# Fileset

[s41467-023-37892-0.pdf](https://mdr.nims.go.jp/filesets/7c769919-e45a-40b9-b491-d86ef34b11ab/download)

## Creator

M. Iqbal Bakti Utama, Hongfei Zeng, Tumpa Sadhukhan, Anushka Dasgupta, S. Carin Gavin, Riddhi Ananth, Dmitry Lebedev, Wei Wang, Jia-Shiang Chen, [Kenji Watanabe](https://orcid.org/0000-0003-3701-8119), [Takashi Taniguchi](https://orcid.org/0000-0002-1467-3105), Tobin J. Marks, Xuedan Ma, Emily A. Weiss, George C. Schatz, Nathaniel P. Stern, Mark C. Hersam

## Rights

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

## Other metadata

[Chemomechanical modification of quantum emission in monolayer WSe2](https://mdr.nims.go.jp/datasets/920e389a-8063-4120-a266-6cac900459c0)

## Fulltext

Chemomechanical modification of quantum emission in monolayer WSe2Article https://doi.org/10.1038/s41467-023-37892-0Chemomechanical modification of quantumemission in monolayer WSe2M. Iqbal Bakti Utama 1,10, Hongfei Zeng2,10, Tumpa Sadhukhan 3,9,10,Anushka Dasgupta 1,10, S. Carin Gavin2, Riddhi Ananth3, Dmitry Lebedev1,Wei Wang 4, Jia-Shiang Chen 4,5, Kenji Watanabe 6, Takashi Taniguchi 7,Tobin J. Marks 1,3, Xuedan Ma 4,5, Emily A. Weiss 3, George C. Schatz 3 ,Nathaniel P. Stern 2 & Mark C. Hersam 1,3,8Two-dimensional (2D) materials have attracted attention for quantum infor-mation science due to their ability to host single-photon emitters (SPEs).Although the properties of atomically thin materials are highly sensitive tosurface modification, chemical functionalization remains unexplored in thedesign and control of 2D material SPEs. Here, we report a chemomechanicalapproach to modify SPEs in monolayer WSe2 through the synergistic combi-nation of localized mechanical strain and noncovalent surface functionaliza-tion with aryl diazonium chemistry. Following the deposition of an aryloligomer adlayer, the spectrally complex defect-related emission of strainedmonolayer WSe2 is simplified into spectrally isolated SPEs with high single-photon purity. Density functional theory calculations reveal energetic align-ment between WSe2 defect states and adsorbed aryl oligomer energy levels,thus providing insight into the observed chemomechanically modified quan-tum emission. By revealing conditions under which chemical functionalizationtunes SPEs, this work broadens the parameter space for controlling quantumemission in 2D materials.Two-dimensional (2D) van der Waals materials, such as hexagonalboron nitride (hBN) and transition metal dichalcogenides (TMDs),have been widely explored as hosts for single-photon emitters(SPEs)1–8. In particular, the combination of SPEs and valley pseudospinphysics9 in monolayer (1L) WSe2 makes this 2D material especiallyattractive for the transduction of quantum information from spin-related degrees of freedom into single photons. Consequently, meth-ods for controlling and modifying SPEs in 1L WSe2 are critical toapplications in quantum information science, such as quantumcommunication10. Thus far, the vast majority of TMD SPE research hasexplored only a single mechanism for manipulating SPEs either byusing localized strain for exciton funneling11–13 (such as nanopillar12,13 ornanoindentation14 arrays) or by performing defect engineering forexciton trapping15,16. Evenwhen these twomechanisms have been usedin tandem17, limited tunability has thus far been achieved in theresulting quantum emission characteristics.One common observation in the SPE properties of WSe2 is acomplicated low-temperature spectrumwithmany emission lines1,3,4,11,18Received: 7 January 2023Accepted: 4 April 2023Check for updates1Department of Materials Science and Engineering and the Materials Research Center, Northwestern University, Evanston, IL 60208, USA. 2Department ofPhysics and Astronomy, Northwestern University, Evanston, IL 60208, USA. 3Department of Chemistry and the Materials Research Center, NorthwesternUniversity, Evanston, IL 60208, USA. 4Center for NanoscaleMaterials, ArgonneNational Laboratory, Lemont, IL 60439, USA. 5Northwestern-Argonne Institute ofScience and Engineering, Northwestern University, Evanston, IL 60208, USA. 6ResearchCenter for FunctionalMaterials, National Institute forMaterials Science,1-1 Namiki, Tsukuba 305-0044, Japan. 7International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan. 8Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL 60208, USA. 9Present address: Department ofChemistry, SRM Institute of Science and Technology, Kattankulathur, Tamil Nadu 603203, India. 10These authors contributed equally: M. Iqbal Bakti Utama,Hongfei Zeng, Tumpa Sadhukhan, Anushka Dasgupta. e-mail: g-schatz@northwestern.edu; n-stern@northwestern.edu; m-hersam@northwestern.eduNature Communications |         (2023) 14:2193 11234567890():,;1234567890():,;http://orcid.org/0000-0002-4454-8348http://orcid.org/0000-0002-4454-8348http://orcid.org/0000-0002-4454-8348http://orcid.org/0000-0002-4454-8348http://orcid.org/0000-0002-4454-8348http://orcid.org/0000-0003-1995-7286http://orcid.org/0000-0003-1995-7286http://orcid.org/0000-0003-1995-7286http://orcid.org/0000-0003-1995-7286http://orcid.org/0000-0003-1995-7286http://orcid.org/0000-0003-3544-8334http://orcid.org/0000-0003-3544-8334http://orcid.org/0000-0003-3544-8334http://orcid.org/0000-0003-3544-8334http://orcid.org/0000-0003-3544-8334http://orcid.org/0000-0003-3890-5561http://orcid.org/0000-0003-3890-5561http://orcid.org/0000-0003-3890-5561http://orcid.org/0000-0003-3890-5561http://orcid.org/0000-0003-3890-5561http://orcid.org/0000-0003-1612-3008http://orcid.org/0000-0003-1612-3008http://orcid.org/0000-0003-1612-3008http://orcid.org/0000-0003-1612-3008http://orcid.org/0000-0003-1612-3008http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0003-3701-8119http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0002-1467-3105http://orcid.org/0000-0001-8771-0141http://orcid.org/0000-0001-8771-0141http://orcid.org/0000-0001-8771-0141http://orcid.org/0000-0001-8771-0141http://orcid.org/0000-0001-8771-0141http://orcid.org/0000-0002-3163-1249http://orcid.org/0000-0002-3163-1249http://orcid.org/0000-0002-3163-1249http://orcid.org/0000-0002-3163-1249http://orcid.org/0000-0002-3163-1249http://orcid.org/0000-0001-5834-463Xhttp://orcid.org/0000-0001-5834-463Xhttp://orcid.org/0000-0001-5834-463Xhttp://orcid.org/0000-0001-5834-463Xhttp://orcid.org/0000-0001-5834-463Xhttp://orcid.org/0000-0001-5837-4740http://orcid.org/0000-0001-5837-4740http://orcid.org/0000-0001-5837-4740http://orcid.org/0000-0001-5837-4740http://orcid.org/0000-0001-5837-4740http://orcid.org/0000-0002-8903-3516http://orcid.org/0000-0002-8903-3516http://orcid.org/0000-0002-8903-3516http://orcid.org/0000-0002-8903-3516http://orcid.org/0000-0002-8903-3516http://orcid.org/0000-0003-4120-1426http://orcid.org/0000-0003-4120-1426http://orcid.org/0000-0003-4120-1426http://orcid.org/0000-0003-4120-1426http://orcid.org/0000-0003-4120-1426http://crossmark.crossref.org/dialog/?doi=10.1038/s41467-023-37892-0&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-023-37892-0&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-023-37892-0&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s41467-023-37892-0&domain=pdfmailto:g-schatz@northwestern.edumailto:n-stern@northwestern.edumailto:m-hersam@northwestern.edudue to the complex defect landscape within samples that trapexcitons18,19. Although spectrally- and spatially isolated emitters inWSe2exhibiting non-classical photonbehavior have been reported even fromcrowded emission spectra, a high density of emission lines around anemitter of interest can create a challenge to completely filter outsignals from neighboring emitters and the broad defect background,impacting the purity of single photons extracted