# Fileset

[Electroluminescence from a phthalocyanine monolayer encapsulated in a van der Waals tunnel diode.pdf](https://mdr.nims.go.jp/filesets/58023a7e-705a-4f54-81b6-715c5a7049a3/download)

## Creator

Tyler James, Jonathan Bradford, James Kerfoot, Vladimir V. Korolkov, Manal Alkhamisi, [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), Anton S. Nizovtsev, Elisa Antolín, Elena Besley, Simon A. Svatek, Peter H. Beton

## Rights

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

## Other metadata

[Electroluminescence from a phthalocyanine monolayer encapsulated in a van der Waals tunnel diode](https://mdr.nims.go.jp/datasets/6111bfab-9700-4d2c-bd14-6065e5f599f3)

## Fulltext

Electroluminescence from a phthalocyanine monolayer encapsulated in a van der Waals tunnel diodeFull Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=tmph20Molecular PhysicsAn International Journal at the Interface Between Chemistry andPhysicsISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tmph20Electroluminescence from a phthalocyaninemonolayer encapsulated in a van der Waals tunneldiodeTyler James, Jonathan Bradford, James Kerfoot, Vladimir V. Korolkov, ManalAlkhamisi, Takashi Taniguchi, Kenji Watanabe, Anton S. Nizovtsev, ElisaAntolín, Elena Besley, Simon A. Svatek & Peter H. BetonTo cite this article: Tyler James, Jonathan Bradford, James Kerfoot, Vladimir V. Korolkov,Manal Alkhamisi, Takashi Taniguchi, Kenji Watanabe, Anton S. Nizovtsev, Elisa Antolín, ElenaBesley, Simon A. Svatek & Peter H. Beton (2023) Electroluminescence from a phthalocyaninemonolayer encapsulated in a van der Waals tunnel diode, Molecular Physics, 121:7-8,e2197081, DOI: 10.1080/00268976.2023.2197081To link to this article:  https://doi.org/10.1080/00268976.2023.2197081© 2023 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup.View supplementary material Published online: 13 Apr 2023. Submit your article to this journal Article views: 369 View related articles View Crossmark datahttps://www.tandfonline.com/action/journalInformation?journalCode=tmph20https://www.tandfonline.com/loi/tmph20https://www.tandfonline.com/action/showCitFormats?doi=10.1080/00268976.2023.2197081https://doi.org/10.1080/00268976.2023.2197081https://www.tandfonline.com/doi/suppl/10.1080/00268976.2023.2197081https://www.tandfonline.com/doi/suppl/10.1080/00268976.2023.2197081https://www.tandfonline.com/action/authorSubmission?journalCode=tmph20&show=instructionshttps://www.tandfonline.com/action/authorSubmission?journalCode=tmph20&show=instructionshttps://www.tandfonline.com/doi/mlt/10.1080/00268976.2023.2197081https://www.tandfonline.com/doi/mlt/10.1080/00268976.2023.2197081http://crossmark.crossref.org/dialog/?doi=10.1080/00268976.2023.2197081&domain=pdf&date_stamp=2023-04-13http://crossmark.crossref.org/dialog/?doi=10.1080/00268976.2023.2197081&domain=pdf&date_stamp=2023-04-13MOLECULAR PHYSICS2023, VOL. 121, NOS. 7–8, e2197081 (9 pages)https://doi.org/10.1080/00268976.2023.2197081RESEARCH ARTICLEElectroluminescence from a phthalocyaninemonolayer encapsulated in a van derWaals tunnel diodeTyler Jamesa, Jonathan Bradford a, James Kerfoota, Vladimir V. Korolkova, Manal Alkhamisia,b,Takashi Taniguchic, Kenji Watanabe d, Anton S. Nizovtseve,f , Elisa Antolíng, Elena Besley e,Simon A. Svatek a,g and Peter H. Beton aaSchool of Physics and Astronomy, University of Nottingham, Nottingham, UK; bPhysics Department, College of Science and Art, KingAbdulaziz University, Rabigh, Saudi Arabia; cInternational Center for Materials Nanoarchitectonics, National Institute for Materials Science,Tsukuba, Japan; dResearch Center for Functional Materials, National Institute for Materials Science, Tsukuba, Japan; eSchool of Chemistry,University of Nottingham, Nottingham, UK; fNikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russian Academy of Sciences,Novosibirsk, Russia; gInstituto de Energía Solar, Universidad Politécnica de Madrid, Madrid, SpainABSTRACTMonolayers of free base phthalocyanine (H2Pc) are grown on monolayer and few-layer exfoliatedflakes of hexagonal boron nitride (hBN) which are subsequently integrated into a van derWaals tun-nel diode. This heterostructure consists of two thin hBN flakes betweenwhich the H2Pcmonolayer issandwiched and also incorporates upper and lower few-layer graphene contacts. When a voltage isapplied between the contacts, a tunnel current flows and the embedded molecules can be excitedresulting in the emission of photons with wavelengths which are close to the peaks observed inphotoluminescence. We also observe electroluminescence at voltages where the energy gained bya tunnellingelectron is lower than theenergyof theemittedphoton implyingamulti-electronexcita-tion pathwaywhichwe attribute to the formation of an intermediate triplet state. Our results provideinsights into the differences in excitation and relaxation of molecules in supramolecular monolayersandbulk crystals andwediscuss how the alignment of the energy levels of themolecules and contactlayers determine the emission process.ARTICLE HISTORYReceived 30 September 2022Accepted 24 March 2023KEYWORDSPhthalocyanine;electroluminescence; van derWaals heterostructure; twodimensional molecular selfassemblyThe integration of monolayers of planar organicmolecules with van der Waals heterostructures enablesthe formation of molecular/2D hybrid tunnel diodesand related structures with novel optoelectronic prop-erties [1–5]. Such vertically stratified structures are ofgreat interest as they further enrich the set of param-eters addressable in the exploration of van der Waalsheterostructures through control over the molecularCONTACT Simon A. Svatek simon.svatek@upm.es School of Physics and Astronomy, University