# Fileset

[ChemNanoMat - 2024 - Yang - Hexagonal Boron Nitride Spacers for Fluorescence Imaging of Biomolecules.pdf](https://mdr.nims.go.jp/filesets/642872c6-8dcf-4547-9f04-d6d57ceb25a1/download)

## Creator

Xiliang Yang, Dong Hoon Shin, Ze Yu, [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), Vitaliy Babenko, Stephan Hofmann, Sabina Caneva

## Rights

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

## Other metadata

[Hexagonal Boron Nitride Spacers for Fluorescence Imaging of Biomolecules](https://mdr.nims.go.jp/datasets/fa43d2ab-6368-4b23-bc7a-bbc1e8bc3b8a)

## Fulltext

Hexagonal Boron Nitride Spacers for Fluorescence Imaging of BiomoleculesHexagonal Boron Nitride Spacers for Fluorescence Imagingof BiomoleculesXiliang Yang,[a] Dong Hoon Shin,[a, d] Ze Yu,[a] Kenji Watanabe,[b] Takashi Taniguchi,[b]Vitaliy Babenko,[c] Stephan Hofmann,[c] and Sabina Caneva*[a]Fluorescence imaging is an invaluable tool to investigatebiomolecular dynamics, mechanics, and interactions in aqueousenvironments. Two-dimensional materials offer large-area,atomically smooth surfaces for wide-field biomolecule imaging.Despite the success of graphene for on-chip biosensing andbiomolecule manipulation, its strong fluorescence-quenchingproperties pose a challenge for biomolecular investigations thatare based on direct optical readouts. Here, we employ few-layerhexagonal boron nitride (hBN) as a precisely tailorablefluorescence spacer between labelled lipid membranes andgraphene substrates. By stacking high-quality hBN crystals inthe 10–20 nm thickness range on monolayer graphene, weobserve distance-dependent fluorescence intensity variations.Remarkably, with hBN spacers as thin as 20 nm, thefluorescence intensity is comparable to bare SiO2/Si substrates,while the intensity was reduced to 60% and 80% with ~10 nmand ~16 nm hBN thicknesses respectively. We confirm that pre-determined hBN thicknesses can be employed to control thenon-radiative energy transfer properties of graphene, withfluorescence quenching following a d� 4 distance-dependentbehaviour. This seamless integration of electronically active anddielectric van der Waals materials into vertical heterostructuresenables multifunctional platforms addressing the manipulation,localization, and visualization of biomolecules for fundamentalbiophysics and biosensing applications.IntroductionFluorescence imaging has emerged as a key tool that can revealaccurate quantitative and mechanistic insights into biomoleculedynamics, mechanics, and interactions down to the single-molecule level.[1–3] This technique has proven invaluable inproviding a detailed understanding of the inner workings ofcomplex biological structures and functions[4] such as ribosomaldynamics and assembly[5] as well as protein folding andbinding.[6] Its simple implementation, molecular specificity andcompatibility with multicolour, multiplexed imaging has led itto be routinely used in biophysics research.Monitoring molecular properties, conformations and spatialarrangements can be done either by measuring freely-diffusingmolecules or molecules that are spatially constrained. Con-straining can include tethering to surfaces and/or beads,[7] andwhile it enables longer observation times during imaging, itmay not fully represent the behaviour of native biomolecules.An ideal scenario is to present biomolecules with a non-stronglyinteracting surface that provides reversible attachment sitesthrough adsorption, while not interfering with biologicalfunction. Such surfaces would offer sufficient binding strengthfor molecules to be localized and tracked during in-planeimaging, yet allow sufficient mobility, thereby still enablingdiffusion, intermolecular interactions and desorption. Two-dimensional (2D) van der Waals (vdW) materials are emergingas promising substrates for biophysical studies. Owing to theiratomically-smooth surfaces free of dangling bonds, biocompat-ibility and transparency in their few-layer form, they representnovel platforms to obtain statistically-relevant datasets of bio-logical properties and interactions over large surface areas.Within the class of vdW materials, graphene has beenextensively studied in the context of biosensor devices. It hasbeen used in various implementations including as an electrodematerial for the dielectrophoretic trapping of DNA,[8] as asurface sensor in FETs,[9] and as a membrane for nanopore andtunnelling based sensing.