# Fileset

[s41467-024-52838-w.pdf](https://mdr.nims.go.jp/filesets/85451460-bbac-4936-acaa-488428179691/download)

## Creator

[Sebastián Castilla](https://orcid.org/0000-0002-8899-0525), Hitesh Agarwal, Ioannis Vangelidis, [Yuliy V. Bludov](https://orcid.org/0000-0001-9648-1459), [David Alcaraz Iranzo](https://orcid.org/0000-0002-2141-569X), Adrià Grabulosa, Matteo Ceccanti, [Mikhail I. Vasilevskiy](https://orcid.org/0000-0003-2930-9434), [Roshan Krishna Kumar](https://orcid.org/0000-0003-0857-4466), Eli Janzen, [James H. Edgar](https://orcid.org/0000-0003-0918-5964), [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), [Nuno M. R. Peres](https://orcid.org/0000-0002-7928-8005), [Elefterios Lidorikis](https://orcid.org/0000-0002-9552-9366), [Frank H. L. Koppens](https://orcid.org/0000-0001-9764-6120)

## Rights

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

## Other metadata

[Electrical spectroscopy of polaritonic nanoresonators](https://mdr.nims.go.jp/datasets/ce6151c9-8ad2-43bd-af18-b6ae6c23584d)

## Fulltext

Electrical spectroscopy of polaritonic nanoresonatorsArticle https://doi.org/10.1038/s41467-024-52838-wElectrical spectroscopy of polaritonicnanoresonatorsSebastián Castilla 1,12 , Hitesh Agarwal1,12, Ioannis Vangelidis2,12,Yuliy V. Bludov 3,4,12, David Alcaraz Iranzo 1, Adrià Grabulosa1,Matteo Ceccanti1, Mikhail I. Vasilevskiy 3,4,5, Roshan Krishna Kumar 1,Eli Janzen6, James H. Edgar 6, Kenji Watanabe 7, Takashi Taniguchi 8,Nuno M. R. Peres 3,4,5,9, Elefterios Lidorikis 2,10 & Frank H. L. Koppens 1,11One of the most captivating properties of polaritons is their capacity to con-fine light at the nanoscale. This confinement is even more extreme in two-dimensional (2D) materials. 2D polaritons have been investigated by opticalmeasurements using an external photodetector. However, their effectivespectrally resolved electrical detection via far-field excitation remains unex-plored. This hinders their exploitation in crucial applications such as sensing,hyperspectral imaging, and optical spectrometry, banking on their potentialfor integration with silicon technologies. Herein, we present the electricalspectroscopy of polaritonic nanoresonators based on a high-quality 2D-material heterostructure, which serves at the same time as the photodetectorand the polaritonic platform. Subsequently, we electrically detect these mid-infrared resonators by near-field coupling to a graphene pn-junction. Thenanoresonators simultaneously exhibit extreme lateral confinement and high-quality factors. This work opens a venue for investigating this tunable andcomplex hybrid system and its use in compact sensing and imaging platforms.Polaritons are coupled excitations of electromagnetic waves withcharged particles (plasmons polaritons)1,2 or lattice vibrations (phononpolaritons)3–5. The polaritonic properties become extreme in two-dimensional (2D) materials, including wavelength confinement byfactors up to 3006,7, ray-like propagating modes8–10, long lifetimes11,12,and capabilities to tune its properties in situ4,13. These polaritons havebeen investigated in near-field studies (e.g., using scanning near-fieldoptical microscopy)8,9,14–21, by electron energy loss spectroscopy(EELS)22–24 and far-field with Fourier Transform Infrared spectroscopy(FTIR)1,6,25–34, which, however, constitute bulky systems that require atypical cooled external detector. In order to achieve a highly compactplatform, 2D polaritons have been electrically detected by using agraphene nanodisk array35 or antennas that launch hyperbolic phononpolaritons (HPPs) of hBN in the detector’s photoactive area36. How-ever, the detection ismainly based on increasing themagnitude of thephotoinduced signal at a fixed incident wavelength35,37–39, at theexpense of spectral information.Electrical spectroscopy of polaritonic nanoresonators is a uniquecapability with prospects for nano-optoelectronic circuits, and mole-cular sensing applications33,40, as they can be strongly coupled toReceived: 1 May 2024Accepted: 23 September 2024Check for updates1ICFO - Institut deCiències Fotòniques, TheBarcelona Institute of Science andTechnology,Castelldefels (Barcelona), Spain. 2Department ofMaterials Scienceand Engineering, University of Ioannina, Ioannina, Greece. 3Centro de Física (CF-UM-UP), Universidade do Minho, Braga, Portugal. 4Departamento de Física,Universidade do Minho, Braga, Portugal. 5International Iberian Nanotechnology Laboratory (INL), Braga, Portugal. 6Tim Taylor Department of ChemicalEngineering, Kansas State University, Manhattan, KS, USA. 7Research Center for Electronic and Optical Materials, National Institute for Materials Science,Tsukuba, Japan. 8Research Center for Materials Nanoarchitectonics, National Institute for Materials Science, Tsukuba, Japan. 9POLIMA-Center for Polariton-driven Light-Matter Interactions, University of Southern Denmark, Odense M, Denmark. 10University Research Center of Ioannina (URCI), Institute of MaterialsScience andComputing, Ioannina, Greece. 11ICREA - InstitucióCatalana deRecerca i EstudisAvançats, Barcelona, Spain. 12These authors contributed equally:Sebastián Castilla, Hitesh Agarwal, Ioannis Vangelidis, Yuliy V. Bludov. e-mail: sebastian.castilla@icfo.eu; frank.koppens@icfo.euNature Communications |         (2024) 15:8635 11234567890():,;1234567890():,;http://orcid.org/0000-0002-8899-0525http://orcid.org/0000-0002-8899-0525http://orcid.org/0000-0002-8899-0525http://orcid.org/0000-0002-8899-0525http://orcid.org/0000-0002-8899-0525http://orcid.org/0000-0001-9648-1459http://orcid.org/0000-0001-9648-1459http://orcid.org/0000-0001-9648-1459http://orcid.org/0000-0001-9648-1459http://orcid.org/0000-0001-9648-1459http://orcid.org/0000-0002-2141-569Xhttp://orcid.org/0000-0002-2141-569Xhttp://orcid.org/0000-0002-2141-569Xhttp://orcid.org/0000-0002-2141-569Xhttp://orcid.org/0000-0002-2141-569Xhttp://orcid.org/0000-0003-2930-9434http://orcid.org/0000-0003-2930-9434http://orcid.org/0000-0003-2930-9434http://orcid.org/0000-0003-2930-9434http://orcid.org/0000-0003-2930-9434http://orcid.org/0000-0003-0857-4466http://orcid.org/0000-0003-0857-4466http://orcid.org/0000-0003-0857-4466http://orcid.org/0000-0003-0857-4466http://orcid.org/0000-0003-0857-4466http://orcid.org/0000-0003-0918-5964http://orcid.org