# Fileset

[Ouaj_2025_2D_Mater._12_015017.pdf](https://mdr.nims.go.jp/filesets/9a3b8dd4-5ea8-44c4-87b3-d18bc3192b75/download)

## Creator

[Taoufiq Ouaj](https://orcid.org/0009-0003-8316-523X), [Christophe Arnold](https://orcid.org/0000-0001-5540-8589), [Jon Azpeitia](https://orcid.org/0000-0003-4542-9735), Sunaja Baltic, [Julien Barjon](https://orcid.org/0000-0003-1749-2980), [José Cascales](https://orcid.org/0009-0005-3433-8063), [Huanyao Cun](https://orcid.org/0000-0002-5225-9861), [David Esteban](https://orcid.org/0009-0006-0167-1545), [Mar Garcia-Hernandez](https://orcid.org/0000-0002-5987-0647), Vincent Garnier, Subodh K Gautam, [Thomas Greber](https://orcid.org/0000-0002-5234-1937), Said Said Hassani, Adrian Hemmi, [Ignacio Jiménez](https://orcid.org/0000-0001-5605-3185), [Catherine Journet](https://orcid.org/0000-0002-3328-317X), [Paul Kögerler](https://orcid.org/0000-0001-7831-3953), [Annick Loiseau](https://orcid.org/0000-0002-1042-5876), [Camille Maestre](https://orcid.org/0000-0002-7911-3758), [Marvin Metzelaars](https://orcid.org/0000-0002-3529-557X), [Philipp Schmidt](https://orcid.org/0000-0002-1278-1727), [Christoph Stampfer](https://orcid.org/0000-0002-4958-7362), [Ingrid Stenger](https://orcid.org/0000-0002-8917-5776), Philippe Steyer, [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), [Bérangère Toury](https://orcid.org/0000-0001-5889-0796), [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Bernd Beschoten](https://orcid.org/0000-0003-2359-2718)

## Rights

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

## Other metadata

[Benchmarking the integration of hexagonal boron nitride crystals and thin films into graphene-based van der Waals heterostructures](https://mdr.nims.go.jp/datasets/138c5538-963c-4c17-863f-ec6b26dec02d)

## Fulltext

Benchmarking the integration of hexagonal boron nitride crystals and thin films into graphene-based van der Waals heterostructures2D Materials     PAPER • OPEN ACCESSBenchmarking the integration of hexagonal boronnitride crystals and thin films into graphene-basedvan der Waals heterostructuresTo cite this article: Taoufiq Ouaj et al 2025 2D Mater. 12 015017 View the article online for updates and enhancements.You may also likeChemically detaching hBN crystals grownat atmospheric pressure and hightemperature for high-performancegraphene devicesTaoufiq Ouaj, Leonard Kramme, MarvinMetzelaars et al.-Spontaneous doping on high quality talc-graphene-hBN van der WaalsheterostructuresE Mania, A B Alencar, A R Cadore et al.-Growth mechanisms of hBN crystallinenanostructures with rf sputteringdeposition: challenges, opportunities, andfuture perspectivesDuc-Quang Hoang, Ngoc-Hai Vu, Thanh-Qui Nguyen et al.-This content was downloaded from IP address 144.213.253.16 on 17/12/2024 at 04:54https://doi.org/10.1088/2053-1583/ad96c9/article/10.1088/1361-6528/acf2a0/article/10.1088/1361-6528/acf2a0/article/10.1088/1361-6528/acf2a0/article/10.1088/1361-6528/acf2a0/article/10.1088/2053-1583/aa76f4/article/10.1088/2053-1583/aa76f4/article/10.1088/2053-1583/aa76f4/article/10.1088/1402-4896/acbe7b/article/10.1088/1402-4896/acbe7b/article/10.1088/1402-4896/acbe7b/article/10.1088/1402-4896/acbe7b2D Mater. 12 (2025) 015017 https://doi.org/10.1088/2053-1583/ad96c9OPEN ACCESSRECEIVED5 September 2024REVISED15 November 2024ACCEPTED FOR PUBLICATION24 November 2024PUBLISHED16 December 2024Original Content fromthis work may be usedunder the terms of theCreative CommonsAttribution 4.0 licence.Any further distributionof this work mustmaintain attribution tothe author(s) and the titleof the work, journalcitation and DOI.PAPERBenchmarking the integration of hexagonal boron nitride crystalsand thin films into graphene-based van der Waals heterostructuresTaoufiq Ouaj1, Christophe Arnold2, Jon Azpeitia3, Sunaja Baltic1, Julien Barjon2, José Cascales3,Huanyao Cun4, David Esteban3, Mar Garcia-Hernandez3, Vincent Garnier5, Subodh K Gautam2,Thomas Greber6, Said Said Hassani2, Adrian Hemmi6, Ignacio Jiménez3, Catherine Journet7,Paul Kögerler8,9, Annick Loiseau10, Camille Maestre7, Marvin Metzelaars1,8, Philipp Schmidt1,Christoph Stampfer1,11, Ingrid Stenger2, Philippe Steyer5, Takashi Taniguchi12, Bérangère Toury7,Kenji Watanabe13 and Bernd Beschoten1,∗1 2nd Institute of Physics and JARA-FIT, RWTH Aachen University, 52074 Aachen, Germany2 GEMaC, UVSQ, CNRS, Université Paris Saclay, 78035 Versailles, France3 Instituto de Ciencia de Materiales de Madrid (ICMM-CSIC), Sor Juana Inés de la Cruz 3, 28049 Madrid, Spain4 Physik-Institut, University of Zürich, 8057 Zürich, Switzerland5 INSA Lyon, Universite Claude Bernard Lyon 1, CNRS, MATEIS, UMR5510, 69621 Villeurbanne, France6 Physik-Institut, University of Zürich, Zürich, Switzerland7 Universite Claude Bernard Lyon 1, CNRS, LMI UMR 5615, F-69100 Villeurbanne, France8 Institute of Inorganic Chemistry, RWTH Aachen University, 52074 Aachen, Germany9 Peter Grünberg Institute (PGI-6), Forschungszentrum Jülich, 52425 Jülich, Germany10 Université Paris Saclay, ONERA, CNRS, Laboratoire d’Etude des Microstructures, 92322 Chatillon, France11 Peter Grünberg Institute (PGI-9) Forschungszentrum Jülich, 52425 Jülich, Germany12 Research Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan13 Research Center for Electronic and Optical Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan∗ Author to whom any correspondence should be addressed.E-mail: bernd.beschoten@physik.rwth-aachen.deKeywords: hBN, graphene, crystal growth, thin film growth, charge carrier mobilityAbstractWe present a benchmarking protocol that combines the characterization of boron nitride (BN)crystals and films with the evaluation of the electronic properties of graphene on these substrates.Our study includes hBN crystals grown under different conditions (atmospheric pressure hightemperature, high pressure high temperature, pressure controlled furnace) and scalable BN filmsdeposited by either chemical or physical vapor deposition (PVD). We explore the complete processfrom boron nitride growth, over its optical characterization by time-resolved cathodoluminescence(TRCL), to the optical and electronic characterization of graphene by Raman spectroscopy afterencapsulation and Hall bar processing. Within our benchmarking protocol we achieve ahomogeneous electronic performance within each Hall bar device through a fast and reproducibleprocessing routine. We find that a free exciton lifetime of 1ns measured on as-grown hBN crystalsby TRCL is sufficient to achieve high graphene room temperature charge carrier mobilities of80000cm2 (Vs)−1 at a carrier density of |n|= 1× 1012 cm−2, while respective exciton lifetimesaround 100ps yield mobilities up to 30000cm2 (Vs)−1. For scalable PVD-grown BN films, wemeasure carrier mobilities exceeding 10000cm2 (Vs)−1 which correlates with a graphene Raman2D peak linewidth of 22cm−1. Our work highlights the importance of the Raman 2D linewidth ofgraphene as a critical metric that effectively assesses the interface quality (i.e. surface roughness) to© 2024 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.https://doi.org/10.1088/2053-1583/ad96c9https://crossmark.crossref.org/dialog/?doi=10.1088/2053-1583/ad96c9&domain=pdf&date_stamp=2024-12-16https://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://orcid.org/0009-0003-8316-523Xhttps://orcid.org/0000-0001-5540-8589https://orcid.org/0000-0003-4542-9735https://orcid.org/0000-0003-1749-2980https://orcid.org/0009-0005-3433-8063https://orcid.org/0000-0002-5225-9861https://orcid.org/0009-0006-0167-1545https://orcid.org/0000-0002-5987-0647https://orcid.org/0000-0002-5234-1937https://orcid.org/0000-0001-5605-3185https://orcid.org/0000-0002-3328-317Xhttps://orcid.org/0000-0001-7831-3953https://orcid.org/0000-0002-1042-5876https://orcid.org/0000-0002-7911-3758https://orcid.org/0000-0002-3529-557Xhttps://orcid.org/0000-0002-1278-1727https://orcid.org/0000-0002-4958-7362https://orcid.org/0000-0002-8917-5776https://orcid.org/0000-0002-1467-3105https://orcid.org/0000-0001-5889-0796https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-2359-2718mailto:bernd.beschoten@physik.rwth-aachen.de2D Mater. 12 (2025) 015017 T Ouaj et althe BN substrate, which directly affects the charge carrier mobility of graphene. Graphene 2Dlinewidth analysis is suitable for all BN substrates and is particularly advantageous when TRCL orBN Raman spectroscopy cannot be applied to specific BN materials such as amorphous or thinfilms. This underlines the superior role of spatially-resolved spectroscopy in the evaluation of BNcrystals and films for the use of high-mobility graphene devices.1. IntroductionBoron nitride (BN) with its remarkable thermalstability, chemical inertness and robust mechanicalproperties has long been used for various applica-tions [1–4]. It has been demonstrated that hexagonalboron nitride (hBN) is of particular importance forapplications in 2D material systems, exhibiting prop-erties crucial for photonics and optoelectronics, suchas efficient deep UV emissions [5, 6] and quantumphotonics capabilities [7–10]. The high thermal con-ductivity [11, 12], the large electronic bandgap [5],and the ultra-flat and inert surface [13] are import-ant prerequisites for the use as a substrate for other2D materials or for interface engineering [14–18].2D materials encapsulated in hBN allow for record-breaking charge carrier mobilites in graphene [19–24], high electronic and optical quality in trans-ition metal dichalcogenides (TMDs) [25–32] or, forexample, bilayer graphene quantum devices withultra-clean tunable bandgaps [33–36].In fundamental research, hBN flakes exfoliatedfrom bulk crystals grown either at high temperatureand high pressure (HPHT) [37–39] or at atmosphericpressure and high temperature (APHT) [6, 24, 40–50] are employed for high-quality device fabricationdue to their superior crystal quality. The synthesisof high-quality hBN crystals in a pressure-controlledfurnace (PCF) is a recent development that offers newopportunities for improvingmaterial quality [51, 52].hBN single crystals are small, a few millimeters atmost, and therefore do not meet industrial manufac-turing requirements. The transition of BN from theuse in fundamental research to industrial applicationsrequires process development capable of providinglarge area single crystal or polycrystalline films thatmeet both device requirements and high volume pro-duction needs.Techniques like chemical vapor deposition (CVD)[53–58], metal-organic CVD (MOCVD) [59–62],molecular beam epitaxy (MBE) [63–66], and phys-ical vapor deposition (PVD) [67, 68] are under devel-opment, offering potential platforms for BN sub-strates with sufficient interface and/or bulk qualitiesfor the desired technological applications. Recently,amorphous (or nanocrystalline) boron nitride (aBN),has gained interest due to its ability to be grown atroom temperature on arbitrary substrates [69] andits low dielectric constant [70–72]. Especially thefull encapsulation of CVD-grown graphene in direct-grown aBN was recently reported to have promisingelectronic properties, showing its potential as a scal-able substrate for graphene and other 2D materi-als [71].The evolving diversity of available BNsubstrates—from high-quality hBN crystals to hBN/-aBN films—underlines the need for comparableand meaningful characterization methods of boththe crystal quality itself and the ability to be usedas substrate in van der Waals heterostructures. Toassess the crystal quality, BN is mostly investigatedby cathodoluminescence (CL) [73–75], photolumin-escence (PL) [64, 76, 77] or Raman spectroscopy[2, 78, 79]. Raman spectroscopy gives a rapid andnon-invasive way to extract the quality of BN filmsand therefore is an indispensable tool to efficientlymonitor the parameter tuning during optimizationof growth processes. On the other side, CL measure-ments and especially time-resolved cathodolumin-escence (TRCL) measurements yield a much moresensitive way to evaluate the crystal quality and togain a deeper understanding of the type of crystaldefects [73–75]. Here, the free exciton lifetime, whichis limited by exciton-defect scattering, yields a sensit-ive benchmark for the bulk crystal quality. However,CL measurements are not suitable for most scalableBN growth approaches as they are only applicableto crystalline and thick (>10µm) hBN. While theseevaluations are highly important for benchmarkingthe quality of hBN crystals for optical applicationswith hBN as the active layer, methods to evaluate thesurface quality become equally important when usedas a substrate [80]. For example, correlations betweenthe amount or type of defects and the surface rough-ness seem possible but remain a topic under investig-ation [81].Graphene, due to its exceptional high chargecarrier mobility, is one of the most interesting2D materials to be used in combination with BN.Additionally to the interest due to its electronic prop-erties, graphene is highly sensitive to charge dis-order and surface roughness of the substrate, drastic-ally limiting the device performance [20, 82]. Dueto both, its huge potential for future high-mobilityapplications and its high sensitivity to the underly-ing substrate, the evaluation of graphene itself on thesubstrate of interest is an appealing way to investig-ate the suitability of various BN films or crystals as a22D Mater. 