[10] The development of graphene forbioimaging is, however, not without challenges. Strongfluorescence quenching of fluorophores by graphene poses asignificant obstacle to its use in fluorescence-based opticalreadouts. While its excellent fluorescence suppression proper-ties have been employed in resonance Raman spectroscopy toweaken the background fluorescence emission and enhancethe Raman peaks from biomolecules,[11] and in the generationof quenching masks for deterministic positioning of hBNquantum emitters, they hinder direct fluorescence-based[a] X. Yang, Dr. D. H. Shin, Dr. Z. Yu, Dr. S. CanevaDepartment of Precision and Microsystems EngineeringDelft University of TechnologyMekelweg 2, 2628 CD, Delft, The NetherlandsE-mail: s.caneva@tudelft.nl[b] Dr. K. Watanabe, Dr. T. TaniguchiNational Institute for Materials Science1-1 Namiki, Tsukuba, Ibaraki 305-0044 Japan[c] Dr. V. Babenko, Prof. Dr. S. HofmannDepartment of EngineeringUniversity of CambridgeCambridge CB3 0FA, UK[d] Dr. D. H. ShinPresent address: Department of Electronics and Information Engineering,Korea University, Sejong 30019, Republic of Korea© 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbH. This isan open access article under the terms of the Creative Commons AttributionLicense, which permits use, distribution and reproduction in any medium,provided the original work is properly cited.Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 76/81] 1ChemNanoMat 2024, 10, e202300592 (1 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Articledoi.org/10.1002/cnma.202300592www.chemnanomat.orghttp://orcid.org/0009-0003-8055-079Xhttp://orcid.org/0000-0002-0438-0835http://orcid.org/0000-0002-0936-8640http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0001-5372-6487http://orcid.org/0000-0001-6375-1459http://orcid.org/0000-0003-3457-7505www.chemnanomat.orghttp://crossmark.crossref.org/dialog/?doi=10.1002%2Fcnma.202300592&domain=pdf&date_stamp=2024-03-08imaging.[12] For example, the measured emitter decay rate forrhodamine molecules is enhanced 90 times (energy transferefficiency of ~99%) at distances d ~5 nm with respect to thedecay in vacuum.[13] This high energy transfer rate is mainly dueto the two-dimensionality and gapless character of themonatomic graphene.[14]To minimize quenching, it is essential to place an inertspacer material between the fluorophore and the graphenelayer. Since fluorescence quenching is highly sensitive to thedistance,[13,15] the spacer‘s thickness needs to be preciselycontrolled. Nanoscale spacers with tunable graphene-dyeseparation have been generated with DNA origami nano-structures anchored on graphene monolayers and functional-ized with single fluorophores (Atto542, Atto647N). Thesestructures show a strong distance-dependent quenching behav-iour proportional to d� 4.[15–16]In general, large-area spacer materials should be of highquality, i. e. free of pinholes, to prevent defect-mediatedquenching.[17–18] Traditional materials for spacer layers includepolymers such as polymethylmethacrylate (PMMA) and ce-ramics, most notably titanium oxide (TiO2), silicon dioxide (SiO2)and silicon nitride (Si3N4).[13,19–21] Polymers, however, are chal-lenging to precisely deposit with atomic level control.[22] Incontrast, ceramic coatings can be generated with high precisionby atomic layer deposition, chemical vapor deposition orelectron beam evaporation.[13,19–20] Experimental results, how-ever, have shown that achieving atomic control over the spacerthickness while retaining uniformity over large areas is challeng-ing with these methods. This can result in unwanted defects,such as pinholes or point defects, that provide fluorescencequenching pathways. To circumvent this issue, materials thatmaintain a high degree of crystallinity when scaled down totheir few-atomic layer form are needed to retain the opticalsignatures from labelled biomolecules. A material that is garner-ing significant attention as an ultraflat, chemically inert andbiocompatible substrate is hexagonal boron nitride (hBN). Dueto its wide bandgap in the visible, hBN does not act as an non-radiative acceptor and can thus be used in measurementsrequiring fluorescence emission. Atomically thin hBN haspreviously been employed as a spacer to improve the metal-enhanced fluorescence of fluorophores on Ag nanoparticles,[23]indicating the suitability of this material as an atomicallysmooth substrate for fluorescence imaging of biomolecules.hBN has recently also been used in single-molecule imaging,enabling the tracking of single-stranded DNA over long timeperiods, which revealed anomalous diffusion along hBNterraces.