of Nottingham, Nottingham, NG7 2RD, UKInstituto de Energía Solar, Universidad Politécnica de Madrid, Avenida Complutense 30, Madrid 28040, Spain; Peter H. Betonpeter.beton@nottingham.ac.uk School of Physics and Astronomy, University of Nottingham, Nottingham, NG7 2RD, UKSupplemental data for this article can be accessed here. https://doi.org/10.1080/00268976.2023.2197081structure and their arrangement. Applications whichhave already been identified include the modification ofmolecular arrangement to favour stronger light–matterinteractions in molecular aggregates [4,6–9], intersystemcrossing in molecular systems [5], environmental pro-tection [10] and modifying the vibrational properties ofmolecules sandwiched at interfaces [11]. The formationof such heterostructures provides myriad directions for© 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the AcceptedManuscript in a repository by the author(s) or with their consent.http://www.tandfonline.comhttps://crossmark.crossref.org/dialog/?doi=10.1080/00268976.2023.2197081&domain=pdf&date_stamp=2023-05-30http://orcid.org/0000-0003-2356-5816http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-9910-7603http://orcid.org/0000-0002-8104-1888http://orcid.org/0000-0002-2120-8033mailto:simon.svatek@upm.esmailto:peter.beton@nottingham.ac.ukhttps://doi.org/10.1080/00268976.2023.2197081http://creativecommons.org/licenses/by/4.0/2 T. JAMES ET AL.research into the detection, harvesting and emission oflight as well as opportunities to manipulate and controlthe optoelectronic properties of nanostructures [12].In this work, we fabricate light-emitting devices basedupon van der Waals heterostructures [13–16] in whichmonolayers of self-assembled free base phthalocyaninemolecules (H2Pc) are incorporated into a hexagonalboron nitride (hBN) tunnel diode [17,18]. In the molec-ular hybrid version of these structures, we have previ-ously shown that the trapped molecules may be excitedinto both spin-singlet and -triplet states which can sub-sequently relax through the emission of photons, thusforming an electroluminescent device [5,19]. The closeproximity (< 1 nm) between the organic emissive layerand the contact layers in these devices is analogous toscanning tunnelling microscopy luminescence (STML)experiments in which electroluminescence from organicmolecules, separated from a metallic substrate by ultra-thin alkali halide layers, is induced by tunnelling betweenthe STM tip and substrate [20–26]. This architectureis distinct from that of conventional organic electrolu-minescent devices, where carriers are typically injectedfrom remote contacts with an engineered work func-tion and form excitons which diffuse to an emissivecentre and recombine [27,28]. In our previous studieswe have investigated electroluminescence from hybriddevices with trapped perylene tetracarboxylic di-imide(PTCDI) molecules [5] and, also, monolayers of poly-thiophene adsorbed with their conjugated backbonealigned parallel to the hBN tunnel barriers [19,29].In our current paper we study a device incorporat-ing a H2Pc monolayer. This molecule differs signifi-cantly from PTCDI since the frontier orbitals, particu-larly the highest occupied molecular orbital (HOMO),lie at much higher energies. We argue that this differ-ence in energy levels leads to an alternative relaxationmechanism for the triplet state of H2Pc. Our work showsthe importance of the alignment of energy levels inthe control of the lifetime of excited states of trappedmolecules.We use polymer stamp-assisted van der Waals assem-bly [30–33] to fabricate our devices. In this approach,flakes of few-layer graphene (FLG) and hBN are sequen-tially picked up to form a van der Waals stack in whicha tunnel barrier is positioned between two FLG layerswhich can be independently contacted. To fabricate adevice we first use a polymer stamp to pick up a large(lateral dimensions 10 s of µm) thick (10 s of nm) hBNflake, which is ultimately used to cap the device. This hBNflake is then used to pick up a FLG flake which servesas the top contact of the completed device, followed byan hBN flake with thickness of, typically, 1–3 monolayerswhich forms the upper tunnel barrier. This van derWaalsstack is then used to pick-up a second hBN tunnel bar-rier, also with a thickness of 1–3 monolayers, on which aH2Pc monolayer has been pre-deposited by sublimationas described previously [34]. This part-formed tunnellingdevice is then released from the stamp onto a second FLGflake which forms the lower contact and has been pre-deposited on a supporting substrate (Si/SiO2); the releasesite is chosen so that the upper and lower graphene layersmake independent contact with two pre-formed metallic(Cr/Au) electrodes which can be connected to externalwiring. Further details are provided in SupplementaryInformation (SI).The completed device consists of a monolayer of H2Pcsandwiched between few layer hBN and overlaid FLGcontacts, as depicted in Figure 1(a) (the upper thick hBNlayer, and a supporting thick hBN flake which provides asupporting substrate for the lower graphene are omittedfor clarity; neither plays an active role in the device oper-ation). The FLG provides transparent top- and bottom-electrodes allowing carrier transport through the H2Pcmonolayer under an applied bias.Images acquired using high-resolution atomic forcemicroscopy (AFM) of a sublimed H2Pc monolayer onan hBN flake are shown in Figure 1(b) and reveal theexpected [34] near-square regular array of flat-lyingmolecules (the deposition and characterisation of H2Pcand preparation of the hBN surface prior to incorpo-ration in the device follow our previous work [34] andare discussed in SI; the structure of the