12 (2025) 015017 T Ouaj et alsubstrate for 2Dmaterials. Spatially-resolved confocalRaman microscopy on graphene encapsulated in BNprovides a very powerful and sensitive way to directlyassess strain, doping, and nm-strain variations [80,83–85] and directly link it to the electronic proper-ties extracted from charge transport measurementson Hall bar devices [22–24, 82].Here, we present a comprehensive evaluation ofvarious BN substrates and present a benchmarkingprotocol covering the characterization of the BN aswell as the evaluation of the electronic propertiesof exfoliated graphene on these BN substrates. Ourstudy includes the growth of both high-quality hBNcrystals grown via APHT or in a PCF and the growthof scalable BN films via PVD or CVD (section 2).We extract the free exciton lifetime from TRCLmeas-urements to compare the crystal quality of BN crys-tals and evaluate both exfoliated flakes and filmsvia Raman spectroscopy (section 3). Using exfoliatedgraphene, we fabricated dry-transferred devices onthe BN substrates to assess the interface quality viaspatially-resolved Raman spectroscopy (section 4). Toestablish reliable benchmarks we focus on the fullwidth at half maximum (FWHM) of the grapheneRaman 2D peak, which we identify as the mostsensitive benchmark for an early-stage evaluationof the suitability for a diverse set of BN substrates(section 5). Following a newly developed fabricationscheme, with a focus on rapid processing, we buildHall bar structures (section 6) to extract the chargecarrier mobility at different charge carrier densities(section 7). We demonstrate that graphene encapsu-lated in APHT hBN crystals compares in electronicquality to graphene encapsulated in HPHT-grownhBN crystals, reaching room temperature charge car-rier mobilities around 80000cm2 (Vs)−1 at a chargecarrier density of n= 1× 1012 cm−2. Importantly,we identify a free exciton lifetime of above 1ns tobe sufficient to achieve these high charge carriermobilities and of 100ps for charge carrier mobilit-ies up to 30000cm2 (Vs)−1. Specifically, we demon-strate that graphene on PVD-grown nanocrystallineboron nitride with a graphene 2D peak FWHMbelow22cm−1 consistently yields charge carrier mobilit-ies exceeding 10000cm2 (Vs)−1. This underscores thepotential of PVD-grown BN films as scalable sub-strates for high-mobility graphene devices.2. Growth and preparation of boronnitrideIn figure 1 the growth and preparation conditions ofhBN crystals (APHT and PCF growth) and of BN thinfilms (PVD and CVD growth) are summarized.2.1. Atmospheric pressure and high temperature(APHT)The hBN crystals in this study were grown froman iron flux (at RWTH) or chromium-nickel (atRWTH and GEMaC) flux via the APHT method (seereference [24] for details on the growth at RWTH). Aschematic illustration of the growth setup is shown infigure 1(a). The boron source is either boron powder(RWTH) mixed with the metal pieces or a pyrolyticBN crucible (GEMaC). The system is first annealedat high temperature under a continuous gas flow ofeither H2 and Ar (RWTH) or N2 (GEMaC) to min-imize contaminations with oxygen and carbon. ThehBNcrystal growth is started upon introduction ofN2while maintaining a constant pressure. After a soak-ing phase at high temperature to saturate the metalflux with B and N, the furnace is cooled down to alower temperature at a slow rate (typically between0.5 ◦C/h and 4 ◦C/h). The system is then quicklyquenched down to room temperature. The resultingthick hBN crystal layer is firmly attached to the under-lying metal ingot, as seen in figure 1(a) (right upperpanel). The hBN crystal sheet can be detached fromthe metal ingot by immersion in hydrochloric acid atroom temperature, see the detached crystal sheet inthe lower right panel of figure 1(a). This step does notaffect the quality of the hBNcrystals and simplifies thefurther processing of the hBN crystals for exfoliationand subsequent dry-transfer [24].2.2. Growth in a pressure-controlled furnace (PCF)In the PCF method, hBN crystals are grown fromthe liquid phase of Li3BN2 −BN at high temperature[52, 86, 87]. The Li3BN2 powder is pre-synthesizedfrom Li3N (Sigma Aldrich, purity> 99.5%) [88] andmixed with commercial hBN powder (20 wt% hBNand 80 wt% Li3BN2) in a molybdenum crucible.Since Li3BN2 is very sensitive to air and moisture, thegrowth preparation is performed under inert condi-tions and careful handling is necessary throughoutthe whole process. Both, hBN powder and cruciblesare pre-treated at 1200 ◦C under vacuum and anAr/H2 gas mix to remove potential contaminations.The growth is performed in a pressure-controlledfurnace (PCF) [51, 52] (schematically shown infigure 1(b)) during a fast cooling after a dwellingtime of 2 h at 1800 ◦C and a pressure of 180 MPaunder Ar atmosphere. The temperature and the pres-sure are increased at a rate of 100 ◦C min−1 and 10MPa min−1. The chamber is initially purged threetimes (Ar filling followed by pumping) to removeoxygen andmoisture. The sample obtained is an ingotcomposed of hBN crystals embedded in a solidifiedLi3BN2 matrix. Li3BN2 dissolution is then performedto retrieve individual crystals. They show a lateral size32D Mater. 12 (2025) 015017 T Ouaj et alFigure 1. Overview of growth techniques for hBN crystals and films. (a) Atmospheric pressure high temperature growth processwith schematic of a gas flow furnace. Optical images of the resulting hBN crystals on top of the iron ingot and of the crystals afterdetaching from the iron ingot. (b) Schematic of the growth of hBN crystals in a pressure controlled furnace and optical image ofhBN crystal after dissolution of Li3BN2. (c) CVD growth of hBN on Pt(111) substrate and an optical image of the transferredhBN film, approximately 1nm thick, on a Si/SiO2 wafer piece. (d) Schematic of the physical vapor deposition growth setup forBN and optical image of Si/SiO2 wafer with the PVD-grown film homogeneously covering the wafer with a thickness of 30nm.ranging from several hundreds of micrometers to fewmillimeters, exemplary shown in figure 1(b), in theright image. The crystals have been previously usedas encapsulants for TMDs and graphene to obtainoptical and electronic devices [52, 89].2.3. Chemical vapor deposition (CVD)For the growth of hBN layers, Pt(111) thin filmswith a thickness of 500nmwere prepared on sapphirewafers [90]. The hBN films were grown via CVD inan ultra-high vacuum cold-wall chamber for wafersup to 4-inch [91, 92]. Prior to all hBN preparations,the Pt/sapphire substrates were cleaned by a series ofargon sputtering, O2 exposure and annealing cycles to1200K until sharp Pt(111) (1×1) LEED patterns wereobserved. Subsequently, hBN layers were prepared attemperatures above 1000K with borazine (HBNH)3as precursor with a partial pressure of 10−7mbar(figure 1(c)). The quality of grown hBN layers wereevaluated with scanning low energy electron diffrac-tion (x-y LEED), x-ray photoelectron spectroscopy(XPS), ultraviolet photoelectron spectroscopy (UPS),scanning tunneling microscopy (STM) and atomicforce microscopy (AFM). The reported thickness isderived from XPS intensity values.The transfer procedure employs the electrochem-ical ‘bubbling’ method [93]. First, the hBN/Pt(111)sample was spin-coated with 4wt% polymethylmethacrylate (PMMA) (495K). Then we put thePMMA/hBN/Pt sample as working electrode and a Ptwire as counter electrode in a 1.0M KCl solution. Anegative voltage between −3V and −5V was appliedto the sample to delaminate the hBN/PMMA filmfrom the substrate. The delaminated hBN/PMMAfilm was then rinsed in ultrapure water (Milli-QAdvantage A10) and transferred onto a clean 280nmSi/SiO2 substrate with gold markers. In the next step,the PMMA was removed via a sequence of acet-one/ethanol baths and gradual annealing in air attemperatures up to 600K for 3h. Figure 1(c) shows arepresentative transferred hBN film with a thicknessof approximately 1nm.2.4. Physical vapor deposition (PVD)Thin nanocrystalline BN films are grown via phys-ical vapor deposition using an ion beam assisteddeposition process (IBAD-PVD). The films exhibithexagonal bonding structure, as assessed by x-rayabsorption near edge spectroscopy (XANES) [94, 95]but lack of x-ray diffraction. The films were growndirectly on Si/SiO2 wafers with an oxide thickness of285nm and pre-defined Cr/Au marker. The growthwas performed at room temperature using nitrogengas and a solid boron source. The IBAD process con-sisted in the interaction of a directional beam of500eV nitrogen ions from a Kaufman source, withconcurrent boron atoms from an electron beam evap-orator. The growth setup is schematically depicted infigure 1(d). The N-ion and B-atom fluxes were care-fully tuned to obtain stoichiometric BN and avoidother BxNy phased and bonding configurations [67].The thickness of the resulting BN film is 30nm andhomogenously covers the whole wafer, as shown infigure 1(d).42D Mater. 