/0000-0003-0918-5964http://orcid.org/0000-0003-0918-5964http://orcid.org/0000-0003-0918-5964http://orcid.org/0000-0003-0918-5964http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-7928-8005http://orcid.org/0000-0002-7928-8005http://orcid.org/0000-0002-7928-8005http://orcid.org/0000-0002-7928-8005http://orcid.org/0000-0002-7928-8005http://orcid.org/0000-0002-9552-9366http://orcid.org/0000-0002-9552-9366http://orcid.org/0000-0002-9552-9366http://orcid.org/0000-0002-9552-9366http://orcid.org/0000-0002-9552-9366http://orcid.org/0000-0001-9764-6120http://orcid.org/0000-0001-9764-6120http://orcid.org/0000-0001-9764-6120http://orcid.org/0000-0001-9764-6120http://orcid.org/0000-0001-9764-6120http://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-52838-w&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-52838-w&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-52838-w&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-024-52838-w&domain=pdfmailto:sebastian.castilla@icfo.eumailto:frank.koppens@icfo.euwww.nature.com/naturecommunicationsmolecular vibrations27,28,41. Here, we merge 2D polaritonic resonatorswith a graphene pn-junction into a single high-quality 2D-materialheterostructure to realize the electrical spectroscopy of deep sub-wavelength polaritonic nanoresonators. The quality factor of thenanoresonators and the high mobility of graphene play key roles, ashigh values enable effective photodetection of polaritonic resonances,in contrast to low-quality factor values in low mobility6,30–32 or pat-ternedgraphene1,25–27,33,42, resulting in adiminisheddetection efficiencyand spectral resolution. Our approach eliminates the need for anexternal detector for spectroscopy and leads to device miniaturiza-tion. Its small photoactive area (comparable to the hot carriers coolinglength of ~ 0.5 to 1μm36,43) is adequate in converting the incoming lightinto an electrical signal36, contrary to FTIR, which requires large opti-cally active device areas (≳ 30 × 30μm2) to obtain a reasonable signal-to-noise ratio6,32.Our methodology has enabled us to investigate the contributionof the hybridized modes present at the underexplored hBN lowerreststrahlen band (RB) in the photocurrent spectrum, which corre-sponds to a different type of hyperbolicity (type I) with respect to thetypically studied (type II) for the upper RB3,4. In fact, at the lower RBspectral range, we have observed the highest Q-factors and lateralconfinement among the whole investigated mid- and long-waveinfrared spectra. The active tunability of the polaritonic nanor-esonators’ spectral photoresponse is explored by gating graphene.Wefind that doped graphene, under certain conditions, also acts effec-tively as a mirror by partially reflecting the polaritons. This modifiesthe hybridized modes, adding extra degrees of freedom in tuning thedevice photoresponse.ResultsDevice configurations and their signal-to-noise ratiosWe investigate five devices with their specifications and fabricationprocedures described in Supplementary Note 1 and Methods, respec-tively. Initially, the optical response of these high-quality polaritonicnanoresonators is studied using FTIR, serving as a control experiment.For this purpose, we fabricate devices 1 and 4 that are used exclusivelyfor FTIR measurements owing to their large area requirement (opticalactive area of ~ 30 × 30μm2, see optical imageof device 1 in the inset ofFig. 1c and Supplementary Fig. 1 for device 4) and with a device con-figuration containing a single backgate to achieve uniform doping ingraphene tomaximize the optical response (see Supplementary Fig. 2,which indicates that the damping rate decreases with the increase inthe Fermi level). However, this gating configuration does not allow thecreation of a pn-junction for efficient photodetection36. Figure 1a (leftpanel) shows the schematic of devices 1 and 4, which consist of tens ofnanometers wide metallic nanorods placed on top of hBN-encapsulated graphene. Upon illumination, scattering at the metallicrod array launches polaritons that propagate across the 2D10 μmxb cdBABA30 μmSimulationnoise levelefSourceniarDGrating Gate 2Grating Gate 1VGG1VGG210 μmExperimentcyxaConfiguration 1(FTIR)Configuration 2(Electrical spectroscopy)mid-IRlighttop gratings tophBNgraphenebottomhBNMCTdetectorbottomgratinggate 2bottomgratinggate 1sourcedrainyDev 2Dev 3Dev 1Dev 5 Dev 41 μmz1-T /T(%)CNP1-T/T(%) CNPFig. 1 | Configurations of devices, transmission and photocurrent measure-ments, and optical simulation. a Schematic representation of the measureddevices (not to scale) consisting of two main configurations depending on theexperiment. The 1st configuration (left panel) corresponds to that used exclusivelyfor transmissionmeasurements in Fourier Transform Infrared spectroscopy (FTIR),with the top metallic nanorods and the 2D stack below. A mercury-cadmium-telluride (MCT) detector is required to perform mid-infrared (mid-IR) spectro-scopy. Devices 1 and 4 have this configuration. The 2nd configuration (right panel)consists of two grating bottom gates with a top 2D stack. Devices 2, 3, and 5comprise this 2nd configuration. Devices 2 and 3 are measured using electricalspectroscopy, whereas device 5 is measured using FTIR. b The signal-to-noise ratio(SNR) of the five devices was measured using the corresponding technique, eitherby FTIR or photocurrent measurements (electrical spectroscopy). The noise leveldashed line corresponds to an SNR of 1. The inset shows a Scanning electronmicroscopy (SEM) image of the metallic nanorod array with a central gap forconfiguration 2, corresponding to devices 2, 3, and 5. c Extinction (1-T/TCNP)spectrumof device 1measured using FTIR,whereT andTCNP are the transmittancesof the device at a certain gate voltage and at charge neutrality point (CNP),respectively. The curves correspond to several Fermi levels, as indicated in thelegend. The inset shows the optical image of the device 1. The white scale barcorresponds to 30μm. The three columns above the 2D stack are arrays of 100 nmwide metal nanorods with a 50nm gap between them. A and B arrows indicate thepolaritonic resonances described in the main text. d Finite-difference time-domain(FDTD) simulated extinction spectra of device 1 for several Fermi levels. e Opticalimage and device circuitry of configuration 2, which corresponds to device 3 usedfor photocurrent measurements. f Scanning photocurrent map (in absolute value)of device3 at the incidentwavelength (λ) of 6.6μm.Thegates are set toGG1 at 0.4 Vand GG2 at −0.25V, thus creating a pn-junction.Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 2www.nature.com/naturecommunicationsheterostructure6,30–32,34. The graphene channel is uniformly doped byusing a silicon backgate. To obtain a higher yield of fabrication of themetallic nanorods, we pattern them prior to the transfer of the 2Dstack in devices 2 and 3, which are used exclusively for photocurrentmeasurements because theydonot require a large optically active area(e.g., device 2 area of ~ 6 × 3μm2, see right panel of Fig. 1a and Sup-plementaryNote 1 formoredetails). Themetal nanostructures of theselatter devices have a central gap (shown in the inset Fig. 1b) to split thearray and gate the two graphene regions independently36, thus creat-ing a pn-junction. However, the grating gates produce a non-uniformelectrostatic profile44–46, which affects the damping rate and opticalresponse, as shown in Supplementary Fig. 2. Device 3 is fabricated onan infrared transparent substrate (CaF2) to investigate the hybridizedpolaritons in more detail since it avoids the presence of phononpolaritons near the hBN RBs spectral regions. We note that device 5 isbased on configuration 2, but it is measured by FTIR, as shown inSupplementary Note 2.The signal-to-noise ratio (SNR) values of the devices, measuredusing FTIR and electrical spectroscopy, are exhibited in Fig. 1b as afunction of the device area. We notice that the devices measured byelectrical spectroscopy show an SNR of 1 to 2 orders of magnitudehigher than those measured by FTIR. For instance, device 5 shows anSNR of 1 because of its small area of 60μm2 for the FTIR requirements,whereas device 2 achieves an SNR of ~ 100 despite having the smallestarea of 18μm2. These findings highlight the advantages of using elec-trical spectroscopy over standard FTIR for measuring polaritonicresonances in small active area devices. Further details of this com-parison are provided in Supplementary Note 2.Optical spectroscopy in the mid-infrared rangeFirstly, the optical response of the polaritonic nanoresonators isexamined using FTIR to determine the extinction 1 − T/TCNP, where Tand TCNP are the transmittances of the device at a certain gate voltageand at charge neutrality point (CNP) respectively6,25,32. Figure 1c dis-plays the extinction spectra for several Fermi energies.We identify twomain peaks that exhibit a graphene plasmonic behavior, whichincrease their amplitude and blue shift (e.g., ≈0.15μm for peak B) forincreasing Fermi level6. Figure 1d depicts the simulated extinctionusing finite-difference time-domain (FDTD) as described in ref. 36 andsemi-analytical rigorous coupled-wave analysis (RCWA) described inSupplementary Note 3. We observe excellent qualitative and quanti-tative agreement with the experimental results that we also supportwith the dispersion relation of the polaritonicmodes present in device1, as shown in Supplementary Fig. 3, Supplementary Note 3, and byshowing the results of device 4 in Supplementary Fig. 5, thereforevalidating our theoreticalmodel that will be explained in further detailbelow. The slightly lower experimental values are likely due to peakbroadening caused by the inhomogeneity of metal rods’ periodicity.This geometrical disorder impacts the scattering time of charge car-riers in graphene, thus causing a decrease in peak intensity (see Sup-plementary Fig. 4). The absorption spatial profiles at the wavelengthsof the measured peaks show a hybridized plasmon-phonon polaritonfor peak A, however, in peak B we do not observe a clear hybridizationsince the graphene plasmon mode resonates at its plane withoutinterfering with the hBN HPPs (see Supplementary Fig. 6).Electrical detection of polaritonic nanoresonatorsAfter determining the optical response and validating the theoreticalmodel, we perform photocurrent spectroscopy measurements forelectrical detection of the polaritonic nanoresonators. First, we per-form amid-infrared scanning photocurrentmapacross the device areashown in the optical image in Fig. 1e with its electrical circuitry. Forthis, we apply the appropriate voltages to dope the graphene regionabove the grating gate 1 (GG1) and 2 (GG2) with opposite polarity,creating a pn-junction in the graphene, for which the photocurrent ismaximum (see Fig. 1f and Supplementary Fig. 7). By tuning the gatevoltages independently, we observe multiple sign changes of thephotocurrent as shown inSupplementary Figs. 8, 9, which is consistentwith the photothermoelectric (PTE) effect15,16,36,47,48 with a responsivityof ~ 10μA/W (see the spectral dependence of the responsivity andnoise equivalent power in Supplementary Figs. 2 and 10 respectively).Further improvements in responsivity can be made through moreoptimized designs36,49.Next, by scanning the wavelength of the source, we spectrallyresolve the resonances of the 2D polaritons from the normalizedphotocurrent spectra of device 2 (see Fig. 2a). The photocurrentspectra are normalized to the spectrum at CNP to probe the Fermienergy-dependent optoelectronic properties. GG1 is set to a fixed lowvoltage (doped region) since the Seebeck coefficient ismaximumcloseto CNP16,36, while GG2 is swept towards high negative voltages (dopedp-type region), thus creating a doping asymmetry in the channel tomaximize the photoresponse. We observe several peaks at the upper(≈ 6–7μm) and lower (≈ 12–13μm) RB of hBN and SiO2 RB (≈ 8–9μm).Some of these peaks at the RBs evolve with the Fermi level, which isascribed to the hybridized plasmon-phononpolaritons.Moreover, twoadditional broader peaks (labeled as 4 and 8) appear at high Fermienergies outside these RBs. Their evolution and amplitude increasewith the Fermi level more pronouncedly than those of the hybridizedones, which is in agreement with previous works6,25.To identify the origin of the resonances in the photocurrentspectrum, we show in Fig. 2b the simulated normalized absorptionusing FDTD and RCWA (see Supplementary Fig. 11). The absorptionand photocurrent are proportionally related via the electronic tem-perature gradient36, as shown in Supplementary Fig. 12. We find anexcellent agreement in terms of spectral position and relative ampli-tude of the peaks. Additional peaks observed in the theoretical curvesare due to resonances of higher-order modes6. We investigate theresonances in more detail by analyzing the field distributions aboveand outside the metal nanorods, as well as between the top and bot-tom layers of hBN, as shown in Fig. 2c–f. It can be seen that the field isconfined to the bottom hBN above themetal, which corresponds to anacoustic graphene plasmon6,18,32. Conversely, between the metalnanorods, the field extends symmetrically in both hBN layers, asexpected for conventional graphene plasmons14,25. Moreover, theelectric field distribution (e.g., the x − component of the field shown inFig. 