from such a spectrum.A high density of these emission lines is hence undesirable forquantum transduction experiments. This issue provides the impetus forinvestigating alternative strategies for controlling SPE properties inWSe2. Since chemical functionalization has been shown to be aneffective strategy for tuning the electronic and optical properties of 2Dsemiconductors20, this approach is also likely to be useful for tuningquantum emission, especially because interfacial modulation is knownto strongly influence excitonic properties21. Despite this promise, sur-face and interface engineering via chemical functionalization has notyet been successfully employed to tune SPEs in TMDs.Here, we report a chemomechanical modification approach for1L WSe2 that produces spectrally isolated SPEs via a synergisticcombination of localized mechanical strain and chemical functiona-lization using aryl diazonium chemistry. In particular, surface mod-ification of strained 1L WSe2 with 4-nitrobenzenediazonium (4-NBD)tetrafluoroborate quenches most strain-induced defect emission,resulting in sharp SPEs with high single-photon purity. Rather thancovalently reactingwithWSe2, the 4-NBD treatment conditions resultin a physisorbed nitrophenyl oligomer layer on the WSe2 surface asconfirmed by X-ray photoelectron spectroscopy, atomic forcemicroscopy, and photoluminescence imaging. First-principles cal-culations show that shallow mid-gap states from the physisorbednitrophenyl oligomer layer are energetically resonant with WSe2defect levels, thereby suppressing most emission pathways andsimplifying the final SPE spectrum. Overall, these results establishchemical functionalization as an effective pathway formodifying SPEin strained 1L WSe2.ResultsDiazonium functionalization and photoluminescencequenchingFigure 1a schematically depicts the spontaneous chemical functiona-lization that occurs upon immersion of 1L WSe2 into an aqueoussolution of 4-NBD tetrafluoroborate. The electrophilic nature of 4-NBDcations is believed to withdraw electrons fromWSe2, releasing N2 andgenerating diazonium radicals in addition to causing hole doping inWSe222. Although aryl diazonium radicals are often assumed to sub-sequently form covalent bonds with the 2D material surface22–29, thehighly reactive diazonium radicals can also react with one another,forming nitrophenyl (NPh) oligomer chains of varying lengths30,including the 2-ring and 3-ring structures illustrated in Fig. 1a (alsosee Supplementary Fig. 1b–d for a general functionalization schemeusing 4-NBD). As will be discussed in more detail below, our reactionconditions favor oligomerization, resulting in a physisorbed NPhadlayer on the 1LWSe2 surface. In particular, following immersion for1.5 h in a 5mM aqueous solution of 4-NBD (Supplementary Fig. 2), 1LWSe2 is fully coated with a ~4–5 nm thick NPh oligomer adlayer thatquenches its room-temperature photoluminescence (PL). To illus-trate this point, a 1L WSe2 sample was prepared that is partiallycovered with an hBN flake (Fig. 1b). Figure 1c shows the PL spectrumcollected from the 1L WSe2 region without the hBN cover before andafter the 4-NBD treatment, where the integrated PL drops to ~15% ofits original level following chemical modification. Meanwhile, PL ofthe hBN-covered WSe2 region is not quenched (SupplementaryFig. 3c) since the hBN cover prevents direct contact between the NPholigomers and the WSe2 surface. This PL quenching effect isalso evident by comparing the spatial map of the peak PL intensityin Fig. 1d, g. In addition, although the PL spectrum of the hBN-covered region shows no significant changes (SupplementaryFig. 3c), the PL of the uncovered WSe2 following 4-NBD treatmentshows a PL peak redshift by ~11meV (Fig. 1e, h) and linewidthbroadening (Fig. 1f, i).Fig. 1 | Room-temperature photoluminescence (PL) quenching following4-nitrobenzenediazonium (4-NBD) treatment. a Illustration of the chemicalfunctionalization scheme.bOptical image of a 1LWSe2 that is partially coveredwitha thin hBN flake. c Room-temperature (T = 296 K) PL spectra of 1L WSe2 before andafter 4-NBD treatment from a location without the hBN cover, showing PL intensityquenching, redshifting, and broadening. The PL spectrum from hBN-coveredWSe2does not show significant changes before and after 4-NBD treatment(Supplementary Fig. 3) and has been used to normalize the PL intensity.d–i PLmapof the sample before (d–f) and after the 4-NBD treatment (g–i). The color in eachmap represents:d, g themax PL count, e,h peakposition, f, i andpeakwidth.Whilethe hBN-covered region shows negligible change with 4-NBD treatment, theuncovered region exhibits PL quenching, redshifting, and broadening. Scale bars in(d–i) correspond to 10 µm.Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 2Low-temperature optical spectroscopyFigure 2 explores the low-temperature optical properties of a 1L WSe2flake on a flat Si/SiO2 substrate. Atomic forcemicroscopy (AFM) showsthat the sample is generally flat, although the flake has wrinkles nearthe edges as indicated by the green circle in Fig. 2a. Before 4-NBDfunctionalization, the PL spectrumcollected from the flat area of the 1LWSe2 sample possesses rich excitonic features (Fig. 2b, blue curve),which are assigned following previous literature31,32 as the neutral Aexciton (X0), biexciton (XX), negatively-charged trion (X−), and darkexciton (XD). Subsequent 4-NBD treatment quenches most of thesespectral features, leaving X0 as the remaining dominant emission fea-ture (Fig. 2b, red curve). Unlike diazonium functionalization of carbonnanotubes33–36, no SPEs or new spectral features are observed on theflat region of the chemically modified 1L WSe2. The absence of lower-energy positively-charged trion (X+) emission features after chemicalmodification, as has beenobserved for 1LWSe2 thatwas hole-dopedbyelectrostatic gating32,37, suggests that the quenching effect of the4-NBD treatment cannot be solely attributed to hole doping. Likewise,in the alternative scenario that 4-NBD treatment changes the dopinglevel only close to charge neutrality, the PL quenching also cannot beattributed to functionalization-induced doping compensation becausethe neutral exciton usually appears with even higher PL intensity(insteadof being quenched)when the doping level is brought closer tocharge neutrality37.Unlike the flat regions, the wrinkled location on the 1L WSe2 flakepreceding chemical modification shows spectrally dense emissionfeatures between 720nm and 780nm that are brighter than the ori-ginal neutral exciton (Fig. 2c, blue). This low-temperature PLenhancement at wrinkles is commonly observed and has been attrib-uted to a strain-assisted hybridization of dark exciton anddefect statesthat increases the radiative recombination yield38. The 4-NBD treat-ment effectively quenches these defect-related emission features(Fig. 2c, red), resulting in significantly fewer emission features that arespectrally isolated (e.g., the sharp peak near 788 nm). As we shall dis-cuss later, it is likely that the emission features remaining after 4-NBDtreatment also originate from pre-existing defects within the WSe2monolayer. A high-resolution spectrum of this sharp feature (Fig. 2d)reveals a zero-phonon linewidth of ~0.5meV. Observing the spectrumover time shows that this emitter is relatively stablewithout significantspectral diffusion (Fig. 2d, inset). The time series of the peak positioncan be constructed into a histogram of the spectral jitter and fittedwith a Gaussian distribution, resulting in jitter FWHM of 200-400 µeV(Supplementary Fig. 4).The second-order correlation function, g 2ð Þ τð Þ, of the emitter stateshows clear antibunching behavior (Fig. 2e). Fitting the measurementwith a two-level model using the equation g 2ð Þ τð Þ= 1� ae�∣τ∣=τ1 , where ais a constant, yields a raw g 2ð Þraw τð Þ= 0.13 ±0.04. Background correctionresults in g 2ð Þ 0ð Þ= 0:01+0:04�0:01 , which is significantly lower than thewidelyaccepted 0.5 threshold for SPE, thus indicating high purity of the pro-duced single photons (the background-corrected g 2ð Þ τð Þ is shown inthe right y-axis). The fitting also allows extraction of the emissionlifetime, τ1 = (2.34 ±0.13) ns, which is comparable to the typicallyFig. 