12 (2025) 015017 T Ouaj et al3. Cathodoluminesence and Ramanspectroscopy on BN crystals and filmsThe as-grown hBN crystals (APHT and PCF) areexamined by means of TRCL measurements andspatially-resolved confocal Raman microscopy.Raman spectroscopy offers a fast and non-invasivetool to spatially probe optical phonons and theirlifetimes, which provide a measure of the crystallin-ity of hBN [2]. Time-resolved cathodoluminescence(TRCL) measurements allow to determine the life-times of the free excitons, which are strongly affectedby scattering with defects, providing a valuable andsensitive tool to locally probe the crystal quality ofhBN crystals.3.1. CathodoluminescenceTRCL measurements were performed on isolatedbulk hBN crystals. For each supplier (GEMaC,RWTH, LMI), crystals from multiple growth batcheswere investigated. The deep UV spectra were recor-ded at room temperature in a JEOL7001F field-emission-gun scanning electron microscope (SEM)coupled to a Horiba Jobin-Yvon cathodolumines-cence (CL) detection system, as described in detail inearlier works [73–75]. To allow for time resolution, acustom-built fast-beam blanker was installed insidethe SEM column, as described in [75]. The dynamicsof the free exciton population is captured by measur-ing the time-dependent CL intensity in a wavelengthrange of 215± 7.5nm with a temporal resolution of100ps. This spectral range corresponds to the mainluminescence feature of high quality hBN crystals.The 215nmCL signal results from the indirect excitonrecombination assisted via optical phonons. To focuson bulk properties and minimize surface recombina-tions, the electron beam acceleration voltage was setto 15kV [73, 96]. The current was maintained at alow value of 85pA to prevent nonlinear effects [97].An exemplary TRCL measurement for each type ofhBN crystal investigated is shown in figure 2. At t= 0,the luminescence peak intensity is normalized to 1,to allow for better comparison of the time evolution.The free exciton lifetime τCL is extracted by fittingthe first decay with a single exponential decay func-tion [75]. We obtain τ = 0.15ns,1.67ns and 3.32nsfor PCF (LMI), APHT (RWTH) andAPHT (GEMaC)crystals, respectively. Statistical evaluation (meanand standard deviation) across different growthbatches and spatial positions on the crystals yieldedτ = (0.11± 0.07)ns,(1.4± 0.3)nsand(3.0± 0.4)nsfor 22, 13 and 8 measured areas on PCF (LMI),APHT (RWTH) and APHT (GEMaC), respectively.We emphasize the need for statistical evaluation dueto notable crystal-to-crystal variations.The variation in free exciton lifetimes is associatedwith differences in defect densities in the crystals.Figure 2. Decay of the free exciton luminescence at 215nmfor hBN crystals grown by APHT (RWTH and GEMaC)and PCF (LMI) measured by time-resolvedcathodoluminescence. The luminescence intensity isnormalized to 1 at t= 0 for all measurements. Arepresentative trace is shown for each crystal type togetherwith an exponential fit to extract the respective lifetime τCL.The table in the inset shows the averaged lifetimes frommeasurements at different crystal positions and batcheswith the total number of measurements.We note that APHT-grown crystals exhibit lifetimessimilar to those produced via the HPHT method[75] which is consistent with a low defect density.In contrast, PCF-grown crystals show significantlyshorter lifetimes. The higher defect density could res-ult from vacancies, impurities, or structural anom-alies, which all may affect the free lifetime. The vari-ations in lifetimes, even within crystals grown bythe same method, highlights the need for carefulcrystal selection for specific experiments or applic-ations. Understanding these defect-induced changesin the optical properties is crucial for the furtherdevelopment of hBN applications in optoelectronicsand quantum technology. The benchmarking of hBNcrystals via TRCL also sets the stage for understand-ing their role as substrates in graphene-based devices.The observed variation in the exciton lifetime of hBNgrown by the different methods is expected to cor-relate with the electronic quality of encapsulated 2Dmaterials. Its impact on the charge carrier mobility inhBN/graphene/hBN Hall bar devices will be detailedfurther below.3.2. Confocal Raman spectroscopyRaman spectroscopy is a practical and widely usedoptical probe for characterizing both hBN crystalsand thin films. Its advantage of accessibility makes itan important tool formonitoring the effect of changesin the growth parameters on the crystal quality ofBN. The primary benchmark for assessing the crystalquality of hBN via Raman spectroscopy is the FWHMΓE2g of the E2g Raman peak, which correlates withthe lifetime τE2g of optical phonons correspondingto intralayer vibrations of B and N atoms [98]. The52D Mater. 12 (2025) 015017 T Ouaj et alFigure 3. Raman spectroscopy of hBN crystals for APHT (RWTH and GEMaC) and PCF (LMI) from left to right. (a)–(c) Opticalimages of exfoliated hBN flakes that were spatially mapped by confocal Raman spectroscopy. (d)–(f) Spatially-resolved Ramanmaps of the FWHM of the E2g mode for each flake shown in (a)–(c). (g)–(i) Representative Raman spectra with the statisticaldistribution of the FWHM of each flake ((d)–(f)) as shown in the respective insets. The scale bars are 10µm.contributions to the phonon linewidth in hBN witha natural isotopic content of boron originate primar-ily from isotopic disorder-induced scattering, anhar-monic phonon decay, or impurity scattering [99].Thus, in hBN with the same crystal structure andisotope distribution, the variations in FWHM aremainly due to the degree of disorder in the crystal [2].Changes in bond lengths due to increased defect dens-ity or not-purely sp2-hybridized bonds might alsoimpact the FWHM, as these factors contribute toaveraging effects over phonons of different frequen-cies. Typically, high quality hBN crystals grown viaHPHT or APHT exhibit a FWHM around 8cm−1 [2,79, 100]. For thin BN films this value can increase upto 40cm−1 [2].3.2.1. Experimental setupRaman measurements were conducted using a com-mercial confocal micro-Raman setup (WITec alpha300 R) at room temperature. We utilized a 532nmexcitation wavelength, a laser power of 2mW anda 100× magnification objective with a numericalaperture of 0.9. The inelastically scattered light wascollected through a fibre (core diameter 100µm)and sent to a CCD through a half meter spectro-meter equipped with a 1200 linesmm−1 grating. Forlinewidth analysis of high-quality hBN crystals, weemployed a grating with 2400 linesmm−1.3.2.2. hBN crystalsWe start by exfoliating thin hBN flakes from the bulkcrystals using tape (Ultron 1007 R) onto Si/SiO2wafer with a 90 nm oxide layer and observe sim-ilar distribution of thicknesses and lateral sizes forflakes from all crystal suppliers. Flakes with thick-nesses between 20nm and 40nm were selected basedon their color contrast towards the substrate [101],as these thicknesses are optimal for building state-of-the-art hBN-encapsulated graphene devices. Infigures 3(a)–(c), we present optical images of rep-resentative flakes from the three suppliers ((a) forAPHT (RWTH), (b) for APHT (GEMaC), and (c)for PCF (LMI)). All flakes look similar in termsof contamination or thickness homogeneity. This isinferred from the optical images in figures 3(a)–(c),where each crystal flake shows a single homogenousoptical contrast to the substrate and no optically vis-ible contaminations.The FWHM of the E2g peak is extracted by fittinga single Lorentzian function to the individual Ramanspectra. Spatially-resolved maps of the FWHM areshown in figures 3(d)–(f). A respective single Raman62D Mater. 12 (2025) 015017 T Ouaj et alFigure 4. Boron nitride films on Si/SiO2. (a) Optical microscope image of a PVD-grown BN film with a thickness of 30nm. (b)Representative Raman spectrum of the PVD-grown BN film (position: red cross in (a)) (c) Surface topography measured byatomic force microscopy. Inset shows a representative line profile along the red horizontal line in the map. (d) Optical microscopeimage of a CVD-grown BN film which was transferred to Si/SiO2 (e) Representative Raman spectrum of the CVD-grown BN film(position: red cross in (d)) (f) Histogram of the FWHM of the E2g hBN mode shown in (d) extracted from single Lorentzian fit.spectrum at a representative position, along with thecorresponding histogram to the FWHM map, areshown in figures 3(g)–(i).There is a narrow hBN Raman peak around1365cm−1 for all flakes. The maps in figures 3(d)–(f)reveal a homogenous and narrow distribution of theFWHM, suggesting uniform crystal quality through-out the exfoliated flakes. A closer inspection of thestatistical distribution (insets of (g)–(i)), reveals aGaussian distribution of the FWHM around 8cm−1,demonstrating high crystallinity for all flakes. Thesevalues are comparable to previous studies on APHTorHPHTgrownhBN [2]. Interestingly, we observe nosignificant difference in the Raman FWHM betweenPCF and APHT crystals. This observation seems sur-prising since the CL lifetime of the PCF-grown hBNflakes is more than an order of magnitude shorterthan the respective lifetimes of the APHT-grown hBNcrystals (see figure 2). It is, however, important toemphasize again that main contributions to the E2gpeak’s FWHM in natural hBN results from isotopicdisorder [99], that is typically the same for all. Whileisotopic disorder is generally the same for all hBNcrystals, variations in defect type and density can sig-nificantly vary between different growth methods.Our studies suggest that the presence of crystal defectsin high-quality hBN crystals can barely be probedby Raman spectroscopy. Analyzing the lifetimes offree excitons, on the other hand, offers a significantlymore sensitive tool for the local probing of crystaldefects.3.2.3. BN filmsWenext evaluate boron nitride films, which are eithergrown directly on the Si/SiO2 substrate (PVD) orgrown by means of CVD and then wet-transferred toa Si/SiO2 substrate. In the case of boron nitride films,cathodoluminescence measurements are not feasible,mainly due to the small thickness of the films. Anoptical image of the PVD-grown film is shown infigure 4(a). We observe a homogeneously grown filmover the entire wafer with some spots where the BNis damaged. In figure 4(b) we additionally show aRaman spectrum at a representative position. In con-trast to the previously shown Raman spectra of flakesfrom exfoliated hBN crystals, we do not observe asingle narrow Raman peak. Instead, a broad responseranging from 1100cm−1 to 1600cm−1 is observed.This can be related to the amorphous nature of the BNfilm, which leads to a strong broadening of the Ramanpeak due to the inclusion of nanocrystalline regionswithin the BN film [102]. The broadening may alsoresult from random strain effects [71]. They lead toan averaging of different bond lengths between theatoms resulting in a statistical averaging of the Ramanresponse due to variations in the phonon frequencies.As the PVD grown BN films will later be used asa substrate for graphene, we next explore their sur-face roughness by atomic force microscopy (AFM).Figure 4(c) displays an AFM image for a small regionof the sample shown in figure 4(a). A root meansquare (RMS) roughness of 0.2nm is extracted fromthis map. This low value is in line with RMS values72D Mater. 12 (2025) 015017 T Ouaj et alof hBN and the 2D semiconductor WSe2, which haveproven to be ideal substrates for graphene [80].In figure 4(d) we show an optical image of theCVD grown BN film which was transferred on SiO2.Due to the wet-transfer process and because multiplelayers of hBN are transferred on top of each other, theBN film does not have a homogenous thickness. FromXPS measurements we estimate an average thicknessof 3 layers of hBN. The Raman spectrum at a repres-entative position is shown in figure 4(e) together witha histogramof the distribution of the FWHM in panel(f). We observe a well-defined hBN Raman peak atωE2g = 1365cm−1 with a FHWM of ΓE2g = 34cm−1.The large FWHM is in striking contrast to the pre-viously discussed crystals but comparable to otherBN films shown in literature [103–105]. We attributethe large FWHM to the wet transfer procedure andthe remaining PMMA residues on the transferred BNfilm.To conclude the pre-characterization of BN crys-tals and films, we note that there is no commonmethod which is either sensitive enough or applicableto all forms of BN, i.e. crystals and films. Especially,for nanocrystalline or amorphousBN films,which arerecognized as potential substrates for scaled devices,the usual characterization methods are not feasible.We therefore proceed with the evaluation of graphenein contact with BN, by using graphene as a sensitivedetector for the suitability of the underlying BN/hBNsubstrate for charge transport.4. Dry-transfer of graphene encapsulatedin BNThe next step in the benchmarking protocol is tobuild van derWaals heterostructures using BNmater-ial to fully encapsulate graphene. The substrate qual-ity of BN is then explored by probing the elec-tronic properties of graphene using both spatially-resolved Raman spectroscopy and charge transportmeasurements.For the stacking of the heterostructures we startby exfoliating hBN and graphene flakes onto 90nmSi/SiO2. The flakes are searched and classified usinga home-built automatic flake detection tool [101].Suitable flakes with a thickness between 20 and 40 nmare identified and stacked on top of each other usingstandard dry-transfer methods with poly(bisphenolA carbonate) (PC) film on top of a drop-shapedpolydimethylsiloxane (PDMS) stamping tool [106].The stacking process is schematically depicted infigure 5. For the benchmarking of hBN crystals(APHT and PCF), graphene is picked up using hBNflakes, which were exfoliated from their respectivebulk crystals while for the evaluation of BN filmsthe graphene is picked up by exfoliated HPHT-grown hBN (figures 5(b)–(d)). In the next step, theFigure 5. Dry transfer of graphene/BN heterostructures. (a)Schematic representation of the used stamp. A PC film isplaced on a self-assembled PDMS droplet on a glass slide.The stamp is placed above the silicon wafer which is placedon a heatable stage. (b) The process starts with heating thesubstrate to T= 110 ◦C. (c) The stamp is brought intocontact with the hBN flake and the temperature is loweredfrom T= 110 ◦C to T= 80 ◦C. (d) The hBN flake ispicked up at T= 80 ◦C. (e) The hBN flake is used to pickup the exfoliated graphene flake at T= 80 ◦C. (f)–(g) Forthe BN films, the hBN/graphene is placed directly on theBN film and released from the stamp at T= 180 ◦C todetach the PC from the PDMS and bond it to the substrate.