2g–j) indicates the resonance’s harmonic order. For instance,peaks 4 and 8 show one sign change of the field in one period, asdisplayed in Fig. 2g, j (see also Supplementary Fig. 13), correspondingto the first harmonic. Different harmonic resonances appear at the topand bottom layers of hBN at peak 6 (Fig. 2i), which implies thesuperposition of the hybridized polaritonic modes. Another interest-ing case occurs at peak 5, where one period of the field matches twoperiods of the grating corresponding to a defect resonance (see Fig. 2hand Supplementary Fig. 14).A complementary assessment involves calculating the dispersionrelation of the polaritonic modes present in device 2 by using thetransfer matrix method (TMM) described in Supplementary Note 3.The metallic rod arrays provide the in-plane effective momentumgiven by: kleff = klx,mw=D+ klx,gg=D= 2πn=D, where D, w, and g are theperiods, width of the nanorod, and the gap between them, respec-tively. The parameter l represents the order number of the mode, andn ≥ 1 is the number of harmonic diffraction orders. The above equationencompasses the combination of both acoustic (kx,m, above themetals) and conventional polaritons (kx,g, above the gap)31,34 asdescribed in Supplementary Note 3 and Supplementary Fig. 15. Thenumber of nodes present in the polaritonic field along the verticaldirectiondetermines l, while thenumber of nodes of thefield along thelateral direction over a period determines n (see also SupplementaryFigs. 13 and 16). Noteworthily, l = 0 outside the RBs.We find a generallyexcellent correspondence with the experimental results as shown inArticle https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 3www.nature.com/naturecommunicationsFig. 2k. In the lower RB, for simplicity we consider the fundamentalmode as the dominant, however, the polaritonic field distributionsuggests a superposition of the fundamental with higher-ordermodes.Gate tunability of the hybridized polaritonic nanoresonatorsThe hybridized polaritonic nanoresonators are investigated inmoredetail with device 3, which contains an infrared transparent sub-strate (CaF2) to avoid phonon polaritons in the substrate near thehBN RBs spectral range, in particular at the lower RB (~ 12 μm)50.Following a similar procedure and gates’ configuration explainedpreviously, we tune the GG1 region to a fixed low n-type doping andsweep GG2 towards high p-type doping. At the upper RB (Fig. 3a),we identify several peaks of photocurrent whose amplitudeincreases and their spectral positions blue shift (e.g., 70 nm or15 cm−1 for peak 1’) with the increase of Fermi level. In fact, at thehighest doping (0.35 eV), the value of normalized photocurrent atpeaks 1’–3’ exceeds the values of the CNP curve, where the Seebeckcoefficient is higher36, thus showing the absorption enhancementcaused by the polaritonic resonances. At the spectral location of theTO phonon of hBN, we observe a pronounced contribution at lowFermi level values which reduces its effect at high doping51. Inthe lower RB (Fig. 3b), we observe two main peaks that boosttheir amplitudes at higher Fermi levels, as well as a small blue shift(e.g., 50 nm or 3 cm−1 for peak 5’), corroborating the theoreticalprediction shown in Supplementary Fig. 17.Thephotocurrent peaks’ locations agreewell with the peaks in thesimulated absorption spectra, shown in Fig. 3c, d. We point out a smallredshift of ~ 0.15μm for the spectral position of peaks at the lower RB,as well as a slight broadening of the experimental peaks compared tothe theoretical ones. The latter is given by the inhomogeneity of themetallic rods’ periodicity, as explained previously. Figure 3e, f showsthat the maximum field occurs at the bottom hBN, which is betweengraphene and metallic nanorods, followed by partial transmission ofHPPs rays towards the upper hBN. Hence, we have two different typesof field distributions in each hBN layer, which are affected by thegraphene doping. The dispersion relation of these hybrid polaritonicmodes are shown in Supplementary Fig. 18 anddemonstrates excellentagreement with experimental results.To analyze the nature of this hBN-confined mode, which is con-trolled by the graphene Fermi energy, we compare two systems at theupper RB. The first one corresponds to the structure described earlierfor device 2 (considering polaritons only abovemetal) shown inFig. 4a,b, which include the spectral and spatial distributions of eigenmodesfor the third and fifth modes, respectively. We notice that the thirdmode is characterized by two nodes in the vertical direction, while thefifth one has four nodes. In both cases, one of the nodes is located inAua b428248kdefect resonancest1  harmonicnd2  harmoniczmode 4Auxairλ = 10.2 μm λ = 12.2 μm      λ = 11.9 μm      λ = 13.3 μm      air g h i j4 5 6 8AuAu Au Au Au Au AuEx+-0Experiment SimulationDispersion relation2Electric field intensity - |E|Electric field x-component - Exc d eλ = 12.2 μm      λ = 10.2 μm λ = 11.9 μm      fλ = 13.3 μm      top hBNbot hBNSiO2grapheneair Au Au Au Au Au Au Au Au4 5 678top hBNbot hBNgrapheneairmaxmin 4381 2675Field intensityFig. 