2 | Partial quenching of defect emitters in locally strained 1LWSe2. aHeightimage from atomic force microscopy (AFM) of a 1L WSe2 flake. The region markedwith the orange square is relatively flat, whereas the region inside the green circlecontains wrinkles with localized strain. b, c Low-temperature (T = 1.7 K) photo-luminescence (PL) spectra of 1L WSe2 at (b) the flat region and (c) the wrinkle,before (blue) and after (red) 4-nitrobenzenediazonium (4-NBD) treatment. Thespectra were measured with ~60 µW excitation power. d High-resolution spectrumof the single-photon emitter (SPE) from (c). The thick red trace is the time averageof the traces shown in pink. Inset: SPE spectral diffusion plot. e Second-ordercorrelation function of the SPE. The left vertical axis denotes the as-measuredgraw2ð Þ τð Þ values, whereas the right vertical axis denotes the background-correctedg 2ð Þ τð Þ values. Fitting of the data (red curve) reveals a graw2ð Þ 0ð Þ of 0.13 ± 0.04, whilebackground correction results in a g 2ð Þ 0ð Þ of 0:01+0:04�0:01 (the details of the back-ground correction are available in “Methods”). The black dashed line marks whereg 2ð Þ τð Þ=0 after background correction. f SPE intensity as a function of excitationpower. The error bars represent the standard deviation from the time averagingand the red solid line is a fit to the data.Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 3observed value for 1L WSe2 SPE in the literature4,5 and the lifetimemeasured using time-resolved fluorescence (Supplementary Fig. 5 andSupplementary Table 1). This emitter also shows a saturating powerdependence that is typical for 1 L WSe2 SPE (Fig. 2f). Fitting the excita-tion power dependence of the PL intensity with the equationI = IsatP= P + Psat� �yields a saturation power (Psat) of (6.3 ±0.7) µW.Because the emitter only appears on the wrinkled regions of the che-mically modified 1L WSe2 flake and has properties consistent withstrained 1L WSe2 SPE, this SPE likely originates from the WSe2itself rather than from the NPh film alone or from interlayer excitonicspecies between themolecular film and theWSe2flake. It should also benoted that resonanceenhancement is observed as the excitation energypasses through the WSe2 A exciton resonance (Supplementary Fig. 6),further confirming that this SPE originates from the 1L WSe2 flake.Spatially deterministic chemomechanically modified SPEIn an effort to translate the chemomechanically modified SPE fromrandom wrinkled regions to spatially deterministic locations, 1L WSe2was transferredonto an array of prefabricated SiOx nanopillars (Fig. 3a,also see Supplementary Figs. 7–8), resulting in mixed-dimensionalheterostructures39 consisting of NPh oligomers (0D), nanopillars(quasi-1D), and 1LWSe2 (2D) that also host the quantum emitters (0D).By comparing the AFM image (Fig. 3b) and low-temperature PLmap ofthe sample (Fig. 3c–e), the location of each nanopillar is determined,thereby enabling comparison of the emission before and after 4-NBDtreatment. Similar to 1L WSe2 on a flat substrate, chemical functiona-lization quenches the majority of dense defect-related emission linesbetween 730nm and 760nm. The 4-NBD treatment also quenches theseries of sharp defect-related emission features that are initiallybrightened by the strain induced by the nanopillars.Comparisonof the PL spectra at the samenanopillar location (e.g.,P1-P4 in Fig. 3f) reveals the simplification of the emission spectra fol-lowing the 4-NBD treatment, where significantly fewer, energeticallyisolated emitter states remain (also see Supplementary Figs. 9–10 forspectra from other locations on the sample). One example is thesharp emission peak at ~759 nm for nanopillar position P1 (Fig. 3g).From g 2ð Þ τð Þ measurements on this emission feature, clear antibunch-ing is observed with a raw g 2ð Þraw 0ð Þ value of 0.178 ±0.015, background-corrected g 2ð Þ 0ð Þ value of 0.132 ± 0.016, and τ1 = (11.5 ± 0.5) ns,thereby confirming high-purity SPE following 4-NBD functionalization.Supplementary Figs. 11 and 12 provide the statistics for the emitterproperties observed with our chemomechanical modificationscheme, including the yield, peak wavelength distribution, peak PLintensity, and narrowest linewidth. Meanwhile, Supplementary Fig. 13provides a measurement of a monolayer WSe2 flake before andafter 4-NBD treatment at low temperature with identical excitationconditions.Noncovalent functionalization following 4-NBD treatmentTo investigate the nature of the bonding between the molecularadlayer and WSe2 following the 4-NBD treatment, surface-sensitivecharacterization was performed using X-ray photoelectron spectro-scopy (XPS). Figure 4a, b shows XPS spectra on WSe2 before and afterthe 4-NBD treatment. For both the W and Se core levels, a chemicalshift of ~0.5 eV to a higher binding energy is observed after chemicalfunctionalization. A similar shift following 4-NBD treatment on WSe2has previously been attributed to hole doping, which causes the Fermilevel to be displaced closer to the valence bandmaximumand the corelevels22. However, the XPS spectra do not show evidence of new che-mical bond formation since no new spectral features nor lineshapechanges are observed (Supplementary Figs. 14 and 15) that can beattributed to the formation of Se–C or W-C bonds. Likewise, no line-shape changes are apparent even when the sample is tilted by asmuchas 40° to improve the surface sensitivity of the XPS measurement(Supplementary Fig. 16). These XPS results suggest that the 4-NBDtreatment results in a physisorbed NPh oligomer adlayer withoutchemical bond formation to the WSe2 surface.Corroborating the physisorbed nature of the molecularadlayer, the NPh oligomer film is easily removed without damagingthe underlying WSe2 using contact-mode AFM. In particular, AFMFig. 3 | Chemomechanically modified emitters on a nanopillar array.a Schematic of the mixed-dimensional heterostructures consisting of consisting ofnitrophenyl (NPh) oligomers (0D), nanopillars (quasi-1D), and 1L WSe2 (2D) thatalso host the quantumemitter (0D).bAtomic forcemicroscopy (AFM) imageof a 1LWSe2 flake on a nanopillar array after 4-nitrobenzenediazonium (4-NBD) treatment.c Low-temperature (T = 1.7 K) photoluminescence (PL) map of integrated intensityof the as-transferredflakebefore the 4-NBD treatment.d, ePLmaps after the 4-NBDtreatment, showing (d) integrated intensity over the measured spectral range and(e)maximumcount forwavelengths between720 nmand800nmonly. f PL spectraat selected nanopillar sites marked in (e) before (blue) and after (red) 4-NBDtreatment. The4-NBD treatment simplifies the emission spectraby quenchingmostof the dense defect-related emission. g Single-photon emitter (SPE) after the 4-NBDtreatment at location P1. h Second-order correlation functionmeasurement on theSPE in (g). The g 2ð Þraw τð Þ values are on the left vertical axis, whereas the background-corrected g 2ð Þ τð Þ values are on the right vertical axis.Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 4scratching with a normal force setpoint between 50 nN and 200 nNresults in nearly complete removal of the molecular adlayer (Fig. 4c),allowing the NPh oligomer film thickness of ~4.4 nm to be directlymeasured (Supplementary Fig. 17). AFM scratching also results inreversal of the effects of the 4-NBD treatment on the room-temperature1LWSe2 PL spectrum as the PL intensity in the scratched region returnsto its higher value relative to the surrounding area that remains coatedwithNPh (Fig. 4d). Furthermore, the PL peak is blue-shifted (Fig. 4e) andthe peak width is decreased (Fig. 4f) following AFM scratching, asexpected for pristine 1LWSe2. The change of the PL lineshape following4-NBD treatment is also reversed upon removal of the NPh film(see Supplementary Figs. 17e, f and Fig. 18c for the PL spectra).To further demonstrate that the PL spectral changes following the4-NBD treatment are not resulting from covalent surfacemodification,a NPh filmwas transferred onto the surfaceofWSe2 using a carrier hBNflake (Fig. 4g, h; also see Supplementary Figs. 19–20). ThisNPh transfermethod circumvents the need for WSe2 immersion into the 4-NBDtetrafluoroborate solution, thereby avoiding WSe2 exposure to dia-zonium radicals and ensuring a noncovalent interaction between theNPh film and the WSe2 surface (Supplementary Fig. 1f, g). The PL mapobtained after NPh film transfer shows quenching of the neutral exci-ton emission on the upper half of the 1L WSe2 that is in direct contactwith the NPh film (Fig. 4i). However, unlike direct exposure of WSe2 tothe 4-NBD solution, theWSe2 PL spectrum for the transferred NPh filmdoes not show an appreciable redshift nor broadening of the linewidth(Fig. 4k; also see Supplementary Fig. 21). The observed PL quenchingcannot be attributed to interference effects or local field changes dueto the variation in hBN thickness, as shown in our calculation of thelight emission outcoupling from the sample (Supplementary Fig. 22).Instead, the weaker quenching in the sample with transferred NPh canbe explained by the interface of WSe2 and transferred NPh film that isless conformal than the typical interface of NPh/WSe2 from direct30203020Si/SiO21L WSe2Si/SiO2hBNNPh-coated hBN10 Max PL count (arb. units)2 μmPeak position (nm)Peak width (nm)765745200Height (nm)10 μm10 Max PL count (arb.units)Peak position (nm)765745Peak width (nm)2515kjihghBN-assisted NPh film transferNPh-coated hBNhBNdcfe42 40 38 36 34 32abW AFM-scratchedNormalized count64 62 60 58 56 54 52Normalized countBinding energy (eV)Binding energy (eV)SeBareAfter 4-NBDBareAfter 4-NBD5p3/24f5/2 4f7/23d3/23d5/2After NPhtransferBeforeFig. 4 | Noncovalent NPh functionalization of WSe2. a, b X-ray photoelectronspectroscopy (XPS) spectra of WSe2 after 4-nitrobenzenediazonium (4-NBD)treatment for (a) W and b Se core levels. While the chemical shift to higher bindingenergies can be explained by hole doping, no lineshape changes are observed thatwould be indicative of bond formation between WSe2 and nitrophenyl (NPh). Red:experimental data points, orange: fitted peaks, blue: sum of fitted peaks, green:baseline. c–f Reversible quenching of the WSe2 PL with AFM scratching. c Atomicforcemicroscopy (AFM) image of 1LWSe2 after removal of a 2 × 2 µm2 region of theNPh film using contact-mode AFM. The PL map of the sample is shown for d themaximumcount, e the peak position, and f the peakwidth. g–k PL quenching on 1LWSe2 functionalized with a NPh film via hBN-assisted transfer. g Schematic of theNPh film transfer process. h Optical micrograph of as-exfoliated WSe2 before andafter NPh film transfer. The photoluminescence (PL) map before and after thetransfer is shown for i peak intensity, j peak position, and k peak width.Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 5solution phase treatment with 4-NBD (Supplementary Fig. 23). Overall,these results suggest that the WSe2 PL quenching following the 4-NBDtreatment cannot be primarily attributed to hole doping or covalentbonding, necessitating the identification of an alternative mechanismusing first-principles calculations.First-principles calculationsDensity functional theory calculations were performed under theassumptions that the NPh film is physisorbed on theWSe2 surface andthat the NPh oligomers are randomly distributed in terms of ringconfiguration and chain length. In addition, the 1L WSe2 substrate wassubjected to a biaxial strain of 1% to simulate the experimental con-ditions when SPE was observed. Figure 5a shows the calculated bandstructure of pristine, defect-free 1L WSe2 in addition to the chargedensity distributions in real-space for the valence band maximum(VBM) and the conduction band minimum (CBM). Multiple NPh con-figurations from 1 to 3 rings and possible open-shell and closed-shellsystems were considered (Supplementary Fig. 24). These results showthat the most energetically stable configuration is with the oligomerrings positioned flat to the 1L WSe2 surface at an equilibrium distanceof ~3.5 Å between the NPh oligomers and the WSe2 surface (Supple-mentary Fig. 25).With the 3-ring NPh oligomer as an example (Fig. 5b), DFT calcu-lations indicate that physisorbed NPh oligomers on WSe2 result innearly flat mid-gap states near the CBM of 1L WSe2. Real-space chargedensity distributions show that these shallowmid-gap states primarilyreside within the NPh oligomers. Comparison to the energy levels ofisolated NPh (Supplementary Fig. 26) suggests that these mid-gapstates originate from the lowest unoccupiedmolecular orbital (LUMO)and higher unoccupied orbitals of NPh. Meanwhile, the valence bandmaximum still resembles that of pristine 1L WSe2, which indicates thata type-II heterojunction is formed between NPh and 1L WSe2. DFTcalculations for oligomers with different ring numbers and config-urations show a diversity in the energies of the mid-gap states, whichgenerally also form type-II heterojunctions withWSe2 (SupplementaryFigs. 26 and 27 and Supplementary Table 2). The formation of type-IIheterojunctions explains the general PL quenching for 1L WSe2 after4-NBD treatment, including in the case without applied strain.Although the physisorbedNPh oligomers form amultilayered film, theNPh oligomers that are in direct contact with the WSe2 surface likelyplay a dominant role in determining the heterojunction behavior. Thelinewidth broadening of the neutral exciton emission after functiona-lization with NPh is also consistent with the formation of a type-IIheterojunction, where the lifetime of excited states is shortened byadditional decay channels that lead to exciton dissociation40.DFT calculations were also performed for 1L WSe2 point defects,with the chalcogen vacancy chosen as a prototypical example since it isbelieved to be one of the defects responsible for SPE17,38,41. For 1L WSe2Fig. 5 | Density functional theory (DFT) calculations for 1L WSe2 and nitro-phenyl (NPh) oligomers.The band structure calculations presentedhere assume a1% biaxial tensile strain for the 1LWSe2 substrate. The Fermi energy (EF) is set to theenergy of valencebandmaximum (VBM).aPristine, defect-free 1LWSe2. Here, blue:electronic bands fromWSe2, purple: spin–orbit split bands ofWSe2 that are closestto the VBM and conduction band maximum (CBM) band edges. b 1L WSe2 with aphysisorbed NPh 3-ring oligomer. The real-space distribution of the density ofstates shows that the valence band maximum retains the WSe2 character while thelowest energy of the weakly dispersive mid-gap states near the conduction bandminimum (CBM) are localized at the NPh oligomer, suggesting the formation of atype-II heterojunction. Although the change in the WSe2 conduction bands isminimal, there is a finite coupling of the bands to the higher lying NPh orbitals, andthus the conduction bands near CBMare also colored in red. c 1LWse2 with a singlemono-vacancy of Se, as a prototypical example of defects that can emit light withlocalized strain. The new mid-gap states near the CBM are localized at the defectsite and are colored in green. d, Calculation for the combination of a Se mono-vacancy and NPh 3-ring oligomer. e Illustration of NPh oligomers physisorbed onthe surface of 1L WSe2. In a typical sample, the 4-NBD treatment results in a poly-disperse mixture of NPh oligomer species. f Illustration of the quenchingmechanism that results in a simplified SPE spectrum following 4-NBD treatment.The represented colors of the band in this illustration follow the color coding usedin the calculation in (a–d). Black dashed arrow indicates NPh orbitals and WSe2defect states that are in resonance and can effectively quench the emission fromthese defects. The strain- and defect-trapped exciton thus only recombine radia-tively from the available defect states with lower energy that are not quenched bythe NPh states (red arrow).Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 6with a Se mono-vacancy (Fig. 5c), new mid-gap states emerge near theCBM. The charge distribution of this state is predominantly localizedon the threeW atoms surrounding the Se vacancy. The combination ofa Se mono-vacancy and physisorbed NPh (Fig. 5d) shows character-istics that combine Fig. 5b and c in the CBM region, with a 0.02 eV shiftin band gap compared to Fig. 5c. It should also be noted that 1% biaxialstrain does not change the band structure significantly compared tothe unstrained case (calculations without strain are also available inSupplementary Figs. 28–30), other than small changes in the band gapvalue (increasing it by ~0.08 eV from the unstrained value) and slightlymore coupling between localized and delocalized states (also seeSupplementary Fig. 31). Overall, the polydispersity of the NPh oligomerstructure in the experimentalmolecular adlayer (Fig. 5e) translates intoa diversity of mid-gap states that are in resonance with most of theWSe2 defect states, resulting in quenching of the vastmajority of strain-activated SPE (Fig. 5f) such that any remaining SPEswill be energeticallyisolated with high photon purity, as is observed experimentally.To elaborate further, new emitters that emerge after the 4-NBDtreatment appear to show properties similar to typical WSe2 emittersand thus likely originate frompre-existing defect states within theWSe2monolayer itself. However, preceding the 4-NBD treatment, many ofthese defectsmay have been outcompeted by other defects in the samevicinity that dominate the emission signal. The insight from the first-principles calculations suggests amechanism forquenchingof emissiondue to charge or energy transfer to the molecular orbitals of nitro-phenyl oligomers, particularly when the orbital energies are resonantwith or energetically positioned lower than the defect states. In addi-tion, at some sites, additional defect states can exist whose energies arelower than themolecular orbitals such that quenching of emission fromthese defect states is not effective. Prior to functionalization, theselower-energy defect states may not produce significant photo-luminescencedue to thepresenceof higher energydefects that providedominant emission pathways, but the lower-energy defect states thenbecome the preferred states for emissive recombination after 4-NBDfunctionalization quenches emission from the higher energy defects.DiscussionThis study has demonstrated that chemomechanical modification sig-nificantly simplifies the low-temperature PL spectrum of strained 1LWSe2. Detailed surface characterization shows that 4-NBD solutionprocessing results in a physisorbed NPh oligomer adlayer, which gen-erates mid-gap states near the CBM of 1L WSe2 as determined by DFT.Since these NPh mid-gap states are in resonance with WSe2 defectstates, the remaining SPEs are energetically isolated with high photonpurity. It is likely that the chemomechanical strategies introduced herecan be applied to other low-dimensional semiconductors, thus allowingthe preparation of high-purity SPEs in other spectral ranges. While the4-NBD treatment shown here is effective for different strain environ-ments ranging from random wrinkles to spatially deterministic nano-pillar arrays, the exact wavelength of the resulting energetically isolatedSPEs varies from sample to sample and even from nanopillar to nano-pillar in a given sample, which suggests that further improvements canbe gained from alternative functionalization chemistries42 and/or bettercontrol over the defects and strain in 1L WSe2. Regarding theformer approach, one key future direction is to achieve control over theenergetic alignment of the functional group orbitals and the WSe2defect levels. For functionalization with 4-NBD, this goal may beachieved by controlling the oligomer formation and configuration. Forexample, 4-NBD functionalization via electrochemical reduction hasproduced controlled nitrophenyl monolayers without spontaneousoligomerization43. The choice of the diazonium aryl group is anotherparameter that can control the energetic alignment of the molecularorbitals44, while also presentingopportunities to exploit unique featuressuch as steric hindrance (e.g., 3,5-bis-tert-butyl benzenediazonium45)that can dictate the molecular configuration.The chemical functionalization approach is compatible with otherschemes and strategies that have been previously described in theliterature to achieve state-of-the art emitters in termsof brightness andlinewidth. This compatibility is facilitated by the fact that the func-tionalization is performed on only one side of the WSe2 monolayersurface. Specific to the 4-NBD treatment, since the linewidth andbrightness of emitters before and after functionalization are compar-able, it is likely that improvements to the emitter properties on thesample before functionalization would be maintained after the 4-NBDtreatment. For example, reduction in the emitter linewidth can beachieved by using alternative substrates and straining structure withlower defect and interface states than that of SiO2 used here. Previouswork that has demonstrated narrow linewidths reaching <100 µeV inWSe2 includes the use of Al2O3 (which can then be proximitized withmetallic structures for plasmonic coupling)46,47, InGaP48 or hBN17,49substrates or encapsulation. Meanwhile, enhancement of the emitterbrightness can be achieved by integrating the chemomechanicallymodified emitters with plasmonic structures46,50 or by relying onenhancement effects with photonic structures51–53. Alternatively, fur-ther permutations of the chemomechanical strategy may also achieveimprovements in the emitter properties by exploring other combina-tions of molecules and 2Dmaterials. For example, selecting moleculeswith precisely aligned energy levels could improve the selectivity ofthe emitter wavelength toward a longer wavelength range or to a verynarrow spectral range, which could be valuable in reducing the inho-mogeneous broadening of these emitters. Likewise, improvement inemitter brightness may be achieved by selecting molecules that formtype-I heterojunctions with 2D materials54.MethodsSample preparationThe WSe2 flakes were micromechanically exfoliated from bulk singlecrystals using the standard scotch tape method. After identifyingmonolayer flakes on PDMS stamps, 1L WSe2 was transferred onto Siwith 285-nm-thick SiO2 (Fig. 1 and Supplementary Fig. 3), PMMA-coated Si/SiO2 (Supplementary Fig. 18), or prefabricated nanopillararrays (Fig. 3) using a transfer stage inside a N2 glove box (<0.1 ppmH2OandO2) following the conventional viscoelastic transfermethodatroom temperature12,13,55.Chemical