(h)–(j) For the full encapsulation in hBN an additionalhBN flake is first picked up (h) and then placed on the finalsubstrate and heated to T= 180 ◦C (i) to release the PCfrom the PDMS and (j) bond it to the substrate.hBN/graphene half stack is either transferred ontocorresponding hBN crystal flakes (figures 5(h)–(j))or placed onto the BN films (figures 5(f)–(g)). Theprotection of graphene from the top by an hBNcrystal is important to ensure heterostructures ofcomparable quality and exclude influences on thegraphene quality and device performance that canbe caused by chemicals or airborne contaminations[107] during the subsequent processing steps. Opticalmicroscope images of the finished stacks are shown infigures 7(a)–(e). The lateral size of the stacks is lim-ited by the size of the exfoliated hBN and grapheneflakes. Within this project, we characterized in totalover 40 dry-transferred samples to obtain a statisticalevaluation of the various BN substrate and to excludesample-to-sample variations.5. Raman spectroscopy on BN-grapheneheterostructures5.1. Extraction of strain, strain variations anddopingWe first give an overview on the key concepts ofgraphene-based Raman spectroscopy. Figure 6(a)82D Mater. 12 (2025) 015017 T Ouaj et alFigure 6. Raman spectrum of graphene and influence onpeak positions. (a) Raman spectrum of grapheneencapsulated in hBN. (b) Schematic presentation of theexpected influence of strain, doping and screening on thepositions of the Raman G and 2D peak of graphene.shows a typical Raman spectrum of grapheneencapsulated in hBN crystals. Three prominent peaksare typically observed corresponding to the aboveanalyzed hBN E2g peak and the graphene G and 2Dpeak. The G peak in graphene results from out-of-phase in-plane vibrations of two carbon atoms ofthe two sublattices and involves phonons from theΓ-point, whereas the double resonant 2D peak cor-responds to a breathing mode, involving phononsnear the K-point [108–110].A crucial and sensitive quantity for the evaluationof the electronic properties of graphene is the FWHMof the 2D peak, which is directly connected to theextent of nm-scale strain variations within the laserspot [84] and therefore also contains information onthe roughness of the substrate [80]. As strain vari-ations locally break the hexagonal symmetry of thelattice, a vector potential is induced which in turnleads to an increased probability of backscattering ofelectrons in charge transport leading to a reducedcharge carrier mobility [82]. The 2D FWHM is there-fore the main quantity of interest in our study as itdirectly connects the interface quality given by the BNwith the electronic quality of the adjacent graphenesheet.The G and 2D peak are both susceptible to strainas well as doping [83] and the 2D peak position isadditionally influenced by dielectric screening fromthe environment [111], which is, however, not relev-ant in the scope of this study. To separate the effectsof strain and doping from spatially-resolved Ramanmaps, the positions of the 2D and G peak are plottedagainst each other, as illustrated in figure 6(b). Sincethe two peaks shift differently as function of dopingand strain, the slopes of the distributions can be usedto qualitatively evaluate the type of disorder in thesystem (strain and/or doping). A distribution paral-lel to the strain axis has a slope of 2.2 and is connec-ted to biaxial strain whereas a distribution along thedoping axis has a slope ranging between 0.3 and 0.7depending on both their charge carrier type and thesubstrate [83, 112].5.2. Results of spatially-resolved RamanspectroscopyRaman measurements were performed with thesame setup as for the characterization of the hBNcrystals and films, using a grating of 1200 linesmm−1.Figures 7(f)–(j) show Raman maps of the graphene2D linewidth for the regions highlighted withblack dashed rectangles in the optical images infigures 7(a)–(e) for each BN source, respectively. Thecorresponding histograms are shown in figures 7(k)–(o) of a selected region of interest, highlighted witha dashed rectangle in the corresponding panel infigures 7(f)–(j). The color scale is the same for allmaps. Regions of a higher 2D linewidth within astack may either result from bubbles (hydrocar-bons) that are trapped at the interface betweenhBN and graphene or may be related to regionswith multilayers. Residual hydrocarbons most likelyoriginate from tape residues during exfoliation orfrom the polymer used for stacking [113–115]. Thelatter is a commonly known challenge when usingpolymer-based dry-transfer techniques. We observethe formation of bubbles for all stacks produced inthis study.Comparison of the contamination-free regions ofthe 2D FWHM maps reveals the lowest Γ2D valuesfor graphene on APHT-hBN (figures 7(f)–(g)), fol-lowed by PCF-grown crystals (figure 7(h)) and thanthe BN films (figures 7(i)–(j)). The correspondinghistograms in figures 7(k)–(o) enable the quantitativeevaluation of the 2D FWHMmaps. The maximum ofthe statistical distribution ranges from 16.5cm−1 forAPHT-grown crystals, over 18cm−1 for PCF-growncrystals to values larger than 20cm−1 for BN films.We identify the peak position of the 2D linewidthdistribution as a robust and sensitive quantity toevaluate the interface quality of the underlying BN, inline with previous works [80, 84]. We conclude thatthe degree of strain variations in graphene is lowestfor the APHT hBN crystals, which shows that theyhave the highest interface quality (flatness) among thestudied BN.The respective ω2D vs ωG scatter plots are shownin figures 7(p)–(t), where the color code correspondsto the FWHM of the 2D peak. For the stack presen-ted in the first row of figure 7 we chose a regionwith a spatially homogeneous and low 2D FWHM.The corresponding 2D vs G peak position distribu-tion shows a strong clustering along the 2.2 strainaxis indicating very small strain variations and neg-ligible doping. For the sample in the second row, thedistribution with the lowest 2D linewidth (blue datapoints) is again mainly distributed along the strainaxis. However, areas with inclusion (bubbles) exhibitlarger 2D linewidths (green, yellow and reddish color)with a distribution outside the strain axis, which isprobably due to larger doping. The effect of doping on92D Mater. 12 (2025) 015017 T Ouaj et alFigure 7. Optical microsopce images and Raman spectroscopy of dry-transferred hBN/graphene heterostructures. (a)–(e) Opticalmicroscope images of one representative stack for each BN source, namely the APHT-hBN from RWTH and GEMaC, thePCF-hBN from LMI, the PVD grown BN film from CSIC and the transferred CVD hBN from UZH. (f)–(j) Spatially-resolvedRaman map of the 2D FWHM of graphene of the region highlighted as a dashed black rectangle in the optical images in (a)–(e).(k)–(o) Statistical representation of the 2D FWHM extracted from the region highlighted as a dashed rectangle in thecorresponding Raman maps shown in (f)–(j). (p)–(t) The 2D peak position vs G peak position. Each point is color coded with theFWHM of the 2D peak. The dashed line corresponds to the expected random strain distribution with a slope of 2.2 (see text),while the solid line corresponds to the expected doping distribution with a slope of 0.7.Figure 8. Distributions of the graphene 2D peak FWHMfor all fabricated and evaluated heterostacks, combined in asingle histogram for each BN source.the peak positions is most clearly seen for the PVD-grown BN shown in the fourth row of figure 7. Thepeak positions show a curved distribution that resultsfrom both strain and doping.To go beyond the evaluation of the comparisonof representative examples, we plot the graphene 2Dlinewidth of high-quality regions of all evaluatedsamples in a combined histogram in figure 8. Forthe APHT-grown crystals we observe narrow distri-butions of the graphene 2D linewidth with the max-imum at 16.5cm−1, demonstrating an excellent andreproducible interface quality between graphene andhBN over a number of 20 different heterostructures102D Mater. 12 (2025) 015017 T Ouaj et alwith hBN crystals taken from different batches. Thehistogram distribution of the PCF-crystals showsa broader distribution ranging from 17.5cm−1 to19cm−1 indicating a larger amount of strain vari-ations, and when evaluating different stacks, we addi-tionally observe a larger sample-to sample variationin the 2D linewidth distribution.While the analysis of the free exciton lifetime τCLin figure 2 shows slightly shorter lifetimes for RWTH-APHT crystals compared to the GEMaC-APHT crys-tals there are no differences in the amount of nm-strain variations of encapsulated graphene as inferredfrom Raman spectroscopy. In contrast, the broaderand shifted graphene 2D linewidth distribution ofheterostacks fabricated by the PCF crystals seems tobe related to their shorter exciton lifetimes. As thegraphene 2D linewidth is connected to nm-strainvariations caused by the roughness of the substratesurface, we conclude that the defect concentration inthe PCF-grown crystals is so high that it affects theelectronic properties of graphene. Further, this quant-ity allows us to compare various substrates independ-ent on their crystal nature to each other.6. Processing into Hall bar structuresWe next determine the key quantity of interest, thecharge carrier mobility of graphene, and link it tothe Raman 2D linewidths of graphene and the freeexciton lifetimes of the BN substrate. For this pur-pose, the fabricated heterostructures are patternedinto Hall bar devices and electrically contacted toperform gate-dependent charge transport measure-ments. For this study, we established a reproduciblefabrication process yielding a high homogeneity ofthe electronic quality of graphene within a device aswell as a high throughput of functioning contacts. Forall devices we applied the same fabrication routine.A simplified overview of the various processingsteps is shown in figure 9(e). First, the Hall barstructure is defined by electron beam lithography(EBL) (step 1). Subsequently, 30nm aluminum (Al)is deposited using electron beam evaporation with arate of 0.1nm s−1 (step 2) and after lift-off we remainwith the final Hall-bar structure protected by the Alhard mask (step 3). The structure is subsequentlyetched using atomic layer etching (Oxford Plasma Pro100) using Ar/SF6 with a flow rate of 5/20 sccm andHF power of 50W and a 5 s oxygen etch pulse. The Alis chemically removed using tetramethylammoniumhydroxide (TMAH) (step 4). The contacts to the Hallbar are defined in a second EBL step (step 5) and5nm/70nm of Cr/Au is evaporated, with a rate of0.2nm s−1 and 0.5nm s−1 (step 6). An optical micro-scope image of a representative, structured and con-tacted device is shown in figure 9(b).At this point, it is important to note that we havetaken particular care to minimize the time betweenthe individual processing steps. The etching, the sub-sequent second lithography step and the evaporationof Cr/Au was performed within the same