[24]Here we experimentally demonstrate the use of high qualityhBN layers of controlled nanoscale thickness as spacers toprevent quenching of the fluorescence emission from labelledlipid membrane deposited on graphene substrates. Grapheneand hBN are highly commensurate, with the lattice constantsdiffering by ~1.6%[25] facilitating the integration of thesematerials in vertical heterostructures. Another significant ad-vantage is that no specific attachment chemistry is needed togenerate the heterostructure and to position molecules atprecise distances from the surfaces, as is typically required forDNA origami/graphene[15] and DNA origami/MXenestructures.[26]We perform our imaging study by deterministically stamp-ing defect-free hBN flakes of various thicknesses on a mono-layer CVD graphene domain. Monolayer graphene was chosenin order to keep a consistent fluorescence baseline, given thatthe energy transfer rate increases significantly with the numberof graphene layers.[14] With this approach we achieve atomic-level control of the donor (fluorophore) and acceptor (gra-phene) distance and reveal a distance-dependent scaling of thefluorescence intensity with hBN thicknesses in the range of 10–20 nm. We measure a significant enhancement in fluorescenceintensity compared to direct molecule deposition on graphene,with an increase of 60% and 80% achieved by using hBNspacers with thicknesses of 10 nm and 16 nm, respectively.Notably, when the thickness reaches 20 nm, the fluorescenceintensity matches that on SiO2/Si substrates where no quench-ing is expected. The advantage of this 2D material stackingapproach is two-fold: 1) it enables the integration of graphene‘sremarkable electronic functionality with 2) hBN’s fluorescenceimaging-compatible properties. Combining these two distinc-tive materials offers a new platform for bio-optoelectronicnanodevices, which can precisely manipulate, localize, andvisualize biomolecules on-chip. Importantly, it can significantlyexpand analytical biosensor capabilities by providing a pathwayto investigate the intricate dynamics of biomolecular interac-tions at the ultimate resolution i. e. at the single-moleculelevel.[24]Results and DiscussionTo image the interaction of fluorescently labelled biomoleculeswith 2D material surfaces, we use fluorescent lipid membranesthat are obtained through the giant unilamellar vesicles (GUVs)splashing method. Figure 1a schematically shows the molecularFigure 1. (a) Chemical composition and schematic of rhodamine labelled PElipids (Rh-PE) used in the preparation of fluorescent GUVs. (b) Fluorescenceimage of a representative GUV obtained with the swelling method. (c)Schematic of the optically accessible flowcell consisting of a liquid chamberbetween a glass slide and a coverslip. Fluorescence images obtained under525 nm illumination for (d) as prepared GUVs, (e) GUVs incubated for 5 minand (f) GUVs incubated for 30 min.Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 77/81] 1ChemNanoMat 2024, 10, e202300592 (2 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Article 2199692x, 2024, 5, Downloaded from https://aces.onlinelibrary.wiley.com/doi/10.1002/cnma.202300592 by Cochrane Japan, Wiley Online Library on [24/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensecomposition of rhodamine-labelled phosphatidylethanolamines(PE) lipids and their incorporation into (diphytanoylphosphati-dylcholine) DPhPC-based GUVs using the PVA-assisted swellingmethod.[27] This approach yields large numbers of GUVs withtypical diameters of ~20–200 μm. Figure 1b shows an epifluor-escence microscopy image of a representative GUV producedwith this method recorded under 525 nm illumination. Toinduce and concurrently image the formation of large area lipidbilayers on glass and SiO2/Si substrates, we prepare a flowcellconsisting of a top glass slide separated from a bottomcoverslip (150 μm thickness) by parafilm spacers (Figure 1c). Thedistance between the two glass layers is ~150 μm. Liquids canbe introduced into the middle chamber through capillary forcesby approaching a pipette tip near the edge of the stack. Theflowcell is first filled with a solution of 30 mM NaCl, and GUVsare then flushed into the pre-filled chamber. The flowcell ismounted onto an inverted fluorescence microscope (NikonTs2R FL) and images of the bilayer formation process areacquired immediately after insertion and after 5 min and30 min incubation time. We used a 60� water immersionobjective (CFI Pan Apochromat VC) and a 10� (CFI PanApochromat Lambda) for imaging. Initially, GUVs