functionalizationThe 4-nitrobenzenediazonium (NBD) tetrafluoroborate powder (97%,Sigma-Aldrich) was purified using recrystallization to remove impu-rities. The WSe2 samples were then immersed into a 5mM 4-NBD tet-rafluoroborate aqueous solution for 30–120min (with the typicalimmersion time being 90min) at standard temperature and pressureinside a glass scintillation vial that was shielded from light with alu-minum foil (see Supplementary Fig. 2). After immersion, the sampleswere rinsed with deionized water and dried with nitrogen flow.CharacterizationConfocal PL spectroscopy, Raman spectroscopy, and PL mapping inambient conditions were performed with a Horiba XploRA Plusinstrument using a 532 nm laser focused with an objective (×100, NA0.9). AFM imagingwas performedwith anAsylumCypher S instrumentin tapping mode using NCHR-W Pointprobe tips. The spring constantof the cantilever was estimated by relating the nominal value of thespring constant and resonance frequency from the manufacturer andthe measured resonance frequency using the following relation14:kestimated = knominal fmeasured=f nominal� �3 ð1ÞThe cantilever was calibrated using thermal tuning to determinethe inverse optical lever sensitivity factor (InvOLS) to allow conversionof the deflection voltage into cantilever deflection distance. With aArticle https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 7contact-mode setpoint of 0.05–0.2 V, AFM scratching was performedwith an estimated normal force setpoint of 50–200nN. XPS measure-ments were performed in high vacuum (~1 × 10−8 mbar base pressure)using a Thermo Scientific ESCALAB 250 Xi instrument in charge com-pensation mode with a nominal spot size diameter of 400 µm. Thebinding energy was calibrated to the adventitious C 1 s level at 248.8 eV.Low-temperature optical spectroscopyLow-temperature optical spectroscopy measurements (Figs. 2 and 3and Supplementary Figs. 6, 9–10, 12) were conducted in a closed-cyclecryostat (Attocube, AttoDRY2100) at a temperature of 1.7 K with asuperconductingmagnet (see Supplementary Fig. 32). For confocal PLmeasurements, a tunable CW laser (M Squared, SolsTis EMM) atλ = 635 nm was used, while a broadband white light source (Thorlabs,SLS201) was used for reflectivity measurements. A 100× magnificationobjective with a 0.82 NA (Attocube, LT-APO/VIS/0.82) was used tofocus the laser beam or white light and collect the PL or reflectionsignal. We estimate the diffraction-limited beam diameter of the PLmeasurement to be D= 1:22λ=N:A:≈ 1 µm. The excitation powers usedwere 60 µW for PL spectroscopy and gð2Þ τð Þ measurements in Fig. 2,10 µW for the PL map and spectra in Fig. 3 before functionalization,45 µWfor PLmap and spectra in Fig. 3 after functionalization, 10 µWforPL spectrum in Fig. 3g, and 10 µW for the gð2Þ τð Þ measurement inFig. 3h. Typically, the emitters both before and after functionalizationachieve intensity saturation for excitation powers below 10 µW.The collected signal was sent to a spectrometer of 750mmfocal length (Andor Shamrock, SR-750-D1-R-SIL) equipped with athermoelectrically-cooled CCD camera (Oxford Instruments,DU420A-BEX2-DD). The PL maps were obtained by scanning thecryogenic non-magnetic linear nanopositioners (Attocube, ANPx101/RES/LT) in the x and y directions. PLEmeasurements were performedusing the TeraScan function of an ultranarrow linewidth CW Ti:Sap-phire laser (M Squared, SolsTis). A Hanbury Brown–Twiss setup wasused tomeasure the second-order correlation function. The emissionsignal was filtered by a band-pass filter (10 nm FWHM bandwidth) toblock the emission other than the desired SPE, where an adjustablewavelength range is achieved by mounting the filter at a tilt angle.The filtered signal was then coupled to a fiber connected with a 1 × 2fiber splitter to split the emission and direct the signal to two ava-lanche photodiodes (APD; PicoQuant, τ-SPAD-100).The raw coincidence data (gð2Þraw τð Þ) were corrected for the back-ground to obtain the correlation function according to the relation56gð2Þ τð Þ= gð2Þraw τð Þ � ð1� ρ2Þρ2ð2Þwhere ρ = S=ðS+BÞ is related to the signal-to-background ratio. Back-ground correction is a commonly performed procedure in the analysisof the second-order autocorrelation function. In the case of solid-statequantum emitters, background correction has been applied to diversesystems, including monolayer WSe25,57,58, hBN59–61, diamond56,62–64, andSiC65. Our method of background correction is consistent with thisprior literature. The correction of the coincidence count was made byaccounting for only the background count (B) that arises fromcontribution of the dark count of the APD detector and environmentalsignal that is sample-independent. This background count was definedas the count measured by the APD while blocking the laser excitationfrom reaching the sample. No other correction of the g 2ð Þ τð Þ data (e.g.,background subtraction from uncorrelated portions of the emissionspectrum, assumptions about instrumental jittering, deconvolutionfrom the instrument response function, or symmetrization) wasperformed.Low-temperature PL lifetime microscopy measurements (Sup-plementary Fig. 5) were performed using an optical setup consisting ofa long-pass dichroic mirror (650nm), a mirror mounted on a scanningS-335 Piezo platform (Physik Instrumente), a scan lens (ThorlabsLSM03-VIS), a 100mm tube lens (Thorlabs TTL100-A), and an opticalcryostat (Montana Instruments Cryostation S100) with a built-inobjective (Zeiss Epiplan-Neofluar 100x/0.90 NA). The sample wasexcited with a picosecond pulsed laser (LDH-P-C-640B, adjusted to 10or 20MHz, Picoquant PDL 800-D) and the collected PL signal wasmeasured with an avalanche photodiode detector (Micro PhotonDevices PDM) in a confocal arrangement. The PL signal was separatedfrom the excitation beamwith a 700nm long-pass filter, while the SPEsignal was filtered using a combination of short- and long-pass filters.The IRF of the system is ~95 ps.First-principles calculations1L WSe2 was modeled using a 4 × 4 × 1 supercell. The NPh oligomerswere modeled using different numbers and geometries of NPh rings asshown in Supplementary Figs. 23 and 24. These isolatedmoleculeswerethen noncovalently interfaced to 1L WSe2. All calculations were per-formed using the Vienna ab initio simulation package (VASP) based onspin-polarized density functional theory (DFT) with a plane-wave basisset and projector-augmented wave (PAW)66 technique. For geometryoptimization, the generalized gradient approximation (GGA) refined byPerdew, Burke, and Ernzerhof (PBE)67 was utilized with Grimme’s DFT-D3BJ correction. The energy and force convergence parameters were1 × 10–6 eV/cell and 1 × 10–2 eV/Å, respectively. The energy cutoff was520 eV, and the Brillouin zone was sampled using the Γ-centeredMonkhorst−Pack k-grid scheme with a 4 × 4 × 1 k-mesh. A 20Å vacuumabove the surface along the c-axis was added to avoid inter-imageinteractions. In the most stable configuration, the NPh oligomers wereflat to the surface. In contrast, the vertical orientation of NPh relative tothe WSe2 surface68 had a Se–C bond length of 2.54Å, which was ener-getically less stable by 0.26 eV compared to the flat configuration. Theisolated molecules were simulated employing a cubic box of50 × 50× 50Å and at the Γ point (1 × 1 × 1). The binding energy of themodeled NPh oligomer on the surface was defined as:Ebinding = Etotal � ðEWSe2+ ENPhÞ ð3Þwhere Etotal, EWSe2 ,and ENPh denote the total energy of the system, theenergy of 1LWSe2, and the isolated NPh oligomer energy, respectively.The binding energy for the different NPh oligomer configurations isprovided in Supplementary Table 2, where negative binding repre-sents an exothermic reaction. Biaxial and isotropic tensile strains of 1%and 3%were applied to the pristine, Semono-vacancy (Sevac),Wmono-vacancy (Wvac), and NPh functionalized (with and without vacancy)surfaces along the a and b axes (comprehensive results are provided inSupplementary Figs. 27–29). All of the band structures along the highsymmetry points in the Brillouin zone and the energy level alignmentswere obtained using the range-separated hybrid functional HSE0669coupled with spin–orbit coupling (HSE06/SOC) on the PBE-D3BJgeometry. Band structures and other post-processing were carried outusing VASPKIT70.Data availabilityRelevant data supporting the key findings of this study are availablewithin the article and the Supplementary Information file. All raw datagenerated during this study are available from the correspondingauthors upon request.References1. Srivastava, A. et al. Optically active quantum dots in monolayerWSe2. Nat. Nanotechnol. 