molecule isshown in Figure 1(a)). An optical image of a com-pleted device is shown in Figure 1(c); the active areaof the device is marked and corresponds to the regionwhere the upper and lower FLG layers overlap. Thedifferent layers are resolved more clearly in the imageacquired using scanning electron microscopy in Figure1(c) inset; here the hBN and FLG are highlighted. Over-all, this device corresponds to a heterostructure with thesequence hBN/FLG/hBN/H2Pc/hBN/FLG/hBN.The bottom and top FLG are contacted to Cr/Au elec-trodes which are shown, respectively, on the left and theright side of the optical image (Figure 1(c)). The cur-rent–voltage characteristics of the device were measuredin an optical cryostat at a temperature T = 6± 2K, andare highly non-linear as expected for a tunnelling device(see Figure 1(d); note the near-exponential rise in currentrevealed in the inset to Figure 1(d)). For this device thecurrent–voltage dependence is highly symmetric withrespect to polarity which indicates that the upper andlower barriers have the same thickness. The effective areaof the device is estimated to be 26± 2 µm2.The photoluminescence from the active region of ourdevice has been measured (see inset in Figure 1(e); exci-tation laser 405 nm focussed using a 50x objective lensMOLECULAR PHYSICS 3Figure 1. FLG/hBN/H2Pc/hBN/FLG heterostructures. a Schematic of device in which a monolayer of H2Pc is encapsulated between twohBN tunnel barriers and charge can be transported between upper and lower FLG contacts under the application of an applied bias;lower left – schematic of the molecular structure of H2Pc. b AFM image of a monolayer coverage of H2Pc on hBN. Inset, high resolutionimagewith schematic overlay. c Optical image of a device showing gold contacts and the van derWaals heterostructure. The dashed lineindicates the active area of the device where the upper and lower FLG layers overlap. Inset top: SEM image of the highlighted area infalse colour showing the FLG (grey) and hBN layers (blue). Inset bottom: scanning electroluminescence image taken at a+2.2 V bias and1 s integration time confirming that light is emitted from the active area of the device. d current–voltage (black) and differential current-voltage (blue) characteristics. Inset: current-voltage on a logarithmic scale. e Electroluminescence spectrum. Inset, black: experimentalPL spectrum, grey: fit using three Lorentzian curves with peak positions 704, 720 and 741 nm. Scale bars b 15 nm, b inset 2 nm, c mainimage and insets 10µm.to a spot with a diameter ∼ 2µm – see SI for fur-ther details) and may be decomposed into three peaksat wavelengths 704± 1, 720± 1 and 741± 2 nm (cor-responding respectively to 1.76, 1.72 and 1.67 eV). Thepeaks at 704 and 741 nm are close to the values recentlydetermined for, respectively, the 0–0 direct (706± 2 nm)and 0–1 vibronic (742± 2 nm) transitions in the pho-toluminescence spectrum of an H2Pc monolayer [34]adsorbed on uncapped hBN. The peak at 720 nm is notobserved in the spectrum for uncapped H2Pc, and wesuggest that this feature arises from molecules which aretrapped between the hBN layers. The red-shift of the 0–0peak of the encapsulated molecules by ∼ 20 nm, com-pared to H2Pc on uncapped hBN, represents a shift inphoton energy ∼ 40meV.It is interesting to compare the energies/wavelengthsof the 0–0 direct transition of H2Pc in the gas phase [35](1.876 eV/661 nm), adsorbed as a flat-lyingmonolayer onhBN [34] (1.76 eV/706 nm) and as a flat-lying monolayersandwiched between two hBN layers (1.72 eV/720 nm) asin the current work. We have shown previously that [4]a red-shift (relative to the gas phase) is expected whena planar organic molecule is absorbed on an insulator.This is due to the dielectric effect of the substrate whichresults in a non-resonant red-shift due to the screen-ing of the electrostatic interactions within the adsorbedmolecule giving rise to changes to the frontier energylevels and relevant electron–hole interactions. The tran-sition energies for molecules in the gas phase, adsorbedon uncapped hBN and trapped between two hBN layerscan be calculated using time-dependent density func-tion theory (TDDFT) as described in SI. The predictedtransition energies are 1.84, 1.75 and 1.68 eV for H2Pcin, respectively, the gas-phase, adsorbed on uncappedhBN and trapped between two hBN layers. These val-ues are in good agreement with experimental observa-tions and reproduce the trend of progressively largerred-shift in the different hBN environments. In our pre-vious work we have also noted [4,34] that additionalresonant shifts occur due to the coupling of the transi-tion dipole moment with image charges in the dielectricsubstrate, and also through coupling with the transitiondipoles of neighbouring molecules. However, for H2Pcon hBN these effects lead to red-shifts which are much4 T. JAMES ET AL.smaller than the non-resonant effects [34] and so are notconsidered further here.The devices emit light when a current flows. Figure1(e) shows the electroluminescence spectrum whichreveals a dominant peak at 723± 1 nm with a width of15 nm (this measurement is derived from light emit-ted from the whole device area). This peak is attributedto electron–hole radiative recombination by electricallyexcited H2Pc molecules. This is further confirmed by amap of the electroluminescence showing the spatial vari-ation of the peak intensity – see Figure 1(c) inset – whichconfirms that the active region of the device, where theFLG/hBN/H2Pc/hBN/FLG overlap, is the source of thephoton emission. The electroluminescence peak emis-sion is close to the position of the peak in PL at 720 nm,which is attributed above to encapsulated molecules.Note that