ohmic contactthrough the linearity of 𝐼d-𝑉sd curves across all 𝑇 ranges in this study (20−300 K), thus rulingout the possibility of photocurrent generation due to a Schottky contact. In contrast to the𝐼ph-𝑉g curves at 𝑇 = 300 K, we observe two distinct 𝐼ph peaks at 𝑇 = 20 K as presented in Fig.3(b). One peak is positive at the CNP (“CNP peak”), while the other is negative at a slightlypositive 𝑉g from the CNP (“negative peak”). The maximum absolute values of these peaksincrease with increasing 𝑃in.6/12Jpn. J. Appl. Phys. REGULAR PAPER123-1 0 1-0.50.00.5-2 -1 0 1 2-200-1000100(b) I d (A)T = 20 KCNP(a)  I d (A)Vsd (mV)Vg = -0.1 V 2.5 W 39 W 126 W 345 W I ph (nA)Vg (V)Fig. 3. (a) Drain current vs 𝑉g at 𝑉sd = 1 mV, 𝑇 = 20 K in the dark. The inset exhibits the 𝐼d-𝑉sd curve at 𝑉g =−0.1 V. (b) Photocurrent as a function of 𝑉g and 𝑃in at 𝑇 = 20 K.To determine whether the peaks originate from the PTE effect, we calculate the Seebeckcoefficient using Eq. 1 for various 𝑇 (Fig. 4(a)). As 𝑇 decreases, the |𝑆| value diminishes,and the 𝑉g positions of maximum positive and negative 𝑆 shift toward the CNP. This shiftcorresponds to the narrowing of the energy region dominated by electron-hole puddles, aresult of reduced thermal excitation of electrons and holes at lower 𝑇 .29) As illustrated in Fig.4(b), the |𝐼ph| values also decrease with 𝑇 in the finite doping regions. It is noteworthy that at 𝑇= 20 K, the photocurrent exhibits strong fluctuations due to the influence of the 𝐼ph peaks. Bytracking the maximum positive 𝑆 (𝑆max) and 𝐼ph values at the same 𝑉g position across various𝑇 , we observe their correlated temperature dependence, as depicted in the inset of Fig. 4(a).This correlation indicates that the photocurrent is dependent on 𝑆, supporting the dominanceof the PTE effect in the photoresponse mechanism for finite doping regions. However, thepositions of the 𝐼ph peaks do not depend on either 𝑇 nor 𝑆. As 𝑇 increases, the intensity of 𝐼phpeaks diminishes, and they disappear for 𝑇 ≥ 200 K (Fig. 4(b)). These observations suggestthat the mechanism behind the 𝐼ph peaks differs from the PTE effect.The device geometry and the size of the laser spot suggest that an edge-excited photocurrent7/12Jpn. J. Appl. Phys. REGULAR PAPERFig. 4. (a-c) Temperature and back-gate voltage dependence of the (a) 𝑆, (b) 𝐼ph at 𝑉sd = 1 mV, and (c) 𝑅.The CNP peak region is shaded in (b,c). The inset of (a) presents the 𝑇 dependence of 𝑆max and 𝐼ph.(EPC) might explain the CNP peak.35) The EPC arises from potential symmetry breaking atgraphene edges and strong electron-electron interaction in charge-neutral graphene. In ourcase, the illumination with a laser spot covering the graphene edges could lead to the generationof the EPC. As the EPC is sensitive to the charge-neutral state of graphene, the width of the CNPpeak should be narrower than that of the resistance peak.35) This characteristic is observablein our device, as indicated by the shaded region in Figs. 4(b) and (c).Conversely, the negative peak, not observed in