10, 491–496 (2015).2. He, Y.-M. et al. Single quantum emitters in monolayer semi-conductors. Nat. Nanotechnol. 10, 497–502 (2015).3. Koperski, M. et al. Single photon emitters in exfoliated WSe2structures. Nat. Nanotechnol. 10, 503–506 (2015).Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 84. Chakraborty, C., Kinnischtzke, L., Goodfellow, K. M., Beams, R. &Vamivakas, A. N. Voltage-controlled quantum light from an atom-ically thin semiconductor. Nat. Nanotechnol. 10, 507–511(2015).5. Tonndorf, P. et al. Single-photonemission from localizedexcitons inan atomically thin semiconductor. Optica 2, 347–352 (2015).6. Tran, T. T., Bray, K., Ford, M. J., Toth, M. & Aharonovich, I. Quantumemission from hexagonal boron nitride monolayers. Nat. Nano-technol. 11, 37–41 (2016).7. Chakraborty, C., Vamivakas, N. & Englund, D. Advances in quantumlight emission from 2Dmaterials. Nanophoton 8, 2017–2032 (2019).8. Toth, M. & Aharonovich, I. Single photon sources in atomically thinmaterials. Annu. Rev. Phys. Chem. 70, 123–142 (2019).9. Xu, X., Yao,W., Xiao, D. &Heinz, T. F. Spin andpseudospins in layeredtransition metal dichalcogenides. Nat. Phys. 10, 343–350 (2014).10. Liu, X. & Hersam, M. C. 2D materials for quantum information sci-ence. Nat. Rev. Mater. 4, 669–684 (2019).11. Kumar, S., Kaczmarczyk, A. & Gerardot, B. D. Strain-induced spatialand spectral isolation of quantum emitters in mono- and bilayerWSe2. Nano Lett. 15, 7567–7573 (2015).12. Branny, A., Kumar, S., Proux, R. & Gerardot, B. D. Deterministicstrain-induced arrays of quantum emitters in a two-dimensionalsemiconductor. Nat. Commun. 8, 15053 (2017).13. Palacios-Berraquero, C. et al. Large-scale quantum-emitter arrays inatomically thin semiconductors. Nat. Commun. 8, 15093 (2017).14. Rosenberger, M. R. et al. Quantum calligraphy: writing single-photon emitters in a two-dimensionalmaterials platform.ACSNano13, 904–912 (2019).15. Klein, J. et al. Site-selectively generated photon emitters in mono-layer MoS2 via local helium ion irradiation. Nat. Commun. 10,2755 (2019).16. Barthelmi, K. et al. Atomistic defects as single-photon emitters inatomically thin MoS2. Appl. Phys. Lett. 117, 070501 (2020).17. Parto, K., Azzam, S. I., Banerjee, K. & Moody, G. Defect and strainengineering of monolayer WSe2 enables site-controlled single-photon emission up to 150 K. Nat. Commun. 12, 3585 (2021).18. Dang, J. et al. Identifying defect-related quantum emitters inmonolayer WSe2. NPJ 2D Mater. Appl. 4, 2 (2020).19. Rhodes, D., Chae, S. H., Ribeiro-Palau, R. & Hone, J. Disorder in vander Waals heterostructures of 2D materials. Nat. Mater. 18,541–549 (2019).20. Ryder, C. R., Wood, J. D., Wells, S. A. & Hersam, M. C.Chemically tailoring semiconducting two-dimensional transitionmetal dichalcogenides and black phosphorus. ACS Nano 10,3900–3917 (2016).21. Liu, X. & Hersam, M. C. Interface characterization and control of 2Dmaterials and heterostructures. Adv. Mater. 30, 1801586(2018).22. Ji, H. G. et al. Chemically tuned p- and n-Type WSe2 monolayerswith high carrier mobility for advanced electronics. Adv. Mater. 31,1903613 (2019).23. Englert, J. M. et al. Covalent bulk functionalization of graphene.Nat.Chem. 3, 279–286 (2011).24. Voiry, D. et al. Covalent functionalization of monolayered transitionmetal dichalcogenides by phase engineering. Nat. Chem. 7,45–49 (2015).25. Knirsch, K. C. et al. Basal-plane functionalization of chemicallyexfoliated molybdenum disulfide by diazonium salts. ACS Nano 9,6018–6030 (2015).26. Ryder, C. R. et al. Covalent functionalization and passivation ofexfoliated black phosphorus via aryl diazonium chemistry. Nat.Chem. 8, 597–602 (2016).27. Chu, X. S. et al. Direct covalent chemical functionalization ofunmodified two-fimensional molybdenum disulfide. Chem. Mater.30, 2112–2128 (2018).28. Li, D. O., Chu, X. S. &Wang, Q. H. Reaction kinetics for the covalentfunctionalization of two-dimensional MoS2 by aryl diazonium salts.Langmuir 35, 5693–5701 (2019).29. Li, D. O. et al. Covalent chemical functionalization of semi-conducting layered chalcogenide nanosheets. Mol. Syst. Des. Eng.4, 962–973 (2019).30. Hossain, M. Z., Walsh, M. A. & Hersam, M. C. Scanning tunnelingmicroscopy, spectroscopy, and nanolithography of epitaxial gra-phene chemically modified with aryl moieties. J. Am. Chem. Soc.132, 15399–15403 (2010).31. Zhou, Y. et al. Probing dark excitons in atomically thin semi-conductors via near-field coupling to surface plasmon polaritons.Nat. Nanotechnol. 12, 856–860 (2017).32. Li, Z. et al. Revealing the biexciton and trion-exciton complexes inBN encapsulated WSe2. Nat. Commun. 9, 3719 (2018).33. Piao, Y. et al. Brightening of carbon nanotube photoluminescencethrough the incorporation of sp3 defects. Nat. Chem. 5,840–845 (2013).34. He, X. et al. Tunable room-temperature single-photon emission attelecom wavelengths from sp3 defects in carbon nanotubes. Nat.Photon. 11, 577–582 (2017).35. Saha, A. et al. Narrow-band single-photon emission throughselective aryl functionalization of zigzag carbon nanotubes. Nat.Chem. 10, 1089–1095 (2018).36. He, X. et al. Carbon nanotubes as emerging quantum-light sources.Nat. Mater. 17, 663–670 (2018).37. Jones, A. M. et al. Optical generation of excitonic valley coherencein monolayer WSe2. Nat. Nanotechnol. 8, 634–638 (2013).38. Linhart, L. et al. Localized intervalley defect excitons as single-photon emitters in WSe2. Phys. Rev. Lett. 123, 146401(2019).39. Jariwala, D., Marks, T. J. & Hersam, M. C. Mixed-dimensional van derWaals heterostructures. Nat. Mater. 16, 170–181 (2017).40. Rigosi, A. F., Hill, H. M., Li, Y., Chernikov, A. & Heinz, T. F. Probinginterlayer interactions in transition metal dichalcogenide hetero-structures by optical spectroscopy: MoS2/WS2 and MoSe2/WSe2.Nano Lett. 15, 5033–5038 (2015).41. Mitterreiter, E. et al. The role of chalcogen vacancies for atomicdefect emission in MoS2. Nat. Commun. 12, 3822 (2021).42. Sadhukhan, T. & Schatz, G. C. Generating bright emissive states bymodulating the bandgap ofmonolayer tungsten diselenide. J. Phys.Chem. C. 126, 5598–5606 (2022).43. Menanteau, T., Levillain, E., Downard, A. J. & Breton, T. Evidence ofmonolayer formation via diazonium grafting with a radical sca-venger: electrochemical, AFM and XPS monitoring. Phys. Chem.Chem. Phys. 17, 13137–13142 (2015).44. Kwon, H. et al. Molecularly tunable fluorescent quantum defects. J.Am. Chem. Soc. 138, 6878–6885 (2016).45. Combellas, C., Kanoufi, F., Pinson, J. & Podvorica, F. I. Stericallyhindered diazonium salts for the grafting of amonolayer onmetals.J. Am. Chem. Soc. 130, 8576–8577 (2008).46. Luo, Y. et al. Deterministic coupling of site-controlled quantumemitters in monolayer WSe2 to plasmonic nanocavities. Nat.Nanotechnol. 