freely diffusein solution (Figure 1d), then they sink to the bottom surfacewhere they rupture due to surface tension effects,[28] and formislands of planar lipid bilayers (Figure 1e). With increasing time,more surface area becomes covered by lipid bilayers, whicheventually fuse to form an almost homogeneous film due tolateral diffusion of lipid molecules (Figure 1f). After 30 min, thechamber is flushed with 30 mM NaCl to remove any unrupturedfreely-diffusing GUVs from the liquid. This ensures that thefluorescence intensity originates only from the deposited film.The same procedure is performed with SiO2/Si substrates withand without graphene. Since the substrates are not transparent,the GUVs are first allowed to rupture on the substrate for30 min, and then the sample is flipped upside down such thatthe lipid bilayer can be imaged from below.Due to their different optical and electronic properties, 2Dmaterials such as graphene and hBN show strikingly differentinteractions with labelled molecules. We demonstrate thedifferences by preparing substrates with both materials andsubsequently depositing fluorescent lipid membranes on them.The graphene/SiO2/Si substrates are prepared by transferringCVD grown graphene domains (typical diameter of ~200 μm) ormonolayer films from the copper foil growth substrate to SiO2/Si chips (0.7�0.7 cm2) using the PMMA-assisted, metal etchingmethod.[29] Figure 2 compares fluorescence images from threedifferent substrates (bare, with graphene domains and with fullcoverage graphene) before (a, c, e) and after (b, d, f) GUVsplashing. As expected, none of the substrates exhibitsfluorescence without GUVs in the flowcell. After GUV incuba-tion, the bare Si substrate displays relatively uniformfluorescence. The substrates with graphene, however, displayfluorescence only over the areas without graphene coverage, asindicated by the dark hexagonal graphene domains against thebright Si background. Similarly, the full coverage graphenesubstrate appears dark all over the surface apart from a cornerwhere the graphene film edge is just visible and beyond whichthe fluorescence intensity is recovered on the SiO2/Si. Theabsence of fluorescence signal on graphene can be attributedto either the absence of lipid molecules on the domains or thequenching of fluorescence on the carbon layer. Previous studieson lipid deposition on graphene established that variousconfigurations, including inverted lipid bilayers[30] and lipidmonolayers,[31–32] rather than bilayers, are formed on graphenedue to the effective hydrophobicity of transferredgraphene.[33–34] Thus the effects observed in our experiments areattributed to quenching. This is in line with reports demonstrat-ing that in the proximity of fluorescent emitters (e.g. fluoro-phores, QDs), graphene acts as an efficient energy sink (i. e.fluorescence quencher), efficiently coupling to the emitter via adistant-dependent, non-radiative energy transfer.[12,19–20,35–36]hBN as an Atomically Precise Fluorescence Quenching BarrierIn order to establish to what extent the fluorescence quenchingof the lipid molecules adsorbed on the graphene can beblocked with a physical distance, we introduce hBN as a spacerlayer of precise thickness to separate the rhodamine-labelledlipids from the graphene. hBN is chosen as a spacer given itstwo-dimensional structure which is highly compatible withstacking onto graphene, thickness control at the atomic level,defect-free quality with surfaces devoid of dangling bonds, highaffinities to aromatic fluorophores via p � p interactions, goodchemical stability, and excellent impermeability.We first transfer a large monolayer graphene domain onSiO2/Si and then use the viscoelastic stamping method[37] todeposit exfoliated hBN flakes (high-temperature high-pressurecrystals, NIMS, Japan) onto the edges of the graphene. Thisallows the subsequent comparison of areas where hBN is indirect contact with the Si and areas where there is a grapheneunderlayer. Figure 3a shows an optical image of the sampleFigure 2. (a, c, e) Fluorescence images of bare SiO2/Si, graphene domains onSiO2/Si and full coverage monolayer graphene on SiO2/Si respectively beforeGUV deposition. (b, d, f) Corresponding images after GUV deposition. Thedark areas correspond to the location of the graphene.Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 78/81] 1ChemNanoMat 2024, 10, e202300592 (3 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Article 2199692x, 2024, 5, Downloaded from https://aces.onlinelibrary.wiley.com/doi/10.1002/cnma.202300592 by Cochrane Japan, Wiley Online Library on [24/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewith three hBN flakes (blue) on graphene (dark pink, right handside). The corresponding image viewed under 525 nm illumina-tion is shown in Figure 3b. Here, we note that the hBN flakesregions that are on the Si have fluorescence levels similar tothat of the bare Si, while the regions on the graphene havelower intensity, however, they are not fully quenched as is thecase on the graphene with no hBN. Figure 3c plots the Ramanspectra of the graphene and hBN displaying the characteristic Gand 2D peaks for monolayer graphene and the E2g peak forhBN, attesting to the high quality of the material after transferand before GUV deposition. The material dependentfluorescence emission and quenching effects of the hBN/Gstacks (Figure 3d) can be understood in the context of theelectronic band structures, schematically shown in Figure 3e.Given its gapless band structure, graphene acts as a non-radiating acceptor that directly quenches the donor (rhod-amine) emission. Conversely, hBN is a wide (~6 eV) bandgapmaterial which does not emit in the visible and to which thefluorescence energy from the rhodamine cannot betransferred.[38] hBN layers can therefore serve as barriers thatinhibit non-radiative energy transfer from the excited rhod-amine molecule to graphene.We perform atomic force microscopy (AFM) imaging todetermine the hBN flake thicknesses and to subsequentlyestablish the fluorescence intensity of the hBN/grapheneregions as a function of hBN thickness. Figure 4a,b,c showtapping mode topography images of the three flakes (F1, F2,F3) and the corresponding height profiles (Figure 4d, e, f) takenat the position of the coloured lines, from which we extract hBNflake thicknesses (t) of ~10, ~16 and ~20 nm.After lipid deposition, we quantify the fluorescence intensitylevels by taking the average intensity from regions withgraphene/SiO2/Si, hBN/graphene/SiO2/Si, and bare SiO2/Si (Fig-ure 5a,b). The normalized intensity histograms from theselected areas range from I=0 i. e. complete quenching ongraphene, to I=1 i. e. maximum fluorescence that can beFigure 3. (a) White light (WL) optical image of three hBN flakes (blue)stamped on the edge of a large monolayer graphene domain (dark pink). (b)Fluorescence (FL) image of the same areas in (a) measured under 525 nmillumination. (c) Raman spectra of hBN and graphene displaying thecharacteristic hBN E2g peak and the graphene G and 2D peaks. (d) Structureand arrangement of the vertical hBN/graphene (hBN/G) stacks. (e) Schematicelectronic band structures of rhodamine, hBN and graphene, showing theprinciple of hBN as a fluorescence quenching barrier.Figure 4. AFM imaging of hBN/graphene stacks on SiO2/Si. Tapping modeAFM topography images of (a) Flake 1, (b) Flake 2 and (c) Flake 3. (d–f) hBNheight profiles measured at the position of the coloured lines in thecorresponding images, with the fitted heights displayed in the graphs.Figure 5. Tunable fluorescence intensities on hBN/graphene stacks. (a,b) Fluorescence (FL) images of Flake 1 and Flake 3 under 525 nm illumination after GUVdeposition. Scar bar: 30 μm. (c,d) Histograms of the average fluorescence intensity from G/SiO2/Si, hBN/G/SiO2/Si, and bare SiO2/Si regions for Flake 1 andFlake 3 respectively. (e) Normalized energy transfer efficiency as a function of the distance of the fluorophores from the graphene, insert figure: Normalizedaverage fluorescence intensity of hBN/G/SiO2/Si as a function of hBN flake thickness.Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 79/81] 1ChemNanoMat 2024, 10, e202300592 (4 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Article 2199692x, 2024, 5, Downloaded from https://aces.onlinelibrary.wiley.com/doi/10.1002/cnma.202300592 by Cochrane Japan, Wiley Online Library on [24/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseachieved on SiO2/Si (Figure 5c,d). On the regions with hBNspacer layers, we observe a fluorescence intensity that isdependent on the thickness of the hBN flakes. To quantify thiseffect, Figure 5e plots the energy transfer efficiency calculatedfrom hBN/G/SiO2/Si intensity as a function of distance fromgraphene to excited rhodamine molecule for the three flakes. Itis crucial to note that the energy transfer distance encompassesnot only the thickness of hBN but also factors in the thicknessof a monolayer lipid membrane (~2 nm)[39] since lipid mono-layers or inverted lipid bilayers are typically formed on 2Dmaterials due to their largely hydrophobic properties.