previous studies, presents a unique case.Depending on the channel symmetry, the EPC mights assume a negative value but wouldtypically appear at the CNP.35) Therefore, neither the PTE effect nor edge excitation canfully explain the negative peak. A plausible cause for the emergence of the negative peakcould be the photogating effect from charge traps, activated at low 𝑇 , due to impurities at thegraphene/hBN or hBN/SiO2 interfaces. However, the precise origin of the negative peak andthe reason for the disappearance of the CNP peak at high temperatures remain unclear and aresubjects for further investigation.In summary, we have fabricated an hBN/graphene/hBN device and investigated its pho-8/12Jpn. J. Appl. Phys. REGULAR PAPERtodetection characteristics for optical communication light. Our analysis of the gate- andtemperature-dependent photoresponse in this device reveals that the predominant photodetec-tion mechanism is the PTE effect. The ability to tune the photoresponse via the gate voltageenables the ambipolar zero-bias switching operations at room temperature. To deepen under-standing of the PTE effect and contribution of other photoresponse mechanisms, investigatinglocal photoresponse by photocurrent mapping on a graphene channel larger than a laser spot isleft as a future task. Since the photoresponse could result from the asymmetricity of the device,the device performance could be improved by introducing asymmetric factors such as channeland electrode structures. It is noteworthy that a simple structure on a typical SiO2/Si substratewas utilized in this study. This structure offers the potential for additional functionalities. Forinstance, integrating local gates into this setup could allow for the electrostatic definition of pnjunctions within in a graphene channel, leading to highly responsive photodetectors withoutcompromising the carrier mobility of graphene. Our study provides the fundamental insightinto the photodetection properties of hBN/graphene/hBN heterostructure devices. Regardingthe role of hBN in addition to improving the quality of graphene, out-of-plane energy transfer,due to the coupling between charge carriers in graphene and hyperbolic phonon polaritonsin hBN, may influence the photoexcitation/hot carrier relaxation process.39,40) This suggeststhat designing proper geometrical parameters, such as the thickness of hBN and the graphenechannel size, could enhance the PTE effect. Given their high carrier mobility and Seebeckcoefficient, hBN/graphene/hBN heterostructures hold promise as the foundation for futurehigh-performance optoelectronic devices.AcknowledgmentThis work was partially supported by JPSJ KAKENHI Grant No. 21H01400, 21H01749,22H01555, 22H01893, 23H00274, 23H05469 and the Advanced Research Infrastructure forMaterials and Nanotechnology in Japan (ARIM) of the Ministry of Education, Culture, Sports,Science and Technology (MEXT). Proposal Number JPMXP1223NM5186.9/12Jpn. J. Appl. Phys. REGULAR PAPERReferences1) M. Romagnoli, V. Sorianello, M. Midrio, F. H. L. Koppens, C. Huyghebaert, D.Neumaier, P. Galli, W. Templ, A. D’Errico, and A. C. Ferrari, Nat. Rev. Mater. 