only the encapsulated molecules would beexpected to couple strongly to the FLG contact layers, sothe additional peaks in PL at 704 and 742 nm would beexpected to appear only in PL and not EL, in agreementwith our observations. It is likely that the PL peaks at 704and 741 nm are either due to emission from moleculesfrom surrounding regions which are excited by scat-tered incident light, ormolecules which are not efficientlytrapped between the hBN layers due to inhomogeneitiesin the fabricated device, so that their emission is charac-teristic of uncapped H2Pc. We also observe a small peakin the EL spectrum at 689± 2 nm which we attribute toan anti-Kasha (1–0) vibronic transition (a similar peakhas recently been reported in STML studies of napthalo-cyanine [36]).We thus attribute the EL emission to radia-tive electron–hole recombination by electrically excitedH2Pc molecules encapsulated with hBN.The spectra recorded at different voltages are pre-sented in the form of a colour map in Figure 2 whichshows that the position and line width of the EL peakis voltage-independent. The integrated EL intensity forthis device is also shown in Figure 2. The EL spectra donot show a strong polarity dependence (in our measure-ments the voltage is applied to the top contact); additionalresults on voltage dependence are included in SI (FiguresS4 and S5).We have also investigated EL at voltages close to theonset for photon emission as shown in Figure 3.Measure-ments at each voltage were performed with acquisitiontimes ranging from 100s to 600 s; the longer acquisitiontimes were required for the detection of weaker EL emis-sion at smaller (absolute) bias voltages. An EL emissionpeak located at 720± 4 nm can be observed in Figure 3for each bias voltage value in the range −1.6V to −1.9V.This peak position is consistent with emission from theencapsulated H2Pc monolayer, as discussed above. Inter-estingly, EL is detected at bias voltages of −1.6V andFigure 2. Electroluminescence in FLG/hBN/H2Pc/hBN/FLG het-erostructures measured at room temperature. Colour map of theEL spectra as a function of applied bias. White circles, right axis:integrated intensity of EL over thewavelength range700–750 nm.Figure 3. Electroluminescence spectra acquired at T = 6 ± 2 Kfrom a FLG/hBN/H2Pc/hBN/FLG heterostructure device for a seriesof bias voltages ranging from −1.5 to −1.9 V. The bias voltageand acquisition time for each spectrum is annotated in the corre-sponding colour. The spectra acquired at−1.6 and−1.7 V exhibitphoton up-conversion.−1.7V (our detection of EL peaks at lower voltage is lim-ited by noise in our CCD detection electronics; a veryweak peak at −1.5V cannot be ruled out). In this voltagerange the energy of the emitted photons, hν > eVSD, theenergy gained by the tunnelling electrons due to the biasvoltage between the contacts. This is an example of pho-ton up-conversion and implies that the excitation of themolecule into the singlet state involves a multi-electronprocess. Up-conversion has previously been reported forsimilar devices containing an encapsulated monolayer ofPTCDI molecules [5] and also in STML studies of H2Pc[24].MOLECULAR PHYSICS 5In similar devices which incorporate a PTCDI mono-layer, it has been shown that the up-conversion process ismediated by an intermediate triplet state; this hypothesiswas based on the fact that the lifetime of any intermediatestate should be longer than the typical time between suc-cessive electron tunnelling events (which can be deducedfrom the current density and the molecular dimensions).The role of triplet intermediates was confirmed by thepresence of an additional peak in the PTCDI EL spec-trum due to triplet emission. In the proposedmechanismthe PTCDI was first excited to a triplet state, and then,in a second inelastic event, excited from the triplet tothe excited singlet state. In the H2Pc analogue deviceconsidered here we observe a similar up-conversion andpropose that a similarmechanism is responsible.We notethat photon emission from H2Pc is observed at a mea-sured current density of 6.5 μA μm−2 (at VSD = −1.6V).Taking the area per molecule to be 2.25 nm2, we estimatea lower bound for the average time interval between elec-trons tunnelling through an effectivemolecular area to be11± 2 ns. The lifetime of the H2Pc triplet state is muchlonger, of the order 130 μs [37], confirming that this statecan participate as an intermediary in a multi-electroninelastic scattering pathway (note that the lifetime ofother excited states, such as vibronic states are expected tobe much shorter). In the proposed mechanism, a 2-stepinelastic tunnelling process leads to up-conversion. Thefirst inelastic scattering event excites the molecule intothe triplet state (T1). This state is sufficiently long-livedthat during the triplet lifetime further inelastic tunnellingevents can occur, promoting the excited electron fromthe T1 state into the first excited singlet state (S1), whichrapidly relaxes via the emission of a photon.As expected for a multi-electron process we observea super-linear dependence of intensity on current, I, forlow voltages (intensity ∝ I2.1 see Figure 4(a)) similar tothat reported in recent STML studies [24] of H2Pc whichwere also attributed to triplet-mediated up-conversion.We also observe a significant change in the dependenceof intensity on current at a voltage of ∼2.54V. This isapparent in Figure 4(a) and, consequently, correspondsto a reduction in the EL efficiency (photons emitted pertunnelling electron) as shown in Figure 4(b). Withinthe model proposed above this implies the appearanceof a voltage-dependent mechanism for the relaxation ofintermediate triplet states. Interestingly, this behaviour