[32–33]Strikingly, for hBN spacers down to 20 nm, the fluorescenceexhibits an intensity level that is comparable to that of lipidsdirectly deposited on SiO2/Si substrates. hBN spacers of ~16 nmdisplay 80% of the SiO2/Si fluorescence level, while this valuedrops to 60% for hBN thicknesses of ~10 nm. We note that thebright spots on the hBN are usually attributed to polymerresidues that are transferred during stamping, which alsoappear bright under 525 nm illumination wavelength.[40–41]These results are in good agreement with previoustheoretical[13] and experimental[15] studies on Förster-like reso-nance energy transfer from an excited state of a donor dyemolecule to a graphene acceptor substrate. The energy transferrate depends on the dimensionality of the donor and theacceptor material. For (i) two single-point dipoles, (ii) one line ofdipoles and one single-point dipole and (iii) a 2D dipole array,the transfer efficiencies are functions of d� 6, d� 5 and d� 4respectively. As expected, our experimental data follows a d� 4distance-dependent behaviour with a 50% energy transferefficiency at a distance (d0) of 11.1 nm.ConclusionsWhile graphene has been successfully integrated as anelectronic component for biosensing and biomolecule manipu-lation, for instance in field effect transistors (FETs),[9,42] plasmonicbiosensors,[43] nanopores[44] and dielectrophoretic (DEP) trappingdevices,[8,10] fluorescence-based sensors have lagged behind,hindered by the quenching properties of the zero-bandgapmaterial. In this work we demonstrate a simple and scalableapproach to circumvent this limitation. We identify hBN as atunable spacer layer that can be stacked on graphene toprevent quenching of fluorescently labelled biomolecules. Thisoccurs due to blocking of the non-radiative energy transferpathway that otherwise takes place between the excitedfluorophore and graphene. A key aspect of this approach is theability to customize the thickness of the hBN spacer with atomicthickness control, which plays a key role defining the degree offluorescence intensity that is measured. By depositing fluores-cent lipid membranes on the hBN/graphene stacks we demon-strate that an hBN spacer thickness down to 20 nm enablesfluorescence emission intensity on a level comparable to thaton SiO2/Si substrates without graphene. Spacers with thick-nesses of ~16 nm and ~10 nm achieve up to 80% and 60% ofthe maximum signal.This work provides valuable insights into optimizing hBNspacer thickness for different biomolecular imaging applica-tions. We foresee particular interest for the single-moleculebiophysics field, where studying the inter- and intra-moleculedynamics and mechanics at individual molecule level could bedone directly on-chip in a high throughput manner using wide-field fluorescence imaging. By preventing fluorescence quench-ing, hBN spacers not only enable optical detection and trackingof biomolecules in space and time in physiological conditionsbut are also compatible with concurrent biomolecule manipu-lation techniques based on other 2D materials e.g. dielectro-phoretic trapping devices with graphene electrodes.[8]In summary, the utilization of hBN spacer layers forfluorescence imaging of biomolecules is a low-cost, low-complexity approach to achieve biomolecule imaging onelectronic materials that are otherwise efficient fluorescentquenchers. The capability to tailor hBN spacer thickness enablesthe tuning of the fluorescent signal intensity and can be usedas a vertical nanoscale ruler in single-molecule studies.Furthermore, the ease of integration with other 2D materialsmakes this approach versatile and compatible with orthogonalcharacterization and manipulation techniques, opening excitingnew prospects in basic and applied biophysics and nanofluidicsresearch.Experimental methodsLipid layer preparation: The swelling method is carried out in thefollowing steps. First, glass coverslips are coated with a film of 5%PVA (Polyvinyl alcohol) and baked at 60 °C for one hour. Afterwards,a solution of fluorescent lipids is pipetted and spread over theslides. This lipid solution is made by adding 1 μL of 0.1 mg/mLfluorescently labelled lipids (Egg Liss Rhod PE, Avanti Polar Lipids)to 25 μL of 5 mg/mL non-labelled diphytanoyl DPhPC lipidsdissolved in chloroform. The fluorescent phosphatidylethanol-amines PE lipids are chemically modified with rhodamine dyemolecules. For each batch of GUVs, 10 μL of fluorescent lipidssolution is used to cover the PVA-coated slides. The