2 | Electrical spectroscopy measurements and simulations. a Normalizedphotocurrent spectrum of device 2 at several Fermi energies. The photocurrentspectra are normalized to the spectrum at the charge neutrality point (CNP). Thepolaritonic peaks are labeled by red arrows. The highlighted spectral regions ingreen correspond to the upper and lower reststrahlen bands (RB) of hBN and, inyellow to the SiO2 RB. The curves are offset for clarity. bOptical (FDTD) simulationof the graphene absorption spectrum for different Fermi energies normalized tothe spectrum at CNP. We label the identified peaks in the same manner as theexperimental ones in panel (a). c–f Cross-sectional view of the simulated electricfield intensity normalized to the incident one across a region containing twometalnanorods, for wavelengths 10.2, 11.9, 12.2, and 13.3μmcorresponding to peaks 4, 5,6, and 8, respectively in panel (a). The x − (horizontal) and z − (vertical) directionsare defined in Fig. 1e. The white scale bar corresponds to 40 nm. The calculationsconsider a non-uniformgraphene Fermi level with a value of0.4 eV above themetal(for a detailed doping profile, see Supplementary Fig. 2). g–j Same as panels (c–f),but the simulations instead show the cross-sectional view of the x-component ofthe electric field normalized to the incident one. The black scale bar corresponds to40nm.kDispersion relation of the polaritonicmodeswith the respective harmonicdiffraction orders (2πn/D). The three horizontal dashed lines correspond to thedefect resonance (n = 1/2), first (n = 1), and second (n = 2) diffraction order reso-nances launched by the metal rod array, respectively. The marked red dotsrepresent the experimental values, which the numeric labels are defined in Fig. 2a.The graphene Fermi level is 0.4 eV. At the hBNRBs (green highlighted regions), theblack, blue, orange, andpurple lines correspond to the 1st, 2nd, 3rd, and4th hybridizedpolaritonic modes, respectively. In yellow is highlighted the SiO2 RB.Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 4www.nature.com/naturecommunicationsthe vicinity of graphene. In addition, the fields are mainly confined inthe bottom hBN for both cases. The second system consists of repla-cing the graphene and top hBN layer with a gold film, as shown inFig. 4c, d, alongwith their respective spatial distributions of fields fromthe two first eigenmodes. We observe similarities between the fielddistributions in the bottom hBN layer and those of the analogousgeometry shown in Fig. 4a, b. This trend is further corroborated bycomparing the dispersion curves of both systems, which overlap(see Fig. 4e). This model demonstrates that doped graphene acts as amirror, partially reflecting polaritons in the bottom hBN layer.Two of the key figures of merit for this resonant polaritonic sys-tem are the Q-factor and the mode volume or wavelength compres-sion, as they are relevant for enhancing the light-matter interactions,for example, for sensing applications. However, the quality factorse fDevice 2 Equivalent systemstEquivalent system - 1  moderdDevice 2 - 3  modetop hBNbot hBNgraphenetop hBNbot hBNgraphenendEquivalent system - 2  modethDevice 2 - 5  modeairAuSiO2airAuSiO2Aubot hBNairAuSiO2airAuSiO2Aubot hBNabcdEx0top hBN bot hBNgrapheneAu AuAuhBNSiO2 SiO2-6.4 6.6 6.8 7Wavelength (μm)6.4 6.6 6.8 7Wavelength (μm)6.2 6.26.4 6.6 6.8 7Wavelength (μm)6.2 6.4 6.6 6.8 7Wavelength (μm)6.2+Fig. 4 | Spatial field distribution and dispersion of hBN equivalent geometryand Q-factor spectrum of the 2D polaritonic nanoresonators. Cross-sectionalview of the spatial x − component of the electric field as a function of the wave-length for (a) 3rd and (b) 5thmodeof the hybridized polariton. The graphene dopingis 0.4 eV. The vertical black scale bar corresponds to 10 nm for panels (a–d). c andd correspond to the first and second order mode, respectively, of the bottom hBN(6nm thick) of the equivalent system (E.S.) geometry consisting of the bottomhBNembedded by two gold layers of 10 nm without the presence of graphene and thetop hBN as shown in the illustrations on the top part of the figure. e Dispersionrelation of the investigated modes of the two systems. f Q-factor spectrum of themeasured 2D polaritonic nanoresonators. The regions highlighted in green corre-spond to the RBs of hBN.6.8 7 7.2 7.4 7.6 7.8 8Wavelength (μm)0.850.90.9511.05Normalizedphotocurrent0.20 eV0.23 eV0.26 eV0.29 eV0.32 eV0.35 eV12.2 12.4 12.6 12.8Wavelength (μm)0.40.60.811.2NormalizedAbsorption0.20 eV0.35 eV12.4 12.6 12.8 13 13.2 13.4Wavelength (μm)0.750.80.850.90.95Normalizedphotocurrent0.26 eV0.29 eV0.32 eV0.35 eV6.8 6.9 7 7.1 7.2 7.3 7.4Wavelength (μm)00.511.5Normalize dAbsorpt ion0.20 eV0.35 eVab decfhBN TO phonon1'2'3'4'5'4'5'top hBNbot hBNgrapheneAu AuairCaF2top hBNbot hBNgrapheneAu AuairzxCaF2zxPeak 1'     λ = 6.91 μm Peak 4'     λ = 12.37 μm 1' 2' 3'Fig. 3 | Normalized photocurrent spectra of device 3 at the hBN RBs.a Normalized photocurrent spectra at the upper RB of hBN for several gate vol-tages. The photocurrent spectra are normalized to the spectrum at CNP. Thepolaritonic peaks are labeled by red arrows. The 0.20 eV curve is slightly shifted(divided by 1.05) for illustration. The Fermi energies are presented in absolutevalues. b Same as panel (a) but for the lower RB range. c Optical simulation ofgraphene absorption at the upper RB spectral region for different Fermi energiesnormalized to the spectrum at CNP. We label the identified peaks in the samemanner as the experimental ones. d Same as panel (c) but for the lower RB range.e Cross-sectional view of the electric field intensity normalized to the incident one.The x − (horizontal) and z − (vertical) directions are defined in Fig. 1e. The simu-lations correspond to a non-uniformgraphene Fermi level at 0.35 eV at wavelength6.91μm (corresponding to peak 1' in panel (a). The white scale bar corresponds to20nm. f Same as panel (e) but for a lower RB range at wavelength 12.37μm (cor-responding to peak 4' in Fig. 3b).Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 5www.nature.com/naturecommunicationsdiminish significantly when shrinking the dimensions of the nanor-esonators to aim for deep subwavelength confinement8,34. The pre-sented polaritonic platform overcomes this limitation by showingsimultaneously resonances with Q-factor values up to ~ 200 inside theRBs (see Fig. 4f and Supplementary Fig. 19 for the Q-factor determi-nation), due to the low loss nature of the hybridized polaritons9,14 andan outstanding value of 330 is achieved for the optical lateral con-finement or effective refractive index34 (neff ≃ kp/kin, see Supplemen-tary Fig. 20). In the case of graphene plasmons, we observe lowerQ-factors up to 50, most likely limited by Ohmic losses, the non-uniform electrostatic potential and inhomogeneities in the metalnanorods’ periodicity that act as additional scattering centers thatreduce the resonant peak linewidth, as shown in Supplementary Fig. 2.In addition, we observe that Q-factor does not show significant varia-tion as a function of Fermi level in agreement with other studies25 (seeSupplementary Fig. 21). We point out that the electrical spectroscopyapproach enables the probing of very small nanoresonators (~ 30 nm),which is highly challenging for conventional techniques such ass-SNOM due to the resolution limitations imposed by the typical tipdiameter (~ 50nm)8 or FTIR due to the sizeable area requirement (e.g.,large arrays of nanoresonators)6,31,32,34 as mentioned previously.DiscussionThe investigated electro-polaritonic platform for performing mid andlong-wave infrared photocurrent spectroscopy can be exploited toenhance photodetectors, hyperspectral and sub-diffractionimaging52,53, and electrical detection of molecular vibrations andgases. The zero-bias operation of our device enables low noise and asignificant reduction in power consumption. In addition, it operates atroom temperature, in contrast to the MCT used in standard FTIR,which requires a voltage bias and liquid nitrogen cooling for optimumperformance. We also highlight that our device is CMOS-compatible54,which enables a highly compact platform that satisfies the size, weight,and power consumption (SWaP) requirements55. We further comparethe electrical spectroscopy with standard FTIR in Supplemen-tary Note 2.These devices can also target other frequencies by tuning the sizeof themetallic nanorods and using different hyperbolicmaterials suchas α-MoO356,57, V2O311, and black phosphorous. In particular, the in-plane anisotropy of MoO3 could potentially change the interferenceand hybridization with graphene plasmons, thus modifying its effecton photodetection58.MethodsDevice fabricationThe fabrication of devices 1 and 4, used for the transmission mea-surements, consists of the following: we first exfoliate the top andbottom hBN and the graphene onto freshly cleaned Si/SiO2 substrates,stack them following the Van der Waals assembly technique59,60 andtransfer the hBN/graphene/hBN stack onto a high resistivity Si/SiO2,which is a 50% transparent and gating capable substrate at the mid-infrared wavelengths6,32. We then use electron beam lithography (EBL)with a PMMA 950K resist film to pattern source and drain electrodesand expose the device to a plasma of CHF3/O2 gases to partially etchthe Van der Waals stack. Subsequently, we deposit side contacts ofchromium (5 nm) / gold (60 nm) and lift off in acetone as described inref. 59. Lastly, we pattern the nanorods with a period of 150 nm usingEBL anddeposit titanium (2 nm) / gold (8 nm)with a subsequent lift-offstep in acetone.Devices 2 and 3 were fabricated and designed for photocurrentmeasurements. On device 3, we first pattern the grating gates by usingEBL and deposit titanium (2 nm) / gold (8 nm), followed by a lift-off stepin acetone. Alternatively, for device 2, we pattern with Ga FIB (gallium-focused ion beam) a thin layer of gold deposited on a Si/SiO2 substrate.The dimensions of the metallic nanorods are described inSupplementary Note 1. Following the previously mentioned procedurewe transfer the hBN/graphene/hBN stack onto the grating gates. Then,we pattern the source and drain electrodes with a PMMA 950K resistfilm using EBL and expose the patterned regions to a plasma of CHF3/O2gases to partially etch the Van der Waals stack. Afterwards, we depositside contacts of chromium (5nm) / gold (80nm) and lift off in acetoneas described in ref. 59. Then, we define the hBN-encapsulatedgraphene channel by patterning a PMMA mask with EBL and etchingit using a CHF3/O2 plasma. By performing electrical measurementsusing a 2-terminal configuration as a function of the gate voltages(varying GG1 and GG2 both at the same potential), we obtain3000–15,000 cm2V−1s−1 as a lower bound of the estimated mobility(see Supplementary Fig. 2).MeasurementsFor the transmission measurements of devices 1, 4 and 5, we use acommercial FTIR (Fourier transform infrared) spectrometer (BrukerTensor FTIR with a Bruker Hyperion 2000 microscope) andnitrogen-cooledmercury-cadmium-telluride (MCT) detector, whichits spectral range goes from 6500 to 650 cm−1 (λ = 1.54 to 15.4 μm)under normal incidence in air with p-polarized light (i.e., with inci-dent polarization perpendicularly oriented respect to the main axisof the metallic gratings)6. We use a spectral resolution of 16 nm(4 cm−1). We normalize the transmission spectrum with a referencesignal at the graphene area with the backgate at the charge neu-trality point (CNP) value (~ 0 V).For the photocurrent spectroscopy measurements, we use aquantumcascade laser (QCL)mid and long-wave infrared laser (MIRcatfrom Daylight Solutions) with tunable wavelength ranges from 6.6 to13.6 μmwith a spectral resolution of < 1 cm−1 and it’s linearly polarized.We modulate the light via an optical chopper at 373Hz, and we mea-sure the photocurrent using a lock-in amplifier (Stanford Research).We scan the device position with a motorized xyz-stage. We focus theinfrared light using a reflective objectivewith anNAof 0.5. To calibratethe incident power, we use a thermopile detector from Thorlabsplaced at the sample location.Data availabilityThedata from this study is provided in themanuscript, SupplementaryInformation figures and from the corresponding authors uponrequest.Code availabilityCodes from this work implementing the semi-analytical rigorouscoupled-wave analysis (RCWA) and transfer matrix methods (TMM)(described in the Supplementary Information) are available uponrequest by contacting the corresponding authors.References1. Low, T. & Avouris, P. Graphene plasmonics for terahertz to mid-infrared applications. ACS Nano 8, 1086–1101 (2014).2. Basov, D. N., Asenjo-Garcia, A., Schuck, P. J., Zhu, X. & Rubio, A.Polariton panorama. Nanophotonics 10, 549–577 (2020).3. Caldwell, J. D. et al. Sub-diffractional volume-confined polaritons inthe natural hyperbolic material hexagonal boron nitride. Nat.Commun. 