isnot observed in the analogue PTCDI device which wasreported previously [5] for which the efficiency increasesmonotonically over the complete voltage range underinvestigation. It is possible that this difference arisesfrom the relative position of the energy levels of the twomolecules. PTCDI is an n-type organic semiconductor,whereas H2Pc is p-type, meaning that the HOMO ofH2Pc lies at a much higher energy compared to PTCDI.Energy level diagrams of the H2Pc and PTCDI devicesunder 0 and 1.6V bias are shown in Figure 5(a–c). Theseenergy level diagrams illustrate the relative energies of theHOMO, LUMO, S1, and T1 states of H2Pc and PTCDIcompared to the Fermi levels of the FLG electrodes. Theenergies of the S1 and T1 states were obtained, respec-tively, from experimental values and the literature [37],and the HOMO/LUMO values determined from theo-retical calculations (see SI). In the absence of an appliedvoltage (Figure 5(a)), both the S1 and T1 energy levels ofH2Pc lie below the Fermi level of the FLG electrodes andno current flows through the device. When the device isunder bias, however, the relative positions of the energylevels of the FLG and H2Pc shift due to the applied volt-age. This is demonstrated schematically in Figure 5(b),while Figure 5(c) shows an equivalent diagram for aPTCDI device. The application of an applied voltage leadsto changes in the carrier concentrations in each electrodeand, consequently, a shift in the Fermi level of the top andbottom (EFB) electrodes and, in addition, a relative shiftFigure 4. a Dependence of EL intensity on current; b Dependence of efficiency on applied voltage. Data are acquired at roomtemperature.6 T. JAMES ET AL.Figure 5. Energy level diagrams for a H2Pc tunnelling device under zero bias (a), 1.6V bias (b), and a PTCDI tunnelling device under 1.6Vbias (c). Inelastic electron scattering (IES) events are labelled. (a) Under zero bias both the H2Pc singlet, S1, and triplet, T1, states lie belowthe Fermi energy of the bottom graphene electrode, EFB . (b) At 1.6V bias, T1 lies 100meV above EFB . Up-converted electroluminescencefrom S1 is observed at 1.6V bias. (c) The lower-lying energy levels of PTCDI mean that under 1.6V bias, both S1 and T1 lie below EFB andemission is observed from both states.of the Dirac point such that the total energy differencebetween the FLG Fermi levels is equal to eVSD(see SI fordetails of these calculations). Themolecular energy levelsare also shifted, and, assuming that the hBN barriers havethe same thickness, the energy levels are shifted by halfthe bias voltage (eVSD/2) relative to their position underzero bias.From a comparison of Figure 5(a,b), it can be seenthat the relative alignment of the H2Pc T1 level andEFBchanges significantly when a 1.6V bias is applied. InFigure 5(a), T1 lies 0.7 eV below EFB , whereas in Figure5(b), the shifting of energy levels results in T1 lying0.1 eV above EFB . In contrast, since the frontier energylevels of PTCDI are much lower-lying than the corre-sponding states in H2Pc, the PTCDI T1 and S1 energylevels lie below EFB(Figure 5(c)) under a bias of 1.6 V. Itis possible that an additional non-radiative pathway fortriplet relaxation could occur for voltages at which thetriplet energy is raised above the Fermi energy of theadjacent electrode, for example due to ionisation, whichmight account for the reduction in efficiency observedat high voltage (Figure 4(b)). From our simple modelpresented in Figure 5 we predict that the triplet state isin alignment with the Fermi level for voltages of 1.4 V(H2Pc) and 3.8V (PTCDI). While this is consistent withpresence (absence) of reduction in efficiency for H2Pc(PTCDI), the measured voltage position of the peak effi-ciency of H2Pc is significantly higher than the calcu-lated threshold for triplet/Fermi level alignment. Thisdifference may be due to uncertainties in the valuesused for the HOMO and triplet energies, but may alsobe affected by additional energy barriers due to elec-trostatic effects, as well as dynamic factors involvingtunnelling rates which are also expected to be voltage-dependent. Implicit in our argument is that, since up-converted emission from the singlet state is observed,the triplet relaxation time of this additional pathway islonger than the time between successive electron traver-sal time (11 ns) but shorter than the lifetime of the tripletstate. This implies a prospective relaxation time, tn−r, forthis non-radiative pathway in the range, 10 ns <∼ tn−r <∼130µs. Notably, this relaxation time is much longer thanthe lifetime of the singlet state (∼250 ps) [38] and sowould not be expected to suppress singlet emission. Wealso note that the efficiency of our H2Pc devices, typ-ically 10−9–10−10 photons per electron, is lower thanthat of analogue PTCDI devices, (10−6–10−8 photons perelectron [5]).Our results show that electroluminescence frommolecular/2D van der Waals heterostructures can occurfor organicmolecules with widely varying frontier orbitalenergies. This suggests a commonmechanism for excita-tion and relaxation of the molecules and the observationof photonup-conversionwhichwe observe for bothH2Pcand PTCDI is suggestive of a long-lifetime intermediatestate which, as we argue above, we attribute to a triplet.Interestingly, for both the PTCDI and the H2Pc devicesdiscussed here, a peak in the EL spectrum which canbe assigned to an anti-Kasha (1–0) vibronic transition[36] is observed. Although the peak has low intensity,its presence suggests that molecules may also be vibra-tionally excited due to inelastic energy transfer from theMOLECULAR PHYSICS 7tunnelling electrons so that there is a non-equilibriumoccupation of vibrational levels; this peak is