slides are left inthe desiccator at vacuum for 40 minutes to dehydrate the samples.Subsequently, 200 μL of sucrose (300 mM) is pipetted on thecovered slides on which the GUVs are allowed to swell in a petridish covered by the aluminium foil for one hour. Finally, the GUVsin sucrose are gently collected with a pipette and stored in an epitube for subsequent imaging.AcknowledgementsD.H.S. and S.C. acknowledge funding from the European Union’sHorizon 2020 research and innovation program (ERC StG,SIMPHONICS, Project No. 101041486). S.C. acknowledges a DelftTechnology Fellowship. X.Y. acknowledges funding from theChinese Scholarship Council (Scholarship No. 202108270002).Z.Y. acknowledges funding from NWO (Project MechanoPore).Conflict of InterestsThe authors declare no conflict of interest.Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 80/81] 1ChemNanoMat 2024, 10, e202300592 (5 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Article 2199692x, 2024, 5, Downloaded from https://aces.onlinelibrary.wiley.com/doi/10.1002/cnma.202300592 by Cochrane Japan, Wiley Online Library on [24/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseData Availability StatementThe data that support the findings of this study are openlyavailable at https://data.4tu.nl/.Keywords: hexagonal boron nitride (hBN) · graphene ·fluorescence · quenching · lipids[1] S. Weiss, Science 1999, 283, 1676–1683.[2] H. Yokota, Biochim. Biophys. Acta Gen. Subj. 2020, 1864, 129362.[3] C. Joo, H. Balci, Y. Ishitsuka, C. Buranachai, T. Ha, Annu. Rev. Biochem.2008, 77, 51–76.[4] M. F. Juette, D. S. Terry, M. R. Wasserman, Z. Zhou, R. B. Altman, Q.Zheng, S. C. Blanchard, Curr. Opin. Chem. Biol. 2014, 20, 103–111.[5] J. B. Munro, R. B. Altman, C.-S. Tung, J. H. D. Cate, K. Y. Sanbonmatsu,S. C. Blanchard, Proc. Natl. Acad. Sci. USA 2010, 107, 709–714.[6] Y. Gambin, A. A. Deniz, Mol. BioSyst. 2010, 6, 1540–1547.[7] M. L. Visnapuu, D. Duzdevich, E. C. Greene, Mol. BioSyst. 2008, 4, 394–403.[8] A. Barik, Y. Zhang, R. Grassi, B. P. Nadappuram, J. B. Edel, T. Low, S. J.Koester, S. H. Oh, Nat. Commun. 2017, 8, 1867.[9] G. Seo, G. Lee, M. J. Kim, S. H. Baek, M. Choi, K. B. Ku, C. S. Lee, S. Jun, D.Park, H. G. Kim, S. J. Kim, J. O. Lee, B. T. Kim, E. C. Park, S. I. Kim, ACSNano 2020, 14, 5135–5142.[10] S. J. Heerema, C. Dekker, Nat. Nanotechnol. 2016, 11, 127–136.[11] L. Xie, X. Ling, Y. Fang, J. Zhang, Z. Liu, J. Am. Chem. Soc. 2009, 131,9890–9891.[12] A. Kasry, A. A. Ardakani, G. S. Tulevski, B. Menges, M. Copel, L. Vyklicky,J. Phys. Chem. C 2012, 116, 2858–2862.[13] L. Gaudreau, K. J. Tielrooij, G. E. Prawiroatmodjo, J. Osmond, F. J.Garcia de Abajo, F. H. Koppens, Nano Lett. 2013, 13, 2030–2035.[14] Z. Chen, S. Berciaud, C. Nuckolls, T. F. Heinz, L. E. Brus, ACS Nano 2010,4, 2964–2968.[15] I. Kaminska, J. Bohlen, S. Rocchetti, F. Selbach, G. P. Acuna, P. Tinnefeld,Nano Lett. 2019, 19, 4257–4262.[16] S. Krause, E. Ploetz, J. Bohlen, P. Schuler, R. Yaadav, F. Selbach, F. Steiner,I. Kaminska, P. Tinnefeld, ACS Nano 2021, 15, 6430–6438.[17] X. Qin, L. Shen, L. Liang, S. Han, Z. Yi, X. Liu, J. Phys. Chem. C 2019, 123,11151–11161.[18] R. Mrad, M. Poggi, R. Ben Chaabane, M. Negrerie, J. Colloid Interface Sci.2020, 571, 368–377.[19] A. Ghosh, A. Sharma, A. I. Chizhik, S. Isbaner, D. Ruhlandt, R. Tsukanov, I.Gregor, N. Karedla, J. Enderlein, Nat. Photonics 2019, 13, 860–865.[20] O. Salihoglu, N. Kakenov, O. Balci, S. Balci, C. Kocabas, Sci. Rep. 2016, 6,33911.[21] K. X. Xie, L. T. Xu, Y. Y. Zhai, Z. C. Wang, M. Chen, X. H. Pan, S. H. Cao,Y. Q. Li, Talanta 2019, 195, 752–756.[22] T. Murakami, Y. Arima, M. Toda, H. Takiguchi, H. Iwata, Anal. Biochem.2012, 421, 632–639.[23] W. Gan, C. Tserkezis, Q. Cai, A. Falin, S. Mateti, M. Nguyen, I.Aharonovich, K. Watanabe, T. Taniguchi, F. Huang, L. Song, L. Kong, Y.Chen, L. H. Li, ACS Nano 2019, 13, 12184–12191.[24] D. H. Shin, S. H. Kim, K. Coshic, K. Watanabe, T. Taniguchi, G. Verbiest, S.Caneva, A. Aksimentiev, P. G. Steeneken, C. Joo, bioRxiv 2023,2023.2011, 2001.565159.[25] L. H. Li, Y. Chen, Adv. Funct. Mater. 2016, 26, 2594–2608.[26] L. Richter, A. M. Szalai, C. L. Manzanares-Palenzuela, I. Kaminska, P.Tinnefeld, Adv. Mater. 2023, 35, e2303152.[27] A. Weinberger, F. C. Tsai, G. H. Koenderink, T. F. Schmidt, R. Itri, W. Meier,T. Schmatko, A. Schroder, C. Marques, Biophys. J. 2013, 105, 154–164.[28] V. N. Ngassam, W. C. Su, D. L. Gettel, Y. Deng, Z. Yang, N. Wang-Tomic,V. P. Sharma, S. Purushothaman, A. N. Parikh, Biophys. J. 2021, 120, 586–597.