day. By fab-ricating many devices, we have clear evidence that thetime window between etching into the Hall bar struc-ture where we expose the edges of graphene to air andthe deposition of the side contacts to graphene shouldbe minimized. For all devices, this time window wasbelow 4h.6.1. Influence of processing on the electronicproperties of grapheneIn this section we discuss the impact of the Hall barprocessing onto themechanical and electronic qualityof the devices by using spatially-resolved Raman spec-troscopy. In figure 9 we show a representative devicePCF (LMI), with an optical image of the stack in panel(a) and the final device in panel (b). Figures 9(c)and (d) depict spatially-resolved Raman maps of thegraphene 2D linewidth of the heterostructure beforeand after processing, respectively. The black rect-angle in figure 9(c) illustrates the region chosen forthe Hall-bar patterning, and only the Raman datafrom this region are used for comparison with thefinal Hall bar. The respective histogram is shown infigure 9(f) (green data). A comparison of the twomaps in figures 9(c) and (d) shows: (i) an overallincreases in the Raman 2D FWHM in the center ofthe Hall bar, which leads to a shift of the respect-ive histogram (red data in figure 9(f)) towards higherwavenumbers and (ii) a strong increase in linewidthtowards the edges of the Hall bar (reddish color infigure 9(d) that is seen as a tail in the histogramextending to values above 20 cm−1. This findingcould be linked to mechanical stress that occurs dur-ing the fabrication steps. The different temperaturesin the fabrication process, e.g. after baking the resistfor lithography or during etching, can lead to stressdue to the different thermal expansion coefficients ofthe materials within the stack and the substrate.Considering the Raman 2D and G peak positionsin figure 9(g), we clearly observe a red shift of thepositions along the 2.2 strain line for the stack afterfabrication. This cloud (red data points) has shif-ted towards phonon frequencies closer to the pointrelated to that of ‘pristine’ graphene [83], suggest-ing that strain release may have occurred during thedevice fabrication. We only show one example here,but this finding is observed inmany different samples,regardless of the type of BN used. A more detailedinvestigation is beyond the scope of this paper andfuture works focusing on the monitoring of differ-ent fabrication steps are necessary to draw clearerconclusions.112D Mater. 12 (2025) 015017 T Ouaj et alFigure 9. Fabrication of Hall-bar structures from van der Waals (vdW) heterostructures. (a) Optical microscope image ofhBN/graphene/hBN vdW heterostructure. The black rectangle denotes the area mapped by Raman spectroscopy shown in panel(c). (b) Optical microscope image of the patterened and contacted Hall bar. (c) Spatially-resolved Raman map of the graphene 2DFWHM. The black rectangle corresponds to the position where the Hall bar is placed. (d) Spatially-resolved Raman map of the2D FWHM of the finished Hall bar. (e) Schematic of the process overview for Hall bar structures. (1) Electron beam lithographyto define the Hall bar structure, followed by (2) electron beam evaporation of aluminum and (3) subsequent atomic layer etching.After chemical etching of the aluminum in TMAH (4) the contacts are defined in a second EBL step (5) and the Hall bar is finallycontacted by Cr/Au evaporation (6). (f) Histogram of the Raman 2D peak FWHM before (green) and after (red) Hall barfabrication. (g) Scatter plot of the graphene 2D vs G peak position before and after Hall bar fabrication. The black line shows theexpected distribution for biaxial strain (slope= 2.2) and the grey line for doping (slope= 0.7).Figure 10. Charge transport measurements on graphene/BN Hall bars. (a) Optical image of a representative Hall bar structurewith a schematical representation of the electrical wiring. (b) Four-terminal resistivity (conductivity) as function of the siliconback gate voltage. (c)–(g) Extracted Drude mobilities as function of charge carrier density for HPHT-NIMS, APHT-RWTH,RWTH-GEMaC, PCF-LMI, PVD-CSIC and CVD-UZH (wet-transferred), respectively. Traces of the same color correspond tomultiple regions measured within the same device. Different colors correspond to different devices.122D Mater. 12 (2025) 015017 T Ouaj et al6.2. Room temperature charge carrier mobilitiesThe individual Hall bars with the different BN sub-strates were fabricated in heterostack regions of thelowest possible and homogeneous graphene Raman2D FWHM (as an example, see black rectangle infigure 9(c)). All charge transport measurementswere taken at room temperature under vacuum.An example of a Hall bar with the measurementscheme is depicted in figure 10(a). We use an ACvoltage V0 = 1V at a frequency of 77 Hz and a seriesresistance of RP = 1MΩ to pass a constant currentof I= 1µA between the source and drain contact.The four-terminal voltage drop is measured for dif-ferent regions along the graphene transport chan-nel, labelled as Vxx in figure 10(a) for the upperregion as an example. This voltage drop converts tothe resistivity (1/conductivity) following ρ= 1/σ =W/L ·Vxx/I, where L is the distance betweenthe contacts and W the width of the transportchannel.Figure 10(b) shows the gate dependent resistiv-ity and conductivity for an APHT device (red tracesin panel (d)). For all measured regions, the con-ductivity σ reaches at least 400e2/h at large gatevoltages, i.e. large charge carrier densities, which ismainly limited by electron–phonon scattering [20].Importantly, and in contrast to previous studies,we observe homogeneous transport properties alongthe graphene channel and a high yield of function-ing contacts (larger than 90 %). While the elec-tronic homogeneity is likely due to the pre-selectionof the regions via Raman mapping, we link thehigh throughput of functioning contacts to thedecreased time between the etching (i.e. exposingof graphene contact areas) and evaporation of theCr/Au.The charge carrier density n is extracted fromHall effect measurements. It is connected to the gatevoltage by n= α(VG −V0G), where V0G is the posi-tion of the charge neutrality point, i.e. the voltageof the Dirac peak, and α is the gate lever arm. Thecarrier mobilities µ= σ/(ne) of graphene with thedifferent BN substrates are shown in figures 10(c)–(h) for each BN source individually. As a reference,we show transport data for a Hall bar device where weused HPHT hBN (NIMS) (see figure 10(c)). For eachdevice, multiple regions were measured. Traces of thesame color are from different regions of the samedevice. There are only small variations in transportcharacteristics within a single device but also betweendifferent devices fabricated from the same BN source.This finding further confirms a robust and reliableprocessing routine, which was developed as part ofthe benchmarking study. For devices built by APHThBN we measure the highest charge carrier mobil-ities exceeding 80000cm2 (Vs)−1 at a charge carrierdensity of |n|= 1× 1012 cm−2 (see figures 10(c) and(d)). These values are fully in line with state of the arthigh-mobility graphene devices using HPHT hBN[20] (see also figure 10(c)) or APHT hBN from othersources [22–24]. We therefore highlight the viab-ility of APHT hBN crystals as a true alternative toHPHT hBN crystals, for high-performance graphenedevices. For the PCF-grown crystals in figure 10(f), weobserve a carrier mobility of up to 30000cm2 (Vs)−1at |n|= 1× 1012 cm−2. The lower charge carriermobility of graphene encapsulated in PCF-grownhBN, when compared to APHT hBN, is fully con-sistent with our two previous observations, a shorterfree exciton lifetime and a higher 2D linewidth ofgraphene encapsulated in PCF-grown hBN crystals.The relation between an increase in graphene 2Dlinewidth and a decrease in charge carrier mobility isunderstood in terms of increased electron backscat-tering due to the stronger nm-strain variations [82].For the PVD grown BN film (figure 10(g)) we extractcharge carrier mobilities over 10000cm2 (Vs)−1at n= 1× 1012 cm−2, while we achieve mobilit-ies around 4000cm2 (Vs)−1 at n= 1× 1012 cm−2for the CVD-grown and wet-transferred films(figure 10(h)).7. DiscussionIn figure 11, we summarize themain results of the BNbenchmarking study: (a) room temperature chargecarriermobility vs carrier density and carriermobilit-ies at n= 1× 1012 cm−2 vs (b) free exciton lifetime ofhBN, (c) ΓE2g of hBN and (d) graphene 2D linewidthof all BN substrates. In figure 11(a) we show thetransport traces for the region of highest mobilityfor each device shown in figures 10(c)–(h). As men-tioned above, APHT grown hBN allows for equallyhigh graphene mobilities as achieved for HPHT-grown hBN crystals. These hBN sources are of highrelevance for many research groups, who are inter-ested in high quality hBN crystals for fundamentalresearch. The PCF-grown crystals, following anotherroute of hBNcrystal growth, allow formobilities up to30000cm2 (Vs)−1 at n= 1× 1012 cm−2, demonstrat-ing the great potential of new synthesis routes for theproduction of high quality hBN crystals. One aim ofthis synthesis route is to satisfy the increasing demandof hBN crystals frommany research groups. However,these approaches to grow high quality hBN crystalsare not scalable, because they cannot be easily com-binedwith technologically relevant substrates and thedesired thicknesses can only be achieved via mech-anical exfoliation. Scalable methods for growing BNare therefore needed to unlock the full potential ofgraphene-based electronics in future nanoelectronicdevices.In this respect, the PVD growth method ismost promising because (i) it allows the growth of132D Mater. 12 (2025) 015017 T Ouaj et alFigure 11. Summary of the main findings of the BN benchmarking study. (a) Charge carrier mobility as a function of Drudemobility for the best transport region of each device for every supplier as a function of charge carrier density. The charge carriermobility of a reference device fully encapsulated in HPHT-hBN is also shown for reference. Charge carrier mobility at a chargecarrier density of 1× 1012 cm−2 as function of (b) the free exciton lifetime, (c) the averaged FWHM of the E2g Raman peak, and(d) the averaged graphene Raman 2D linewidth of the Hall bar after fabrication.tens of nanometer thick films with very low sur-face roughness and (ii) it can be deposited dir-ectly onto Si/SiO2 substrates. The large BN thicknessscreens disorder from the silicon substrates, while thedeposition on the target substrates prevents the needfor large scale layer transfer. Most importantly, thePVD-grown BN allows for room temperature carriermobilities of graphene exceeding 10000cm2 (Vs)−1at n= 1× 1012 cm−2. We conclude that the lowtemperature PVD growth process of BN on SiO2is a promising platform for achieving scalable BNsubstrates not only for graphene, but also for other 2Dmaterials.If we compare the charge carriermobility with theRaman 2D FWHM we see a clear trend of decreas-ing mobility with increasing 2D FWHM. This isin good agreement with the finding that nm-scalestrain variations are the limitation for high chargecarrier mobilities [82] and shows that the grapheneRaman 2D FWHM is a good measure for bench-marking as used in the ICE TS 62 607-6-6 key controlcharacteristics [116]. Whereas we do not find a cor-relation between the FWHM of the hBN Raman peakand the charge carrier mobility we observe a clearcorrelation between the CL lifetime and the mobil-ity, as shown in figure 11(b). We observe that weneedCL lifetimes of over 1ns to achieve charge carriermobilities in the range of 80000cm2 (Vs)−1 at n=1× 1012 cm−2. For CL lifetimes of 100ps we achievecharge carriermobilities up to 30000cm2 (Vs)−1. Theinterface quality is therefore connected to the hBNcrystal quality, i.e. the number of defects, in a sens-itive way.In conclusion, we have presented a comprehens-ive study of the electronic properties of grapheneon different