not observedin the PL spectra of either molecule.Despite the similarities between the properties ofour H2Pc and previously published PTCDI devices, theabsence of emission from the triplet state in the currentstudy represents a significant difference. In our previouswork [5] we observed a clear peak in the EL spectrumwhich was assigned to triplet emission. It is possible thatthe additional relaxation pathways for triplets (see above)may account for the absence of a triplet emission peakin the current work, although we cannot rule out otherexplanations such as an intrinsically lower rate for tripletemission fromH2Pcmolecules in the device structure, orthat the signal is beyond the detection limits of our exper-imental arrangement (note the lower efficiency even forsinglet emission discussed above).Moreover, it has recently been suggested that theassignment of emission from a triplet in a related STMLstudy [39] is actually due to trion emission froma chargedmolecule. Although not directly relevant to the resultsdiscussed here, it is interesting to consider whether tri-ons might play a role in the EL of molecular/van derWaals tunnel devices. Trion emission occurswhen singly-charged molecular anions are excited into a doublet stateand subsequently relax through the emission of a photon;this transition conserves spin and is optically-allowed.We would expect that both trion emission, and the pres-ence of long-lived charged molecules in our deviceswould be detected in changes in the voltage-dependentPL of our device structures. In fact, PL measurementsreveal a complete absence of any voltage dependence.These data were previously reported for the PTCDI vari-ation of the device (Ref 5 SI Figure S6) and similar mea-surements for the current study on H2Pc are shown in SI(Figure S6); for both molecules the PL peak positions areclose to those previously reported for the samemoleculeson thick hBNflakes [4,34] and confirm that themoleculesremain neutral over the measured voltage range. Theseresults show charged molecules cannot account for thelong-lived intermediary excited states which are neces-sary for up-conversion and which have been previouslyattributed to the presence of triplets [5,24]. However,it is not possible to rule out (or in) a possible path-way in which a long-lived triplet acquires a charge viaelectron capture followed by rapid relaxation via trionemission. To determine whether this process might berelevant would require the identification of the spectralposition for trion emission for PTCDI which would, inturn, require a means of forming an ensemble of chargedmolecules for investigation using PL. Interestingly thismay be possible using an asymmetric variation of ourdevice in which the upper and lower barrier thicknessesare significantly different and we plan to undertake suchstudies in the near future.Currently there is no detailed theoretical framework tounderpin the phenomenological explanations reportedhere and our workmotivates further studies of the mech-anisms of energy transfer from tunnelling electrons toorganic molecules, including transitions to the tripletstate which are optically-forbidden. While there remainsome uncertainties in the detailed explanations of theoptoelectronic properties of our devices, we stress thatthe current work provides insights into the possiblerole of energy level alignment on the generation of sta-ble intermediary states of relevance to the fundamentalstudy of long-livedmolecular excited stateswith potentialapplications in quantum spin-based technologies and,through photon up-conversion, low-voltage light emit-ting devices.AcknowledgementsWe are grateful to the Engineering and Physical SciencesResearch Council for support through grant EP/N033906/1.A.S.N. is grateful to the Ministry of Science and Higher Edu-cation (Agreement No. 121031700313-8). We thank BjarkeSørensen Jessen for helpful discussions. K.W. and T.T. acknowl-edge support from the JSPS KAKENHI (Grant Numbers19H05790, 20H00354 and 21H05233). P.H.B thanks the Lev-erhulme Trust for the award of a Research Fellowship [RF-2019-460]. This work evolved from the study of the influenceof absorption on the transition energies of organic moleculeswhich was initiated in collaboration with Professor Nick Besleyto whom this paper is dedicated.Data availability statementThe authors declare that the data supporting the findings ofthis study are available within the article and its Supplemen-tary Information files, or from the corresponding author onreasonable request.Disclosure statementNo potential conflict of interest was reported by the author(s).FundingThis work was supported by Engineering and Physical Sci-ences Research Council [grant number EP/N033906/1]; JapanSociety for the Promotion of Science Kakenhi [grant numbers19H05790, 20H00354 and 21H05233]; The Leverhulme Trust[grant number RF2019-460].ORCIDJonathan Bradford http://orcid.org/0000-0003-2356-5816Kenji Watanabe http://orcid.org/0000-0003-3701-8119Elena Besley http://orcid.org/0000-0002-9910-7603Simon A. Svatek http://orcid.org/0000-0002-8104-1888Peter H. Beton http://orcid.org/0000-0002-2120-8033http://orcid.org/0000-0003-2356-5816http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-9910-7603http://orcid.org/0000-0002-8104-1888http://orcid.org/0000-0002-2120-80338 T. JAMES ET AL.References[1] S.H. Amsterdam, T.J. Marks and M.C. Hersam, J. Phys.Chem. Lett. 12, 4543 (2021). doi:10.1021/acs.jpclett.1c00799[2] J.H. Park, A. Sanne, Y. Guo, M. Amani, K. Zhang, H.C.P.Movva, J.A. Robinson, A. Javey, J. Robertson, S.K. Baner-jee and A.C. Kummel, Sci. Adv. 3, e1701661 (2017).doi:10.1126/sciadv.1701661[3] Y.L. Huang, Y.J. Zheng, Z. Song, D. Chi, A.T.S. Weeand S.Y. Quek, Chem. Soc. Rev. 47, 3241 (2018).doi:10.1039/C8CS00159F[4] J. Kerfoot, V.V. Korolkov, A.S. Nizovtsev, R. Jones,T. Taniguchi, K. Watanabe, I. Lesanovsky, B. Olmos,N.A. Besley, E. Besley and P.H. Beton, J. Chem. Phys. 149,054701 (2018). doi:10.1063/1.5041418[5] S.A. Svatek, J. Kerfoot, A. Summerfield, A.S. Nizovtsev,V.V.V. Korolkov, T. Taniguchi, K. Watanabe, E. Antolín,E. Besley and P.H. Beton, Nano Lett. 20, 278 (2020).doi:10.1021/acs.nanolett.9b03787[6] A.S. Davydov, Sov. Phys. Uspekhi. 