5, 1–9 (2014).4. Basov, D. N., Fogler, M. M. & García De Abajo, F. J. Polaritons in vander Waals materials. Science 354, https://doi.org/10.1126/science.aag1992 (2016).5. Giles, A. J. et al. Ultralow-loss polaritons in isotopically pure boronnitride. Nat. Mater. 17, 134–139 (2018).6. Iranzo, D. A. et al. Probing the ultimate plasmon confinement limitswith a van der waals heterostructure. Science 360, 291–295 (2018).7. Tamagnone, M. et al. Ultra-confined mid-infrared resonant phononpolaritons in van derWaals nanostructures.Sci. Adv. 4, 4–10 (2018).Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 6https://doi.org/10.1126/science.aag1992https://doi.org/10.1126/science.aag1992www.nature.com/naturecommunications8. Herzig Sheinfux, H. et al. High-quality nanocavities through multi-modal confinement of hyperbolic polaritons in hexagonal boronnitride. Nat. Mater. 23, 499–506 (2024).9. Dai, S. et al. Graphene on hexagonal boron nitride as a tunablehyperbolic metamaterial. Nat. Nanotechnol. 10, 682–686 (2015).10. Dai, S. et al. Phonon Polaritons in Monolayers of Hexagonal BoronNitride. Adv. Mater. 31, 1–5 (2019).11. Taboada-Gutiérrez, J. et al. Broad spectral tuning of ultra-low-losspolaritons in a van derWaals crystal by intercalation.Nat. Mater. 19,964–968 (2020).12. Ni, G. et al. Long-lived phonon polaritons in hyperbolic materials.Nano Lett. 21, 5767–5773 (2021).13. Sternbach, A. J. et al. Programmable hyperbolic polaritons in vander Waals semiconductors. Science 371, 617–620 (2021).14. Woessner, A. et al. Highly confined low-loss plasmons in graphene-boron nitride heterostructures. Nat. Mater. 14, 421–425 (2015).15. Woessner, A. et al. Electrical detection of hyperbolic phonon-polaritons in heterostructures of graphene and boron nitride. Npj2D Mater. Appl. 1, 25 (2017).16. Lundeberg, M. B. et al. Thermoelectric detection and imaging ofpropagating graphene plasmons. Nat. Mater. 16, 204–207 (2017).17. Lundeberg, M. B. et al. Tuning quantum nonlocal effects in gra-phene plasmonics. Science 357, 187–191 (2017).18. Alonso-González, P. et al. Acoustic terahertz graphene plasmonsrevealed by photocurrent nanoscopy. Nat. Nano 12, 31–35 (2017).19. Dai, S. et al. Subdiffractional focusing and guiding of polaritonicrays in a natural hyperbolic material. Nat. Commun. 6, 1–7 (2015).20. Ni, G. X. et al. Fundamental limits to graphene plasmonics. Nature557, 530–533 (2018).21. Moore, S. L. et al. Nanoscale lattice dynamics in hexagonal boronnitride moiré superlattices. Nat. Commun. 12, 1–7 (2021).22. Lin, X. et al. Splashing transients of 2D plasmons launched by swiftelectrons. Conference on Lasers and Electro-Optics, CLEO 2017 -Proceedings 1–2 (2017).23. Wachsmuth, P., Hambach, R., Benner, G. & Kaiser, U. Plasmonbands in multilayer graphene. Phys. Rev. B. Condens. Matter Mater.Phys. 90, 1–5 (2014).24. Li, N. et al. Direct observation of highly confined phonon polaritonsin suspended monolayer hexagonal boron nitride. Nat. Mater. 20,43–48 (2021).25. Brar, V. W., Jang, M. S., Sherrott, M., Lopez, J. J. & Atwater, H. A.Highly confined tunable mid-infrared plasmonics in graphenenanoresonators. Nano Lett. 13, 2541–2547 (2013).26. Brar, V. W. et al. Hybrid surface-phonon-plasmon polariton modesin graphene/monolayer h-BN heterostructures. Nano Lett. 14,3876–3880 (2014).27. Rodrigo, D. et al. Mid-infrared plasmonic biosensingwith graphene.Science 349, 165–168 (2015).28. Autore,M. et al. Boronnitride nanoresonators for Phonon-Enhancedmolecular vibrational spectroscopy at the strong coupling limit.Light Sci. Appl. 7, 17172–17178 (2018).29. Kim, S. et al. Electronically tunable perfect absorption in graphene.Nano Lett. 18, 971–979 (2018).30. Bylinkin, A. et al. Tight-binding terahertz plasmons in chemical-vapor-deposited graphene. Phys. Rev. Appl. 11, 1 (2019).31. Lee, I. H., Yoo, D., Avouris, P., Low, T. &Oh, S. H. Graphene acousticplasmon resonator for ultrasensitive infrared spectroscopy. Nat.Nanotechnol. 14, 313–319 (2019).32. Epstein, I. et al. Far-field excitation of single graphene plasmoncavities with ultracompressed mode volumes. Science 368,1219–1223 (2020).33. Hu, H. et al. Gas identification with graphene plasmons. Nat.Commun. 10, 1–7 (2019).34. Lee, I. H. et al. Image polaritons in boron nitride for extremepolariton confinement with low losses. Nat. Commun. 11,3649 (2020).35. Guo,Q. et al. Efficient electrical detection ofmid-infrared grapheneplasmons at room temperature. Nat. Mater. 17, 986–992 (2018).36. Castilla, S. et al. Plasmonic antenna coupling tohyperbolic phonon-polaritons for sensitive and fast mid-infrared photodetection withgraphene. Nat. Commun. 11, 4872 (2020).37. Freitag, M. et al. Photocurrent in graphene harnessed by tunableintrinsic plasmons. Nat. Commun. 4, 1–8 (2013).38. Bandurin, D. A. et al. Resonant terahertz detection using grapheneplasmons. Nat. Commun. 9, 4–11 (2018).39. Safaei, A., Chandra, S., Shabbir, M. W., Leuenberger, M. N. &Chanda, D. Dirac plasmon-assisted asymmetric hot carrier gen-eration for room-temperature infrared detection. Nat. Commun.10,1–7 (2019).40. Bareza, N. J., Gopalan, K. K., Alani, R., Paulillo, B. & Pruneri, V. Mid-infrared gas sensing using graphene plasmons tuned by reversiblechemical doping. ACS Photon. 7, 879–884 (2020).41. Bylinkin, A. et al. Real-space observation of vibrational strong cou-pling between propagating phonon polaritons and organic mole-cules. Nat. Photon. 15, 197–202 (2021).42. Luxmoore, I. J., Liu, P. Q., Li, P., Faist, J. & Nash, G. R. Graphene-metamaterial photodetectors for integrated infrared sensing. ACSPhoton. 3, 936–941 (2016).43. Tielrooij, K. J. et al. Out-of-plane heat transfer in van der Waalsstacks through electron-hyperbolic phonon coupling. Nat. Nano-technol. 