boron nitride substrates using a newlydeveloped reproducible processing routine. We haveshown the complete process from boron nitridesynthesis, over its optical characterization, to theoptical and electronic characterization of grapheneafter encapsulation and Hall bar fabrication. Weidentify the Raman spectrum of BN as a valuablemeasure for distinguishing hBN in the high crys-tallinity limit from BN films, but we also point outthe limitations of the Raman analysis when compar-ing high-quality hBN crystals. In this respect, time-resolved cathodoluminescence has a clear advant-age over Raman spectroscopy when evaluating theas-grown quality of hBN, as the probing of thefree exciton lifetime is very sensitive to the defectsin hBN. The fabrication of graphene-based hetero-structures on BN substrates demonstrates the highsensitivity of graphene to the environment, allowinggraphene to be used as a sensitive detector of the sub-strate and interface quality. Variations in the qual-ity of the graphene-BN interface are directly reflec-ted in a broadening of the graphene Raman 2Dpeak. This broadening has a direct effect on the car-rier mobility, i.e. the mobility is inversely propor-tional to the peak of the 2D linewidth distributionof graphene. It is therefore advisable to characterizethe Raman 2D linewidth distribution of the finishedheterostructure prior to any processing. In terms ofbenchmarking we find that a CL lifetime larger than1ns is sufficient for high hBN crystal quality andhigh graphene-hBN interface qualities with low nmstrain variations in graphene, which is essential forfundamental studies on highest mobility graphene-based devices. For scalable approaches we see thata graphene Raman 2D linewidth below 22cm−1 isnecessary to achieve charge carrier mobilities over10000cm2 (Vs)−1. PVD-grown BN films, therefore,offer a promising platform for scalable high mobilitygraphene devices.142D Mater. 12 (2025) 015017 T Ouaj et alData availability statementThe data supporting the findings of this study areavailable in a Zenodo repository under, https://doi.org/10.5281/zenodo.13684712 [117].AcknowledgmentsThis project has received funding from the EuropeanUnion’s Horizon 2020 research and innovationprogramme under Grant Agreement No. 881603(Graphene Flagship), T O, S B, P S, C S, and B Backnowledge support from the European ResearchCouncil (ERC) under Grant Agreement No. 820254,and the Deutsche Forschungsgemeinschaft (DFG,German Research Foundation) under Germany’sExcellence Strategy—Cluster of Excellence Matterand Light for Quantum Computing (ML4Q) EXC2004/1—390534769. H Y C was supported bya SPARK grant of the Swiss National ScienceFoundation (Grant No. CRSK-2_220582). A Hacknowledges a Forschungskredit of the Universityof Zürich (Grant No. FK-20-206 114). K W and TT acknowledge support from the JSPS KAKENHI(Grant Numbers 21H05233 and 23H02052) andWorld Premier International Research CenterInitiative (WPI), MEXT, Japan.ORCID iDsTaoufiq Ouaj https://orcid.org/0009-0003-8316-523XChristophe Arnold https://orcid.org/0000-0001-5540-8589Jon Azpeitia https://orcid.org/0000-0003-4542-9735Julien Barjon https://orcid.org/0000-0003-1749-2980José Cascales https://orcid.org/0009-0005-3433-8063Huanyao Cun https://orcid.org/0000-0002-5225-9861David Esteban https://orcid.org/0009-0006-0167-1545Mar Garcia-Hernandez https://orcid.org/0000-0002-5987-0647Thomas Greber https://orcid.org/0000-0002-5234-1937Ignacio Jiménez https://orcid.org/0000-0001-5605-3185Catherine Journet https://orcid.org/0000-0002-3328-317XPaul Kögerler https://orcid.org/0000-0001-7831-3953Annick Loiseau https://orcid.org/0000-0002-1042-5876Camille Maestre https://orcid.org/0000-0002-7911-3758Marvin Metzelaars https://orcid.org/0000-0002-3529-557XPhilipp Schmidt https://orcid.org/0000-0002-1278-1727Christoph Stampfer https://orcid.org/0000-0002-4958-7362Ingrid Stenger https://orcid.org/0000-0002-8917-5776Takashi Taniguchi https://orcid.org/0000-0002-1467-3105Bérangère Toury https://orcid.org/0000-0001-5889-0796Kenji Watanabe https://orcid.org/0000-0003-3701-8119Bernd Beschoten https://orcid.org/0000-0003-2359-2718References[1] Meng J, Wang D, Cheng L, Gao M and Zhang X 2019Recent progress in synthesis, properties and applications ofhexagonal boron nitride-based heterostructuresNanotechnology 30 074003[2] Schué L, Stenger I, Fossard F, Loiseau A and Barjon J 2016Characterization methods dedicated to nanometer-thickhBN layers 2D Mater. 4 1–11[3] Backes C et al 2020 Production and processing of grapheneand related materials 2D Mater. 7 022001[4] Naclerio A E and Kidambi P R 2023 A review of scalablehexagonal boron Nitride (h-BN) synthesis for present andfuture applications Adv. Mater. 35 2207374[5] Watanabe K, Taniguchi T and Kanda H 2004Direct-bandgap properties and evidence for ultravioletlasing of hexagonal boron nitride single crystal Nat. Mater.3 404–9[6] Kubota Y, Watanabe K, Tsuda O and Taniguchi T 2007Deep ultraviolet light-emitting hexagonal boron nitridesynthesized at atmospheric pressure Science 317 932–4[7] Bourrellier R, Meuret S, Tararan A, Stéphan O, Kociak M,Tizei L H G and Zobelli A 2016 Bright UV single photonemission at point defects in h-BN Nano Lett. 16 4317–21[8] Grosso G, Moon H, Lienhard B, Ali S, Efetov D K,Furchi M M, Jarillo-Herrero P, Ford M J, Aharonovich Iand Englund D 2017 Tunable and high-purity roomtemperature single-photon emission from atomic defects inhexagonal boron nitride Nat. Commun. 8 705[9] Martínez L J, Pelini T, Waselowski V, Maze J R, Gil B,Cassabois G and Jacques V 2016 Efficient single photonemission from a high-purity hexagonal boron nitridecrystal Phys. Rev. B 94 121405(R)[10] Fournier C et al 2021 Position-controlled quantumemitters with reproducible emission wavelength inhexagonal boron nitride Nat. Commun. 12 3779[11] Lindsay L and Broido D A 2011 Enhanced thermalconductivity and isotope effect in single-layer hexagonalboron nitride Phys. Rev. B 84 155421[12] Yuan C, Li J, Lindsay L, Cherns D, Pomeroy J W, Liu S,Edgar J H and Kuball M 2019 Modulating the thermalconductivity in hexagonal boron nitride via controlledboron isotope concentration Commun. Phys. 2 43[13] Xue J Sanchez-Yamagishi J Bulmash D, Jacquod P,Deshpande A, Watanabe K, Taniguchi T, Jarillo-Herrero Pand LeRoy B J 2011 Scanning tunnelling microscopy andspectroscopy of ultra-flat graphene on hexagonal boronnitride Nat. Mater. 10 282–515https://doi.org/10.5281/zenodo.13684712https://doi.org/10.5281/zenodo.13684712https://orcid.org/0009-0003-8316-523Xhttps://orcid.org/0009-0003-8316-523Xhttps://orcid.org/0009-0003-8316-523Xhttps://orcid.org/0000-0001-5540-8589https://orcid.org/0000-0001-5540-8589https://orcid.org/0000-0001-5540-8589https://orcid.org/0000-0003-4542-9735https://orcid.org/0000-0003-4542-9735https://orcid.org/0000-0003-4542-9735https://orcid.org/0000-0003-1749-2980https://orcid.org/0000-0003-1749-2980https://orcid.org/0000-0003-1749-2980https://orcid.org/0009-0005-3433-8063https://orcid.org/0009-0005-3433-8063https://orcid.org/0009-0005-3433-8063https://orcid.org/0000-0002-5225-9861https://orcid.org/0000-0002-5225-9861https://orcid.org/0000-0002-5225-9861https://orcid.org/0009-0006-0167-1545https://orcid.org/0009-0006-0167-1545https://orcid.org/0009-0006-0167-1545https://orcid.org/0000-0002-5987-0647https://orcid.org/0000-0002-5987-0647https://orcid.org/0000-0002-5987-0647https://orcid.org/0000-0002-5234-1937https://orcid.org/0000-0002-5234-1937https://orcid.org/0000-0002-5234-1937https://orcid.org/0000-0001-5605-3185https://orcid.org/0000-0001-5605-3185https://orcid.org/0000-0001-5605-3185https://orcid.org/0000-0002-3328-317Xhttps://orcid.org/0000-0002-3328-317Xhttps://orcid.org/0000-0002-3328-317Xhttps://orcid.org/0000-0001-7831-3953https://orcid.org/0000-0001-7831-3953https://orcid.org/0000-0001-7831-3953https://orcid.org/0000-0002-1042-5876https://orcid.org/0000-0002-1042-5876https://orcid.org/0000-0002-1042-5876https://orcid.org/0000-0002-7911-3758https://orcid.org/0000-0002-7911-3758https://orcid.org/0000-0002-7911-3758https://orcid.org/0000-0002-3529-557Xhttps://orcid.org/0000-0002-3529-557Xhttps://orcid.org/0000-0002-3529-557Xhttps://orcid.org/0000-0002-1278-1727https://orcid.org/0000-0002-1278-1727https://orcid.org/0000-0002-1278-1727https://orcid.org/0000-0002-4958-7362https://orcid.org/0000-0002-4958-7362https://orcid.org/0000-0002-4958-7362https://orcid.org/0000-0002-8917-5776https://orcid.org/0000-0002-8917-5776https://orcid.org/0000-0002-8917-5776https://orcid.org/0000-0002-1467-3105https://orcid.org/0000-0002-1467-3105https://orcid.org/0000-0002-1467-3105https://orcid.org/0000-0001-5889-0796https://orcid.org/0000-0001-5889-0796https://orcid.org/0000-0001-5889-0796https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-3701-8119https://orcid.org/0000-0003-2359-2718https://orcid.org/0000-0003-2359-2718https://orcid.org/0000-0003-2359-2718https://doi.org/10.1088/1361-6528/aaf301https://doi.org/10.1088/1361-6528/aaf301https://doi.org/10.1088/2053-1583/4/1/015028https://doi.org/10.1088/2053-1583/4/1/015028https://doi.org/10.1088/2053-1583/ab1e0ahttps://doi.org/10.1088/2053-1583/ab1e0ahttps://doi.org/10.1002/adma.202207374https://doi.org/10.1002/adma.202207374https://doi.org/10.1038/nmat1134https://doi.org/10.1038/nmat1134https://doi.org/10.1126/science.1144216https://doi.org/10.1126/science.1144216https://doi.org/10.1021/acs.nanolett.6b01368https://doi.org/10.1021/acs.nanolett.6b01368https://doi.org/10.1038/s41467-017-00810-2https://doi.org/10.1038/s41467-017-00810-2https://doi.org/10.1103/PhysRevB.94.121405https://doi.org/10.1103/PhysRevB.94.121405https://doi.org/10.1038/s41467-021-24019-6https://doi.org/10.1038/s41467-021-24019-6https://doi.org/10.1103/PhysRevB.84.155421https://doi.org/10.1103/PhysRevB.84.155421https://doi.org/10.1038/s42005-019-0145-5https://doi.org/10.1038/s42005-019-0145-5https://doi.org/10.1038/nmat2968https://doi.org/10.1038/nmat29682D Mater. 12 (2025) 015017 T Ouaj et al[14] Woods C R et al 2014 Commensurate–incommensuratetransition in graphene on hexagonal boron nitride Nat.Phys. 10 451–6[15] Britnell L et al 2012 Electron tunneling through ultrathinboron nitride crystalline barriers Nano Lett. 12 1707–10[16] Sharpe A L, Fox E J, Barnard A W, Finney J, Watanabe K,Taniguchi T, Kastner M A and Goldhaber-Gordon D 2019Emergent ferromagnetism near three-quarters filling intwisted bilayer graphene Science 365 605–8[17] Tebbe D, Schütte M, Watanabe K, Taniguchi T, Stampfer C,Beschoten B and Waldecker L 2023 Tailoring the dielectricscreening in WS2 graphene heterostructures npj 2D Mater.Appl. 7 29[18] Tebbe D, Schütte M, Watanabe K, Taniguchi T, Stampfer C,Beschoten B and Waldecker L 2024 Distance dependence ofthe energy transfer mechanism in WS2-grapheneheterostructures Phys. Rev. Lett. 132 196902[19] Dean C R et al 2010 Boron nitride substrates forhigh-quality graphene electronics Nat. Nanotechnol.5 722–6[20] Wang L et al 2013 One-dimensional electrical contact to atwo-dimensional material Science 342 614–7[21] Banszerus L, Schmitz M, Engels S, Dauber J, Oellers M,Haupt F, Watanabe K, Taniguchi T, Beschoten B andStampfer C 2015 Ultrahigh-mobility graphene devicesfrom chemical vapor deposition on reusable copper Sci.Adv. 1 e1500222[22] Onodera M, Taniguchi T, Watanabe K, Isayama M,Masubuchi S, Moriya R and Machida T 2020 hexagonalboron nitride synthesized at atmospheric pressure usingmetal alloy solvents: evaluation as a substrate for 2Dmaterials Nano Lett. 20 735–40[23] Sonntag J Li J, Plaud A Loiseau A, Barjon J Edgar J H andStampfer C 2020 Excellent electronic transport inheterostructures of graphene and monoisotopicboron-nitride grown at atmospheric pressure 2D Mater.7 031009[24] Ouaj T, Kramme L, Metzelaars M, Li J, Watanabe K,Taniguchi T, Edgar J H, Beschoten B, Kögerler P andStampfer C 2023 Chemically detaching hBN crystals grownat atmospheric pressure and high temperature forhigh-performance graphene devices Nanotechnology34 475703[25] Ajayi O A et al 2017 Approaching the intrinsicphotoluminescence linewidth in transition metaldichalcogenide monolayers 2D Mater. 