82, 393–448 (1964).doi:10.3367/ufnr.0082.196403a.0393.[7] F.C. Spano and C. Silva, Annu. Rev. Phys. Chem.65, 477 (2014). doi:10.1146/annurev-physchem-040513-103639[8] A. Eisfeld, C.Marquardt, A. Paulheim andM. Sokolowski,Phys. Rev. Lett. 119, 097402 (2017). doi:10.1103/PhysRevLett.119.097402[9] H. Zhao, Y. Zhao, Y. Song, M. Zhou, W. Lv, L. Tao,Y. Feng, B. Song, Y. Ma, J. Zhang, J. Xiao, Y. Wang,D.-H. Lien, M. Amani, H. Kim, X. Chen, Z. Wu, Z. Ni,P. Wang, Y. Shi, H. Ma, X. Zhang, J.-B. Xu, A. Troisi,A. Javey and X. Wang, Nat. Commun. 10, 5589 (2019).doi:10.1038/s41467-019-13581-9[10] S. Koo, I. Park, K. Watanabe, T. Taniguchi, J.H. Shim andS. Ryu, Nano Lett. 21, 6600 (2021). doi:10.1021/acs.nanolett.1c02009[11] K.S. Vasu, E. Prestat, J. Abraham, J. Dix, R.J. Kashtiban,J. Beheshtian, J. Sloan, P. Carbone, M. Neek-Amal,S.J. Haigh, A.K. Geim and R.R. Nair, Nat. Commun. 7,12168 (2016). doi:10.1038/ncomms12168[12] Y. Zhao, V. Wang and A. Javey, Matter. 3, 1832 (2020).doi:10.1016/j.matt.2020.09.009[13] F. Withers, O. Del Pozo-Zamudio, A. Mishchenko,A.P. Rooney, A. Gholinia, K. Watanabe, T. Taniguchi,S.J. Haigh,A.K.Geim,A.I. Tartakovskii andK.S.Novoselov,Nat. Mater. 14, 301 (2015). doi:10.1038/nmat4205[14] C. Palacios-Berraquero,M. Barbone, D.M. Kara, X. Chen,I. Goykhman, D. Yoon, A.K. Ott, J. Beitner, K. Watanabe,T. Taniguchi, A.C. Ferrari andM. Atatüre, Nat. Commun.7, 12978 (2016). doi:10.1038/ncomms12978[15] D. Kwak, M. Paur, K. Watanabe, T. Taniguchi andT. Mueller, Adv. Mater. Technol. 7, 2100915 (2022).doi:10.1002/admt.202100915[16] M. Paur, A.J.Molina-Mendoza, R. Bratschitsch, K.Watan-abe, T. Taniguchi and T. Mueller, Nat. Commun. 10, 1709(2019). doi:10.1038/s41467-019-09781-y[17] L. Britnell, R.V. Gorbachev, R. Jalil, B.D. Belle, F.Schedin, A. Mishchenko, T. Georgiou, M.I. Katsnelson,L. Eaves, S.V. Morozov, N.M.R. Peres, J. Leist, A.K. Geim,K.S. Novoselov and L.A. Ponomarenko, Science. 335, 947(2012). doi:10.1126/science.1218461[18] A. Mishchenko, J.S. Tu, Y. Cao, R.V. Gorbachev,J.R. Wallbank, M.T. Greenaway, V.E. Morozov, S.V. Moro-zov, M.J. Zhu, S.L. Wong, F. Withers, C.R. Woods,Y.-J. Kim, K. Watanabe, T. Taniguchi, E.E. Vdovin,O. Makarovsky, T.M. Fromhold, V.I. Fal’ko, A.K. Geim,L. Eaves and K.S. Novoselov, Nat. Nanotechnol. 9, 808(2014). doi:10.1038/nnano.2014.187[19] J. Kerfoot, S.A. Svatek, V.V. Korolkov, T. Taniguchi,K. Watanabe, E. Antolin and P.H. Beton, ACS Nano. 14,13886 (2020). doi:10.1021/acsnano.0c06280[20] Y. Zhang, Y. Luo, Y. Zhang, Y.-J. Yu, Y.-M. Kuang,L. Zhang, Q.-S. Meng, Y. Luo, J.-L. Yang, Z.-C. Dong andJ.G.Hou,Nature. 531, 623 (2016). doi:10.1038/nature17428[21] H. Imada, K. Miwa, M. Imai-Imada, S. Kawahara,K. Kimura and Y. Kim, Nature. 538, 364 (2016).doi:10.1038/nature19765[22] B. Doppagne, T. Neuman, R. Soria-Martinez, L.E.P.López, H. Bulou, M. Romeo, S. Berciaud, F. Scheurer, J.Aizpurua andG. Schull, Nat.Nanotechnol. 15, 207 (2020).doi:10.1038/s41565-019-0620-x[23] B. Doppagne, M.C. Chong, H. Bulou, A. Boeglin,F. Scheurer and G. Schull, Science. 361, 251 (2018).doi:10.1126/science.aat1603[24] G. Chen, Y. Luo,H.Gao, J. Jiang, Y. Yu, L. Zhang, Y. Zhang,X. Li, Z. Zhang and Z. Dong, Phys. Rev. Lett. 122, 177401(2019). doi:10.1103/PhysRevLett.122.177401[25] K. Kuhnke, C. Große, P. Merino and K. Kern, Chem. Rev.117, 5174 (2017). doi:10.1021/acs.chemrev.6b00645[26] P. Merino, C. Große, A. Rosławska, K. Kuhnke and K.Kern, Nat. Commun. 6, 8461 (2015). doi:10.1038/ncomms9461[27] M. Nothaft, S. Höhla, F. Jelezko, N. Frühauf, J. Pflaum andJ. Wrachtrup, Nat. Commun. 3, 628 (2012). doi:10.1038/ncomms1637[28] S.-J. Zou, Y. Shen, F.-M. Xie, J.-D. Chen, Y.-Q. Liand J.-X. Tang, Mater. Chem. Front. 4, 788 (2020).doi:10.1039/C9QM00716D[29] V.V. Korolkov, A. Summerfield, A. Murphy, D.B. Ama-bilino, K. Watanabe, T. Taniguchi and P.H. Beton,Nat. Commun. 10, 1537 (2019). doi:10.1038/s41467-019-09571-6[30] F. Pizzocchero, L. Gammelgaard, B.S. Jessen, J.M. Cari-dad, L. Wang, J. Hone, P. Bøggild and T.J. Booth, Nat.Commun. 7, 11894 (2016). doi:10.1038/ncomms11894[31] L. Wang, I. Meric, P.Y. Huang, Q. Gao, Y. Gao, H. Tran,T. Taniguchi, K. Watanabe, L.M. Campos, D.A. Muller,J. Guo, P. Kim, J. Hone, K.L. Shepard and C.R. Dean,Science. 342, 614 (2013). doi:10.1126/science.1244358[32] A.V. Kretinin, Y. Cao, J.S. Tu, G.L. Yu, R. Jalil,K.S. Novoselov, S.J. Haigh, A. Gholinia, A. Mishchenko,M. Lozada, T. Georgiou, C.R.Woods, F. Withers, P. Blake,G. Eda, A. Wirsig, C. Hucho, K. Watanabe, T. Taniguchi,A.K. Geim and R.V. Gorbachev, Nano Lett. 14, 3270(2014). doi:10.1021/nl5006542[33] A. Castellanos-Gomez, M. Buscema, R. Molenaar, V.Singh, L. Janssen, H.S.J. van der Zant and G.A. Steele,2DMater. 1, 011002 (2014). doi:10.1088/2053-1583/1/1/011002[34] M. Alkhamisi, V.V. Korolkov, A.S. Nizovtsev, J. Ker-foot, T. Taniguchi, K. Watanabe, N.A. Besley, E. Besleyand P.H. Beton, Chem. Commun. 