13, 41–46 (2018).44. Delgado-Notario, J. A. et al. Enhanced terahertz detection of mul-tigate graphene nanostructures.Nanophotonics 11, 519–529 (2022).45. Ryzhii, V., Otsuji, T. & Shur, M. Graphene based plasma-wavedevices for terahertz applications. Appl. Phys. Lett. 116,140501 (2020).46. Boubanga-Tombet, S. et al. Room-temperature amplification ofterahertz radiation by grating-gate graphene structures. Phys. Rev.X 10, 031004 (2020).47. Massicotte, M., Soavi, G., Principi, A. & Tielrooij, K. J. Hot carriers ingraphene-fundamentals and applications. Nanoscale 13,8376–8411 (2021).48. Viti, L. et al. Thermoelectric graphene photodetectors with sub-nanosecond response times at terahertz frequencies. Nanopho-tonics 10, 89–98 (2021).49. Castilla, S. et al. Fast and sensitive terahertz detection using anantenna-integrated graphene pn junction. Nano Lett. 19,2765–2773 (2019).50. Kischkat, J. et al. Mid-infrared optical properties of thin films ofaluminum oxide, titanium dioxide, silicon dioxide, aluminumnitride, and silicon nitride. Appl. Opt. 51, 6789–6798 (2012).51. Badioli, M. et al. Phonon-mediated mid-infrared photoresponse ofgraphene. Nano Lett. 14, 6374–6381 (2014).52. Ju, L. et al. Tunable excitons in bilayer graphene. Science 358,907–910 (2017).53. Shen, D. et al. High-performance mid-IR to deep-UV van der Waalsphotodetectors capable of local spectroscopy at room tempera-ture. Nano Lett. 22, 3425–3432 (2022).54. Goossens, S. et al. Broadband image sensor array based ongraphene-CMOS integration. Nat. Photon. 11, 366–371 (2017).55. Rogalski, A. Graphene-basedmaterials in the infrared and terahertzdetector families: a tutorial. Adv. Opt. Photon. 11, 314 (2019).56. Ma, W. et al. In-plane anisotropic and ultra-low-loss polaritons in anatural van der Waals crystal. Nature 562, 557–562 (2018).57. Matveeva, O. G. et al. Twist-tunable polaritonic nanoresonators in avan der Waals crystal. Npj 2D Mater. Appl. 7, 1–7 (2023).Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 7www.nature.com/naturecommunications58. Álvarez-Pérez, G. et al. Active tuning of highly anisotropic phononpolaritons in van der Waals crystal slabs by gated graphene. ACSPhoton. 9, 383–390 (2022).59. Wang, L. One-dimensional electrical contact to. Science 342,614–617 (2013).60. Pizzocchero, F. et al. The hot pick-up technique for batch assemblyof van der Waals heterostructures. Nat. Commun. 7, 11894 (2016).AcknowledgementsThe authors thank Hanan Herzig Sheinfux, Krystian Nowakowski, andIacopo Torre for fruitful discussions. F.H.L.K. acknowledges financialsupport from the Spanish Ministry of Economy and Competitiveness,through the “Severo Ochoa” Program for Centers of Excellence in R&D(SEV-2015-0522), support by Fundacio Cellex Barcelona, Generalitat deCatalunya through the CERCA program, and the Agency for Manage-ment of University and Research Grants (AGAUR) 2017 SGR 1656. Fur-thermore, the research leading to these results has received fundingfrom the European Union Seventh Framework Program under grantagreementsno.785219 andno. 881603GrapheneFlagship forCore2andCore3. S.C. acknowledgesfinancial support from the Barcelona Instituteof Science andTechnology (BIST), theSecretaria d’Universitats i Recercadel Departament d’Empresa i Coneixement de la Generalitat de Cata-lunya and the European Social Fund (L’FSE inverteix en el teu futur) -FEDER. N.M.R.P. acknowledges support from the Independent ResearchFund Denmark (grant no. 2032-00045B) and the Danish NationalResearch Foundation (Project No. DNRF165). Y.V.B., M.I.V., and N.M.R.P.acknowledge support by the Portuguese Foundation for Science andTechnology (FCT) in the framework of the Strategic Funding UIDB/04650/2020. K.W. and T.T. acknowledge support from the JSPSKAKENHI (Grant Numbers 21H05233 and 23H02052) andWorld PremierInternational Research Center Initiative (WPI), MEXT, Japan, for thegrowthof h-BNcrystals. Funding for hBNcrystal growthby E.J. and J.H.E.was provided by the Office of Naval Research, Award no. N00014-22-1-2582. F.H.L.K. and S.C. acknowledge funding from the European Union(ERC, POLARSENSE, 101123421). Views and opinions expressed are,however, those of the author(s) only and do not necessarily reflect thoseof the European Union or the European Research Council ExecutiveAgency. Neither the European Union nor the granting authority can beheld responsible for them.Author contributionsS.C., D.A.I., and F.H.L.K. conceived the project. S.C. and H.A. fabricatedthe devices and performed the experiments. D.A.I. and A.G. assisted inthe experiments. M.C. and R.K.K. supported the device fabrication. I.V.,Y.V.B., N.M.R.P., and E.L. performed the simulations and developed thetheoretical model. S.C. and M.I.V. assisted in the modeling. S.C., I.V.,Y.V.B., E.L., and F.H.L.K. wrote themanuscript. K.W. and T.T. synthesizedthe hBN crystals. E.J. and J.H.E. synthesized the isotopically enrichedhBN crystals. N.M.R.P., E.L., and F.H.L.K. supervised the work and dis-cussed the results. All authors contributed to the scientific discussionand manuscript revisions. S.C., H.A., I.V., and Y.V.B. contributed equallyto the work.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s41467-024-52838-w.Correspondence and requests for materials should be addressed toSebastián. Castilla or Frank H. L. Koppens.Peer review information Nature Communications thanks the anon-ymous reviewers for their contribution to the peer review of this work. Apeer review file is available.Reprints and permissions information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard to jur-isdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format, aslong as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in thisarticle are included in the article’s Creative Commons licence, unlessindicated otherwise in a credit line to the material. If material is notincluded in the article’s Creative Commons licence and your intendeduse is not permitted by statutory regulation or exceeds the permitteduse, you will need to obtain permission directly from the copyrightholder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.© The Author(s) 2024Article https://doi.org/10.1038/s41467-024-52838-wNature Communications |         (2024) 15:8635 8https://doi.org/10.1038/s41467-024-52838-whttp://www.nature.com/reprintshttp://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/www.nature.com/naturecommunications Electrical spectroscopy of polaritonic nanoresonators Results Device configurations and their signal-to-noise ratios Optical spectroscopy in the mid-infrared range Electrical detection of polaritonic nanoresonators Gate tunability of the hybridized polaritonic nanoresonators Discussion Methods Device fabrication Measurements Data availability Code availability References Acknowledgements Author contributions Competing interests Additional information