3, 392(2018).2) F. Xia, T. Mueller, Y. Lin, A. Valdes-Garcia, and P. Avouris, Nat. Nanotechnol. 4, 839(2009).3) T. Mueller, F. Xia, and P. Avouris, Nat. Photon. 4, 297 (2010).4) A. Urich, K. Unterrainer, and T. Mueller, Nano Lett. 11, 2804 (2011).5) T. J. Echtermeyer, P. S. Nene, M. Trushin, R. V. Gorbachev, A. L. Eiden, S. Milana, Z.Sun, J. Schliemann, E. Lidorikis, K. S. Novoselov, and A. C. Ferrari, Nano Lett. 14,3733 (2014).6) Y. Liu, F. Wang, X. Wang, X. Wang, E. Flahaut, X. Liu, Y. Li, X. Wang, Y. Xu, Y. Shi,and R. Zhang, Nat. Commun. 6, 8589 (2015).7) X. Gan, R.-J. Shiue, Y. Gao, I. Meric, T. F. Heinz, K. Shepard, J. Hone, S. Assefa, andD. Englund, Nat. Photon. 7, 883 (2013).8) X. Wang, Z. Cheng, K. Xu, H. K. Tsang, and J.-B. Xu, Nat. Photon. 7, 888 (2013).9) A. Pospischil, M. Humer, M. M. Furchi, D. Bachmann, R. Guider, T. Fromherz, and T.Mueller, Nat. Photon. 7, 892 (2013).10) D. Schall, D. Neumaier, M. Mohsin, B. Chmielak, J. Bolten, C. Porschatis, A. Prinzen,C. Matheisen, W. Kuebart, B. Junginger, W. Templ, A. L. Giesecke, and H. Kurz, ACSPhotonics 1, 781 (2014).11) R.-J. Shiue, Y. Gao, Y. Wang, C. Peng, A. D. Robertson, D. K. Efetov, S. Assefa, F. H.L. Koppens, J. Hone, and D. Englund, Nano Lett. 15, 7288 (2015).12) S. Schuler, D. Schall, D. Neumaier, L. Dobusch, O. Bethge, B. Schwarz, M. Krall, andT. Mueller, Nano Lett. 16, 7107 (2016).13) S. Schuler, D. Schall, D. Neumaier, B. Schwarz, K. Watanabe, T. Taniguchi, and T.Mueller, ACS Photonics 5, 4758 (2018).14) S. Marconi, M. A. Giambra, A. Montanaro, V. Mišeikis, S. Soresi, S. Tirelli, P. Galli, F.Buchali, W. Templ, C. Coletti, V. Sorianello, M. Romagnoli, Nat. Commun. 12, 806(2021).15) J. E. Muench, A. Ruocco, M. A. Giambra, V. Miseikis, D. Zhang, J. Wang, H. F. Y.Watson, G. C. Park, S. Akhavan, V. Sorianello, M. Midrio, A. Tomadin, C. Coletti, M.Romagnoli, A. C. Ferrari, and I. Goykhman, Nano Lett. 19, 7632 (2019).10/12Jpn. J. Appl. Phys. REGULAR PAPER16) V. Mišeikis, S. Marconi, M. A. Giambra, A. Montanaro, L. Martini, F. Fabbri, S.Pezzini, G. Piccinini, S. Forti, B. Terrés, I. Goykhman, L. Hamidouche, P. Legagneux,V. Sorianello, A. C. Ferrari, F. H. L. Koppens, M. Romagnoli, and C. Coletti, ACSNano 14, 11190 (2020).17) Z. Chen, X. Li, J. Wang, L. Tao, M. Long, S.-J. Liang, L. K. Ang, C. Shu, H. Ki Tsang,and J.-B. Xu, ACS Nano 11, 430 (2017).18) S. M. Koepfli, M. Baumann, Y. Koyaz, R. Gadola, A. Güngör, K. Keller, Y. Horst, S.Nashashibi, R. Schwanninger, M. Doderer, E. Passerini, Y. Fedoryshyn, and J.Leuthold, Science 380, 1169 (2023).19) C. R. Dean, A. F. Young, I. Meric, C. Lee, L. Wang, S. Sorgenfrei, K. Watanabe, T.Taniguchi, P. Kim, K. L. Shepard, and J. Hone, Nat. Nanotechnol. 5, 722 (2010).20) K. Yoshioka, T. Wakamura, M. Hashisaka, K. Watanabe, T. Taniguchi, and N. Kumada,Nat. Photon. 