4 031011[26] Ye Z et al 2018 Efficient generation of neutral and chargedbiexcitons in encapsulated WSe2 monolayers Nat.Commun. 9 3718[27] Raja A et al 2019 Dielectric disorder in two-dimensionalmaterials Nat. Nanotechnol. 14 832–7[28] Cadiz F et al 2017 Excitonic Linewidth Approaching theHomogeneous Limit in MoS2-Based van der WaalsHeterostructures Phys. Rev. X 7 021026[29] Ersfeld M et al 2020 Unveiling Valley Lifetimes of FreeCharge Carriers in Monolayer WSe2 Nano Lett. 20 3147–54[30] Shi Q, Shih E-M, Gustafsson M V, Rhodes D A, Kim B,Watanabe K, Taniguchi T, Papíc Z, Hone J and Dean C R2020 Odd- and even-denominator fractional quantum Hallstates in monolayer WSe2 Nat. Nanotechnol. 15 569–73[31] Volmer F et al 2023 Twist angle dependent interlayertransfer of valley polarization from excitons to free chargecarriers in WSe2/MoSe2 heterobilayers npj 2D Mater. Appl.7 58[32] Kundu B et al 2024 Electrically controlled excitons, chargetransfer induced trions and narrowband emitters inMoSe2–WSe2 lateral heterostructure Nano Lett.2024 14615–24[33] Eich M et al 2018 Coupled quantum dots in bilayergraphene Nano Lett. 18 5042–8[34] Banszerus L, Frohn B, Epping A, Neumaier D, Watanabe K,Taniguchi T and Stampfer C 2018 Gate-definedelectron–hole double dots in bilayer graphene Nano Lett.18 4785–90[35] Icking E et al 2022 Transport spectroscopy of ultracleantunable band gaps in bilayer graphene Adv. Electron. Mater.8 2200510[36] Icking E, Emmerich D, Watanabe K, Taniguchi T,Beschoten B, Lemme M C, Knoch J and Stampfer C 2024Ultrasteep slope cryogenic FETs based on bilayer grapheneNano Lett. 24 11454–61[37] Taniguchi T and Watanabe K 2007 Synthesis of high-purityboron nitride single crystals under high pressure by usingBa–BN solvent J. Cryst. Growth 303 525–9[38] Zhigadlo N D 2014 Crystal growth of hexagonal boronnitride (hBN) fromMg–B–N solvent system under highpressure J. Cryst. Growth 402 308–11[39] Fukunaga O, Nakano S and Taniguchi T 2022 Experimentalresults on the high-pressure phase diagram of boronnitride Jpn. J. Appl. Phys. 61 125502[40] Kubota Y, Watanabe K, Tsuda O and Taniguchi T 2008Hexagonal boron nitride single crystal growth atatmospheric pressure using Ni-Cr solvent Chem. Mater.20 1661–3[41] Hoffman T B, Clubine B, Zhang Y, Snow K and Edgar J H2014 Optimization of Ni–Cr flux growth for hexagonalboron nitride single crystals J. Cryst. Growth393 114–8[42] Edgar J H, Hoffman T B, Clubine B, Currie M andJiang H X 2014 Characterization of bulk hexagonal boronnitride single crystals grown by the metal flux technique J.Cryst. Growth 403 110–3[43] Liu S, He R, Ye Z, Du X, Lin J, Jiang H, Liu B and Edgar J H2017 Large-scale growth of high-quality hexagonal boronnitride crystals at atmospheric pressure from an Fe–Cr fluxCryst. Growth Des. 17 4932–5[44] Liu S, He R, Xue L, Li J, Liu B and Edgar J H 2018 Singlecrystal growth of millimeter-sized monoisotopic hexagonalboron nitride Chem. Mater. 30 6222–5[45] Zhang S-Y, Xu K, Zhao X-K, Shao Z-Y and Wan N 2019Improved hBN single-crystal growth by adding carbon inthe metal flux Cryst. Growth Des. 19 6252–7[46] Li J et al 2020 Hexagonal boron nitride single crystalgrowth from solution with a temperature gradient Chem.Mater. 32 5066–72[47] Li J et al 2021 Hexagonal boron nitride crystal growth fromiron, a single component flux ACS Nano 15 7032–9[48] Li J et al 2020 Single crystal growth of monoisotopichexagonal boron nitride from a Fe–Cr flux J. Mater. Chem.C 8 9931–5[49] Li Y et al 2021 Synthesis of centimeter-scale high-qualitypolycrystalline hexagonal boron nitride films from Fefluxes Nanoscale 13 11223–31[50] Zhang N Yang N, Wang W, Zhong X and Chen X 2021Growth of hexagonal boron nitride crystals at atmosphericpressure from CuCr flux J. Cryst. Growth 562 126074[51] Li Y, Garnier V, Steyer P, Journet C and Toury B 2020Millimeter-scale hexagonal boron nitride single crystalsfor nanosheet generation ACS Appl. Nano Mater.3 1508–15[52] Maestre C et al 2022 From the synthesis of hBN crystals totheir use as nanosheets in van der Waals heterostructures2D Mater. 9 035008[53] Chubarov M, Pedersen H Högberg H Filippov S,Engelbrecht J O’Connel J and Henry A 2013Characterization of boron nitride thin films 2013 Conf. onLasers and Electro-Optics Pacific Rim (CLEOPR) (IEEE) pp2013–04[54] Jang A-R et al 2016 Wafer-scale and Wrinkle-Free epitaxialgrowth of single-orientated multilayer hexagonal boronnitride on sapphire Nano Lett. 16 3360–6[55] Lee J S et al 2018 Wafer-scale single-crystal hexagonalboron nitride film via self-collimated grain formationScience 362 817–2116https://doi.org/10.1038/nphys2954https://doi.org/10.1038/nphys2954https://doi.org/10.1021/nl3002205https://doi.org/10.1021/nl3002205https://doi.org/10.1126/science.aaw3780https://doi.org/10.1126/science.aaw3780https://doi.org/10.1038/s41699-023-00394-0https://doi.org/10.1038/s41699-023-00394-0https://doi.org/10.1103/PhysRevLett.132.196902https://doi.org/10.1103/PhysRevLett.132.196902https://doi.org/10.1038/nnano.2010.172https://doi.org/10.1038/nnano.2010.172https://doi.org/10.1126/science.1244358https://doi.org/10.1126/science.1244358https://doi.org/10.1126/sciadv.1500222https://doi.org/10.1126/sciadv.1500222https://doi.org/10.1021/acs.nanolett.9b04641https://doi.org/10.1021/acs.nanolett.9b04641https://doi.org/10.1088/2053-1583/ab89e5https://doi.org/10.1088/2053-1583/ab89e5https://doi.org/10.1088/1361-6528/acf2a0https://doi.org/10.1088/1361-6528/acf2a0https://doi.org/10.1088/2053-1583/aa6aa1https://doi.org/10.1088/2053-1583/aa6aa1https://doi.org/10.1038/s41467-018-05917-8https://doi.org/10.1038/s41467-018-05917-8https://doi.org/10.1038/s41565-019-0520-0https://doi.org/10.1038/s41565-019-0520-0https://doi.org/10.1103/PhysRevX.7.021026https://doi.org/10.1103/PhysRevX.7.021026https://doi.org/10.1021/acs.nanolett.9b05138https://doi.org/10.1021/acs.nanolett.9b05138https://doi.org/10.1038/s41565-020-0685-6https://doi.org/10.1038/s41565-020-0685-6https://doi.org/10.1038/s41699-023-00420-1https://doi.org/10.1038/s41699-023-00420-1https://doi.org/10.1021/acs.nanolett.4c03464https://doi.org/10.1021/acs.nanolett.4c03464https://doi.org/10.1021/acs.nanolett.8b01859https://doi.org/10.1021/acs.nanolett.8b01859https://doi.org/10.1021/acs.nanolett.8b01303https://doi.org/10.1021/acs.nanolett.8b01303https://doi.org/10.1002/aelm.202200510https://doi.org/10.1002/aelm.202200510https://doi.org/10.1021/acs.nanolett.4c02463https://doi.org/10.1021/acs.nanolett.4c02463https://doi.org/10.1016/j.jcrysgro.2006.12.061https://doi.org/10.1016/j.jcrysgro.2006.12.061https://doi.org/10.1016/j.jcrysgro.2014.06.038https://doi.org/10.1016/j.jcrysgro.2014.06.038https://doi.org/10.35848/1347-4065/ac9dd5https://doi.org/10.35848/1347-4065/ac9dd5https://doi.org/10.1021/cm7028382https://doi.org/10.1021/cm7028382https://doi.org/10.1016/j.jcrysgro.2013.09.030https://doi.org/10.1016/j.jcrysgro.2013.09.030https://doi.org/10.1016/j.jcrysgro.2014.06.006https://doi.org/10.1016/j.jcrysgro.2014.06.006https://doi.org/10.1021/acs.cgd.7b00871https://doi.org/10.1021/acs.cgd.7b00871https://doi.org/10.1021/acs.chemmater.8b02589https://doi.org/10.1021/acs.chemmater.8b02589https://doi.org/10.1021/acs.cgd.9b00712https://doi.org/10.1021/acs.cgd.9b00712https://doi.org/10.1021/acs.chemmater.0c00830https://doi.org/10.1021/acs.chemmater.0c00830https://doi.org/10.1021/acsnano.1c00115https://doi.org/10.1021/acsnano.1c00115https://doi.org/10.1039/D0TC02143Ahttps://doi.org/10.1039/D0TC02143Ahttps://doi.org/10.1039/D1NR02408Fhttps://doi.org/10.1039/D1NR02408Fhttps://doi.org/10.1016/j.jcrysgro.2021.126074https://doi.org/10.1016/j.jcrysgro.2021.126074https://doi.org/10.1021/acsanm.9b02315https://doi.org/10.1021/acsanm.9b02315https://doi.org/10.1088/2053-1583/ac6c31https://doi.org/10.1088/2053-1583/ac6c31https://doi.org/10.1021/acs.nanolett.6b01051https://doi.org/10.1021/acs.nanolett.6b01051https://doi.org/10.1126/science.aau2132https://doi.org/10.1126/science.aau21322D Mater. 12 (2025) 015017 T Ouaj et al[56] Asgari M et al 2022 Terahertz photodetection in scalablesingle-layer-graphene and hexagonal boron nitrideheterostructures Appl. Phys. Lett. 121 031103[57] Calandrini E et al 2023 Near- and far-field observation ofphonon polaritons in wafer-scale multilayer hexagonalboron nitride prepared by chemical vapor deposition Adv.Mater. 35 2302045[58] Tailpied L, Andrieux-Ledier A, Fossard F, Mérot J-S,Decams J-M and Loiseau A 2024 Ni(111) Substrateengineering for the epitaxial chemical vapor depositiongrowth of wrinkle-free multilayer rhombohedral boronnitride films Cryst. Growth Des. 24 7432–44[59] Kobayashi Y and Akasaka T 2008 Hexagonal BN epitaxialgrowth on (0001) sapphire substrate by MOVPE J. Cryst.Growth 310 5044–7[60] Kobayashi Y, Kumakura K, Akasaka T and Makimoto T2012 Layered boron nitride as a release layer for mechanicaltransfer of GaN-based devices Nature 484 223–7[61] Li X, Sundaram S, El Gmili Y, Ayari T, Puybaret R,Patriarche G, Voss P L, Salvestrini J P and Ougazzaden A2016 Large-area two-dimensional layered hexagonal boronnitride grown on sapphire by metalorganic vapor phaseepitaxy Cryst. Growth Des. 16 3409–15[62] Jeong H et al 2019 Wafer-scale and selective-area growth ofhigh-quality hexagonal boron nitride on Ni(111) bymetal-organic chemical vapor deposition Sci. Rep.9 5736[63] Cho Y-J et al 2016 Hexagonal boron nitride tunnel barriersgrown on graphite by high temperature molecular beamepitaxy Sci. Rep. 6 34474[64] Elias C et al 2019 Direct band-gap crossover in epitaxialmonolayer boron nitride Nat. Commun. 10 2639[65] Heilmann M, Prikhodko A S, Hanke M, Sabelfeld A,Borgardt N I and Lopes J M J 2020 Influence of proximityto supporting substrate on van der Waals epitaxy ofatomically thin graphene/hexagonal boron nitrideheterostructures ACS Appl. Mater. Interfaces 12 8897–907[66] Rousseau A et al 2024 Spatially-resolved UV-C emission inepitaxial monolayer boron nitride 2D Mater. 11 025026[67] Caretti I and Jiménez I 2011 Composition and bondingstructure of boron nitride B1−xNx thin films grown byion-beam assisted evaporation Chem. Phys. Lett.511 235–40[68] Jiménez I, Torres R, Caretti I, Gago R and Albella J M 2012A review of monolithic and multilayer coatings within theboron–carbon–nitrogen system by ion-beam-assisteddeposition J. Mater. Res. 27 743–64[69] Torres R, Caretti I, Serin V, Brun N, Radnóczic G andJiménez I 2014 Reversed texture in nanometriccarbon/boron nitride multilayers Carbon 74 374–8[70] Hong S et al 2020 Ultralow-dielectric-constant amorphousboron nitride Nature 582 511–4[71] Sattari-Esfahlan S M et al 2023 Low-temperature directgrowth of amorphous boron nitride films forhigh-performance nanoelectronic device applications ACSAppl. Mater. Interfaces 15 7274–81[72] Martini L et al 2023 Scalable high-mobility Graphene/hBNheterostructures ACS Appl. Mater. Interfaces 15 37794–801[73] Schué L, Berini B, Betz A C, Plaçais B, Ducastelle F, Barjon Jand Loiseau A 2016 Dimensionality effects on theluminescence properties of hBN Nanoscale 8 6986–93[74] Schué L Sponza L, Plaud A, Bensalah H, Watanabe K,Taniguchi T, Ducastelle F, Loiseau A and Barjon J 2019Bright luminescence from indirect and strongly boundexcitons in h-BN Phys. Rev. Lett. 122 067401[75] Roux S et al 2021 Radiative lifetime of free excitons inhexagonal boron nitride Phys. Rev. B 104 L161203[76] Cassabois G, Valvin P and Gil B 2016 Hexagonal boronnitride is an indirect bandgap semiconductor Nat. Photon.10 262–6[77] Rousseau A et al 2021 Monolayer boron nitride:hyperspectral imaging in the deep ultraviolet Nano Lett.21 10133–8[78] Reich S, Ferrari A C, Arenal R, Loiseau A, Bello I andRobertson J 2005 Resonant Raman scattering in cubic andhexagonal boron nitride Phys. Rev. B 71 205201[79] Stenger I, Schué L, Boukhicha M, Berini B Plaçais B,Loiseau A and Barjon J 2017 Low frequency Ramanspectroscopy of few-atomic-layer thick hBN crystals 2DMater. 