13, 1137–1142 (2018).47. Tripathi, L. N. et al. Spontaneous emission enhancement in strain-inducedWSe2monolayer-based quantum light sources onmetallicsurfaces. ACS Photon. 5, 1919–1926 (2018).48. Iff, O. et al. Substrate engineering for high-quality emission of freeand localized excitons from atomic monolayers in hybrid archi-tectures. Optica 4, 669–673 (2017).49. Lu, X. et al. Optical initialization of a single spin-valley in chargedWSe2 quantum dots. Nat. Nanotechnol. 14, 426–431 (2019).50. Cai, T. et al. Radiative enhancement of single quantum emitters inWSe2 monolayers using site-controlled metallic nanopillars. ACSPhoton. 5, 3466–3471 (2018).Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 951. Iff, O. et al. Purcell-enhanced single photon source based on adeterministically placedWSe2monolayer quantum dot in a circularBragg grating cavity. Nano Lett. 21, 4715–4720 (2021).52. Sortino, L. et al. Bright single photon emitters with enhancedquantum efficiency in a two-dimensional semiconductor coupledwith dielectric nano-antennas. Nat. Commun. 12, 6063 (2021).53. Flatten, L. C. et al. Microcavity enhanced single photon emissionfrom two-dimensional WSe2. Appl. Phys. Lett. 112, 191105 (2018).54. Zhang, L. et al. Efficient and layer-dependent exciton pumpingacross atomically thin organic–inorganic type-I heterostructures.Adv. Mater. 30, 1803986 (2018).55. Castellanos-Gomez, A. et al. Deterministic transfer of two-dimensional materials by all-dry viscoelastic stamping. 2D Mater. 1,011002 (2014).56. Brouri, R., Beveratos, A., Poizat, J.-P. & Grangier, P. Photon anti-bunching in the fluorescence of individual color centers in dia-mond. Opt. Lett. 25, 1294–1296 (2000).57. Peyskens, F., Chakraborty, C., Muneeb, M., Van Thourhout, D. &Englund, D. Integration of single photon emitters in 2D layeredmaterials with a silicon nitride photonic chip. Nat. Commun. 10,4435 (2019).58. Kumar, S. et al. Resonant laser spectroscopyof localized excitons inmonolayer WSe2. Optica 3, 882–886 (2016).59. Jha, P. K., Akbari, H., Kim, Y., Biswas, S. & Atwater, H. A. Nanoscaleaxial position and orientation measurement of hexagonal boronnitride quantum emitters using a tunable nanophotonic environ-ment. Nanotechnology 33, 015001 (2022).60. Tran, T. T. et al. Deterministic coupling of quantum emitters in 2Dmaterials to plasmonic nanocavity arrays. Nano Lett. 17,2634–2639 (2017).61. Dietrich, A. et al. Observation of Fourier transform limited lines inhexagonal boron nitride. Phys. Rev. B 98, 081414 (2018).62. Huang, T.-Y. et al. A monolithic immersion metalens for imagingsolid-state quantum emitters. Nat. Commun. 10, 2392 (2019).63. Tizei, L. H. G. et al. Spatially and spectrally resolved cath-odoluminescence with fast electrons: a tool for background sub-traction in luminescence intensity second-order correlationmeasurements applied to subwavelength inhomogeneous dia-mond nanocrystals. Phys. Status Solidi A 210, 2060–2065 (2013).64. Beveratos, A. et al. Room temperature stable single-photon source.Eur. Phys. J. D. 18, 191–196 (2002).65. Christle, D. J. et al. Isolated electron spins in silicon carbide withmillisecond coherence times. Nat. Mater. 14, 160–163 (2015).66. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the pro-jector augmented-wave method. Phys. Rev. B 59, 1758–1775 (1999).67. Paier, J., Hirschl, R., Marsman, M. & Kresse, G. ThePerdew–Burke–Ernzerhof exchange-correlation functional appliedto theG2-1 test set using a plane-wave basis set. J. Chem. Phys. 122,234102 (2005).68. Jones, L.O.,Mosquera,M.A., Ratner,M.A. &Schatz,G.C.Control ofcharge carriers and band structure in 2D monolayer molybdenumdisulfide via covalent functionalization. ACS Appl. Mater. Interfaces12, 4607–4615 (2020).69. Heyd, J., Scuseria, G. E. & Ernzerhof,M. Hybrid functionals based ona screened Coulomb potential. J. Chem. Phys. 118,8207–8215 (2003).70. Wang, V., Xu, N., Liu, J.-C., Tang, G. & Geng, W.-T. VASPKIT: a user-friendly interface facilitating high-throughput computing and ana-lysis using VASP code. Comput. Phys. Commun. 267,108033 (2021).AcknowledgementsThis research was primarily supported by the Center for MolecularQuantum Transduction, an Energy Frontier Research Center funded bythe U.S. Department of Energy, Office of Science, Office of Basic EnergySciences, under Award No. DE-SC0021314. This work was performed, inpart, at theNational Science FoundationMaterials ResearchScience andEngineering Center at Northwestern University under Award No. DMR-1720319. Work performed at the Center for Nanoscale Materials, a U.S.Department of EnergyOffice of Science User Facility, was supported bythe U.S. DOE, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. This work also made use of the Keck-II and EPICfacilities of the Northwestern University NUANCE Center, which hasreceived support from the SHyNE Resource (NSF ECCS-2025633), theIIN, and the Northwestern MRSEC program (NSF DMR-1720139). K.W.and T.T. acknowledge support from JSPS KAKENHI (Grant Numbers19H05790, 20H00354 and 21H05233). A.D. acknowledges a NationalScience Foundation Graduate Research Fellowship. D.L. acknowledgessupport from the Swiss National Science Foundation for an Early Post-DocMobility Fellowship (P2EZP2_181614). The authors also acknowledgeRoel Tempelaar, Teodor K. Stanev, and Pufan Liu for useful discussions.Author contributionsM.I.B.U., A.D., D.L., W.W., K.W., and T.T. contributed materials and sam-ple preparation.M.I.B.U. andA.D. performedmaterial characterization atroom temperature. H.Z., S.C.G., R.A., J-S.C., and W.W. performed low-temperature optical spectroscopy. T.S. performed first-principles cal-culations. M.I.B.U. and H.Z. analyzed the data. M.I.B.U. prepared themanuscript with input from all authors. M.C.H., N.P.S., G.C.S., E.A.W.,X.M., and T.J.M. supervised the project.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s41467-023-37892-0.Correspondence and requests for materials should be addressed toGeorge C. Schatz, Nathaniel P. Stern or Mark C. Hersam.Peer review information Nature Communications thanks the anon-ymous reviewer(s) for their contribution to the peer review of thiswork. Peer reviewer reports are available.Reprints and permissions information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard to jur-isdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format, aslong as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons license, and indicate ifchanges were made. The images or other third party material in thisarticle are included in the article’s Creative Commons license, unlessindicated otherwise in a credit line to the material. If material is notincluded in the article’s Creative Commons license and your intendeduse is not permitted by statutory regulation or exceeds the permitteduse, you will need to obtain permission directly from the copyrightholder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.© The Author(s) 2023Article https://doi.org/10.1038/s41467-023-37892-0Nature Communications |         (2023) 14:2193 10https://doi.org/10.1038/s41467-023-37892-0http://www.nature.com/reprintshttp://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/ Chemomechanical modification of quantum emission in monolayer WSe2 Results Diazonium functionalization and photoluminescence quenching Low-temperature optical spectroscopy Spatially deterministic chemomechanically modified SPE Noncovalent functionalization following 4-NBD treatment First-principles calculations Discussion Methods Sample preparation Chemical functionalization Characterization Low-temperature optical spectroscopy First-principles calculations Data availability References Acknowledgements Author contributions Competing interests Additional information