54, 12021 (2018).doi:10.1039/C8CC06304Dhttps://doi.org/10.1021/acs.jpclett.1c00799https://doi.org/10.1126/sciadv.1701661https://doi.org/10.1039/C8CS00159Fhttps://doi.org/10.1063/1.5041418https://doi.org/10.1021/acs.nanolett.9b03787https://doi.org/doi:10.3367/ufnr.0082.196403a.0393https://doi.org/10.1146/annurev-physchem-040513-103639https://doi.org/10.1103/PhysRevLett.119.097402https://doi.org/10.1038/s41467-019-13581-9https://doi.org/10.1021/acs.nanolett.1c02009https://doi.org/10.1038/ncomms12168https://doi.org/10.1016/j.matt.2020.09.009https://doi.org/10.1038/nmat4205https://doi.org/10.1038/ncomms12978https://doi.org/10.1002/admt.202100915https://doi.org/10.1038/s41467-019-09781-yhttps://doi.org/10.1126/science.1218461https://doi.org/10.1038/nnano.2014.187https://doi.org/10.1021/acsnano.0c06280https://doi.org/10.1038/nature17428https://doi.org/10.1038/nature19765https://doi.org/10.1038/s41565-019-0620-xhttps://doi.org/10.1126/science.aat1603https://doi.org/10.1103/PhysRevLett.122.177401https://doi.org/10.1021/acs.chemrev.6b00645https://doi.org/10.1038/ncomms9461https://doi.org/10.1038/ncomms1637https://doi.org/10.1039/C9QM00716Dhttps://doi.org/10.1038/s41467-019-09571-6https://doi.org/10.1038/ncomms11894https://doi.org/10.1126/science.1244358https://doi.org/10.1021/nl5006542https://doi.org/10.1088/2053-1583/1/1/011002https://doi.org/10.1039/C8CC06304DMOLECULAR PHYSICS 9[35] C. Murray, N. Dozova, J.G. McCaffrey, N. Shafizadeh, W.Chin, M. Broquier and C. Crépin, Phys. Chem. Chem.Phys. 13, 17543 (2011). doi:10.1039/c1cp22039j[36] H. Imada, M. Imai-Imada, X. Ouyang, A. Muranaka andY. Kim, J. Chem. Phys. 157 (2022). doi:10.1063/5.0102087[37] J. McVie, R.S. Sinclair and T.G. Truscott, J. Chem. Soc.Faraday Trans. 2. 74, 1870 (1978). doi:10.1039/f29787401870[38] B.W. Caplins, T.K. Mullenbach, R.J. Holmes andD.A. Blank, J. Phys. Chem. C. 119, 27340 (2015).doi:10.1021/acs.jpcc.5b09817[39] K. Kimura, K. Miwa, H. Imada, M. Imai-Imada,S. Kawahara, J. Takeya, M. Kawai, M. Galperin and Y.Kim, Nature. 570, 210 (2019). doi:10.1038/s41586-019-1284-2https://doi.org/10.1039/c1cp22039jhttps://doi.org/10.1063/5.0102087https://doi.org/10.1039/f29787401870https://doi.org/10.1021/acs.jpcc.5b09817https://doi.org/10.1038/s41586-019-1284-2 Acknowledgements Data availability statement Disclosure statement Funding ORCID References<<  /ASCII85EncodePages false  /AllowTransparency false  /AutoPositionEPSFiles false  /AutoRotatePages /PageByPage  /Binding /Left  /CalGrayProfile ()  /CalRGBProfile (Adobe RGB \0501998\051)  /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2)  /sRGBProfile (sRGB IEC61966-2.1)  /CannotEmbedFontPolicy /Error  /CompatibilityLevel 1.3  /CompressObjects /Off  /CompressPages true  /ConvertImagesToIndexed true  /PassThroughJPEGImages false  /CreateJobTicket false  /DefaultRenderingIntent /Default  /DetectBlends true  /DetectCurves 0.1000  /ColorConversionStrategy /sRGB  /DoThumbnails true  /EmbedAllFonts true  /EmbedOpenType false  /ParseICCProfilesInComments true  /EmbedJobOptions true  /DSCReportingLevel 0  /EmitDSCWarnings false  /EndPage -1  /ImageMemory 524288  /LockDistillerParams true  /MaxSubsetPct 100  /Optimize true  /OPM 1  /ParseDSCComments false  /ParseDSCCommentsForDocInfo true  /PreserveCopyPage true  /PreserveDICMYKValues true  /PreserveEPSInfo false  /PreserveFlatness true  /PreserveHalftoneInfo false  /PreserveOPIComments false  /PreserveOverprintSettings false  /StartPage 1  /SubsetFonts true  /TransferFunctionInfo /Remove  /UCRandBGInfo /Remove  /UsePrologue false  /ColorSettingsFile ()  /AlwaysEmbed [ true  ]  /NeverEmbed [ true  ]  /AntiAliasColorImages false  /CropColorImages true  /ColorImageMinResolution 150  /ColorImageMinResolutionPolicy /OK  /DownsampleColorImages true  /ColorImageDownsampleType /Bicubic  /ColorImageResolution 300  /ColorImageDepth -1  /ColorImageMinDownsampleDepth 1  /ColorImageDownsampleThreshold 1.50000  /EncodeColorImages true  /ColorImageFilter /DCTEncode  /AutoFilterColorImages false  /ColorImageAutoFilterStrategy /JPEG  /ColorACSImageDict <<    /QFactor 0.90    /HSamples [2 1 1 2] /VSamples [2 1 1 2]  >>  /ColorImageDict <<    /QFactor 0.40    /HSamples [1 1 1 1] /VSamples [1 1 1 1]  >>  /JPEG2000ColorACSImageDict <<    /TileWidth 256    /TileHeight 256    /Quality 15  >>  /JPEG2000ColorImageDict <<    /TileWidth 256    /TileHeight 256    /Quality 15  >>  /AntiAliasGrayImages false  /CropGrayImages true  /GrayImageMinResolution 150  /GrayImageMinResolutionPolicy /OK  /DownsampleGrayImages true  /GrayImageDownsampleType /Bicubic  /GrayImageResolution 300  /GrayImageDepth -1  /GrayImageMinDownsampleDepth 2  /GrayImageDownsampleThreshold 1.50000  /EncodeGrayImages true  /GrayImageFilter /DCTEncode  /AutoFilterGrayImages false  /GrayImageAutoFilterStrategy /JPEG  /GrayACSImageDict <<    /QFactor 0.90    /HSamples [2 1 1 2] /VSamples [2 1 1 2]  >>  /GrayImageDict <<    /QFactor 0.40    /HSamples [1 1 1 1] /VSamples [1 1 1 1]  >>  /JPEG2000GrayACSImageDict <<    /TileWidth 256    /TileHeight 256    /Quality 15  >>  /JPEG2000GrayImageDict <<    /TileWidth 256    /TileHeight 256    /Quality 15  >>  /AntiAliasMonoImages false  /CropMonoImages true  /MonoImageMinResolution 1200  /MonoImageMinResolutionPolicy /OK  /DownsampleMonoImages true  /MonoImageDownsampleType /Average  /MonoImageResolution 300  /MonoImageDepth -1  /MonoImageDownsampleThreshold 1.50000  /EncodeMonoImages true  /MonoImageFilter /CCITTFaxEncode  /MonoImageDict <<    /K -1  >>  /AllowPSXObjects true  /CheckCompliance [    /None  ]  /PDFX1aCheck false  /PDFX3Check false  /PDFXCompliantPDFOnly false  /PDFXNoTrimBoxError true  /PDFXTrimBoxToMediaBoxOffset [    0.00000    0.00000    0.00000    0.00000  ]  /PDFXSetBleedBoxToMediaBox true  /PDFXBleedBoxToTrimBoxOffset [    0.00000    0.00000    0.00000    0.00000  ]  /PDFXOutputIntentProfile (None)  /PDFXOutputConditionIdentifier ()  /PDFXOutputCondition ()  /PDFXRegistryName ()  /PDFXTrapped /False  /Description <<    /ENU ()  >>>> setdistillerparams<<  /HWResolution [600 600]  /PageSize [609.704 794.013]>> setpagedevice