16, 718 (2022).21) X. Xu, N. M. Gabor, J. S. Alden, A. M. van der Zande, and P. L. McEuen, Nano Lett.10, 562 (2010).22) N. M. Gabor, J. C. W. Song, Q. Ma, N. L. Nair, T. Taychatanapat, K. Watanabe, T.Taniguchi, L. S. Levitov, and P. Jarillo-Herrero, Science 334, 648 (2011).23) K. J. Tielrooij, J. C. W. Song, S. A. Jensen, A. Centeno, A. Pesquera, A. ZurutuzaElorza, M. Bonn, L. S. Levitov, and F. H. L. Koppens, Nat. Phys. 9, 248 (2013).24) R. Bistritzer and A. H. MacDonald, Phys. Rev. Lett. 102, 206410 (2009).25) A. Tomadin, S. M. Hornett, H. I. Wang, E. M. Alexeev, A. Candini, C. Coletti, D.Turchinovich, M. Kläui, M. Bonn, F. H. L. Koppens, E. Hendry, M. Polini, and K.-J.Tielrooij, Sci. Adv. 4, eaar5313 (2018).26) T. Iwasaki, K. Endo, E. Watanabe, D. Tsuya, Y. Morita, S. Nakaharai, Y. Noguchi, Y.Wakayama, K. Watanabe, T. Taniguchi, and S. Moriyama, ACS Appl. Mater. Interfaces12, 8533 (2020).27) B. S. Jessen, L. Gammelgaard, M. R. Thomsen, D. M. A. Mackenzie, J. D. Thomsen, J.M. Caridad, E. Duegaard, K. Watanabe, T. Taniguchi, T. J. Booth, T. G. Pedersen, A.-P.Jauho, and P. Bøggild, Nat. Nanotechnol.14, 340 (2019).28) Y. M. Zuev, W. Chang, and P. Kim, Phys. Rev. Lett. 102, 096807 (2009).29) J. Duan, X. Wang, X. Lai, G. Li, K. Watanabe, T. Taniguchi, M. Zebarjadi, and E. Y.Andrei, Proc. Natl. Acad. Sci. USA 113, 14272 (2016).30) M. Cutler and N. F. Mott, Phys. Rev. 181, 1336 (1969).31) S. Das Sarma, S. Adam, E. H. Hwang, and E. Rossi, Rev. Mod. Phys. 83, 407 (2011).11/12Jpn. J. Appl. Phys. REGULAR PAPER32) T. Mueller, F. Xia, M. Freitag, J. Tsang, and P. Avouris, Phys. Rev. B 79, 245130 (2009).33) A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, and C. N. Lau,Nano Lett. 8, 902 (2008).34) F. Xia, T. Mueller, R. Golizadeh-Mojarad, M. Freitag, Y.-M. Lin, J. Tsang, V.Perebeinos, and P. Avouris, Nano Lett. 9, 1039 (2009).35) Q. Ma, C. H. Lui, J. C. W. Song, Y. Lin, J. F. Kong, Y. Cao, T. H. Dinh, N. L. Nair, W.Fang, K. Watanabe, T. Taniguchi, S.-Y. Xu, J. Kong, T. Palacios, N. Gedik, N. M. Gabor,and P. Jarillo-Herrero, Nat. Nanotechnol. 14, 145 (2019).36) J. Yan, M-H. Kim, J. A. Elle, A. B. Sushkov, G. S. Jenkins, H. M. Milchberg, M. S.Fuhrer, and H. D. Drew, Nat. Nanotechnol. 7, 472 (2012).37) F. Luo, M. Zhu, Y. Tan, H. Sun, W. Luo, G. Peng, Z. Zhu, X.-A. Zhang, and S. Qin, AIPAdv. 8, 115106 (2018).38) S. Castilla, B. Terrés, M. Autore, L. Viti, J. Li, A. Y. Nikitin, I. Vangelidis, K.Watanabe, T. Taniguchi, E. Lidorikis, M. S. Vitiello, R. Hillenbrand, K.-J. Tielrooij,and F. H. L. Koppens, Nano Lett. 19, 2765 (2019).39) K.-J. Tielrooij, N. C. H. Hesp, A. Principi, M. B. Lundeberg, E. A. A. Pogna, L.Banszerus, Z. Mics, M. Massicotte, P. Schmidt, D. Davydovskaya, D. G. Purdie, I.Goykhman, G. Soavi, A. Lombardo, K. Watanabe, T. Taniguchi, M. Bonn, D.Turchinovich, C. Stampfer, A. C. Ferrari, G. Cerullo, M. Polini, and F. H. L. Koppens,Nat. Nanotechnol. 3, 41 (2018).40) A. Woessner, R. Parret, D. Davydovskaya, Y. Gao, J.-S. Wu, M. B. Lundeberg, S.Nanot, P. Alonso-González, K. Watanabe, T. Taniguchi, R. Hillenbrand, M. M. Fogler,J. Hone, and F. H. L. Koppens, npj 2D Mater. Appl. 1, 25 (2017).12/12