[29] S. Hofmann, P. Braeuninger-Weimer, R. S. Weatherup, J. Phys. Chem. Lett.2015, 6, 2714–2721.[30] M. Hirtz, A. Oikonomou, T. Georgiou, H. Fuchs, A. Vijayaraghavan, Nat.Commun. 2013, 4, 2591.[31] S. R. Tabaei, W. B. Ng, S. J. Cho, N. J. Cho, ACS Appl. Mater. Interfaces2016, 8, 11875–11880.[32] L. M. Lima, W. Fu, L. Jiang, A. Kros, G. F. Schneider, Nanoscale 2016, 8,18646–18653.[33] B. M. Blaschke, P. Bohm, S. Drieschner, B. Nickel, J. A. Garrido, Langmuir2018, 34, 4224–4233.[34] J. Zhang, K. Jia, Y. Huang, Y. Wang, N. Liu, Y. Chen, X. Liu, X. Liu, Y. Zhu,L. Zheng, H. Chen, F. Liang, M. Zhang, X. Duan, H. Wang, L. Lin, H. Peng,Z. Liu, Nano Lett. 2021, 21, 9587–9593.[35] M. Lorenzoni, F. Brandi, S. Dante, A. Giugni, B. Torre, Sci. Rep. 2013, 3,1954.[36] J. Kim, L. J. Cote, F. Kim, J. Huang, J. Am. Chem. Soc. 2010, 132, 260–267.[37] A. Castellanos-Gomez, M. Buscema, R. Molenaar, V. Singh, L. Janssen,H. S. J. van der Zant, G. A. Steele, 2D Mater. 2014, 1, 011002.[38] S. Castelletto, F. A. Inam, S. I. Sato, A. Boretti, Beilstein J. Nanotechnol.2020, 11, 740–769.[39] T. Chen, A. Ghosh, J. Enderlein, Nano Lett. 2023, 23, 2421–2426.[40] D. G. Purdie, N. M. Pugno, T. Taniguchi, K. Watanabe, A. C. Ferrari, A.Lombardo, Nat. Commun. 2018, 9, 5387.[41] M. R. Rosenberger, H. J. Chuang, K. M. McCreary, A. T. Hanbicki, S. V.Sivaram, B. T. Jonker, ACS Appl. Mater. Interfaces 2018, 10, 10379–10387.[42] M. T. Hwang, M. Heiranian, Y. Kim, S. You, J. Leem, A. Taqieddin, V.Faramarzi, Y. Jing, I. Park, A. M. van der Zande, S. Nam, N. R. Aluru, R.Bashir, Nat. Commun. 2020, 11, 1543.[43] D. Rodrigo, O. Limaj, D. Janner, D. Etezadi, J. García de Abajo, V. Pruneri,H. Altug, Science 2015, 349, 165–168.[44] S. J. Heerema, L. Vicarelli, S. Pud, R. N. Schouten, H. W. Zandbergen, C.Dekker, ACS Nano 2018, 12, 2623–2633.Manuscript received: December 1, 2023Revised manuscript received: February 17, 2024Accepted manuscript online: February 19, 2024Version of record online: March 8, 2024Wiley VCH Dienstag, 14.05.20242405 / 344021 [S. 81/81] 1ChemNanoMat 2024, 10, e202300592 (6 of 6) © 2024 The Authors. ChemNanoMat published by Wiley-VCH GmbHResearch Article 2199692x, 2024, 5, Downloaded from https://aces.onlinelibrary.wiley.com/doi/10.1002/cnma.202300592 by Cochrane Japan, Wiley Online Library on [24/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://data.4tu.nl/https://doi.org/10.1126/science.283.5408.1676https://doi.org/10.1016/j.bbagen.2019.05.005https://doi.org/10.1146/annurev.biochem.77.070606.101543https://doi.org/10.1146/annurev.biochem.77.070606.101543https://doi.org/10.1016/j.cbpa.2014.05.010https://doi.org/10.1073/pnas.0908597107https://doi.org/10.1039/c003024dhttps://doi.org/10.1039/b800444ghttps://doi.org/10.1039/b800444ghttps://doi.org/10.1021/acsnano.0c02823https://doi.org/10.1021/acsnano.0c02823https://doi.org/10.1038/nnano.2015.307https://doi.org/10.1021/ja9037593https://doi.org/10.1021/ja9037593https://doi.org/10.1021/jp207972fhttps://doi.org/10.1021/nl400176bhttps://doi.org/10.1021/nn1005107https://doi.org/10.1021/nn1005107https://doi.org/10.1021/acs.nanolett.9b00172https://doi.org/10.1021/acsnano.0c08383https://doi.org/10.1021/acs.jpcc.9b02596https://doi.org/10.1021/acs.jpcc.9b02596https://doi.org/10.1016/j.jcis.2020.03.058https://doi.org/10.1016/j.jcis.2020.03.058https://doi.org/10.1038/s41566-019-0510-7https://doi.org/10.1016/j.talanta.2018.11.112https://doi.org/10.1016/j.ab.2011.12.008https://doi.org/10.1016/j.ab.2011.12.008https://doi.org/10.1021/acsnano.9b06858https://doi.org/10.1002/adfm.201504606https://doi.org/10.1016/j.bpj.2013.05.024https://doi.org/10.1016/j.bpj.2021.01.006https://doi.org/10.1016/j.bpj.2021.01.006https://doi.org/10.1021/acs.jpclett.5b01052https://doi.org/10.1021/acs.jpclett.5b01052https://doi.org/10.1021/acsami.6b02837https://doi.org/10.1021/acsami.6b02837https://doi.org/10.1039/C6NR05706Chttps://doi.org/10.1039/C6NR05706Chttps://doi.org/10.1021/acs.langmuir.8b00162https://doi.org/10.1021/acs.langmuir.8b00162https://doi.org/10.1021/acs.nanolett.1c03344https://doi.org/10.1088/2053-1583/1/1/011002https://doi.org/10.3762/bjnano.11.61https://doi.org/10.3762/bjnano.11.61https://doi.org/10.1021/acs.nanolett.2c04635https://doi.org/10.1021/acsami.8b01224https://doi.org/10.1126/science.aab2051https://doi.org/10.1021/acsnano.7b08635 Hexagonal Boron Nitride Spacers for Fluorescence Imaging of Biomolecules Introduction Results and Discussion hBN as an Atomically Precise Fluorescence Quenching Barrier Conclusions Experimental methods Acknowledgements Conflict of Interests Data Availability Statement