4 031003[80] Banszerus L, Janssen H, Otto M, Epping A, Taniguchi T,Watanabe K, Beschoten B, Neumaier D and Stampfer C2017 Identifying suitable substrates for high-qualitygraphene-based heterostructures 2D Mater. 4 025030[81] Caretti I and Jiménez I 2012 Influence of carbon contentand nitrogen vacancies on the bonding structure andmechanical performance of graphite-like BCxN thin filmsJ. Appl. Phys. 112 063525[82] Couto N J G, Costanzo D, Engels S, Ki D-K, Watanabe K,Taniguchi T, Stampfer C, Guinea F and Morpurgo A F 2014Random strain fluctuations as dominant disorder sourcefor high-quality on-substrate graphene devices Phys. Rev. X4 041019[83] Lee J E, Ahn G, Shim J, Lee Y S and Ryu S 2012 Opticalseparation of mechanical strain from charge doping ingraphene Nat. Commun. 3 1024[84] Neumann C et al 2015 Raman spectroscopy as probe ofnanometre-scale strain variations in graphene Nat.Commun. 6 8429[85] Vincent T, Panchal V, Booth T, Power S R, Jauho A-P,Antonov V and Kazakova O 2018 Probing the nanoscaleorigin of strain and doping in graphene-hBNheterostructures 2D Mater. 6 015022[86] Solozhenko V L and Turkevich V Z 1997 High pressurephase equilibria in the Li3N-BN system: in situ studiesMater. Lett. 32 179–84[87] Maestre C, Toury B, Steyer P, Garnier V and Journet C 2021Hexagonal boron nitride: a review on selfstanding crystalssynthesis towards 2D nanosheets J. Phys. Mater. 4 044018[88] Sahni K, Ashuri M, Emani S, Kaduk J A, Németh K andShaw L L 2018 On the synthesis of lithium boron nitride(Li3BN2) Ceram. Int. 44 7734–40[89] Schmitt A et al 2023 High-field 1/f noise inhBN-encapsulated graphene transistors Phys. Rev. B107 L161104[90] Verguts K et al 2017 Controlling water intercalation is keyto a direct graphene transfer ACS Appl. Mater. Interfaces9 37484–92[91] Hemmi A et al 2014 High quality single atomic layerdeposition of hexagonal boron nitride on singlecrystalline Rh(111) four-inch wafers Rev. Sci. Instrum.85 035101[92] Hemmi A, Cun H, Brems S, Huyghebaert C and Greber T2021 Wafer-scale, epitaxial growth of single layer hexagonalboron nitride on Pt(111) J. Phys. Mater. 4 044012[93] Cun H et al 2018 Centimeter-sized single-orientationmonolayer hexagonal boron nitride with or withoutnanovoids Nano Lett. 18 1205–12[94] Jiménez I, Jankowski A F, Terminello L J, Sutherland D G J,Carlisle J A, Doll G L, Tong WM, Shuh D K andHimpsel F J 1997 Core-level photoabsorption study ofdefects and metastable bonding configurations in boronnitride Phys. Rev. B 55 12025–37[95] Caretti I and Jiménez I 2011 Point defects in hexagonal BN,BC3 and BCxN compounds studied by x-ray absorptionnear-edge structure J. Appl. Phys. 110 023511[96] Roux S et al 2024 Surface recombination and out-of-planediffusivity of free excitons in hexagonal boron nitride Phys.Rev. B 109 155305[97] Plaud A, Schué L, Watanabe K, Taniguchi T, Fossard FDucastelle F, Loiseau A and Barjon J 2019 Exciton-excitonannihilation in hBN Appl. Phys. Lett. 114 232103[98] Geick R, Perry C H and Rupprecht G 1966 Normal modesin hexagonal boron nitride Phys. Rev. 146 543–7[99] Cuscó R, Edgar J H, Liu S, Li J and Artús L 2020 Isotopicdisorder: the prevailing mechanism in limiting the17https://doi.org/10.1063/5.0097726https://doi.org/10.1063/5.0097726https://doi.org/10.1002/adma.202302045https://doi.org/10.1002/adma.202302045https://doi.org/10.1021/acs.cgd.4c00478https://doi.org/10.1021/acs.cgd.4c00478https://doi.org/10.1016/j.jcrysgro.2008.07.010https://doi.org/10.1016/j.jcrysgro.2008.07.010https://doi.org/10.1038/nature10970https://doi.org/10.1038/nature10970https://doi.org/10.1021/acs.cgd.6b00398https://doi.org/10.1021/acs.cgd.6b00398https://doi.org/10.1038/s41598-019-42236-4https://doi.org/10.1038/s41598-019-42236-4https://doi.org/10.1038/srep34474https://doi.org/10.1038/srep34474https://doi.org/10.1038/s41467-019-10610-5https://doi.org/10.1038/s41467-019-10610-5https://doi.org/10.1021/acsami.9b21490https://doi.org/10.1021/acsami.9b21490https://doi.org/10.1088/2053-1583/ad2f45https://doi.org/10.1088/2053-1583/ad2f45https://doi.org/10.1016/j.cplett.2011.06.001https://doi.org/10.1016/j.cplett.2011.06.001https://doi.org/10.1557/jmr.2011.398https://doi.org/10.1557/jmr.2011.398https://doi.org/10.1016/j.carbon.2014.03.027https://doi.org/10.1016/j.carbon.2014.03.027https://doi.org/10.1038/s41586-020-2375-9https://doi.org/10.1038/s41586-020-2375-9https://doi.org/10.1021/acsami.2c18706https://doi.org/10.1021/acsami.2c18706https://doi.org/10.1021/acsami.3c06120https://doi.org/10.1021/acsami.3c06120https://doi.org/10.1039/C6NR01253Ahttps://doi.org/10.1039/C6NR01253Ahttps://doi.org/10.1103/PhysRevLett.122.067401https://doi.org/10.1103/PhysRevLett.122.067401https://doi.org/10.1103/PhysRevB.104.L161203https://doi.org/10.1103/PhysRevB.104.L161203https://doi.org/10.1038/nphoton.2015.277https://doi.org/10.1038/nphoton.2015.277https://doi.org/10.1021/acs.nanolett.1c02531https://doi.org/10.1021/acs.nanolett.1c02531https://doi.org/10.1103/PhysRevB.71.205201https://doi.org/10.1103/PhysRevB.71.205201https://doi.org/10.1088/2053-1583/aa77d4https://doi.org/10.1088/2053-1583/aa77d4https://doi.org/10.1088/2053-1583/aa5b0fhttps://doi.org/10.1088/2053-1583/aa5b0fhttps://doi.org/10.1063/1.4752757https://doi.org/10.1063/1.4752757https://doi.org/10.1103/PhysRevX.4.041019https://doi.org/10.1103/PhysRevX.4.041019https://doi.org/10.1038/ncomms2022https://doi.org/10.1038/ncomms2022https://doi.org/10.1038/ncomms9429https://doi.org/10.1038/ncomms9429https://doi.org/10.1088/2053-1583/aaf1dchttps://doi.org/10.1088/2053-1583/aaf1dchttps://doi.org/10.1016/S0167-577X(97)00025-6https://doi.org/10.1016/S0167-577X(97)00025-6https://doi.org/10.1088/2515-7639/ac2b87https://doi.org/10.1088/2515-7639/ac2b87https://doi.org/10.1016/j.ceramint.2018.01.200https://doi.org/10.1016/j.ceramint.2018.01.200https://doi.org/10.1103/PhysRevB.107.L161104https://doi.org/10.1103/PhysRevB.107.L161104https://doi.org/10.1021/acsami.7b12573https://doi.org/10.1021/acsami.7b12573https://doi.org/10.1063/1.4866648https://doi.org/10.1063/1.4866648https://doi.org/10.1088/2515-7639/ac0d9ehttps://doi.org/10.1088/2515-7639/ac0d9ehttps://doi.org/10.1021/acs.nanolett.7b04752https://doi.org/10.1021/acs.nanolett.7b04752https://doi.org/10.1103/PhysRevB.55.12025https://doi.org/10.1103/PhysRevB.55.12025https://doi.org/10.1063/1.3602996https://doi.org/10.1063/1.3602996https://doi.org/10.1103/PhysRevB.109.155305https://doi.org/10.1103/PhysRevB.109.155305https://doi.org/10.1063/1.5090218https://doi.org/10.1063/1.5090218https://doi.org/10.1103/PhysRev.146.543https://doi.org/10.1103/PhysRev.146.5432D Mater. 12 (2025) 015017 T Ouaj et alphonon lifetime in hexagonal BN Phys. Rev. Lett.124 167402[100] Gorbachev R V et al 2011 Hunting for monolayer boronnitride: optical and Raman signatures Small 7 465–8[101] Uslu J-L et al 2024 An open-source robust machinelearning platform for real-time detection and classificationof 2D material flakesMach. Learn.: Sci. Technol. 5 015027[102] Nemanich R J, Solin S A and Martin R M 1981 Lightscattering study of boron nitride microcrystals Phys. Rev. B23 6348–56[103] Song L et al 2010 Large scale growth and characterizationof atomic hexagonal boron nitride layers Nano Lett.10 3209–15[104] Shi Y et al 2010 Synthesis of few-layer hexagonal boronnitride thin film by chemical Vapor deposition Nano Lett.10 4134–9[105] Hadid J et al 2022 Molecular beam epitaxial growth ofmultilayer 2D-boron nitride on Ni substrates fromborazine and plasma-activated nitrogen Nanotechnology34 035601[106] Bisswanger T, Winter Z, Schmidt A, Volmer F, Watanabe K,Taniguchi T, Stampfer C and Beschoten B 2022 CVDbilayer graphene spin valves with 26 µm spin diffusionlength at room temperature Nano Lett. 22 4949–55[107] Pálinkás A et al 2022 The composition and structure of theubiquitous hydrocarbon contamination on van der Waalsmaterials Nat. Commun. 13 6770[108] Ferrari A C et al 2006 Raman spectrum of graphene andgraphene layers Phys. Rev. Lett. 97 187401[109] Graf D, Molitor F, Ensslin K, Stampfer C, Jungen A,Hierold C and Wirtz L 2007 Spatially resolved ramanspectroscopy of single- and few-layer graphene Nano Lett.7 238–42[110] Ferrari A C and Basko D M 2013 Raman spectroscopy as aversatile tool for studying the properties of graphene Nat.Nanotechnol. 8 235–46[111] Forster F, Molina-Sanchez A, Engels S, Epping A,Watanabe K, Taniguchi T, Wirtz L and Stampfer C 2013Dielectric screening of the Kohn anomaly of grapheneon hexagonal boron nitride Phys. Rev. B88 085419[112] Sonntag J, Watanabe K, Taniguchi T, Beschoten B andStampfer C 2023 Charge carrier density dependent Ramanspectra of graphene encapsulated in hexagonal boronnitride Phys. Rev. B 107 075420[113] Lin Y-C, Lu C-C, Yeh C-H, Jin C, Suenaga K and Chiu P-W2012 Graphene annealing: how clean can it be? Nano Lett.12 414–9[114] Schwartz J, Chuang H-J, Rosenberger M R, Sivaram S V,McCreary K M, Jonker B T and Centrone A 2019 Chemicalidentification of interlayer contaminants within van derWaals heterostructures ACS Appl. Mater. Interfaces11 25578–85[115] Volmer F, Seidler I, Bisswanger T, Tu J-S, Schreiber L R,Stampfer C and Beschoten B 2021 How to solve problemsin micro- and nanofabrication caused by the emission ofelectrons and charged metal atoms during e-beamevaporation J. Appl. Phys. 54 225304[116] IEC TS 62607-6-6:2021(E) 2021 Nanomanufacturing—Keycontrol characteristics—Part 6-6: “Graphene - Uniformityof strain in graphene analyzed by Raman spectroscopy[117] Ouaj Taoufiq et al 2024 Supporting data for“Benchmarking the integration of hexagonal boron nitridecrystals and thin films into graphene-based van der Waalsheterostructures” Zenodo (https://doi.org/10.5281/zenodo.13684712)18https://doi.org/10.1103/PhysRevLett.124.167402https://doi.org/10.1103/PhysRevLett.124.167402https://doi.org/10.1002/smll.201001628https://doi.org/10.1002/smll.201001628https://doi.org/10.1088/2632-2153/ad2287https://doi.org/10.1088/2632-2153/ad2287https://doi.org/10.1103/PhysRevB.23.6348https://doi.org/10.1103/PhysRevB.23.6348https://doi.org/10.1021/nl1022139https://doi.org/10.1021/nl1022139https://doi.org/10.1021/nl1023707https://doi.org/10.1021/nl1023707https://doi.org/10.1088/1361-6528/ac99e5https://doi.org/10.1088/1361-6528/ac99e5https://doi.org/10.1021/acs.nanolett.2c01119https://doi.org/10.1021/acs.nanolett.2c01119https://doi.org/10.1038/s41467-022-34641-7https://doi.org/10.1038/s41467-022-34641-7https://doi.org/10.1103/PhysRevLett.97.187401https://doi.org/10.1103/PhysRevLett.97.187401https://doi.org/10.1021/nl061702ahttps://doi.org/10.1021/nl061702ahttps://doi.org/10.1038/nnano.2013.46https://doi.org/10.1038/nnano.2013.46https://doi.org/10.1103/PhysRevB.88.085419https://doi.org/10.1103/PhysRevB.88.085419https://doi.org/10.1103/PhysRevB.107.075420https://doi.org/10.1103/PhysRevB.107.075420https://doi.org/10.1021/nl203733rhttps://doi.org/10.1021/nl203733rhttps://doi.org/10.1021/acsami.9b06594https://doi.org/10.1021/acsami.9b06594https://doi.org/10.1088/1361-6463/abe89bhttps://doi.org/10.1088/1361-6463/abe89bhttps://doi.org/10.5281/zenodo.13684712https://doi.org/10.5281/zenodo.13684712 Benchmarking the integration of hexagonal boron nitride crystals and thin films into graphene-based van der Waals heterostructures 1. Introduction 2. Growth and preparation of boron nitride 2.1. Atmospheric pressure and high temperature (APHT) 2.2. Growth in a pressure-controlled furnace (PCF) 2.3. Chemical vapor deposition (CVD) 2.4. Physical vapor deposition (PVD) 3. Cathodoluminesence and Raman spectroscopy on BN crystals and films 3.1. Cathodoluminescence 3.2. Confocal Raman spectroscopy 3.2.1. Experimental setup 3.2.2. hBN crystals 3.2.3. BN films 4. Dry-transfer of graphene encapsulated in BN 5. Raman spectroscopy on BN-graphene heterostructures 5.1. Extraction of strain, strain variations and doping 5.2. Results of spatially-resolved Raman spectroscopy 6. Processing into Hall bar structures 6.1. Influence of processing on the electronic properties of graphene 6.2. Room temperature charge carrier mobilities 7. Discussion References